Compare commits

..

42 Commits

Author SHA1 Message Date
Ahmed Ibrahim
8aa5e7770c fmt + clippy: codex-core deterministic shell tool tests, conflict cleanup 2025-09-10 23:42:57 -07:00
Ahmed Ibrahim
162e1235a8 Change forking to read the rollout from file (#3440)
This PR changes get history op to get path. Then, forking will use a
path. This will help us have one unified codepath for resuming/forking
conversations. Will also help in having rollout history in order. It
also fixes a bug where you won't see the UI when resuming after forking.
2025-09-10 17:42:54 -07:00
jif-oai
c09ed74a16 Unified execution (#3288)
## Unified PTY-Based Exec Tool

Note: this requires to have this flag in the config:
`use_experimental_unified_exec_tool=true`

- Adds a PTY-backed interactive exec feature (“unified_exec”) with
session reuse via
  session_id, bounded output (128 KiB), and timeout clamping (≤ 60 s).
- Protocol: introduces ResponseItem::UnifiedExec { session_id,
arguments, timeout_ms }.
- Tools: exposes unified_exec as a function tool (Responses API);
excluded from Chat
  Completions payload while still supported in tool lists.
- Path handling: resolves commands via PATH (or explicit paths), with
UTF‑8/newline‑aware
  truncation (truncate_middle).
- Tests: cover command parsing, path resolution, session
persistence/cleanup, multi‑session
  isolation, timeouts, and truncation behavior.
2025-09-10 17:38:11 -07:00
Michael Bolin
65f3528cad feat: add UserInfo request to JSON-RPC server (#3428)
This adds a simple endpoint that provides the email address encoded in
`$CODEX_HOME/auth.json`.

As noted, for now, we do not hit the server to verify this is the user's
true email address.
2025-09-10 17:03:35 -07:00
Michael Bolin
44262d8fd8 fix: ensure output of codex-rs/mcp-types/generate_mcp_types.py matches codex-rs/mcp-types/src/lib.rs (#3439)
https://github.com/openai/codex/pull/3395 updated `mcp-types/src/lib.rs`
by hand, but that file is generated code that is produced by
`mcp-types/generate_mcp_types.py`. Unfortunately, we do not have
anything in CI to verify this right now, but I will address that in a
subsequent PR.

#3395 ended up introducing a change that added a required field when
deserializing `InitializeResult`, breaking Codex when used as an MCP
client, so the quick fix in #3436 was to make the new field `Optional`
with `skip_serializing_if = "Option::is_none"`, but that did not address
the problem that `mcp-types/generate_mcp_types.py` and
`mcp-types/src/lib.rs` are out of sync.

This PR gets things back to where they are in sync. It removes the
custom `mcp_types::McpClientInfo` type that was added to
`mcp-types/src/lib.rs` and forces us to use the generated
`mcp_types::Implementation` type. Though this PR also updates
`generate_mcp_types.py` to generate the additional `user_agent:
Optional<String>` field on `Implementation` so that we can continue to
specify it when Codex operates as an MCP server.

However, this also requires us to specify `user_agent: None` when Codex
operates as an MCP client.

We may want to introduce our own `InitializeResult` type that is
specific to when we run as a server to avoid this in the future, but my
immediate goal is just to get things back in sync.
2025-09-10 16:14:41 -07:00
Jeremy Rose
95a9938d3a fix trampling projects table when accepting trusted dirs (#3434)
Co-authored-by: Codex <199175422+chatgpt-codex-connector[bot]@users.noreply.github.com>
2025-09-10 23:01:31 +00:00
Jeremy Rose
f69f07b028 put workspace roots in the environment context (#3375)
to keep the tool description constant when the writable roots change.
2025-09-10 15:10:52 -07:00
Gabriel Peal
8d766088e6 Make user_agent optional (#3436)
# External (non-OpenAI) Pull Request Requirements

Currently, mcp server fail to start with:
```
🖐  MCP client for `<CLIENT>` failed to start: missing field `user_agent`
````

It isn't clear to me yet why this is happening. My understanding is that
this struct is simply added as a new field to the response but this
should fix it until I figure out the full story here.

<img width="714" height="262" alt="CleanShot 2025-09-10 at 13 58 59"
src="https://github.com/user-attachments/assets/946b1313-5c1c-43d3-8ae8-ecc3de3406fc"
/>
2025-09-10 14:15:02 -07:00
dedrisian-oai
87654ec0b7 Persist model & reasoning changes (#2799)
Persists `/model` changes across both general and profile-specific
sessions.
2025-09-10 20:53:46 +00:00
Michael Bolin
51d9e05de7 Back out "feat: POSIX unification and snapshot sessions (#3179)" (#3430)
This reverts https://github.com/openai/codex/pull/3179.

#3179 appears to introduce a regression where sourcing dotfiles causes a
bunch of activity in the title bar (and potentially slows things down?)


https://github.com/user-attachments/assets/a68f7fb3-0749-4e0e-a321-2aa6993e01da

Verified this no longer happens after backing out #3179.

Original commit changeset: 62bd0e3d9d
2025-09-10 12:40:24 -07:00
Jeremy Rose
8068cc75f8 replace tui_markdown with a custom markdown renderer (#3396)
Also, simplify the streaming behavior.

This fixes a number of display issues with streaming markdown, and paves
the way for better markdown features (e.g. customizable styles, syntax
highlighting, markdown-aware wrapping).

Not currently supported:
- footnotes
- tables
- reference-style links
2025-09-10 12:13:53 -07:00
Eric Traut
acb28bf914 Improved resiliency of two auth-related tests (#3427)
This PR improves two existing auth-related tests. They were failing when
run in an environment where an `OPENAI_API_KEY` env variable was
defined. The change makes them more resilient.
2025-09-10 11:46:02 -07:00
Kazuhiro Sera
97338de578 Remove a broken link to prompting_guide.md in docs/getting-started.md (#2858)
The file no longer exists. We've been receiving this feedback several
times.
- https://github.com/openai/codex/issues/2374
- https://github.com/openai/codex/issues/2810
- https://github.com/openai/codex/issues/2826

My previous PR https://github.com/openai/codex/pull/2413 for this issue
restored the file but now it's compatible with the current file
structure. Thus, let's simply delete the link.
2025-09-10 10:52:50 -07:00
katyhshi
5200b7a95d docs: fix codex exec heading typo (#2703)
# External (non-OpenAI) Pull Request Requirements

Before opening this Pull Request, please read the "Contributing" section
of the README or your PR may be closed:
https://github.com/openai/codex#contributing

If your PR conforms to our contribution guidelines, replace this text
with a detailed and high quality description of your changes.
2025-09-10 10:39:53 -07:00
Michael Bolin
64e6c4afbb fix: remove empty file: chatwidget_stream_tests.rs (#3356)
Originally added in https://github.com/openai/codex/pull/2029.
2025-09-10 10:35:24 -07:00
Eric Traut
39db113cc9 Added images to UserMessageEvent (#3400)
This PR adds an `images` field to the existing `UserMessageEvent` so we
can encode zero or more images associated with a user message. This
allows images to be restored when conversations are restored.
2025-09-10 10:18:43 -07:00
Ahmed Ibrahim
45bd5ca4b9 Move initial history to protocol (#3422)
To fix an edge case of forking then resuming

#3419
2025-09-10 10:17:24 -07:00
Michael Bolin
c13c3dadbf fix: remove unnecessary #[allow(dead_code)] annotation (#3357) 2025-09-10 08:19:05 -07:00
Gabriel Peal
8636bff46d Set a user agent suffix when used as a mcp server (#3395)
This automatically adds a user agent suffix whenever the CLI is used as
a MCP server
2025-09-10 02:32:57 +00:00
Ahmed Ibrahim
43809a454e Introduce rollout items (#3380)
This PR introduces Rollout items. This enable us to rollout eventmsgs
and session meta.

This is mostly #3214 with rebase on main
2025-09-09 23:52:33 +00:00
dank-openai
5c48600bb3 alt+delete deletes the word to the right of the cursor (delete_forward_word) (#3394)
This mirrors alt+backspace, which deletes to the left of the cursor.
2025-09-09 22:41:23 +00:00
Andrew Tan
de6559f2ab Include apply_patch tool for oss models from gpt-oss providers with different naming convention (e.g. openai/gpt-oss-*) (#2811)
Model providers like Groq, Openrouter, AWS Bedrock, VertexAI and others
typically prefix the name of gpt-oss models with `openai`, e.g.
`openai/gpt-oss-120b`.

This PR is to match the model name slug using `contains` instead of
`starts_with` to ensure that the `apply_patch` tool is included in the
tools for models names like `openai/gpt-oss-120b`

Without this, the gpt-oss models will often try to call the
`apply_patch` tool directly instead of via the `shell` command, leading
to validation errors.

I have run all the local checks.

Note: The gpt-oss models from non-Ollama providers are typically run via
a profile with a different base_url (instead of with the `--oss` flag)

---------

Co-authored-by: Andrew Tan <andrewtan@Andrews-Mac.local>
2025-09-09 15:02:02 -07:00
pakrym-oai
5bcc9d8b77 Do not send reasoning item IDs (#3390)
Response API doesn't require IDs on reasoning items anymore. 

Fixes: https://github.com/openai/codex/issues/3292
2025-09-09 14:47:06 -07:00
Gabriel Peal
5eab4c7ab4 Replace config.responses_originator_header_internal_override with CODEX_INTERNAL_ORIGINATOR_OVERRIDE_ENV_VAR (#3388)
The previous config approach had a few issues:
1. It is part of the config but not designed to be used externally
2. It had to be wired through many places (look at the +/- on this PR
3. It wasn't guaranteed to be set consistently everywhere because we
don't have a super well defined way that configs stack. For example, the
extension would configure during newConversation but anything that
happened outside of that (like login) wouldn't get it.

This env var approach is cleaner and also creates one less thing we have
to deal with when coming up with a better holistic story around configs.

One downside is that I removed the unit test testing for the override
because I don't want to deal with setting the global env or spawning
child processes and figuring out how to introspect their originator
header. The new code is sufficiently simple and I tested it e2e that I
feel as if this is still worth it.
2025-09-09 17:23:23 -04:00
jif-oai
f656e192bf No fail fast (#3387)
Add --no-fail-fast to the new `nextest`
2025-09-09 13:17:14 -07:00
Jeremy Rose
ee5ecae7c0 tweak "failed to find expected lines" message in apply_patch (#3374)
It was hard for me to read the expected lines as a `["one", "two",
"three"]` array, maybe not so hard for the model but probably not having
to un-escape in its head would help it out :)

Co-authored-by: Codex <199175422+chatgpt-codex-connector[bot]@users.noreply.github.com>
2025-09-09 12:27:50 -07:00
Michael Bolin
58bb2048ac fix: LoginChatGptCompleteNotification does not need to be listed explicitly in protocol-ts (#3222)
I verified that the output of `protocol-ts$ cargo run` is unchanged by
removing this line..

Added a comment on `ServerNotification` with justification to make this
clear.
2025-09-09 11:06:59 -07:00
Wang
ac8a3155d6 feat(core): re-export InitialHistory from conversation_manager (#3270)
This commit adds a re-export for InitialHistory from the internal
conversation_manager module in codex-core's lib.rs.

The `RolloutRecorder::get_rollout_history` method (exposed via `pub use
rollout::RolloutRecorder;`, already present in lib.rs) returns an
`InitialHistory` type, which is defined in the private
conversation_manager module. Without this re-export, consumers of the
public RolloutRecorder API would not be able to directly use the return
type, as they cannot access the private module. This would result in an
inconvenient experience where the method's return value cannot be
handled without additional, non-obvious imports.

By adding `pub use conversation_manager::InitialHistory;`, we make
InitialHistory available as `codex_core::InitialHistory`, improving API
ergonomics for users of the rollout functionality while keeping the
conversation_manager module internal.

No functional changes are made; this is a pure re-export for better
usability.

Signed-off-by: M4n5ter <m4n5terrr@gmail.com>
2025-09-09 10:37:08 -07:00
Michael Bolin
ace14e8d36 feat: add ArchiveConversation to ClientRequest (#3353)
Adds support for `ArchiveConversation` in the JSON-RPC server that takes
a `(ConversationId, PathBuf)` pair and:

- verifies the `ConversationId` corresponds to the rollout id at the
`PathBuf`
- if so, invokes
`ConversationManager.remove_conversation(ConversationId)`
- if the `CodexConversation` was in memory, send `Shutdown` and wait for
`ShutdownComplete` with a timeout
- moves the `.jsonl` file to `$CODEX_HOME/archived_sessions`

---------

Co-authored-by: Gabriel Peal <gabriel@openai.com>
2025-09-09 11:39:00 -04:00
Michael Bolin
2a76a08a9e fix: include rollout_path in NewConversationResponse (#3352)
Adding the `rollout_path` to the `NewConversationResponse` makes it so a
client can perform subsequent operations on a `(ConversationId,
PathBuf)` pair. #3353 will introduce support for `ArchiveConversation`.

---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/3352).
* #3353
* __->__ #3352
2025-09-09 00:11:48 -07:00
Michael Bolin
16309d6b68 chore: try switching to cargo nextest to speed up CI builds (#3323)
I started looking at https://nexte.st/ because I was interested in a
test harness that lets a test dynamically declare itself "skipped,"
which would be a nice alternative to this pattern:


4c46490e53/codex-rs/core/tests/suite/cli_stream.rs (L22-L27)

ChatGPT pointed me at https://nexte.st/, which also claims to be "up to
3x as fast as cargo test." Locally, in `codex-rs`, I see

- `cargo nextest run` finishes in 19s
- `cargo test` finishes in 37s

Though looking at CI, the wins are quite as big, presumably because my
laptop has more cores than our GitHub runners (which is a separate
issue...). Comparing the [CI jobs from this
PR](https://github.com/openai/codex/actions/runs/17561325162/job/49878216246?pr=3323)
with that of a [recent open
PR](https://github.com/openai/codex/actions/runs/17561066581/job/49877342753?pr=3321):

| | `cargo test` | `cargo nextest` |
| ----------------------------------------------- | ------------ |
--------------- |
| `macos-14 - aarch64-apple-darwin` | 2m16s | 1m51s |
| `macos-14 - aarch64-apple-darwin` | 5m04s | 3m44s |
| `ubuntu-24.04 - x86_64-unknown-linux-musl` | 2m02s | 1m56s |
| `ubuntu-24.04-arm - aarch64-unknown-linux-musl` | 2m01s | 1m35s |
| `windows-latest - x86_64-pc-windows-msvc` | 3m07s | 2m53s |
| `windows-11-arm - aarch64-pc-windows-msvc` | 3m10s | 2m45s |

I thought that, to start, we would only make this change in CI before
declaring it the "official" way for the team to run the test suite.

Though unfortunately, I do not believe that `cargo nextest` _actually_
supports a dynamic skip feature, so I guess I'll have to keep looking?
Some related discussions:

- https://internals.rust-lang.org/t/pre-rfc-skippable-tests/14611
- https://internals.rust-lang.org/t/skippable-tests/21260
2025-09-08 21:39:18 -07:00
jif-oai
62bd0e3d9d feat: POSIX unification and snapshot sessions (#3179)
## Session snapshot
For POSIX shell, the goal is to take a snapshot of the interactive shell
environment, store it in a session file located in `.codex/` and only
source this file for every command that is run.
As a result, if a snapshot files exist, `bash -lc <CALL>` get replaced
by `bash -c <CALL>`.

This also fixes the issue that `bash -lc` does not source `.bashrc`,
resulting in missing env variables and aliases in the codex session.
## POSIX unification
Unify `bash` and `zsh` shell into a POSIX shell. The rational is that
the tool will not use any `zsh` specific capabilities.

---------

Co-authored-by: Michael Bolin <mbolin@openai.com>
2025-09-08 18:09:45 -07:00
jif-oai
a9c68ea270 feat: Run cargo shear during CI (#3338)
Run cargo shear as part of the CI to ensure no unused dependencies
2025-09-09 01:05:08 +00:00
Jeremy Rose
ac58749bd3 allow mach-lookup for com.apple.system.opendirectoryd.libinfo (#3334)
in the base sandbox policy. this is [allowed in Chrome
renderers](https://source.chromium.org/chromium/chromium/src/+/main:sandbox/policy/mac/common.sb;l=266;drc=7afa0043cfcddb3ef9dafe5acbfc01c2f7e7df01),
so I feel it's fairly safe.
2025-09-08 16:28:52 -07:00
Robert
79cbd2ab1b Improve explanation of how the shell handles quotes in config.md (#3169)
* Clarify how the shell's handling of quotes affects the interpretation
of TOML values in `--config`/`-c`
* Provide examples of the right way to pass complex TOML values
* The previous explanation incorrectly demonstrated how to pass TOML
values to `--config`/`-c` (misunderstanding how the shell’s handling of
quotes affects things) and would result in invalid invocations of
`codex`.
2025-09-08 15:58:25 -07:00
Gabriel Peal
5eaaf307e1 Generate more typescript types and return conversation id with ConversationSummary (#3219)
This PR does multiple things that are necessary for conversation resume
to work from the extension. I wanted to make sure everything worked so
these changes wound up in one PR:
1. Generate more ts types
2. Resume rollout history files rather than create a new one every time
it is resumed so you don't see a duplicate conversation in history for
every resume. Chatted with @aibrahim-oai to verify this
3. Return conversation_id in conversation summaries
4. [Cleanup] Use serde and strong types for a lot of the rollout file
parsing
2025-09-08 17:54:47 -04:00
Justin Lebar
18330c2362 Format large numbers in a more readable way. (#2046)
- In the bottom line of the TUI, print the number of tokens to 3 sigfigs
  with an SI suffix, e.g. "1.23K".
- Elsewhere where we print a number, I figure it's worthwhile to print
  the exact number, because e.g. it's a summary of your session. Here we print
  the numbers comma-separated.
2025-09-08 21:48:48 +00:00
Jeremy Rose
4c46490e53 Highlight Proposed Command preview (#3319)
#### Summary
- highlight proposed command previews with the shared bash syntax
highlighter
- keep the Proposed Command section consistent with other execution
renderings
2025-09-08 10:48:41 -07:00
Gabriel Peal
5c1416d99b Add a getUserAgent MCP method (#3320)
This will allow the extension to pass this user agent + a suffix for its
requests
2025-09-08 13:30:13 -04:00
Michael Bolin
0525b48baa chore: upgrade to actions/setup-node@v5 (#3316)
Dependabot tried to automatically upgrade us to `actions/setup-node@v5`
in https://github.com/openai/codex/pull/3293, but it broke our CI. Note
this upgrade has breaking changes:

https://github.com/actions/setup-node/releases/tag/v5.0.0

I think the problem was that `v5` was correctly reading our
`packageManager` line here:


e2b3053b2b/package.json (L24)

and then tried to run `pnpm`, but couldn't because it wasn't available
yet. This PR:

- moves `pnpm/action-setup` before `actions/setup-node`
- drops `version` from our `pnpm/action-setup` step because it is not
necessary when it is specified in `package.json` (which it is in our
case), so leaving it here ran the risk of the two getting out of sync
- upgrades `actions/setup-node` from `v4` to `v5`
- deletes the two custom steps we had to enable Node.js caching since
`v5` claims to do this for us now
- adds `--frozen-lockfile` to our `pnpm install` invocation, which
seemed like something we should have always had there
2025-09-08 09:34:59 -07:00
Jeremy Rose
1f4f9cde8e tui: paste with ctrl+v checks file_list (#3211)
I found that pasting images from Finder with Ctrl+V was resulting in
incorrect results; this seems to work better.
2025-09-08 09:31:42 -07:00
Biturd
cad37009e1 fix: improve MCP server initialization error handling #3196 #2346 #2555 (#3243)
• I have signed the CLA by commenting the required sentence and
triggered recheck.
• Local checks are all green (fmt / clippy / test).
• Could you please approve the pending GitHub Actions workflows
(first-time contributor), and when convenient, help with one approving
review so I can proceed? Thanks!

  ## Summary
- Catch and log task panics during server initialization instead of
propagating JoinError
- Handle tool listing failures gracefully, allowing partial server
initialization
- Improve error resilience on macOS where init timeouts are more common

  ## Test plan
  - [x] Test MCP server initialization with timeout scenarios
  - [x] Verify graceful handling of tool listing failures
  - [x] Confirm improved error messages and logging
  - [x] Test on macOS 

 ## Fix issue  #3196 #2346 #2555
### fix before:
<img width="851" height="363" alt="image"
src="https://github.com/user-attachments/assets/e1f9c749-71fd-4873-a04f-d3fc4cbe0ae6"
/>

<img width="775" height="108" alt="image"
src="https://github.com/user-attachments/assets/4e4748bd-9dd6-42b5-b38b-8bfe9341a441"
/>

### fix improved:
<img width="966" height="528" alt="image"
src="https://github.com/user-attachments/assets/418324f3-e37a-4a3c-8bdd-934f9ff21dfb"
/>

---------

Co-authored-by: Michael Bolin <mbolin@openai.com>
2025-09-08 09:28:12 -07:00
117 changed files with 6985 additions and 3377 deletions

View File

@@ -14,33 +14,18 @@ jobs:
- name: Checkout repository
uses: actions/checkout@v5
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: 22
- name: Setup pnpm
uses: pnpm/action-setup@v4
with:
version: 10.8.1
run_install: false
- name: Get pnpm store directory
id: pnpm-cache
shell: bash
run: |
echo "store_path=$(pnpm store path --silent)" >> $GITHUB_OUTPUT
- name: Setup pnpm cache
uses: actions/cache@v4
- name: Setup Node.js
uses: actions/setup-node@v5
with:
path: ${{ steps.pnpm-cache.outputs.store_path }}
key: ${{ runner.os }}-pnpm-store-${{ hashFiles('**/pnpm-lock.yaml') }}
restore-keys: |
${{ runner.os }}-pnpm-store-
node-version: 22
- name: Install dependencies
run: pnpm install
run: pnpm install --frozen-lockfile
# Run all tasks using workspace filters

View File

@@ -63,6 +63,24 @@ jobs:
- name: cargo fmt
run: cargo fmt -- --config imports_granularity=Item --check
cargo_shear:
name: cargo shear
runs-on: ubuntu-24.04
needs: changed
if: ${{ needs.changed.outputs.codex == 'true' || needs.changed.outputs.workflows == 'true' || github.event_name == 'push' }}
defaults:
run:
working-directory: codex-rs
steps:
- uses: actions/checkout@v5
- uses: dtolnay/rust-toolchain@1.89
- uses: taiki-e/install-action@0c5db7f7f897c03b771660e91d065338615679f4 # v2
with:
tool: cargo-shear
version: 1.5.1
- name: cargo shear
run: cargo shear
# --- CI to validate on different os/targets --------------------------------
lint_build_test:
name: ${{ matrix.runner }} - ${{ matrix.target }}${{ matrix.profile == 'release' && ' (release)' || '' }}
@@ -160,12 +178,17 @@ jobs:
find . -name Cargo.toml -mindepth 2 -maxdepth 2 -print0 \
| xargs -0 -n1 -I{} bash -c 'cd "$(dirname "{}")" && cargo check --profile ${{ matrix.profile }}'
- name: cargo test
- uses: taiki-e/install-action@0c5db7f7f897c03b771660e91d065338615679f4 # v2
with:
tool: nextest
version: 0.9.103
- name: tests
id: test
# `cargo test` takes too long for release builds to run them on every PR
# Tests take too long for release builds to run them on every PR.
if: ${{ matrix.profile != 'release' }}
continue-on-error: true
run: cargo test --all-features --target ${{ matrix.target }} --profile ${{ matrix.profile }}
run: cargo nextest run --all-features --no-fail-fast --target ${{ matrix.target }}
env:
RUST_BACKTRACE: 1
@@ -182,7 +205,7 @@ jobs:
# --- Gatherer job that you mark as the ONLY required status -----------------
results:
name: CI results (required)
needs: [changed, general, lint_build_test]
needs: [changed, general, cargo_shear, lint_build_test]
if: always()
runs-on: ubuntu-24.04
steps:
@@ -190,6 +213,7 @@ jobs:
shell: bash
run: |
echo "general: ${{ needs.general.result }}"
echo "shear : ${{ needs.cargo_shear.result }}"
echo "matrix : ${{ needs.lint_build_test.result }}"
# If nothing relevant changed (PR touching only root README, etc.),
@@ -201,4 +225,5 @@ jobs:
# Otherwise require the jobs to have succeeded
[[ '${{ needs.general.result }}' == 'success' ]] || { echo 'general failed'; exit 1; }
[[ '${{ needs.cargo_shear.result }}' == 'success' ]] || { echo 'cargo_shear failed'; exit 1; }
[[ '${{ needs.lint_build_test.result }}' == 'success' ]] || { echo 'matrix failed'; exit 1; }

874
codex-rs/Cargo.lock generated

File diff suppressed because it is too large Load Diff

View File

@@ -35,7 +35,7 @@ npx @modelcontextprotocol/inspector codex mcp
You can enable notifications by configuring a script that is run whenever the agent finishes a turn. The [notify documentation](../docs/config.md#notify) includes a detailed example that explains how to get desktop notifications via [terminal-notifier](https://github.com/julienXX/terminal-notifier) on macOS.
### `codex exec` to run Codex programmatially/non-interactively
### `codex exec` to run Codex programmatically/non-interactively
To run Codex non-interactively, run `codex exec PROMPT` (you can also pass the prompt via `stdin`) and Codex will work on your task until it decides that it is done and exits. Output is printed to the terminal directly. You can set the `RUST_LOG` environment variable to see more about what's going on.

View File

@@ -726,9 +726,9 @@ fn compute_replacements(
line_index = start_idx + pattern.len();
} else {
return Err(ApplyPatchError::ComputeReplacements(format!(
"Failed to find expected lines {:?} in {}",
chunk.old_lines,
path.display()
"Failed to find expected lines in {}:\n{}",
path.display(),
chunk.old_lines.join("\n"),
)));
}
}

View File

@@ -12,7 +12,6 @@ clap = { version = "4", features = ["derive"] }
codex-common = { path = "../common", features = ["cli"] }
codex-core = { path = "../core" }
codex-protocol = { path = "../protocol" }
reqwest = { version = "0.12", features = ["json", "stream"] }
serde = { version = "1", features = ["derive"] }
serde_json = "1"
tokio = { version = "1", features = ["full"] }

View File

@@ -31,7 +31,7 @@ pub async fn run_apply_command(
ConfigOverrides::default(),
)?;
init_chatgpt_token_from_auth(&config.codex_home, &config.responses_originator_header).await?;
init_chatgpt_token_from_auth(&config.codex_home).await?;
let task_response = get_task(&config, apply_cli.task_id).await?;
apply_diff_from_task(task_response, cwd).await

View File

@@ -13,10 +13,10 @@ pub(crate) async fn chatgpt_get_request<T: DeserializeOwned>(
path: String,
) -> anyhow::Result<T> {
let chatgpt_base_url = &config.chatgpt_base_url;
init_chatgpt_token_from_auth(&config.codex_home, &config.responses_originator_header).await?;
init_chatgpt_token_from_auth(&config.codex_home).await?;
// Make direct HTTP request to ChatGPT backend API with the token
let client = create_client(&config.responses_originator_header);
let client = create_client();
let url = format!("{chatgpt_base_url}{path}");
let token =

View File

@@ -19,11 +19,8 @@ pub fn set_chatgpt_token_data(value: TokenData) {
}
/// Initialize the ChatGPT token from auth.json file
pub async fn init_chatgpt_token_from_auth(
codex_home: &Path,
originator: &str,
) -> std::io::Result<()> {
let auth = CodexAuth::from_codex_home(codex_home, AuthMode::ChatGPT, originator)?;
pub async fn init_chatgpt_token_from_auth(codex_home: &Path) -> std::io::Result<()> {
let auth = CodexAuth::from_codex_home(codex_home, AuthMode::ChatGPT)?;
if let Some(auth) = auth {
let token_data = auth.get_token_data().await?;
set_chatgpt_token_data(token_data);

View File

@@ -12,8 +12,8 @@ use codex_protocol::mcp_protocol::AuthMode;
use std::env;
use std::path::PathBuf;
pub async fn login_with_chatgpt(codex_home: PathBuf, originator: String) -> std::io::Result<()> {
let opts = ServerOptions::new(codex_home, CLIENT_ID.to_string(), originator);
pub async fn login_with_chatgpt(codex_home: PathBuf) -> std::io::Result<()> {
let opts = ServerOptions::new(codex_home, CLIENT_ID.to_string());
let server = run_login_server(opts)?;
eprintln!(
@@ -27,12 +27,7 @@ pub async fn login_with_chatgpt(codex_home: PathBuf, originator: String) -> std:
pub async fn run_login_with_chatgpt(cli_config_overrides: CliConfigOverrides) -> ! {
let config = load_config_or_exit(cli_config_overrides);
match login_with_chatgpt(
config.codex_home,
config.responses_originator_header.clone(),
)
.await
{
match login_with_chatgpt(config.codex_home).await {
Ok(_) => {
eprintln!("Successfully logged in");
std::process::exit(0);
@@ -65,11 +60,7 @@ pub async fn run_login_with_api_key(
pub async fn run_login_status(cli_config_overrides: CliConfigOverrides) -> ! {
let config = load_config_or_exit(cli_config_overrides);
match CodexAuth::from_codex_home(
&config.codex_home,
config.preferred_auth_method,
&config.responses_originator_header,
) {
match CodexAuth::from_codex_home(&config.codex_home, config.preferred_auth_method) {
Ok(Some(auth)) => match auth.mode {
AuthMode::ApiKey => match auth.get_token().await {
Ok(api_key) => {

View File

@@ -40,7 +40,6 @@ pub async fn run_main(opts: ProtoCli) -> anyhow::Result<()> {
let conversation_manager = ConversationManager::new(AuthManager::shared(
config.codex_home.clone(),
config.preferred_auth_method,
config.responses_originator_header.clone(),
));
let NewConversation {
conversation_id: _,

View File

@@ -26,14 +26,12 @@ eventsource-stream = "0.2.3"
futures = "0.3"
libc = "0.2.175"
mcp-types = { path = "../mcp-types" }
mime_guess = "2.0"
os_info = "3.12.0"
portable-pty = "0.9.0"
rand = "0.9"
regex-lite = "0.1.7"
reqwest = { version = "0.12", features = ["json", "stream"] }
serde = { version = "1", features = ["derive"] }
serde_bytes = "0.11"
serde_json = "1"
sha1 = "0.10.6"
shlex = "1.3.0"
@@ -56,7 +54,7 @@ tracing = { version = "0.1.41", features = ["log"] }
tree-sitter = "0.25.9"
tree-sitter-bash = "0.25.0"
uuid = { version = "1", features = ["serde", "v4"] }
whoami = "1.6.1"
which = "6"
wildmatch = "2.4.0"
@@ -72,9 +70,6 @@ openssl-sys = { version = "*", features = ["vendored"] }
[target.aarch64-unknown-linux-musl.dependencies]
openssl-sys = { version = "*", features = ["vendored"] }
[target.'cfg(target_os = "windows")'.dependencies]
which = "6"
[dev-dependencies]
assert_cmd = "2"
core_test_support = { path = "tests/common" }
@@ -85,3 +80,6 @@ tempfile = "3"
tokio-test = "0.4"
walkdir = "2.5.0"
wiremock = "0.6"
[package.metadata.cargo-shear]
ignored = ["openssl-sys"]

View File

@@ -75,9 +75,8 @@ impl CodexAuth {
pub fn from_codex_home(
codex_home: &Path,
preferred_auth_method: AuthMode,
originator: &str,
) -> std::io::Result<Option<CodexAuth>> {
load_auth(codex_home, true, preferred_auth_method, originator)
load_auth(codex_home, true, preferred_auth_method)
}
pub async fn get_token_data(&self) -> Result<TokenData, std::io::Error> {
@@ -173,7 +172,7 @@ impl CodexAuth {
mode: AuthMode::ChatGPT,
auth_file: PathBuf::new(),
auth_dot_json,
client: crate::default_client::create_client("codex_cli_rs"),
client: crate::default_client::create_client(),
}
}
@@ -188,10 +187,7 @@ impl CodexAuth {
}
pub fn from_api_key(api_key: &str) -> Self {
Self::from_api_key_with_client(
api_key,
crate::default_client::create_client(crate::default_client::DEFAULT_ORIGINATOR),
)
Self::from_api_key_with_client(api_key, crate::default_client::create_client())
}
}
@@ -232,13 +228,12 @@ fn load_auth(
codex_home: &Path,
include_env_var: bool,
preferred_auth_method: AuthMode,
originator: &str,
) -> std::io::Result<Option<CodexAuth>> {
// First, check to see if there is a valid auth.json file. If not, we fall
// back to AuthMode::ApiKey using the OPENAI_API_KEY environment variable
// (if it is set).
let auth_file = get_auth_file(codex_home);
let client = crate::default_client::create_client(originator);
let client = crate::default_client::create_client();
let auth_dot_json = match try_read_auth_json(&auth_file) {
Ok(auth) => auth,
// If auth.json does not exist, try to read the OPENAI_API_KEY from the
@@ -473,7 +468,7 @@ mod tests {
auth_dot_json,
auth_file: _,
..
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT, "codex_cli_rs")
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT)
.unwrap()
.unwrap();
assert_eq!(None, api_key);
@@ -525,7 +520,7 @@ mod tests {
auth_dot_json,
auth_file: _,
..
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT, "codex_cli_rs")
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT)
.unwrap()
.unwrap();
assert_eq!(None, api_key);
@@ -576,7 +571,7 @@ mod tests {
auth_dot_json,
auth_file: _,
..
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT, "codex_cli_rs")
} = super::load_auth(codex_home.path(), false, AuthMode::ChatGPT)
.unwrap()
.unwrap();
assert_eq!(Some("sk-test-key".to_string()), api_key);
@@ -596,7 +591,7 @@ mod tests {
)
.unwrap();
let auth = super::load_auth(dir.path(), false, AuthMode::ChatGPT, "codex_cli_rs")
let auth = super::load_auth(dir.path(), false, AuthMode::ChatGPT)
.unwrap()
.unwrap();
assert_eq!(auth.mode, AuthMode::ApiKey);
@@ -680,7 +675,6 @@ mod tests {
#[derive(Debug)]
pub struct AuthManager {
codex_home: PathBuf,
originator: String,
inner: RwLock<CachedAuth>,
}
@@ -689,13 +683,12 @@ impl AuthManager {
/// preferred auth method. Errors loading auth are swallowed; `auth()` will
/// simply return `None` in that case so callers can treat it as an
/// unauthenticated state.
pub fn new(codex_home: PathBuf, preferred_auth_mode: AuthMode, originator: String) -> Self {
let auth = CodexAuth::from_codex_home(&codex_home, preferred_auth_mode, &originator)
pub fn new(codex_home: PathBuf, preferred_auth_mode: AuthMode) -> Self {
let auth = CodexAuth::from_codex_home(&codex_home, preferred_auth_mode)
.ok()
.flatten();
Self {
codex_home,
originator,
inner: RwLock::new(CachedAuth {
preferred_auth_mode,
auth,
@@ -712,7 +705,6 @@ impl AuthManager {
};
Arc::new(Self {
codex_home: PathBuf::new(),
originator: "codex_cli_rs".to_string(),
inner: RwLock::new(cached),
})
}
@@ -734,7 +726,7 @@ impl AuthManager {
/// whether the auth value changed.
pub fn reload(&self) -> bool {
let preferred = self.preferred_auth_method();
let new_auth = CodexAuth::from_codex_home(&self.codex_home, preferred, &self.originator)
let new_auth = CodexAuth::from_codex_home(&self.codex_home, preferred)
.ok()
.flatten();
if let Ok(mut guard) = self.inner.write() {
@@ -755,12 +747,8 @@ impl AuthManager {
}
/// Convenience constructor returning an `Arc` wrapper.
pub fn shared(
codex_home: PathBuf,
preferred_auth_mode: AuthMode,
originator: String,
) -> Arc<Self> {
Arc::new(Self::new(codex_home, preferred_auth_mode, originator))
pub fn shared(codex_home: PathBuf, preferred_auth_mode: AuthMode) -> Arc<Self> {
Arc::new(Self::new(codex_home, preferred_auth_mode))
}
/// Attempt to refresh the current auth token (if any). On success, reload

View File

@@ -84,7 +84,7 @@ impl ModelClient {
summary: ReasoningSummaryConfig,
conversation_id: ConversationId,
) -> Self {
let client = create_client(&config.responses_originator_header);
let client = create_client();
Self {
config,

File diff suppressed because it is too large Load Diff

View File

@@ -38,9 +38,7 @@ const OPENAI_DEFAULT_MODEL: &str = "gpt-5";
/// the context window.
pub(crate) const PROJECT_DOC_MAX_BYTES: usize = 32 * 1024; // 32 KiB
const CONFIG_TOML_FILE: &str = "config.toml";
const DEFAULT_RESPONSES_ORIGINATOR_HEADER: &str = "codex_cli_rs";
pub(crate) const CONFIG_TOML_FILE: &str = "config.toml";
/// Application configuration loaded from disk and merged with overrides.
#[derive(Debug, Clone, PartialEq)]
@@ -169,16 +167,20 @@ pub struct Config {
pub tools_web_search_request: bool,
/// The value for the `originator` header included with Responses API requests.
pub responses_originator_header: String,
/// If set to `true`, the API key will be signed with the `originator` header.
pub preferred_auth_method: AuthMode,
pub use_experimental_streamable_shell_tool: bool,
/// If set to `true`, used only the experimental unified exec tool.
pub use_experimental_unified_exec_tool: bool,
/// Include the `view_image` tool that lets the agent attach a local image path to context.
pub include_view_image_tool: bool,
/// The active profile name used to derive this `Config` (if any).
pub active_profile: Option<String>,
/// When true, disables burst-paste detection for typed input entirely.
/// All characters are inserted as they are received, and no buffering
/// or placeholder replacement will occur for fast keypress bursts.
@@ -262,17 +264,7 @@ pub fn load_config_as_toml(codex_home: &Path) -> std::io::Result<TomlValue> {
}
}
/// Patch `CODEX_HOME/config.toml` project state.
/// Use with caution.
pub fn set_project_trusted(codex_home: &Path, project_path: &Path) -> anyhow::Result<()> {
let config_path = codex_home.join(CONFIG_TOML_FILE);
// Parse existing config if present; otherwise start a new document.
let mut doc = match std::fs::read_to_string(config_path.clone()) {
Ok(s) => s.parse::<DocumentMut>()?,
Err(e) if e.kind() == std::io::ErrorKind::NotFound => DocumentMut::new(),
Err(e) => return Err(e.into()),
};
fn set_project_trusted_inner(doc: &mut DocumentMut, project_path: &Path) -> anyhow::Result<()> {
// Ensure we render a human-friendly structure:
//
// [projects]
@@ -288,14 +280,26 @@ pub fn set_project_trusted(codex_home: &Path, project_path: &Path) -> anyhow::Re
// Ensure top-level `projects` exists as a non-inline, explicit table. If it
// exists but was previously represented as a non-table (e.g., inline),
// replace it with an explicit table.
let mut created_projects_table = false;
{
let root = doc.as_table_mut();
let needs_table = !root.contains_key("projects")
|| root.get("projects").and_then(|i| i.as_table()).is_none();
if needs_table {
root.insert("projects", toml_edit::table());
created_projects_table = true;
// If `projects` exists but isn't a standard table (e.g., it's an inline table),
// convert it to an explicit table while preserving existing entries.
let existing_projects = root.get("projects").cloned();
if existing_projects.as_ref().is_none_or(|i| !i.is_table()) {
let mut projects_tbl = toml_edit::Table::new();
projects_tbl.set_implicit(true);
// If there was an existing inline table, migrate its entries to explicit tables.
if let Some(inline_tbl) = existing_projects.as_ref().and_then(|i| i.as_inline_table()) {
for (k, v) in inline_tbl.iter() {
if let Some(inner_tbl) = v.as_inline_table() {
let new_tbl = inner_tbl.clone().into_table();
projects_tbl.insert(k, toml_edit::Item::Table(new_tbl));
}
}
}
root.insert("projects", toml_edit::Item::Table(projects_tbl));
}
}
let Some(projects_tbl) = doc["projects"].as_table_mut() else {
@@ -304,12 +308,6 @@ pub fn set_project_trusted(codex_home: &Path, project_path: &Path) -> anyhow::Re
));
};
// If we created the `projects` table ourselves, keep it implicit so we
// don't render a standalone `[projects]` header.
if created_projects_table {
projects_tbl.set_implicit(true);
}
// Ensure the per-project entry is its own explicit table. If it exists but
// is not a table (e.g., an inline table), replace it with an explicit table.
let needs_proj_table = !projects_tbl.contains_key(project_key.as_str())
@@ -328,6 +326,21 @@ pub fn set_project_trusted(codex_home: &Path, project_path: &Path) -> anyhow::Re
};
proj_tbl.set_implicit(false);
proj_tbl["trust_level"] = toml_edit::value("trusted");
Ok(())
}
/// Patch `CODEX_HOME/config.toml` project state.
/// Use with caution.
pub fn set_project_trusted(codex_home: &Path, project_path: &Path) -> anyhow::Result<()> {
let config_path = codex_home.join(CONFIG_TOML_FILE);
// Parse existing config if present; otherwise start a new document.
let mut doc = match std::fs::read_to_string(config_path.clone()) {
Ok(s) => s.parse::<DocumentMut>()?,
Err(e) if e.kind() == std::io::ErrorKind::NotFound => DocumentMut::new(),
Err(e) => return Err(e.into()),
};
set_project_trusted_inner(&mut doc, project_path)?;
// ensure codex_home exists
std::fs::create_dir_all(codex_home)?;
@@ -477,9 +490,7 @@ pub struct ConfigToml {
pub experimental_instructions_file: Option<PathBuf>,
pub experimental_use_exec_command_tool: Option<bool>,
/// The value for the `originator` header included with Responses API requests.
pub responses_originator_header_internal_override: Option<String>,
pub experimental_use_unified_exec_tool: Option<bool>,
pub projects: Option<HashMap<String, ProjectConfig>>,
@@ -661,7 +672,11 @@ impl Config {
tools_web_search_request: override_tools_web_search_request,
} = overrides;
let config_profile = match config_profile_key.as_ref().or(cfg.profile.as_ref()) {
let active_profile_name = config_profile_key
.as_ref()
.or(cfg.profile.as_ref())
.cloned();
let config_profile = match active_profile_name.as_ref() {
Some(key) => cfg
.profiles
.get(key)
@@ -773,10 +788,6 @@ impl Config {
Self::get_base_instructions(experimental_instructions_path, &resolved_cwd)?;
let base_instructions = base_instructions.or(file_base_instructions);
let responses_originator_header: String = cfg
.responses_originator_header_internal_override
.unwrap_or(DEFAULT_RESPONSES_ORIGINATOR_HEADER.to_owned());
let config = Self {
model,
model_family,
@@ -826,12 +837,15 @@ impl Config {
include_plan_tool: include_plan_tool.unwrap_or(false),
include_apply_patch_tool: include_apply_patch_tool.unwrap_or(false),
tools_web_search_request,
responses_originator_header,
preferred_auth_method: cfg.preferred_auth_method.unwrap_or(AuthMode::ChatGPT),
use_experimental_streamable_shell_tool: cfg
.experimental_use_exec_command_tool
.unwrap_or(false),
use_experimental_unified_exec_tool: cfg
.experimental_use_unified_exec_tool
.unwrap_or(true),
include_view_image_tool,
active_profile: active_profile_name,
disable_paste_burst: cfg.disable_paste_burst.unwrap_or(false),
};
Ok(config)
@@ -1203,10 +1217,11 @@ model_verbosity = "high"
include_plan_tool: false,
include_apply_patch_tool: false,
tools_web_search_request: false,
responses_originator_header: "codex_cli_rs".to_string(),
preferred_auth_method: AuthMode::ChatGPT,
use_experimental_streamable_shell_tool: false,
use_experimental_unified_exec_tool: true,
include_view_image_tool: true,
active_profile: Some("o3".to_string()),
disable_paste_burst: false,
},
o3_profile_config
@@ -1260,10 +1275,11 @@ model_verbosity = "high"
include_plan_tool: false,
include_apply_patch_tool: false,
tools_web_search_request: false,
responses_originator_header: "codex_cli_rs".to_string(),
preferred_auth_method: AuthMode::ChatGPT,
use_experimental_streamable_shell_tool: false,
use_experimental_unified_exec_tool: true,
include_view_image_tool: true,
active_profile: Some("gpt3".to_string()),
disable_paste_burst: false,
};
@@ -1332,10 +1348,11 @@ model_verbosity = "high"
include_plan_tool: false,
include_apply_patch_tool: false,
tools_web_search_request: false,
responses_originator_header: "codex_cli_rs".to_string(),
preferred_auth_method: AuthMode::ChatGPT,
use_experimental_streamable_shell_tool: false,
use_experimental_unified_exec_tool: true,
include_view_image_tool: true,
active_profile: Some("zdr".to_string()),
disable_paste_burst: false,
};
@@ -1390,10 +1407,11 @@ model_verbosity = "high"
include_plan_tool: false,
include_apply_patch_tool: false,
tools_web_search_request: false,
responses_originator_header: "codex_cli_rs".to_string(),
preferred_auth_method: AuthMode::ChatGPT,
use_experimental_streamable_shell_tool: false,
use_experimental_unified_exec_tool: true,
include_view_image_tool: true,
active_profile: Some("gpt5".to_string()),
disable_paste_burst: false,
};
@@ -1404,17 +1422,14 @@ model_verbosity = "high"
#[test]
fn test_set_project_trusted_writes_explicit_tables() -> anyhow::Result<()> {
let codex_home = TempDir::new().unwrap();
let project_dir = TempDir::new().unwrap();
let project_dir = Path::new("/some/path");
let mut doc = DocumentMut::new();
// Call the function under test
set_project_trusted(codex_home.path(), project_dir.path())?;
set_project_trusted_inner(&mut doc, project_dir)?;
// Read back the generated config.toml and assert exact contents
let config_path = codex_home.path().join(CONFIG_TOML_FILE);
let contents = std::fs::read_to_string(&config_path)?;
let contents = doc.to_string();
let raw_path = project_dir.path().to_string_lossy();
let raw_path = project_dir.to_string_lossy();
let path_str = if raw_path.contains('\\') {
format!("'{raw_path}'")
} else {
@@ -1432,12 +1447,10 @@ trust_level = "trusted"
#[test]
fn test_set_project_trusted_converts_inline_to_explicit() -> anyhow::Result<()> {
let codex_home = TempDir::new().unwrap();
let project_dir = TempDir::new().unwrap();
let project_dir = Path::new("/some/path");
// Seed config.toml with an inline project entry under [projects]
let config_path = codex_home.path().join(CONFIG_TOML_FILE);
let raw_path = project_dir.path().to_string_lossy();
let raw_path = project_dir.to_string_lossy();
let path_str = if raw_path.contains('\\') {
format!("'{raw_path}'")
} else {
@@ -1449,13 +1462,12 @@ trust_level = "trusted"
{path_str} = {{ trust_level = "untrusted" }}
"#
);
std::fs::create_dir_all(codex_home.path())?;
std::fs::write(&config_path, initial)?;
let mut doc = initial.parse::<DocumentMut>()?;
// Run the function; it should convert to explicit tables and set trusted
set_project_trusted(codex_home.path(), project_dir.path())?;
set_project_trusted_inner(&mut doc, project_dir)?;
let contents = std::fs::read_to_string(&config_path)?;
let contents = doc.to_string();
// Assert exact output after conversion to explicit table
let expected = format!(
@@ -1470,5 +1482,37 @@ trust_level = "trusted"
Ok(())
}
// No test enforcing the presence of a standalone [projects] header.
#[test]
fn test_set_project_trusted_migrates_top_level_inline_projects_preserving_entries()
-> anyhow::Result<()> {
let initial = r#"toplevel = "baz"
projects = { "/Users/mbolin/code/codex4" = { trust_level = "trusted", foo = "bar" } , "/Users/mbolin/code/codex3" = { trust_level = "trusted" } }
model = "foo""#;
let mut doc = initial.parse::<DocumentMut>()?;
// Approve a new directory
let new_project = Path::new("/Users/mbolin/code/codex2");
set_project_trusted_inner(&mut doc, new_project)?;
let contents = doc.to_string();
// Since we created the [projects] table as part of migration, it is kept implicit.
// Expect explicit per-project tables, preserving prior entries and appending the new one.
let expected = r#"toplevel = "baz"
model = "foo"
[projects."/Users/mbolin/code/codex4"]
trust_level = "trusted"
foo = "bar"
[projects."/Users/mbolin/code/codex3"]
trust_level = "trusted"
[projects."/Users/mbolin/code/codex2"]
trust_level = "trusted"
"#;
assert_eq!(contents, expected);
Ok(())
}
}

View File

@@ -0,0 +1,582 @@
use crate::config::CONFIG_TOML_FILE;
use anyhow::Result;
use std::path::Path;
use tempfile::NamedTempFile;
use toml_edit::DocumentMut;
pub const CONFIG_KEY_MODEL: &str = "model";
pub const CONFIG_KEY_EFFORT: &str = "model_reasoning_effort";
/// Persist overrides into `config.toml` using explicit key segments per
/// override. This avoids ambiguity with keys that contain dots or spaces.
pub async fn persist_overrides(
codex_home: &Path,
profile: Option<&str>,
overrides: &[(&[&str], &str)],
) -> Result<()> {
let config_path = codex_home.join(CONFIG_TOML_FILE);
let mut doc = match tokio::fs::read_to_string(&config_path).await {
Ok(s) => s.parse::<DocumentMut>()?,
Err(e) if e.kind() == std::io::ErrorKind::NotFound => {
tokio::fs::create_dir_all(codex_home).await?;
DocumentMut::new()
}
Err(e) => return Err(e.into()),
};
let effective_profile = if let Some(p) = profile {
Some(p.to_owned())
} else {
doc.get("profile")
.and_then(|i| i.as_str())
.map(|s| s.to_string())
};
for (segments, val) in overrides.iter().copied() {
let value = toml_edit::value(val);
if let Some(ref name) = effective_profile {
if segments.first().copied() == Some("profiles") {
apply_toml_edit_override_segments(&mut doc, segments, value);
} else {
let mut seg_buf: Vec<&str> = Vec::with_capacity(2 + segments.len());
seg_buf.push("profiles");
seg_buf.push(name.as_str());
seg_buf.extend_from_slice(segments);
apply_toml_edit_override_segments(&mut doc, &seg_buf, value);
}
} else {
apply_toml_edit_override_segments(&mut doc, segments, value);
}
}
let tmp_file = NamedTempFile::new_in(codex_home)?;
tokio::fs::write(tmp_file.path(), doc.to_string()).await?;
tmp_file.persist(config_path)?;
Ok(())
}
/// Persist overrides where values may be optional. Any entries with `None`
/// values are skipped. If all values are `None`, this becomes a no-op and
/// returns `Ok(())` without touching the file.
pub async fn persist_non_null_overrides(
codex_home: &Path,
profile: Option<&str>,
overrides: &[(&[&str], Option<&str>)],
) -> Result<()> {
let filtered: Vec<(&[&str], &str)> = overrides
.iter()
.filter_map(|(k, v)| v.map(|vv| (*k, vv)))
.collect();
if filtered.is_empty() {
return Ok(());
}
persist_overrides(codex_home, profile, &filtered).await
}
/// Apply a single override onto a `toml_edit` document while preserving
/// existing formatting/comments.
/// The key is expressed as explicit segments to correctly handle keys that
/// contain dots or spaces.
fn apply_toml_edit_override_segments(
doc: &mut DocumentMut,
segments: &[&str],
value: toml_edit::Item,
) {
use toml_edit::Item;
if segments.is_empty() {
return;
}
let mut current = doc.as_table_mut();
for seg in &segments[..segments.len() - 1] {
if !current.contains_key(seg) {
current[*seg] = Item::Table(toml_edit::Table::new());
if let Some(t) = current[*seg].as_table_mut() {
t.set_implicit(true);
}
}
let maybe_item = current.get_mut(seg);
let Some(item) = maybe_item else { return };
if !item.is_table() {
*item = Item::Table(toml_edit::Table::new());
if let Some(t) = item.as_table_mut() {
t.set_implicit(true);
}
}
let Some(tbl) = item.as_table_mut() else {
return;
};
current = tbl;
}
let last = segments[segments.len() - 1];
current[last] = value;
}
#[cfg(test)]
mod tests {
use super::*;
use pretty_assertions::assert_eq;
use tempfile::tempdir;
/// Verifies model and effort are written at top-level when no profile is set.
#[tokio::test]
async fn set_default_model_and_effort_top_level_when_no_profile() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
persist_overrides(
codex_home,
None,
&[
(&[CONFIG_KEY_MODEL], "gpt-5"),
(&[CONFIG_KEY_EFFORT], "high"),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"model = "gpt-5"
model_reasoning_effort = "high"
"#;
assert_eq!(contents, expected);
}
/// Verifies values are written under the active profile when `profile` is set.
#[tokio::test]
async fn set_defaults_update_profile_when_profile_set() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed config with a profile selection but without profiles table
let seed = "profile = \"o3\"\n";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
persist_overrides(
codex_home,
None,
&[
(&[CONFIG_KEY_MODEL], "o3"),
(&[CONFIG_KEY_EFFORT], "minimal"),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"profile = "o3"
[profiles.o3]
model = "o3"
model_reasoning_effort = "minimal"
"#;
assert_eq!(contents, expected);
}
/// Verifies profile names with dots/spaces are preserved via explicit segments.
#[tokio::test]
async fn set_defaults_update_profile_with_dot_and_space() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed config with a profile name that contains a dot and a space
let seed = "profile = \"my.team name\"\n";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
persist_overrides(
codex_home,
None,
&[
(&[CONFIG_KEY_MODEL], "o3"),
(&[CONFIG_KEY_EFFORT], "minimal"),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"profile = "my.team name"
[profiles."my.team name"]
model = "o3"
model_reasoning_effort = "minimal"
"#;
assert_eq!(contents, expected);
}
/// Verifies explicit profile override writes under that profile even without active profile.
#[tokio::test]
async fn set_defaults_update_when_profile_override_supplied() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// No profile key in config.toml
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), "")
.await
.expect("seed write");
// Persist with an explicit profile override
persist_overrides(
codex_home,
Some("o3"),
&[(&[CONFIG_KEY_MODEL], "o3"), (&[CONFIG_KEY_EFFORT], "high")],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"[profiles.o3]
model = "o3"
model_reasoning_effort = "high"
"#;
assert_eq!(contents, expected);
}
/// Verifies nested tables are created as needed when applying overrides.
#[tokio::test]
async fn persist_overrides_creates_nested_tables() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
persist_overrides(
codex_home,
None,
&[
(&["a", "b", "c"], "v"),
(&["x"], "y"),
(&["profiles", "p1", CONFIG_KEY_MODEL], "gpt-5"),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"x = "y"
[a.b]
c = "v"
[profiles.p1]
model = "gpt-5"
"#;
assert_eq!(contents, expected);
}
/// Verifies a scalar key becomes a table when nested keys are written.
#[tokio::test]
async fn persist_overrides_replaces_scalar_with_table() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
let seed = "foo = \"bar\"\n";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
persist_overrides(codex_home, None, &[(&["foo", "bar", "baz"], "ok")])
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"[foo.bar]
baz = "ok"
"#;
assert_eq!(contents, expected);
}
/// Verifies comments and spacing are preserved when writing under active profile.
#[tokio::test]
async fn set_defaults_preserve_comments() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed a config with comments and spacing we expect to preserve
let seed = r#"# Global comment
# Another line
profile = "o3"
# Profile settings
[profiles.o3]
# keep me
existing = "keep"
"#;
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
// Apply defaults; since profile is set, it should write under [profiles.o3]
persist_overrides(
codex_home,
None,
&[(&[CONFIG_KEY_MODEL], "o3"), (&[CONFIG_KEY_EFFORT], "high")],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"# Global comment
# Another line
profile = "o3"
# Profile settings
[profiles.o3]
# keep me
existing = "keep"
model = "o3"
model_reasoning_effort = "high"
"#;
assert_eq!(contents, expected);
}
/// Verifies comments and spacing are preserved when writing at top level.
#[tokio::test]
async fn set_defaults_preserve_global_comments() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed a config WITHOUT a profile, containing comments and spacing
let seed = r#"# Top-level comments
# should be preserved
existing = "keep"
"#;
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
// Since there is no profile, the defaults should be written at top-level
persist_overrides(
codex_home,
None,
&[
(&[CONFIG_KEY_MODEL], "gpt-5"),
(&[CONFIG_KEY_EFFORT], "minimal"),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"# Top-level comments
# should be preserved
existing = "keep"
model = "gpt-5"
model_reasoning_effort = "minimal"
"#;
assert_eq!(contents, expected);
}
/// Verifies errors on invalid TOML propagate and file is not clobbered.
#[tokio::test]
async fn persist_overrides_errors_on_parse_failure() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Write an intentionally invalid TOML file
let invalid = "invalid = [unclosed";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), invalid)
.await
.expect("seed write");
// Attempting to persist should return an error and must not clobber the file.
let res = persist_overrides(codex_home, None, &[(&["x"], "y")]).await;
assert!(res.is_err(), "expected parse error to propagate");
// File should be unchanged
let contents = read_config(codex_home).await;
assert_eq!(contents, invalid);
}
/// Verifies changing model only preserves existing effort at top-level.
#[tokio::test]
async fn changing_only_model_preserves_existing_effort_top_level() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed with an effort value only
let seed = "model_reasoning_effort = \"minimal\"\n";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
// Change only the model
persist_overrides(codex_home, None, &[(&[CONFIG_KEY_MODEL], "o3")])
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"model_reasoning_effort = "minimal"
model = "o3"
"#;
assert_eq!(contents, expected);
}
/// Verifies changing effort only preserves existing model at top-level.
#[tokio::test]
async fn changing_only_effort_preserves_existing_model_top_level() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed with a model value only
let seed = "model = \"gpt-5\"\n";
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
// Change only the effort
persist_overrides(codex_home, None, &[(&[CONFIG_KEY_EFFORT], "high")])
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"model = "gpt-5"
model_reasoning_effort = "high"
"#;
assert_eq!(contents, expected);
}
/// Verifies changing model only preserves existing effort in active profile.
#[tokio::test]
async fn changing_only_model_preserves_effort_in_active_profile() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// Seed with an active profile and an existing effort under that profile
let seed = r#"profile = "p1"
[profiles.p1]
model_reasoning_effort = "low"
"#;
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
persist_overrides(codex_home, None, &[(&[CONFIG_KEY_MODEL], "o4-mini")])
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"profile = "p1"
[profiles.p1]
model_reasoning_effort = "low"
model = "o4-mini"
"#;
assert_eq!(contents, expected);
}
/// Verifies changing effort only preserves existing model in a profile override.
#[tokio::test]
async fn changing_only_effort_preserves_model_in_profile_override() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
// No active profile key; we'll target an explicit override
let seed = r#"[profiles.team]
model = "gpt-5"
"#;
tokio::fs::write(codex_home.join(CONFIG_TOML_FILE), seed)
.await
.expect("seed write");
persist_overrides(
codex_home,
Some("team"),
&[(&[CONFIG_KEY_EFFORT], "minimal")],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"[profiles.team]
model = "gpt-5"
model_reasoning_effort = "minimal"
"#;
assert_eq!(contents, expected);
}
/// Verifies `persist_non_null_overrides` skips `None` entries and writes only present values at top-level.
#[tokio::test]
async fn persist_non_null_skips_none_top_level() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
persist_non_null_overrides(
codex_home,
None,
&[
(&[CONFIG_KEY_MODEL], Some("gpt-5")),
(&[CONFIG_KEY_EFFORT], None),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = "model = \"gpt-5\"\n";
assert_eq!(contents, expected);
}
/// Verifies no-op behavior when all provided overrides are `None` (no file created/modified).
#[tokio::test]
async fn persist_non_null_noop_when_all_none() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
persist_non_null_overrides(
codex_home,
None,
&[(&["a"], None), (&["profiles", "p", "x"], None)],
)
.await
.expect("persist");
// Should not create config.toml on a pure no-op
assert!(!codex_home.join(CONFIG_TOML_FILE).exists());
}
/// Verifies entries are written under the specified profile and `None` entries are skipped.
#[tokio::test]
async fn persist_non_null_respects_profile_override() {
let tmpdir = tempdir().expect("tmp");
let codex_home = tmpdir.path();
persist_non_null_overrides(
codex_home,
Some("team"),
&[
(&[CONFIG_KEY_MODEL], Some("o3")),
(&[CONFIG_KEY_EFFORT], None),
],
)
.await
.expect("persist");
let contents = read_config(codex_home).await;
let expected = r#"[profiles.team]
model = "o3"
"#;
assert_eq!(contents, expected);
}
// Test helper moved to bottom per review guidance.
async fn read_config(codex_home: &Path) -> String {
let p = codex_home.join(CONFIG_TOML_FILE);
tokio::fs::read_to_string(p).await.unwrap_or_default()
}
}

View File

@@ -1,12 +1,5 @@
use std::collections::HashMap;
use std::path::PathBuf;
use std::sync::Arc;
use crate::AuthManager;
use crate::CodexAuth;
use codex_protocol::mcp_protocol::ConversationId;
use tokio::sync::RwLock;
use crate::codex::Codex;
use crate::codex::CodexSpawnOk;
use crate::codex::INITIAL_SUBMIT_ID;
@@ -18,13 +11,14 @@ use crate::protocol::Event;
use crate::protocol::EventMsg;
use crate::protocol::SessionConfiguredEvent;
use crate::rollout::RolloutRecorder;
use codex_protocol::mcp_protocol::ConversationId;
use codex_protocol::models::ResponseItem;
#[derive(Debug, Clone, PartialEq)]
pub enum InitialHistory {
New,
Resumed(Vec<ResponseItem>),
}
use codex_protocol::protocol::InitialHistory;
use codex_protocol::protocol::RolloutItem;
use std::collections::HashMap;
use std::path::PathBuf;
use std::sync::Arc;
use tokio::sync::RwLock;
/// Represents a newly created Codex conversation, including the first event
/// (which is [`EventMsg::SessionConfigured`]).
@@ -77,7 +71,7 @@ impl ConversationManager {
let CodexSpawnOk {
codex,
conversation_id,
} = { Codex::spawn(config, auth_manager, InitialHistory::New).await? };
} = Codex::spawn(config, auth_manager, InitialHistory::New).await?;
self.finalize_spawn(codex, conversation_id).await
}
}
@@ -139,8 +133,15 @@ impl ConversationManager {
self.finalize_spawn(codex, conversation_id).await
}
pub async fn remove_conversation(&self, conversation_id: ConversationId) {
self.conversations.write().await.remove(&conversation_id);
/// Removes the conversation from the manager's internal map, though the
/// conversation is stored as `Arc<CodexConversation>`, it is possible that
/// other references to it exist elsewhere. Returns the conversation if the
/// conversation was found and removed.
pub async fn remove_conversation(
&self,
conversation_id: &ConversationId,
) -> Option<Arc<CodexConversation>> {
self.conversations.write().await.remove(conversation_id)
}
/// Fork an existing conversation by dropping the last `drop_last_messages`
@@ -149,13 +150,13 @@ impl ConversationManager {
/// caller's `config`). The new conversation will have a fresh id.
pub async fn fork_conversation(
&self,
conversation_history: Vec<ResponseItem>,
num_messages_to_drop: usize,
config: Config,
path: PathBuf,
) -> CodexResult<NewConversation> {
// Compute the prefix up to the cut point.
let history =
truncate_after_dropping_last_messages(conversation_history, num_messages_to_drop);
let history = RolloutRecorder::get_rollout_history(&path).await?;
let history = truncate_after_dropping_last_messages(history, num_messages_to_drop);
// Spawn a new conversation with the computed initial history.
let auth_manager = self.auth_manager.clone();
@@ -170,31 +171,37 @@ impl ConversationManager {
/// Return a prefix of `items` obtained by dropping the last `n` user messages
/// and all items that follow them.
fn truncate_after_dropping_last_messages(items: Vec<ResponseItem>, n: usize) -> InitialHistory {
fn truncate_after_dropping_last_messages(history: InitialHistory, n: usize) -> InitialHistory {
if n == 0 {
return InitialHistory::Resumed(items);
return InitialHistory::Forked(history.get_rollout_items());
}
// Walk backwards counting only `user` Message items, find cut index.
let mut count = 0usize;
let mut cut_index = 0usize;
for (idx, item) in items.iter().enumerate().rev() {
if let ResponseItem::Message { role, .. } = item
// Work directly on rollout items, and cut the vector at the nth-from-last user message input.
let items: Vec<RolloutItem> = history.get_rollout_items();
// Find indices of user message inputs in rollout order.
let mut user_positions: Vec<usize> = Vec::new();
for (idx, item) in items.iter().enumerate() {
if let RolloutItem::ResponseItem(ResponseItem::Message { role, .. }) = item
&& role == "user"
{
count += 1;
if count == n {
// Cut everything from this user message to the end.
cut_index = idx;
break;
}
user_positions.push(idx);
}
}
if cut_index == 0 {
// No prefix remains after dropping; start a new conversation.
// If fewer than n user messages exist, treat as empty.
if user_positions.len() < n {
return InitialHistory::New;
}
// Cut strictly before the nth-from-last user message (do not keep the nth itself).
let cut_idx = user_positions[user_positions.len() - n];
let rolled: Vec<RolloutItem> = items.into_iter().take(cut_idx).collect();
if rolled.is_empty() {
InitialHistory::New
} else {
InitialHistory::Resumed(items.into_iter().take(cut_index).collect())
InitialHistory::Forked(rolled)
}
}
@@ -249,13 +256,30 @@ mod tests {
assistant_msg("a4"),
];
let truncated = truncate_after_dropping_last_messages(items.clone(), 1);
// Wrap as InitialHistory::Forked with response items only.
let initial: Vec<RolloutItem> = items
.iter()
.cloned()
.map(RolloutItem::ResponseItem)
.collect();
let truncated = truncate_after_dropping_last_messages(InitialHistory::Forked(initial), 1);
let got_items = truncated.get_rollout_items();
let expected_items = vec![
RolloutItem::ResponseItem(items[0].clone()),
RolloutItem::ResponseItem(items[1].clone()),
RolloutItem::ResponseItem(items[2].clone()),
];
assert_eq!(
truncated,
InitialHistory::Resumed(vec![items[0].clone(), items[1].clone(), items[2].clone(),])
serde_json::to_value(&got_items).unwrap(),
serde_json::to_value(&expected_items).unwrap()
);
let truncated2 = truncate_after_dropping_last_messages(items, 2);
assert_eq!(truncated2, InitialHistory::New);
let initial2: Vec<RolloutItem> = items
.iter()
.cloned()
.map(RolloutItem::ResponseItem)
.collect();
let truncated2 = truncate_after_dropping_last_messages(InitialHistory::Forked(initial2), 2);
assert!(matches!(truncated2, InitialHistory::New));
}
}

View File

@@ -1,38 +1,123 @@
pub const DEFAULT_ORIGINATOR: &str = "codex_cli_rs";
use reqwest::header::HeaderValue;
use std::sync::LazyLock;
use std::sync::Mutex;
pub fn get_codex_user_agent(originator: Option<&str>) -> String {
/// Set this to add a suffix to the User-Agent string.
///
/// It is not ideal that we're using a global singleton for this.
/// This is primarily designed to differentiate MCP clients from each other.
/// Because there can only be one MCP server per process, it should be safe for this to be a global static.
/// However, future users of this should use this with caution as a result.
/// In addition, we want to be confident that this value is used for ALL clients and doing that requires a
/// lot of wiring and it's easy to miss code paths by doing so.
/// See https://github.com/openai/codex/pull/3388/files for an example of what that would look like.
/// Finally, we want to make sure this is set for ALL mcp clients without needing to know a special env var
/// or having to set data that they already specified in the mcp initialize request somewhere else.
///
/// A space is automatically added between the suffix and the rest of the User-Agent string.
/// The full user agent string is returned from the mcp initialize response.
/// Parenthesis will be added by Codex. This should only specify what goes inside of the parenthesis.
pub static USER_AGENT_SUFFIX: LazyLock<Mutex<Option<String>>> = LazyLock::new(|| Mutex::new(None));
pub const CODEX_INTERNAL_ORIGINATOR_OVERRIDE_ENV_VAR: &str = "CODEX_INTERNAL_ORIGINATOR_OVERRIDE";
#[derive(Debug, Clone)]
pub struct Originator {
pub value: String,
pub header_value: HeaderValue,
}
pub static ORIGINATOR: LazyLock<Originator> = LazyLock::new(|| {
let default = "codex_cli_rs";
let value = std::env::var(CODEX_INTERNAL_ORIGINATOR_OVERRIDE_ENV_VAR)
.unwrap_or_else(|_| default.to_string());
match HeaderValue::from_str(&value) {
Ok(header_value) => Originator {
value,
header_value,
},
Err(e) => {
tracing::error!("Unable to turn originator override {value} into header value: {e}");
Originator {
value: default.to_string(),
header_value: HeaderValue::from_static(default),
}
}
}
});
pub fn get_codex_user_agent() -> String {
let build_version = env!("CARGO_PKG_VERSION");
let os_info = os_info::get();
format!(
let prefix = format!(
"{}/{build_version} ({} {}; {}) {}",
originator.unwrap_or(DEFAULT_ORIGINATOR),
ORIGINATOR.value.as_str(),
os_info.os_type(),
os_info.version(),
os_info.architecture().unwrap_or("unknown"),
crate::terminal::user_agent()
)
);
let suffix = USER_AGENT_SUFFIX
.lock()
.ok()
.and_then(|guard| guard.clone());
let suffix = suffix
.as_deref()
.map(str::trim)
.filter(|value| !value.is_empty())
.map_or_else(String::new, |value| format!(" ({value})"));
let candidate = format!("{prefix}{suffix}");
sanitize_user_agent(candidate, &prefix)
}
/// Sanitize the user agent string.
///
/// Invalid characters are replaced with an underscore.
///
/// If the user agent fails to parse, it falls back to fallback and then to ORIGINATOR.
fn sanitize_user_agent(candidate: String, fallback: &str) -> String {
if HeaderValue::from_str(candidate.as_str()).is_ok() {
return candidate;
}
let sanitized: String = candidate
.chars()
.map(|ch| if matches!(ch, ' '..='~') { ch } else { '_' })
.collect();
if !sanitized.is_empty() && HeaderValue::from_str(sanitized.as_str()).is_ok() {
tracing::warn!(
"Sanitized Codex user agent because provided suffix contained invalid header characters"
);
sanitized
} else if HeaderValue::from_str(fallback).is_ok() {
tracing::warn!(
"Falling back to base Codex user agent because provided suffix could not be sanitized"
);
fallback.to_string()
} else {
tracing::warn!(
"Falling back to default Codex originator because base user agent string is invalid"
);
ORIGINATOR.value.clone()
}
}
/// Create a reqwest client with default `originator` and `User-Agent` headers set.
pub fn create_client(originator: &str) -> reqwest::Client {
pub fn create_client() -> reqwest::Client {
use reqwest::header::HeaderMap;
use reqwest::header::HeaderValue;
let mut headers = HeaderMap::new();
let originator_value = HeaderValue::from_str(originator)
.unwrap_or_else(|_| HeaderValue::from_static(DEFAULT_ORIGINATOR));
headers.insert("originator", originator_value);
let ua = get_codex_user_agent(Some(originator));
headers.insert("originator", ORIGINATOR.header_value.clone());
let ua = get_codex_user_agent();
match reqwest::Client::builder()
reqwest::Client::builder()
// Set UA via dedicated helper to avoid header validation pitfalls
.user_agent(ua)
.default_headers(headers)
.build()
{
Ok(client) => client,
Err(_) => reqwest::Client::new(),
}
.unwrap_or_else(|_| reqwest::Client::new())
}
#[cfg(test)]
@@ -41,7 +126,7 @@ mod tests {
#[test]
fn test_get_codex_user_agent() {
let user_agent = get_codex_user_agent(None);
let user_agent = get_codex_user_agent();
assert!(user_agent.starts_with("codex_cli_rs/"));
}
@@ -53,8 +138,7 @@ mod tests {
use wiremock::matchers::method;
use wiremock::matchers::path;
let originator = "test_originator";
let client = create_client(originator);
let client = create_client();
// Spin up a local mock server and capture a request.
let server = MockServer::start().await;
@@ -82,21 +166,43 @@ mod tests {
let originator_header = headers
.get("originator")
.expect("originator header missing");
assert_eq!(originator_header.to_str().unwrap(), originator);
assert_eq!(originator_header.to_str().unwrap(), "codex_cli_rs");
// User-Agent matches the computed Codex UA for that originator
let expected_ua = get_codex_user_agent(Some(originator));
let expected_ua = get_codex_user_agent();
let ua_header = headers
.get("user-agent")
.expect("user-agent header missing");
assert_eq!(ua_header.to_str().unwrap(), expected_ua);
}
#[test]
fn test_invalid_suffix_is_sanitized() {
let prefix = "codex_cli_rs/0.0.0";
let suffix = "bad\rsuffix";
assert_eq!(
sanitize_user_agent(format!("{prefix} ({suffix})"), prefix),
"codex_cli_rs/0.0.0 (bad_suffix)"
);
}
#[test]
fn test_invalid_suffix_is_sanitized2() {
let prefix = "codex_cli_rs/0.0.0";
let suffix = "bad\0suffix";
assert_eq!(
sanitize_user_agent(format!("{prefix} ({suffix})"), prefix),
"codex_cli_rs/0.0.0 (bad_suffix)"
);
}
#[test]
#[cfg(target_os = "macos")]
fn test_macos() {
use regex_lite::Regex;
let user_agent = get_codex_user_agent(None);
let user_agent = get_codex_user_agent();
let re = Regex::new(
r"^codex_cli_rs/\d+\.\d+\.\d+ \(Mac OS \d+\.\d+\.\d+; (x86_64|arm64)\) (\S+)$",
)

View File

@@ -26,6 +26,7 @@ pub(crate) struct EnvironmentContext {
pub approval_policy: Option<AskForApproval>,
pub sandbox_mode: Option<SandboxMode>,
pub network_access: Option<NetworkAccess>,
pub writable_roots: Option<Vec<PathBuf>>,
pub shell: Option<Shell>,
}
@@ -57,6 +58,16 @@ impl EnvironmentContext {
}
None => None,
},
writable_roots: match sandbox_policy {
Some(SandboxPolicy::WorkspaceWrite { writable_roots, .. }) => {
if writable_roots.is_empty() {
None
} else {
Some(writable_roots.clone())
}
}
_ => None,
},
shell,
}
}
@@ -72,6 +83,7 @@ impl EnvironmentContext {
/// <cwd>...</cwd>
/// <approval_policy>...</approval_policy>
/// <sandbox_mode>...</sandbox_mode>
/// <writable_roots>...</writable_roots>
/// <network_access>...</network_access>
/// <shell>...</shell>
/// </environment_context>
@@ -94,6 +106,16 @@ impl EnvironmentContext {
" <network_access>{network_access}</network_access>"
));
}
if let Some(writable_roots) = self.writable_roots {
lines.push(" <writable_roots>".to_string());
for writable_root in writable_roots {
lines.push(format!(
" <root>{}</root>",
writable_root.to_string_lossy()
));
}
lines.push(" </writable_roots>".to_string());
}
if let Some(shell) = self.shell
&& let Some(shell_name) = shell.name()
{
@@ -115,3 +137,77 @@ impl From<EnvironmentContext> for ResponseItem {
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use pretty_assertions::assert_eq;
fn workspace_write_policy(writable_roots: Vec<&str>, network_access: bool) -> SandboxPolicy {
SandboxPolicy::WorkspaceWrite {
writable_roots: writable_roots.into_iter().map(PathBuf::from).collect(),
network_access,
exclude_tmpdir_env_var: false,
exclude_slash_tmp: false,
}
}
#[test]
fn serialize_workspace_write_environment_context() {
let context = EnvironmentContext::new(
Some(PathBuf::from("/repo")),
Some(AskForApproval::OnRequest),
Some(workspace_write_policy(vec!["/repo", "/tmp"], false)),
None,
);
let expected = r#"<environment_context>
<cwd>/repo</cwd>
<approval_policy>on-request</approval_policy>
<sandbox_mode>workspace-write</sandbox_mode>
<network_access>restricted</network_access>
<writable_roots>
<root>/repo</root>
<root>/tmp</root>
</writable_roots>
</environment_context>"#;
assert_eq!(context.serialize_to_xml(), expected);
}
#[test]
fn serialize_read_only_environment_context() {
let context = EnvironmentContext::new(
None,
Some(AskForApproval::Never),
Some(SandboxPolicy::ReadOnly),
None,
);
let expected = r#"<environment_context>
<approval_policy>never</approval_policy>
<sandbox_mode>read-only</sandbox_mode>
<network_access>restricted</network_access>
</environment_context>"#;
assert_eq!(context.serialize_to_xml(), expected);
}
#[test]
fn serialize_full_access_environment_context() {
let context = EnvironmentContext::new(
None,
Some(AskForApproval::OnFailure),
Some(SandboxPolicy::DangerFullAccess),
None,
);
let expected = r#"<environment_context>
<approval_policy>on-failure</approval_policy>
<sandbox_mode>danger-full-access</sandbox_mode>
<network_access>enabled</network_access>
</environment_context>"#;
assert_eq!(context.serialize_to_xml(), expected);
}
}

View File

@@ -25,31 +25,56 @@ pub(crate) fn map_response_item_to_event_messages(
return Vec::new();
}
let events: Vec<EventMsg> = content
.iter()
.filter_map(|content_item| match content_item {
ContentItem::OutputText { text } => {
Some(EventMsg::AgentMessage(AgentMessageEvent {
message: text.clone(),
}))
}
let mut events: Vec<EventMsg> = Vec::new();
let mut message_parts: Vec<String> = Vec::new();
let mut images: Vec<String> = Vec::new();
let mut kind: Option<InputMessageKind> = None;
for content_item in content.iter() {
match content_item {
ContentItem::InputText { text } => {
let trimmed = text.trim_start();
let kind = if trimmed.starts_with("<environment_context>") {
Some(InputMessageKind::EnvironmentContext)
} else if trimmed.starts_with("<user_instructions>") {
Some(InputMessageKind::UserInstructions)
} else {
Some(InputMessageKind::Plain)
};
Some(EventMsg::UserMessage(UserMessageEvent {
message: text.clone(),
kind,
}))
if kind.is_none() {
let trimmed = text.trim_start();
kind = if trimmed.starts_with("<environment_context>") {
Some(InputMessageKind::EnvironmentContext)
} else if trimmed.starts_with("<user_instructions>") {
Some(InputMessageKind::UserInstructions)
} else {
Some(InputMessageKind::Plain)
};
}
message_parts.push(text.clone());
}
_ => None,
})
.collect();
ContentItem::InputImage { image_url } => {
images.push(image_url.clone());
}
ContentItem::OutputText { text } => {
events.push(EventMsg::AgentMessage(AgentMessageEvent {
message: text.clone(),
}));
}
}
}
if !message_parts.is_empty() || !images.is_empty() {
let message = if message_parts.is_empty() {
String::new()
} else {
message_parts.join("")
};
let images = if images.is_empty() {
None
} else {
Some(images)
};
events.push(EventMsg::UserMessage(UserMessageEvent {
message,
kind,
images,
}));
}
events
}
@@ -96,3 +121,47 @@ pub(crate) fn map_response_item_to_event_messages(
| ResponseItem::Other => Vec::new(),
}
}
#[cfg(test)]
mod tests {
use super::map_response_item_to_event_messages;
use crate::protocol::EventMsg;
use crate::protocol::InputMessageKind;
use codex_protocol::models::ContentItem;
use codex_protocol::models::ResponseItem;
use pretty_assertions::assert_eq;
#[test]
fn maps_user_message_with_text_and_two_images() {
let img1 = "https://example.com/one.png".to_string();
let img2 = "https://example.com/two.jpg".to_string();
let item = ResponseItem::Message {
id: None,
role: "user".to_string(),
content: vec![
ContentItem::InputText {
text: "Hello world".to_string(),
},
ContentItem::InputImage {
image_url: img1.clone(),
},
ContentItem::InputImage {
image_url: img2.clone(),
},
],
};
let events = map_response_item_to_event_messages(&item, false);
assert_eq!(events.len(), 1, "expected a single user message event");
match &events[0] {
EventMsg::UserMessage(user) => {
assert_eq!(user.message, "Hello world");
assert!(matches!(user.kind, Some(InputMessageKind::Plain)));
assert_eq!(user.images, Some(vec![img1.clone(), img2.clone()]));
}
other => panic!("expected UserMessage, got {other:?}"),
}
}
}

View File

@@ -24,6 +24,9 @@ pub(crate) struct ExecCommandSession {
/// JoinHandle for the child wait task.
wait_handle: StdMutex<Option<JoinHandle<()>>>,
/// Tracks whether the underlying process has exited.
exit_status: std::sync::Arc<std::sync::atomic::AtomicBool>,
}
impl ExecCommandSession {
@@ -34,6 +37,7 @@ impl ExecCommandSession {
reader_handle: JoinHandle<()>,
writer_handle: JoinHandle<()>,
wait_handle: JoinHandle<()>,
exit_status: std::sync::Arc<std::sync::atomic::AtomicBool>,
) -> Self {
Self {
writer_tx,
@@ -42,6 +46,7 @@ impl ExecCommandSession {
reader_handle: StdMutex::new(Some(reader_handle)),
writer_handle: StdMutex::new(Some(writer_handle)),
wait_handle: StdMutex::new(Some(wait_handle)),
exit_status,
}
}
@@ -52,6 +57,10 @@ impl ExecCommandSession {
pub(crate) fn output_receiver(&self) -> broadcast::Receiver<Vec<u8>> {
self.output_tx.subscribe()
}
pub(crate) fn has_exited(&self) -> bool {
self.exit_status.load(std::sync::atomic::Ordering::SeqCst)
}
}
impl Drop for ExecCommandSession {

View File

@@ -6,6 +6,7 @@ mod session_manager;
pub use exec_command_params::ExecCommandParams;
pub use exec_command_params::WriteStdinParams;
pub(crate) use exec_command_session::ExecCommandSession;
pub use responses_api::EXEC_COMMAND_TOOL_NAME;
pub use responses_api::WRITE_STDIN_TOOL_NAME;
pub use responses_api::create_exec_command_tool_for_responses_api;

View File

@@ -3,6 +3,7 @@ use std::io::ErrorKind;
use std::io::Read;
use std::sync::Arc;
use std::sync::Mutex as StdMutex;
use std::sync::atomic::AtomicBool;
use std::sync::atomic::AtomicU32;
use portable_pty::CommandBuilder;
@@ -19,6 +20,7 @@ use crate::exec_command::exec_command_params::ExecCommandParams;
use crate::exec_command::exec_command_params::WriteStdinParams;
use crate::exec_command::exec_command_session::ExecCommandSession;
use crate::exec_command::session_id::SessionId;
use crate::truncate::truncate_middle;
use codex_protocol::models::FunctionCallOutputPayload;
#[derive(Debug, Default)]
@@ -327,11 +329,14 @@ async fn create_exec_command_session(
// Keep the child alive until it exits, then signal exit code.
let (exit_tx, exit_rx) = oneshot::channel::<i32>();
let exit_status = Arc::new(AtomicBool::new(false));
let wait_exit_status = exit_status.clone();
let wait_handle = tokio::task::spawn_blocking(move || {
let code = match child.wait() {
Ok(status) => status.exit_code() as i32,
Err(_) => -1,
};
wait_exit_status.store(true, std::sync::atomic::Ordering::SeqCst);
let _ = exit_tx.send(code);
});
@@ -343,116 +348,11 @@ async fn create_exec_command_session(
reader_handle,
writer_handle,
wait_handle,
exit_status,
);
Ok((session, exit_rx))
}
/// Truncate the middle of a UTF-8 string to at most `max_bytes` bytes,
/// preserving the beginning and the end. Returns the possibly truncated
/// string and `Some(original_token_count)` (estimated at 4 bytes/token)
/// if truncation occurred; otherwise returns the original string and `None`.
fn truncate_middle(s: &str, max_bytes: usize) -> (String, Option<u64>) {
// No truncation needed
if s.len() <= max_bytes {
return (s.to_string(), None);
}
let est_tokens = (s.len() as u64).div_ceil(4);
if max_bytes == 0 {
// Cannot keep any content; still return a full marker (never truncated).
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
// Helper to truncate a string to a given byte length on a char boundary.
fn truncate_on_boundary(input: &str, max_len: usize) -> &str {
if input.len() <= max_len {
return input;
}
let mut end = max_len;
while end > 0 && !input.is_char_boundary(end) {
end -= 1;
}
&input[..end]
}
// Given a left/right budget, prefer newline boundaries; otherwise fall back
// to UTF-8 char boundaries.
fn pick_prefix_end(s: &str, left_budget: usize) -> usize {
if let Some(head) = s.get(..left_budget)
&& let Some(i) = head.rfind('\n')
{
return i + 1; // keep the newline so suffix starts on a fresh line
}
truncate_on_boundary(s, left_budget).len()
}
fn pick_suffix_start(s: &str, right_budget: usize) -> usize {
let start_tail = s.len().saturating_sub(right_budget);
if let Some(tail) = s.get(start_tail..)
&& let Some(i) = tail.find('\n')
{
return start_tail + i + 1; // start after newline
}
// Fall back to a char boundary at or after start_tail.
let mut idx = start_tail.min(s.len());
while idx < s.len() && !s.is_char_boundary(idx) {
idx += 1;
}
idx
}
// Refine marker length and budgets until stable. Marker is never truncated.
let mut guess_tokens = est_tokens; // worst-case: everything truncated
for _ in 0..4 {
let marker = format!("{guess_tokens} tokens truncated…");
let marker_len = marker.len();
let keep_budget = max_bytes.saturating_sub(marker_len);
if keep_budget == 0 {
// No room for any content within the cap; return a full, untruncated marker
// that reflects the entire truncated content.
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
let left_budget = keep_budget / 2;
let right_budget = keep_budget - left_budget;
let prefix_end = pick_prefix_end(s, left_budget);
let mut suffix_start = pick_suffix_start(s, right_budget);
if suffix_start < prefix_end {
suffix_start = prefix_end;
}
let kept_content_bytes = prefix_end + (s.len() - suffix_start);
let truncated_content_bytes = s.len().saturating_sub(kept_content_bytes);
let new_tokens = (truncated_content_bytes as u64).div_ceil(4);
if new_tokens == guess_tokens {
let mut out = String::with_capacity(marker_len + kept_content_bytes + 1);
out.push_str(&s[..prefix_end]);
out.push_str(&marker);
// Place marker on its own line for symmetry when we keep line boundaries.
out.push('\n');
out.push_str(&s[suffix_start..]);
return (out, Some(est_tokens));
}
guess_tokens = new_tokens;
}
// Fallback: use last guess to build output.
let marker = format!("{guess_tokens} tokens truncated…");
let marker_len = marker.len();
let keep_budget = max_bytes.saturating_sub(marker_len);
if keep_budget == 0 {
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
let left_budget = keep_budget / 2;
let right_budget = keep_budget - left_budget;
let prefix_end = pick_prefix_end(s, left_budget);
let suffix_start = pick_suffix_start(s, right_budget);
let mut out = String::with_capacity(marker_len + prefix_end + (s.len() - suffix_start) + 1);
out.push_str(&s[..prefix_end]);
out.push_str(&marker);
out.push('\n');
out.push_str(&s[suffix_start..]);
(out, Some(est_tokens))
}
#[cfg(test)]
mod tests {
use super::*;
@@ -616,50 +516,4 @@ Output:
abc"#;
assert_eq!(expected, text);
}
#[test]
fn truncate_middle_no_newlines_fallback() {
// A long string with no newlines that exceeds the cap.
let s = "abcdefghijklmnopqrstuvwxyz0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
let max_bytes = 16; // force truncation
let (out, original) = truncate_middle(s, max_bytes);
// For very small caps, we return the full, untruncated marker,
// even if it exceeds the cap.
assert_eq!(out, "…16 tokens truncated…");
// Original string length is 62 bytes => ceil(62/4) = 16 tokens.
assert_eq!(original, Some(16));
}
#[test]
fn truncate_middle_prefers_newline_boundaries() {
// Build a multi-line string of 20 numbered lines (each "NNN\n").
let mut s = String::new();
for i in 1..=20 {
s.push_str(&format!("{i:03}\n"));
}
// Total length: 20 lines * 4 bytes per line = 80 bytes.
assert_eq!(s.len(), 80);
// Choose a cap that forces truncation while leaving room for
// a few lines on each side after accounting for the marker.
let max_bytes = 64;
// Expect exact output: first 4 lines, marker, last 4 lines, and correct token estimate (80/4 = 20).
assert_eq!(
truncate_middle(&s, max_bytes),
(
r#"001
002
003
004
…12 tokens truncated…
017
018
019
020
"#
.to_string(),
Some(20)
)
);
}
}

View File

@@ -3,6 +3,7 @@ use std::path::Path;
use std::path::PathBuf;
use codex_protocol::mcp_protocol::GitSha;
use codex_protocol::protocol::GitInfo;
use futures::future::join_all;
use serde::Deserialize;
use serde::Serialize;
@@ -43,19 +44,6 @@ pub fn get_git_repo_root(base_dir: &Path) -> Option<PathBuf> {
/// Timeout for git commands to prevent freezing on large repositories
const GIT_COMMAND_TIMEOUT: TokioDuration = TokioDuration::from_secs(5);
#[derive(Serialize, Deserialize, Clone, Debug)]
pub struct GitInfo {
/// Current commit hash (SHA)
#[serde(skip_serializing_if = "Option::is_none")]
pub commit_hash: Option<String>,
/// Current branch name
#[serde(skip_serializing_if = "Option::is_none")]
pub branch: Option<String>,
/// Repository URL (if available from remote)
#[serde(skip_serializing_if = "Option::is_none")]
pub repository_url: Option<String>,
}
#[derive(Serialize, Deserialize, Clone, Debug)]
pub struct GitDiffToRemote {
pub sha: GitSha,

View File

@@ -16,6 +16,7 @@ mod codex_conversation;
pub mod token_data;
pub use codex_conversation::CodexConversation;
pub mod config;
pub mod config_edit;
pub mod config_profile;
pub mod config_types;
mod conversation_history;
@@ -34,6 +35,8 @@ mod mcp_tool_call;
mod message_history;
mod model_provider_info;
pub mod parse_command;
mod truncate;
mod unified_exec;
mod user_instructions;
pub use model_provider_info::BUILT_IN_OSS_MODEL_PROVIDER_ID;
pub use model_provider_info::ModelProviderInfo;
@@ -42,6 +45,7 @@ pub use model_provider_info::built_in_model_providers;
pub use model_provider_info::create_oss_provider_with_base_url;
mod conversation_manager;
mod event_mapping;
pub use codex_protocol::protocol::InitialHistory;
pub use conversation_manager::ConversationManager;
pub use conversation_manager::NewConversation;
// Re-export common auth types for workspace consumers
@@ -61,7 +65,10 @@ pub mod spawn;
pub mod terminal;
mod tool_apply_patch;
pub mod turn_diff_tracker;
pub use rollout::ARCHIVED_SESSIONS_SUBDIR;
pub use rollout::RolloutRecorder;
pub use rollout::SESSIONS_SUBDIR;
pub use rollout::SessionMeta;
pub use rollout::list::ConversationItem;
pub use rollout::list::ConversationsPage;
pub use rollout::list::Cursor;

View File

@@ -163,6 +163,10 @@ impl McpConnectionManager {
name: "codex-mcp-client".to_owned(),
version: env!("CARGO_PKG_VERSION").to_owned(),
title: Some("Codex".into()),
// This field is used by Codex when it is an MCP
// server: it should not be used when Codex is
// an MCP client.
user_agent: None,
},
protocol_version: mcp_types::MCP_SCHEMA_VERSION.to_owned(),
};
@@ -187,7 +191,13 @@ impl McpConnectionManager {
let mut clients: HashMap<String, ManagedClient> = HashMap::with_capacity(join_set.len());
while let Some(res) = join_set.join_next().await {
let (server_name, client_res) = res?; // JoinError propagation
let (server_name, client_res) = match res {
Ok((server_name, client_res)) => (server_name, client_res),
Err(e) => {
warn!("Task panic when starting MCP server: {e:#}");
continue;
}
};
match client_res {
Ok((client, startup_timeout)) => {
@@ -205,7 +215,13 @@ impl McpConnectionManager {
}
}
let all_tools = list_all_tools(&clients).await?;
let all_tools = match list_all_tools(&clients).await {
Ok(tools) => tools,
Err(e) => {
warn!("Failed to list tools from some MCP servers: {e:#}");
Vec::new()
}
};
let tools = qualify_tools(all_tools);
@@ -270,8 +286,19 @@ async fn list_all_tools(clients: &HashMap<String, ManagedClient>) -> Result<Vec<
let mut aggregated: Vec<ToolInfo> = Vec::with_capacity(join_set.len());
while let Some(join_res) = join_set.join_next().await {
let (server_name, list_result) = join_res?;
let list_result = list_result?;
let (server_name, list_result) = if let Ok(result) = join_res {
result
} else {
warn!("Task panic when listing tools for MCP server: {join_res:#?}");
continue;
};
let list_result = if let Ok(result) = list_result {
result
} else {
warn!("Failed to list tools for MCP server '{server_name}': {list_result:#?}");
continue;
};
for tool in list_result.tools {
let tool_info = ToolInfo {

View File

@@ -23,7 +23,6 @@ use std::path::PathBuf;
use serde::Deserialize;
use serde::Serialize;
use codex_protocol::mcp_protocol::ConversationId;
use std::time::Duration;
use tokio::fs;
use tokio::io::AsyncReadExt;
@@ -31,6 +30,7 @@ use tokio::io::AsyncReadExt;
use crate::config::Config;
use crate::config_types::HistoryPersistence;
use codex_protocol::mcp_protocol::ConversationId;
#[cfg(unix)]
use std::os::unix::fs::OpenOptionsExt;
#[cfg(unix)]

View File

@@ -103,7 +103,7 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
slug, "gpt-4.1",
needs_special_apply_patch_instructions: true,
)
} else if slug.starts_with("gpt-oss") {
} else if slug.starts_with("gpt-oss") || slug.starts_with("openai/gpt-oss") {
model_family!(slug, "gpt-oss", apply_patch_tool_type: Some(ApplyPatchToolType::Function))
} else if slug.starts_with("gpt-4o") {
simple_model_family!(slug, "gpt-4o")

View File

@@ -8,7 +8,6 @@ use std::collections::HashMap;
use crate::model_family::ModelFamily;
use crate::plan_tool::PLAN_TOOL;
use crate::protocol::AskForApproval;
use crate::protocol::SandboxPolicy;
use crate::tool_apply_patch::ApplyPatchToolType;
use crate::tool_apply_patch::create_apply_patch_freeform_tool;
use crate::tool_apply_patch::create_apply_patch_json_tool;
@@ -58,7 +57,7 @@ pub(crate) enum OpenAiTool {
#[derive(Debug, Clone)]
pub enum ConfigShellToolType {
DefaultShell,
ShellWithRequest { sandbox_policy: SandboxPolicy },
ShellWithRequest,
LocalShell,
StreamableShell,
}
@@ -70,17 +69,18 @@ pub(crate) struct ToolsConfig {
pub apply_patch_tool_type: Option<ApplyPatchToolType>,
pub web_search_request: bool,
pub include_view_image_tool: bool,
pub experimental_unified_exec_tool: bool,
}
pub(crate) struct ToolsConfigParams<'a> {
pub(crate) model_family: &'a ModelFamily,
pub(crate) approval_policy: AskForApproval,
pub(crate) sandbox_policy: SandboxPolicy,
pub(crate) include_plan_tool: bool,
pub(crate) include_apply_patch_tool: bool,
pub(crate) include_web_search_request: bool,
pub(crate) use_streamable_shell_tool: bool,
pub(crate) include_view_image_tool: bool,
pub(crate) experimental_unified_exec_tool: bool,
}
impl ToolsConfig {
@@ -88,12 +88,12 @@ impl ToolsConfig {
let ToolsConfigParams {
model_family,
approval_policy,
sandbox_policy,
include_plan_tool,
include_apply_patch_tool,
include_web_search_request,
use_streamable_shell_tool,
include_view_image_tool,
experimental_unified_exec_tool,
} = params;
let mut shell_type = if *use_streamable_shell_tool {
ConfigShellToolType::StreamableShell
@@ -103,9 +103,7 @@ impl ToolsConfig {
ConfigShellToolType::DefaultShell
};
if matches!(approval_policy, AskForApproval::OnRequest) && !use_streamable_shell_tool {
shell_type = ConfigShellToolType::ShellWithRequest {
sandbox_policy: sandbox_policy.clone(),
}
shell_type = ConfigShellToolType::ShellWithRequest;
}
let apply_patch_tool_type = match model_family.apply_patch_tool_type {
@@ -126,6 +124,7 @@ impl ToolsConfig {
apply_patch_tool_type,
web_search_request: *include_web_search_request,
include_view_image_tool: *include_view_image_tool,
experimental_unified_exec_tool: *experimental_unified_exec_tool,
}
}
}
@@ -200,7 +199,56 @@ fn create_shell_tool() -> OpenAiTool {
})
}
fn create_shell_tool_for_sandbox(sandbox_policy: &SandboxPolicy) -> OpenAiTool {
fn create_unified_exec_tool() -> OpenAiTool {
let mut properties = BTreeMap::new();
properties.insert(
"input".to_string(),
JsonSchema::Array {
items: Box::new(JsonSchema::String { description: None }),
description: Some(
"When no session_id is provided, treat the array as the command and arguments \
to launch. When session_id is set, concatenate the strings (in order) and write \
them to the session's stdin."
.to_string(),
),
},
);
properties.insert(
"session_id".to_string(),
JsonSchema::String {
description: Some(
"Identifier for an existing interactive session. If omitted, a new command \
is spawned."
.to_string(),
),
},
);
properties.insert(
"timeout_ms".to_string(),
JsonSchema::Number {
description: Some(
"Maximum time in milliseconds to wait for output after writing the input."
.to_string(),
),
},
);
OpenAiTool::Function(ResponsesApiTool {
name: "unified_exec".to_string(),
description:
"Runs a command in a PTY. Provide a session_id to reuse an existing interactive session.".to_string(),
strict: false,
parameters: JsonSchema::Object {
properties,
required: Some(vec!["input".to_string()]),
additional_properties: Some(false),
},
})
}
const SHELL_TOOL_DESCRIPTION: &str = r#"Runs a shell command and returns its output"#;
fn create_shell_tool_for_sandbox() -> OpenAiTool {
let mut properties = BTreeMap::new();
properties.insert(
"command".to_string(),
@@ -212,82 +260,29 @@ fn create_shell_tool_for_sandbox(sandbox_policy: &SandboxPolicy) -> OpenAiTool {
properties.insert(
"workdir".to_string(),
JsonSchema::String {
description: Some("The working directory to execute the command in".to_string()),
description: Some("Working directory to execute the command in.".to_string()),
},
);
properties.insert(
"timeout_ms".to_string(),
JsonSchema::Number {
description: Some("The timeout for the command in milliseconds".to_string()),
description: Some("Timeout for the command in milliseconds.".to_string()),
},
);
if matches!(sandbox_policy, SandboxPolicy::WorkspaceWrite { .. }) {
properties.insert(
properties.insert(
"with_escalated_permissions".to_string(),
JsonSchema::Boolean {
description: Some("Whether to request escalated permissions. Set to true if command needs to be run without sandbox restrictions".to_string()),
description: Some("Request escalated permissions, only for when a command would otherwise be blocked by the sandbox.".to_string()),
},
);
properties.insert(
properties.insert(
"justification".to_string(),
JsonSchema::String {
description: Some("Only set if with_escalated_permissions is true. 1-sentence explanation of why we want to run this command.".to_string()),
description: Some("Required if and only if with_escalated_permissions == true. One sentence explaining why escalation is needed (e.g., write outside CWD, network fetch, git commit).".to_string()),
},
);
}
let description = match sandbox_policy {
SandboxPolicy::WorkspaceWrite {
network_access,
writable_roots,
..
} => {
format!(
r#"
The shell tool is used to execute shell commands.
- When invoking the shell tool, your call will be running in a sandbox, and some shell commands will require escalated privileges:
- Types of actions that require escalated privileges:
- Writing files other than those in the writable roots
- writable roots:
{}{}
- Examples of commands that require escalated privileges:
- git commit
- npm install or pnpm install
- cargo build
- cargo test
- When invoking a command that will require escalated privileges:
- Provide the with_escalated_permissions parameter with the boolean value true
- Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter."#,
writable_roots.iter().map(|wr| format!(" - {}", wr.to_string_lossy())).collect::<Vec<String>>().join("\n"),
if !network_access {
"\n - Commands that require network access\n"
} else {
""
}
)
}
SandboxPolicy::DangerFullAccess => {
"Runs a shell command and returns its output.".to_string()
}
SandboxPolicy::ReadOnly => {
r#"
The shell tool is used to execute shell commands.
- When invoking the shell tool, your call will be running in a sandbox, and some shell commands (including apply_patch) will require escalated permissions:
- Types of actions that require escalated privileges:
- Writing files
- Applying patches
- Examples of commands that require escalated privileges:
- apply_patch
- git commit
- npm install or pnpm install
- cargo build
- cargo test
- When invoking a command that will require escalated privileges:
- Provide the with_escalated_permissions parameter with the boolean value true
- Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter"#.to_string()
}
};
let description = SHELL_TOOL_DESCRIPTION.to_string();
OpenAiTool::Function(ResponsesApiTool {
name: "shell".to_string(),
@@ -300,7 +295,6 @@ The shell tool is used to execute shell commands.
},
})
}
fn create_view_image_tool() -> OpenAiTool {
// Support only local filesystem path.
let mut properties = BTreeMap::new();
@@ -534,23 +528,27 @@ pub(crate) fn get_openai_tools(
) -> Vec<OpenAiTool> {
let mut tools: Vec<OpenAiTool> = Vec::new();
match &config.shell_type {
ConfigShellToolType::DefaultShell => {
tools.push(create_shell_tool());
}
ConfigShellToolType::ShellWithRequest { sandbox_policy } => {
tools.push(create_shell_tool_for_sandbox(sandbox_policy));
}
ConfigShellToolType::LocalShell => {
tools.push(OpenAiTool::LocalShell {});
}
ConfigShellToolType::StreamableShell => {
tools.push(OpenAiTool::Function(
crate::exec_command::create_exec_command_tool_for_responses_api(),
));
tools.push(OpenAiTool::Function(
crate::exec_command::create_write_stdin_tool_for_responses_api(),
));
if config.experimental_unified_exec_tool {
tools.push(create_unified_exec_tool());
} else {
match &config.shell_type {
ConfigShellToolType::DefaultShell => {
tools.push(create_shell_tool());
}
ConfigShellToolType::ShellWithRequest => {
tools.push(create_shell_tool_for_sandbox());
}
ConfigShellToolType::LocalShell => {
tools.push(OpenAiTool::LocalShell {});
}
ConfigShellToolType::StreamableShell => {
tools.push(OpenAiTool::Function(
crate::exec_command::create_exec_command_tool_for_responses_api(),
));
tools.push(OpenAiTool::Function(
crate::exec_command::create_write_stdin_tool_for_responses_api(),
));
}
}
}
@@ -577,10 +575,8 @@ pub(crate) fn get_openai_tools(
if config.include_view_image_tool {
tools.push(create_view_image_tool());
}
if let Some(mcp_tools) = mcp_tools {
// Ensure deterministic ordering to maximize prompt cache hits.
// HashMap iteration order is non-deterministic, so sort by fully-qualified tool name.
let mut entries: Vec<(String, mcp_types::Tool)> = mcp_tools.into_iter().collect();
entries.sort_by(|a, b| a.0.cmp(&b.0));
@@ -636,18 +632,18 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: true,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(&config, Some(HashMap::new()));
assert_eq_tool_names(
&tools,
&["local_shell", "update_plan", "web_search", "view_image"],
&["unified_exec", "update_plan", "web_search", "view_image"],
);
}
@@ -657,18 +653,18 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: true,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(&config, Some(HashMap::new()));
assert_eq_tool_names(
&tools,
&["shell", "update_plan", "web_search", "view_image"],
&["unified_exec", "update_plan", "web_search", "view_image"],
);
}
@@ -678,12 +674,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(
&config,
@@ -726,7 +722,7 @@ mod tests {
assert_eq_tool_names(
&tools,
&[
"shell",
"unified_exec",
"web_search",
"view_image",
"test_server/do_something_cool",
@@ -783,12 +779,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: false,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
// Intentionally construct a map with keys that would sort alphabetically.
@@ -841,11 +837,11 @@ mod tests {
]);
let tools = get_openai_tools(&config, Some(tools_map));
// Expect shell first, followed by MCP tools sorted by fully-qualified name.
// Expect unified_exec first, followed by MCP tools sorted by fully-qualified name.
assert_eq_tool_names(
&tools,
&[
"shell",
"unified_exec",
"view_image",
"test_server/cool",
"test_server/do",
@@ -860,12 +856,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(
@@ -893,7 +889,7 @@ mod tests {
assert_eq_tool_names(
&tools,
&["shell", "web_search", "view_image", "dash/search"],
&["unified_exec", "web_search", "view_image", "dash/search"],
);
assert_eq!(
@@ -922,12 +918,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(
@@ -953,7 +949,7 @@ mod tests {
assert_eq_tool_names(
&tools,
&["shell", "web_search", "view_image", "dash/paginate"],
&["unified_exec", "web_search", "view_image", "dash/paginate"],
);
assert_eq!(
tools[3],
@@ -979,12 +975,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(
@@ -1008,7 +1004,10 @@ mod tests {
)])),
);
assert_eq_tool_names(&tools, &["shell", "web_search", "view_image", "dash/tags"]);
assert_eq_tool_names(
&tools,
&["unified_exec", "web_search", "view_image", "dash/tags"],
);
assert_eq!(
tools[3],
OpenAiTool::Function(ResponsesApiTool {
@@ -1036,12 +1035,12 @@ mod tests {
let config = ToolsConfig::new(&ToolsConfigParams {
model_family: &model_family,
approval_policy: AskForApproval::Never,
sandbox_policy: SandboxPolicy::ReadOnly,
include_plan_tool: false,
include_apply_patch_tool: false,
include_web_search_request: true,
use_streamable_shell_tool: false,
include_view_image_tool: true,
experimental_unified_exec_tool: true,
});
let tools = get_openai_tools(
@@ -1065,7 +1064,10 @@ mod tests {
)])),
);
assert_eq_tool_names(&tools, &["shell", "web_search", "view_image", "dash/value"]);
assert_eq_tool_names(
&tools,
&["unified_exec", "web_search", "view_image", "dash/value"],
);
assert_eq!(
tools[3],
OpenAiTool::Function(ResponsesApiTool {
@@ -1086,13 +1088,7 @@ mod tests {
#[test]
fn test_shell_tool_for_sandbox_workspace_write() {
let sandbox_policy = SandboxPolicy::WorkspaceWrite {
writable_roots: vec!["workspace".into()],
network_access: false,
exclude_tmpdir_env_var: false,
exclude_slash_tmp: false,
};
let tool = super::create_shell_tool_for_sandbox(&sandbox_policy);
let tool = super::create_shell_tool_for_sandbox();
let OpenAiTool::Function(ResponsesApiTool {
description, name, ..
}) = &tool
@@ -1101,29 +1097,13 @@ mod tests {
};
assert_eq!(name, "shell");
let expected = r#"
The shell tool is used to execute shell commands.
- When invoking the shell tool, your call will be running in a sandbox, and some shell commands will require escalated privileges:
- Types of actions that require escalated privileges:
- Writing files other than those in the writable roots
- writable roots:
- workspace
- Commands that require network access
- Examples of commands that require escalated privileges:
- git commit
- npm install or pnpm install
- cargo build
- cargo test
- When invoking a command that will require escalated privileges:
- Provide the with_escalated_permissions parameter with the boolean value true
- Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter."#;
let expected = super::SHELL_TOOL_DESCRIPTION;
assert_eq!(description, expected);
}
#[test]
fn test_shell_tool_for_sandbox_readonly() {
let tool = super::create_shell_tool_for_sandbox(&SandboxPolicy::ReadOnly);
let tool = super::create_shell_tool_for_sandbox();
let OpenAiTool::Function(ResponsesApiTool {
description, name, ..
}) = &tool
@@ -1132,27 +1112,13 @@ The shell tool is used to execute shell commands.
};
assert_eq!(name, "shell");
let expected = r#"
The shell tool is used to execute shell commands.
- When invoking the shell tool, your call will be running in a sandbox, and some shell commands (including apply_patch) will require escalated permissions:
- Types of actions that require escalated privileges:
- Writing files
- Applying patches
- Examples of commands that require escalated privileges:
- apply_patch
- git commit
- npm install or pnpm install
- cargo build
- cargo test
- When invoking a command that will require escalated privileges:
- Provide the with_escalated_permissions parameter with the boolean value true
- Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter"#;
let expected = super::SHELL_TOOL_DESCRIPTION;
assert_eq!(description, expected);
}
#[test]
fn test_shell_tool_for_sandbox_danger_full_access() {
let tool = super::create_shell_tool_for_sandbox(&SandboxPolicy::DangerFullAccess);
let tool = super::create_shell_tool_for_sandbox();
let OpenAiTool::Function(ResponsesApiTool {
description, name, ..
}) = &tool
@@ -1161,6 +1127,7 @@ The shell tool is used to execute shell commands.
};
assert_eq!(name, "shell");
assert_eq!(description, "Runs a shell command and returns its output.");
let expected = super::SHELL_TOOL_DESCRIPTION;
assert_eq!(description, expected);
}
}

View File

@@ -10,6 +10,9 @@ use time::macros::format_description;
use uuid::Uuid;
use super::SESSIONS_SUBDIR;
use crate::protocol::EventMsg;
use codex_protocol::protocol::RolloutItem;
use codex_protocol::protocol::RolloutLine;
/// Returned page of conversation summaries.
#[derive(Debug, Default, PartialEq)]
@@ -34,7 +37,7 @@ pub struct ConversationItem {
}
/// Hard cap to bound worstcase work per request.
const MAX_SCAN_FILES: usize = 10_000;
const MAX_SCAN_FILES: usize = 100;
const HEAD_RECORD_LIMIT: usize = 10;
/// Pagination cursor identifying a file by timestamp and UUID.
@@ -167,10 +170,16 @@ async fn traverse_directories_for_paths(
if items.len() == page_size {
break 'outer;
}
let head = read_first_jsonl_records(&path, HEAD_RECORD_LIMIT)
.await
.unwrap_or_default();
items.push(ConversationItem { path, head });
// Read head and simultaneously detect message events within the same
// first N JSONL records to avoid a second file read.
let (head, saw_session_meta, saw_user_event) =
read_head_and_flags(&path, HEAD_RECORD_LIMIT)
.await
.unwrap_or((Vec::new(), false, false));
// Apply filters: must have session meta and at least one user message event
if saw_session_meta && saw_user_event {
items.push(ConversationItem { path, head });
}
}
}
}
@@ -273,16 +282,19 @@ fn parse_timestamp_uuid_from_filename(name: &str) -> Option<(OffsetDateTime, Uui
Some((ts, uuid))
}
async fn read_first_jsonl_records(
async fn read_head_and_flags(
path: &Path,
max_records: usize,
) -> io::Result<Vec<serde_json::Value>> {
) -> io::Result<(Vec<serde_json::Value>, bool, bool)> {
use tokio::io::AsyncBufReadExt;
let file = tokio::fs::File::open(path).await?;
let reader = tokio::io::BufReader::new(file);
let mut lines = reader.lines();
let mut head: Vec<serde_json::Value> = Vec::new();
let mut saw_session_meta = false;
let mut saw_user_event = false;
while head.len() < max_records {
let line_opt = lines.next_line().await?;
let Some(line) = line_opt else { break };
@@ -290,9 +302,29 @@ async fn read_first_jsonl_records(
if trimmed.is_empty() {
continue;
}
if let Ok(v) = serde_json::from_str::<serde_json::Value>(trimmed) {
head.push(v);
let parsed: Result<RolloutLine, _> = serde_json::from_str(trimmed);
let Ok(rollout_line) = parsed else { continue };
match rollout_line.item {
RolloutItem::SessionMeta(session_meta_line) => {
if let Ok(val) = serde_json::to_value(session_meta_line) {
head.push(val);
saw_session_meta = true;
}
}
RolloutItem::ResponseItem(item) => {
if let Ok(val) = serde_json::to_value(item) {
head.push(val);
}
}
RolloutItem::EventMsg(ev) => {
if matches!(ev, EventMsg::UserMessage(_)) {
saw_user_event = true;
}
}
}
}
Ok(head)
Ok((head, saw_session_meta, saw_user_event))
}

View File

@@ -1,13 +1,15 @@
//! Rollout module: persistence and discovery of session rollout files.
pub(crate) const SESSIONS_SUBDIR: &str = "sessions";
pub const SESSIONS_SUBDIR: &str = "sessions";
pub const ARCHIVED_SESSIONS_SUBDIR: &str = "archived_sessions";
pub mod list;
pub(crate) mod policy;
pub mod recorder;
pub use codex_protocol::protocol::SessionMeta;
pub use recorder::RolloutRecorder;
pub use recorder::SessionStateSnapshot;
pub use recorder::RolloutRecorderParams;
#[cfg(test)]
pub mod tests;

View File

@@ -1,8 +1,21 @@
use crate::protocol::EventMsg;
use crate::protocol::RolloutItem;
use codex_protocol::models::ResponseItem;
/// Whether a rollout `item` should be persisted in rollout files.
#[inline]
pub(crate) fn is_persisted_response_item(item: &RolloutItem) -> bool {
match item {
RolloutItem::ResponseItem(item) => should_persist_response_item(item),
RolloutItem::EventMsg(ev) => should_persist_event_msg(ev),
// Always persist session meta
RolloutItem::SessionMeta(_) => true,
}
}
/// Whether a `ResponseItem` should be persisted in rollout files.
#[inline]
pub(crate) fn is_persisted_response_item(item: &ResponseItem) -> bool {
pub(crate) fn should_persist_response_item(item: &ResponseItem) -> bool {
match item {
ResponseItem::Message { .. }
| ResponseItem::Reasoning { .. }
@@ -14,3 +27,44 @@ pub(crate) fn is_persisted_response_item(item: &ResponseItem) -> bool {
ResponseItem::WebSearchCall { .. } | ResponseItem::Other => false,
}
}
/// Whether an `EventMsg` should be persisted in rollout files.
#[inline]
pub(crate) fn should_persist_event_msg(ev: &EventMsg) -> bool {
match ev {
EventMsg::UserMessage(_)
| EventMsg::AgentMessage(_)
| EventMsg::AgentReasoning(_)
| EventMsg::AgentReasoningRawContent(_)
| EventMsg::TokenCount(_) => true,
EventMsg::Error(_)
| EventMsg::TaskStarted(_)
| EventMsg::TaskComplete(_)
| EventMsg::AgentMessageDelta(_)
| EventMsg::AgentReasoningDelta(_)
| EventMsg::AgentReasoningRawContentDelta(_)
| EventMsg::AgentReasoningSectionBreak(_)
| EventMsg::SessionConfigured(_)
| EventMsg::McpToolCallBegin(_)
| EventMsg::McpToolCallEnd(_)
| EventMsg::WebSearchBegin(_)
| EventMsg::WebSearchEnd(_)
| EventMsg::ExecCommandBegin(_)
| EventMsg::ExecCommandOutputDelta(_)
| EventMsg::ExecCommandEnd(_)
| EventMsg::ExecApprovalRequest(_)
| EventMsg::ApplyPatchApprovalRequest(_)
| EventMsg::BackgroundEvent(_)
| EventMsg::StreamError(_)
| EventMsg::PatchApplyBegin(_)
| EventMsg::PatchApplyEnd(_)
| EventMsg::TurnDiff(_)
| EventMsg::GetHistoryEntryResponse(_)
| EventMsg::McpListToolsResponse(_)
| EventMsg::ListCustomPromptsResponse(_)
| EventMsg::PlanUpdate(_)
| EventMsg::TurnAborted(_)
| EventMsg::ShutdownComplete
| EventMsg::ConversationPath(_) => false,
}
}

View File

@@ -4,6 +4,7 @@ use std::fs::File;
use std::fs::{self};
use std::io::Error as IoError;
use std::path::Path;
use std::path::PathBuf;
use codex_protocol::mcp_protocol::ConversationId;
use serde::Deserialize;
@@ -25,25 +26,15 @@ use super::list::Cursor;
use super::list::get_conversations;
use super::policy::is_persisted_response_item;
use crate::config::Config;
use crate::conversation_manager::InitialHistory;
use crate::git_info::GitInfo;
use crate::default_client::ORIGINATOR;
use crate::git_info::collect_git_info;
use codex_protocol::models::ResponseItem;
#[derive(Serialize, Deserialize, Clone, Default)]
pub struct SessionMeta {
pub id: ConversationId,
pub timestamp: String,
pub instructions: Option<String>,
}
#[derive(Serialize)]
struct SessionMetaWithGit {
#[serde(flatten)]
meta: SessionMeta,
#[serde(skip_serializing_if = "Option::is_none")]
git: Option<GitInfo>,
}
use codex_protocol::protocol::InitialHistory;
use codex_protocol::protocol::ResumedHistory;
use codex_protocol::protocol::RolloutItem;
use codex_protocol::protocol::RolloutLine;
use codex_protocol::protocol::SessionMeta;
use codex_protocol::protocol::SessionMetaLine;
#[derive(Serialize, Deserialize, Default, Clone)]
pub struct SessionStateSnapshot {}
@@ -70,16 +61,45 @@ pub struct SavedSession {
#[derive(Clone)]
pub struct RolloutRecorder {
tx: Sender<RolloutCmd>,
pub(crate) rollout_path: PathBuf,
}
#[derive(Clone)]
pub enum RolloutRecorderParams {
Create {
conversation_id: ConversationId,
instructions: Option<String>,
},
Resume {
path: PathBuf,
},
}
enum RolloutCmd {
AddItems(Vec<ResponseItem>),
UpdateState(SessionStateSnapshot),
Shutdown { ack: oneshot::Sender<()> },
AddItems(Vec<RolloutItem>),
/// Ensure all prior writes are processed; respond when flushed.
Flush {
ack: oneshot::Sender<()>,
},
Shutdown {
ack: oneshot::Sender<()>,
},
}
impl RolloutRecorderParams {
pub fn new(conversation_id: ConversationId, instructions: Option<String>) -> Self {
Self::Create {
conversation_id,
instructions,
}
}
pub fn resume(path: PathBuf) -> Self {
Self::Resume { path }
}
}
impl RolloutRecorder {
#[allow(dead_code)]
/// List conversations (rollout files) under the provided Codex home directory.
pub async fn list_conversations(
codex_home: &Path,
@@ -92,24 +112,49 @@ impl RolloutRecorder {
/// Attempt to create a new [`RolloutRecorder`]. If the sessions directory
/// cannot be created or the rollout file cannot be opened we return the
/// error so the caller can decide whether to disable persistence.
pub async fn new(
config: &Config,
conversation_id: ConversationId,
instructions: Option<String>,
) -> std::io::Result<Self> {
let LogFileInfo {
file,
conversation_id: session_id,
timestamp,
} = create_log_file(config, conversation_id)?;
pub async fn new(config: &Config, params: RolloutRecorderParams) -> std::io::Result<Self> {
let (file, rollout_path, meta) = match params {
RolloutRecorderParams::Create {
conversation_id,
instructions,
} => {
let LogFileInfo {
file,
path,
conversation_id: session_id,
timestamp,
} = create_log_file(config, conversation_id)?;
let timestamp_format: &[FormatItem] = format_description!(
"[year]-[month]-[day]T[hour]:[minute]:[second].[subsecond digits:3]Z"
);
let timestamp = timestamp
.to_offset(time::UtcOffset::UTC)
.format(timestamp_format)
.map_err(|e| IoError::other(format!("failed to format timestamp: {e}")))?;
let timestamp_format: &[FormatItem] = format_description!(
"[year]-[month]-[day]T[hour]:[minute]:[second].[subsecond digits:3]Z"
);
let timestamp = timestamp
.to_offset(time::UtcOffset::UTC)
.format(timestamp_format)
.map_err(|e| IoError::other(format!("failed to format timestamp: {e}")))?;
(
tokio::fs::File::from_std(file),
path,
Some(SessionMeta {
id: session_id,
timestamp,
cwd: config.cwd.clone(),
originator: ORIGINATOR.value.clone(),
cli_version: env!("CARGO_PKG_VERSION").to_string(),
instructions,
}),
)
}
RolloutRecorderParams::Resume { path } => (
tokio::fs::OpenOptions::new()
.append(true)
.open(&path)
.await?,
path,
None,
),
};
// Clone the cwd for the spawned task to collect git info asynchronously
let cwd = config.cwd.clone();
@@ -122,21 +167,12 @@ impl RolloutRecorder {
// Spawn a Tokio task that owns the file handle and performs async
// writes. Using `tokio::fs::File` keeps everything on the async I/O
// driver instead of blocking the runtime.
tokio::task::spawn(rollout_writer(
tokio::fs::File::from_std(file),
rx,
Some(SessionMeta {
timestamp,
id: session_id,
instructions,
}),
cwd,
));
tokio::task::spawn(rollout_writer(file, rx, meta, cwd));
Ok(Self { tx })
Ok(Self { tx, rollout_path })
}
pub(crate) async fn record_items(&self, items: &[ResponseItem]) -> std::io::Result<()> {
pub(crate) async fn record_items(&self, items: &[RolloutItem]) -> std::io::Result<()> {
let mut filtered = Vec::new();
for item in items {
// Note that function calls may look a bit strange if they are
@@ -155,55 +191,85 @@ impl RolloutRecorder {
.map_err(|e| IoError::other(format!("failed to queue rollout items: {e}")))
}
pub(crate) async fn record_state(&self, state: SessionStateSnapshot) -> std::io::Result<()> {
/// Flush all queued writes and wait until they are committed by the writer task.
pub async fn flush(&self) -> std::io::Result<()> {
let (tx, rx) = oneshot::channel();
self.tx
.send(RolloutCmd::UpdateState(state))
.send(RolloutCmd::Flush { ack: tx })
.await
.map_err(|e| IoError::other(format!("failed to queue rollout state: {e}")))
.map_err(|e| IoError::other(format!("failed to queue rollout flush: {e}")))?;
rx.await
.map_err(|e| IoError::other(format!("failed waiting for rollout flush: {e}")))
}
pub async fn get_rollout_history(path: &Path) -> std::io::Result<InitialHistory> {
pub(crate) async fn get_rollout_history(path: &Path) -> std::io::Result<InitialHistory> {
info!("Resuming rollout from {path:?}");
tracing::error!("Resuming rollout from {path:?}");
let text = tokio::fs::read_to_string(path).await?;
let mut lines = text.lines();
let _ = lines
.next()
.ok_or_else(|| IoError::other("empty session file"))?;
let mut items = Vec::new();
if text.trim().is_empty() {
return Err(IoError::other("empty session file"));
}
for line in lines {
let mut items: Vec<RolloutItem> = Vec::new();
let mut conversation_id: Option<ConversationId> = None;
for line in text.lines() {
if line.trim().is_empty() {
continue;
}
let v: Value = match serde_json::from_str(line) {
Ok(v) => v,
Err(_) => continue,
};
if v.get("record_type")
.and_then(|rt| rt.as_str())
.map(|s| s == "state")
.unwrap_or(false)
{
continue;
}
match serde_json::from_value::<ResponseItem>(v.clone()) {
Ok(item) => {
if is_persisted_response_item(&item) {
items.push(item);
}
}
Err(e) => {
warn!("failed to parse item: {v:?}, error: {e}");
warn!("failed to parse line as JSON: {line:?}, error: {e}");
continue;
}
};
// Parse the rollout line structure
match serde_json::from_value::<RolloutLine>(v.clone()) {
Ok(rollout_line) => match rollout_line.item {
RolloutItem::SessionMeta(session_meta_line) => {
// Use the FIRST SessionMeta encountered in the file as the canonical
// conversation id and main session information. Keep all items intact.
if conversation_id.is_none() {
conversation_id = Some(session_meta_line.meta.id);
}
items.push(RolloutItem::SessionMeta(session_meta_line));
}
RolloutItem::ResponseItem(item) => {
items.push(RolloutItem::ResponseItem(item));
}
RolloutItem::EventMsg(_ev) => {
items.push(RolloutItem::EventMsg(_ev));
}
},
Err(e) => {
warn!("failed to parse rollout line: {v:?}, error: {e}");
}
}
}
info!("Resumed rollout successfully from {path:?}");
tracing::error!(
"Resumed rollout with {} items, conversation ID: {:?}",
items.len(),
conversation_id
);
let conversation_id = conversation_id
.ok_or_else(|| IoError::other("failed to parse conversation ID from rollout file"))?;
if items.is_empty() {
Ok(InitialHistory::New)
} else {
Ok(InitialHistory::Resumed(items))
return Ok(InitialHistory::New);
}
info!("Resumed rollout successfully from {path:?}");
Ok(InitialHistory::Resumed(ResumedHistory {
conversation_id,
history: items,
rollout_path: path.to_path_buf(),
}))
}
pub(crate) fn get_rollout_path(&self) -> PathBuf {
self.rollout_path.clone()
}
pub async fn shutdown(&self) -> std::io::Result<()> {
@@ -226,6 +292,9 @@ struct LogFileInfo {
/// Opened file handle to the rollout file.
file: File,
/// Full path to the rollout file.
path: PathBuf,
/// Session ID (also embedded in filename).
conversation_id: ConversationId,
@@ -265,6 +334,7 @@ fn create_log_file(
Ok(LogFileInfo {
file,
path,
conversation_id,
timestamp,
})
@@ -281,13 +351,15 @@ async fn rollout_writer(
// If we have a meta, collect git info asynchronously and write meta first
if let Some(session_meta) = meta.take() {
let git_info = collect_git_info(&cwd).await;
let session_meta_with_git = SessionMetaWithGit {
let session_meta_line = SessionMetaLine {
meta: session_meta,
git: git_info,
};
// Write the SessionMeta as the first item in the file
writer.write_line(&session_meta_with_git).await?;
// Write the SessionMeta as the first item in the file, wrapped in a rollout line
writer
.write_rollout_item(RolloutItem::SessionMeta(session_meta_line))
.await?;
}
// Process rollout commands
@@ -296,23 +368,17 @@ async fn rollout_writer(
RolloutCmd::AddItems(items) => {
for item in items {
if is_persisted_response_item(&item) {
writer.write_line(&item).await?;
writer.write_rollout_item(item).await?;
}
}
}
RolloutCmd::UpdateState(state) => {
#[derive(Serialize)]
struct StateLine<'a> {
record_type: &'static str,
#[serde(flatten)]
state: &'a SessionStateSnapshot,
RolloutCmd::Flush { ack } => {
// Ensure underlying file is flushed and then ack.
if let Err(e) = writer.file.flush().await {
let _ = ack.send(());
return Err(e);
}
writer
.write_line(&StateLine {
record_type: "state",
state: &state,
})
.await?;
let _ = ack.send(());
}
RolloutCmd::Shutdown { ack } => {
let _ = ack.send(());
@@ -328,10 +394,24 @@ struct JsonlWriter {
}
impl JsonlWriter {
async fn write_rollout_item(&mut self, rollout_item: RolloutItem) -> std::io::Result<()> {
let timestamp_format: &[FormatItem] = format_description!(
"[year]-[month]-[day]T[hour]:[minute]:[second].[subsecond digits:3]Z"
);
let timestamp = OffsetDateTime::now_utc()
.format(timestamp_format)
.map_err(|e| IoError::other(format!("failed to format timestamp: {e}")))?;
let line = RolloutLine {
timestamp,
item: rollout_item,
};
self.write_line(&line).await
}
async fn write_line(&mut self, item: &impl serde::Serialize) -> std::io::Result<()> {
let mut json = serde_json::to_string(item)?;
json.push('\n');
let _ = self.file.write_all(json.as_bytes()).await;
self.file.write_all(json.as_bytes()).await?;
self.file.flush().await?;
Ok(())
}

View File

@@ -42,10 +42,30 @@ fn write_session_file(
let meta = serde_json::json!({
"timestamp": ts_str,
"id": uuid.to_string()
"type": "session_meta",
"payload": {
"id": uuid,
"timestamp": ts_str,
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
}
});
writeln!(file, "{meta}")?;
// Include at least one user message event to satisfy listing filters
let user_event = serde_json::json!({
"timestamp": ts_str,
"type": "event_msg",
"payload": {
"type": "user_message",
"message": "Hello from user",
"kind": "plain"
}
});
writeln!(file, "{user_event}")?;
for i in 0..num_records {
let rec = serde_json::json!({
"record_type": "response",
@@ -93,24 +113,30 @@ async fn test_list_conversations_latest_first() {
.join("01")
.join(format!("rollout-2025-01-01T12-00-00-{u1}.jsonl"));
let head_3 = vec![
serde_json::json!({"timestamp": "2025-01-03T12-00-00", "id": u3.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
serde_json::json!({"record_type": "response", "index": 1}),
serde_json::json!({"record_type": "response", "index": 2}),
];
let head_2 = vec![
serde_json::json!({"timestamp": "2025-01-02T12-00-00", "id": u2.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
serde_json::json!({"record_type": "response", "index": 1}),
serde_json::json!({"record_type": "response", "index": 2}),
];
let head_1 = vec![
serde_json::json!({"timestamp": "2025-01-01T12-00-00", "id": u1.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
serde_json::json!({"record_type": "response", "index": 1}),
serde_json::json!({"record_type": "response", "index": 2}),
];
let head_3 = vec![serde_json::json!({
"id": u3,
"timestamp": "2025-01-03T12-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let head_2 = vec![serde_json::json!({
"id": u2,
"timestamp": "2025-01-02T12-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let head_1 = vec![serde_json::json!({
"id": u1,
"timestamp": "2025-01-01T12-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let expected_cursor: Cursor =
serde_json::from_str(&format!("\"2025-01-01T12-00-00|{u1}\"")).unwrap();
@@ -170,14 +196,22 @@ async fn test_pagination_cursor() {
.join("03")
.join("04")
.join(format!("rollout-2025-03-04T09-00-00-{u4}.jsonl"));
let head_5 = vec![
serde_json::json!({"timestamp": "2025-03-05T09-00-00", "id": u5.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
];
let head_4 = vec![
serde_json::json!({"timestamp": "2025-03-04T09-00-00", "id": u4.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
];
let head_5 = vec![serde_json::json!({
"id": u5,
"timestamp": "2025-03-05T09-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let head_4 = vec![serde_json::json!({
"id": u4,
"timestamp": "2025-03-04T09-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let expected_cursor1: Cursor =
serde_json::from_str(&format!("\"2025-03-04T09-00-00|{u4}\"")).unwrap();
let expected_page1 = ConversationsPage {
@@ -212,14 +246,22 @@ async fn test_pagination_cursor() {
.join("03")
.join("02")
.join(format!("rollout-2025-03-02T09-00-00-{u2}.jsonl"));
let head_3 = vec![
serde_json::json!({"timestamp": "2025-03-03T09-00-00", "id": u3.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
];
let head_2 = vec![
serde_json::json!({"timestamp": "2025-03-02T09-00-00", "id": u2.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
];
let head_3 = vec![serde_json::json!({
"id": u3,
"timestamp": "2025-03-03T09-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let head_2 = vec![serde_json::json!({
"id": u2,
"timestamp": "2025-03-02T09-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let expected_cursor2: Cursor =
serde_json::from_str(&format!("\"2025-03-02T09-00-00|{u2}\"")).unwrap();
let expected_page2 = ConversationsPage {
@@ -248,10 +290,14 @@ async fn test_pagination_cursor() {
.join("03")
.join("01")
.join(format!("rollout-2025-03-01T09-00-00-{u1}.jsonl"));
let head_1 = vec![
serde_json::json!({"timestamp": "2025-03-01T09-00-00", "id": u1.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
];
let head_1 = vec![serde_json::json!({
"id": u1,
"timestamp": "2025-03-01T09-00-00",
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let expected_cursor3: Cursor =
serde_json::from_str(&format!("\"2025-03-01T09-00-00|{u1}\"")).unwrap();
let expected_page3 = ConversationsPage {
@@ -287,11 +333,14 @@ async fn test_get_conversation_contents() {
.join("04")
.join("01")
.join(format!("rollout-2025-04-01T10-30-00-{uuid}.jsonl"));
let expected_head = vec![
serde_json::json!({"timestamp": ts, "id": uuid.to_string()}),
serde_json::json!({"record_type": "response", "index": 0}),
serde_json::json!({"record_type": "response", "index": 1}),
];
let expected_head = vec![serde_json::json!({
"id": uuid,
"timestamp": ts,
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})];
let expected_cursor: Cursor = serde_json::from_str(&format!("\"{ts}|{uuid}\"")).unwrap();
let expected_page = ConversationsPage {
items: vec![ConversationItem {
@@ -305,10 +354,15 @@ async fn test_get_conversation_contents() {
assert_eq!(page, expected_page);
// Entire file contents equality
let meta = serde_json::json!({"timestamp": ts, "id": uuid.to_string()});
let meta = serde_json::json!({"timestamp": ts, "type": "session_meta", "payload": {"id": uuid, "timestamp": ts, "instructions": null, "cwd": ".", "originator": "test_originator", "cli_version": "test_version"}});
let user_event = serde_json::json!({
"timestamp": ts,
"type": "event_msg",
"payload": {"type": "user_message", "message": "Hello from user", "kind": "plain"}
});
let rec0 = serde_json::json!({"record_type": "response", "index": 0});
let rec1 = serde_json::json!({"record_type": "response", "index": 1});
let expected_content = format!("{meta}\n{rec0}\n{rec1}\n");
let expected_content = format!("{meta}\n{user_event}\n{rec0}\n{rec1}\n");
assert_eq!(content, expected_content);
}
@@ -341,7 +395,14 @@ async fn test_stable_ordering_same_second_pagination() {
.join("01")
.join(format!("rollout-2025-07-01T00-00-00-{u2}.jsonl"));
let head = |u: Uuid| -> Vec<serde_json::Value> {
vec![serde_json::json!({"timestamp": ts, "id": u.to_string()})]
vec![serde_json::json!({
"id": u,
"timestamp": ts,
"instructions": null,
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
})]
};
let expected_cursor1: Cursor = serde_json::from_str(&format!("\"{ts}|{u2}\"")).unwrap();
let expected_page1 = ConversationsPage {

View File

@@ -69,3 +69,8 @@
; Added on top of Chrome profile
; Needed for python multiprocessing on MacOS for the SemLock
(allow ipc-posix-sem)
; needed to look up user info, see https://crbug.com/792228
(allow mach-lookup
(global-name "com.apple.system.opendirectoryd.libinfo")
)

View File

@@ -0,0 +1,180 @@
//! Utilities for truncating large chunks of output while preserving a prefix
//! and suffix on UTF-8 boundaries.
/// Truncate the middle of a UTF-8 string to at most `max_bytes` bytes,
/// preserving the beginning and the end. Returns the possibly truncated
/// string and `Some(original_token_count)` (estimated at 4 bytes/token)
/// if truncation occurred; otherwise returns the original string and `None`.
pub(crate) fn truncate_middle(s: &str, max_bytes: usize) -> (String, Option<u64>) {
if s.len() <= max_bytes {
return (s.to_string(), None);
}
let est_tokens = (s.len() as u64).div_ceil(4);
if max_bytes == 0 {
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
fn truncate_on_boundary(input: &str, max_len: usize) -> &str {
if input.len() <= max_len {
return input;
}
let mut end = max_len;
while end > 0 && !input.is_char_boundary(end) {
end -= 1;
}
&input[..end]
}
fn pick_prefix_end(s: &str, left_budget: usize) -> usize {
if let Some(head) = s.get(..left_budget)
&& let Some(i) = head.rfind('\n')
{
return i + 1;
}
truncate_on_boundary(s, left_budget).len()
}
fn pick_suffix_start(s: &str, right_budget: usize) -> usize {
let start_tail = s.len().saturating_sub(right_budget);
if let Some(tail) = s.get(start_tail..)
&& let Some(i) = tail.find('\n')
{
return start_tail + i + 1;
}
let mut idx = start_tail.min(s.len());
while idx < s.len() && !s.is_char_boundary(idx) {
idx += 1;
}
idx
}
let mut guess_tokens = est_tokens;
for _ in 0..4 {
let marker = format!("{guess_tokens} tokens truncated…");
let marker_len = marker.len();
let keep_budget = max_bytes.saturating_sub(marker_len);
if keep_budget == 0 {
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
let left_budget = keep_budget / 2;
let right_budget = keep_budget - left_budget;
let prefix_end = pick_prefix_end(s, left_budget);
let mut suffix_start = pick_suffix_start(s, right_budget);
if suffix_start < prefix_end {
suffix_start = prefix_end;
}
let kept_content_bytes = prefix_end + (s.len() - suffix_start);
let truncated_content_bytes = s.len().saturating_sub(kept_content_bytes);
let new_tokens = (truncated_content_bytes as u64).div_ceil(4);
if new_tokens == guess_tokens {
let mut out = String::with_capacity(marker_len + kept_content_bytes + 1);
out.push_str(&s[..prefix_end]);
out.push_str(&marker);
out.push('\n');
out.push_str(&s[suffix_start..]);
return (out, Some(est_tokens));
}
guess_tokens = new_tokens;
}
let marker = format!("{guess_tokens} tokens truncated…");
let marker_len = marker.len();
let keep_budget = max_bytes.saturating_sub(marker_len);
if keep_budget == 0 {
return (format!("{est_tokens} tokens truncated…"), Some(est_tokens));
}
let left_budget = keep_budget / 2;
let right_budget = keep_budget - left_budget;
let prefix_end = pick_prefix_end(s, left_budget);
let suffix_start = pick_suffix_start(s, right_budget);
let mut out = String::with_capacity(marker_len + prefix_end + (s.len() - suffix_start) + 1);
out.push_str(&s[..prefix_end]);
out.push_str(&marker);
out.push('\n');
out.push_str(&s[suffix_start..]);
(out, Some(est_tokens))
}
#[cfg(test)]
mod tests {
use super::truncate_middle;
#[test]
fn truncate_middle_no_newlines_fallback() {
let s = "abcdefghijklmnopqrstuvwxyz0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ*";
let max_bytes = 32;
let (out, original) = truncate_middle(s, max_bytes);
assert!(out.starts_with("abc"));
assert!(out.contains("tokens truncated"));
assert!(out.ends_with("XYZ*"));
assert_eq!(original, Some((s.len() as u64).div_ceil(4)));
}
#[test]
fn truncate_middle_prefers_newline_boundaries() {
let mut s = String::new();
for i in 1..=20 {
s.push_str(&format!("{i:03}\n"));
}
assert_eq!(s.len(), 80);
let max_bytes = 64;
let (out, tokens) = truncate_middle(&s, max_bytes);
assert!(out.starts_with("001\n002\n003\n004\n"));
assert!(out.contains("tokens truncated"));
assert!(out.ends_with("017\n018\n019\n020\n"));
assert_eq!(tokens, Some(20));
}
#[test]
fn truncate_middle_handles_utf8_content() {
let s = "😀😀😀😀😀😀😀😀😀😀\nsecond line with ascii text\n";
let max_bytes = 32;
let (out, tokens) = truncate_middle(s, max_bytes);
assert!(out.contains("tokens truncated"));
assert!(!out.contains('\u{fffd}'));
assert_eq!(tokens, Some((s.len() as u64).div_ceil(4)));
}
#[test]
fn truncate_middle_prefers_newline_boundaries_2() {
// Build a multi-line string of 20 numbered lines (each "NNN\n").
let mut s = String::new();
for i in 1..=20 {
s.push_str(&format!("{i:03}\n"));
}
// Total length: 20 lines * 4 bytes per line = 80 bytes.
assert_eq!(s.len(), 80);
// Choose a cap that forces truncation while leaving room for
// a few lines on each side after accounting for the marker.
let max_bytes = 64;
// Expect exact output: first 4 lines, marker, last 4 lines, and correct token estimate (80/4 = 20).
assert_eq!(
truncate_middle(&s, max_bytes),
(
r#"001
002
003
004
…12 tokens truncated…
017
018
019
020
"#
.to_string(),
Some(20)
)
);
}
}

View File

@@ -1,4 +1,3 @@
use std::collections::BTreeSet;
use std::collections::HashMap;
use std::fs;
use std::path::Path;
@@ -250,64 +249,6 @@ impl TurnDiffTracker {
}
}
pub fn build_undo_patch(&mut self) -> Result<Option<String>> {
let mut delete_paths: BTreeSet<PathBuf> = BTreeSet::new();
let mut add_entries: Vec<(PathBuf, String)> = Vec::new();
let mut baseline_file_names: Vec<String> =
self.baseline_file_info.keys().cloned().collect();
baseline_file_names.sort();
for internal in baseline_file_names {
let Some(info) = self.baseline_file_info.get(&internal) else {
continue;
};
let current_path = self
.get_path_for_internal(&internal)
.unwrap_or(info.path.clone());
if current_path.exists() {
delete_paths.insert(current_path);
}
if info.oid.as_str() != ZERO_OID {
let content = String::from_utf8(info.content.clone()).map_err(|_| {
anyhow!(
"undo is not supported for non-UTF8 baseline file {}",
info.path.display()
)
})?;
add_entries.push((info.path.clone(), content));
}
}
if delete_paths.is_empty() && add_entries.is_empty() {
return Ok(None);
}
add_entries.sort_by(|(left_path, _), (right_path, _)| left_path.cmp(right_path));
let mut patch = String::from("*** Begin Patch\n");
for path in delete_paths {
patch.push_str(&format!("*** Delete File: {}\n", path.display()));
}
for (path, content) in add_entries {
patch.push_str(&format!("*** Add File: {}\n", path.display()));
if !content.is_empty() {
for line in content.split_terminator('\n') {
patch.push('+');
patch.push_str(line);
patch.push('\n');
}
if !content.ends_with('\n') {
patch.push_str("+\n");
}
}
}
patch.push_str("*** End Patch\n");
Ok(Some(patch))
}
fn get_file_diff(&mut self, internal_file_name: &str) -> String {
let mut aggregated = String::new();
@@ -562,146 +503,6 @@ mod tests {
out
}
fn normalize_patch_for_test(input: &str, root: &Path) -> String {
let root_str = root.display().to_string().replace('\\', "/");
let mut replaced = input.replace('\\', "/");
replaced = replaced.replace(&root_str, "<TMP>");
if let Some(root_name) = root.file_name().and_then(|name| name.to_str()) {
let marker = format!("/{root_name}");
let mut normalized = String::with_capacity(replaced.len());
let mut search_start = 0;
while let Some(relative_pos) = replaced[search_start..].find(&marker) {
let absolute_pos = search_start + relative_pos;
let path_start = replaced[..absolute_pos]
.rfind(['\n', ' '])
.map(|idx| idx + 1)
.unwrap_or(0);
let prefix_end = replaced[path_start..absolute_pos]
.find('/')
.map(|idx| path_start + idx + 1)
.unwrap_or(path_start);
normalized.push_str(&replaced[search_start..prefix_end]);
normalized.push_str("<TMP>");
let after_marker = absolute_pos + marker.len();
let mut rest_start = after_marker;
if after_marker < replaced.len() && replaced.as_bytes()[after_marker] == b'/' {
normalized.push('/');
rest_start += 1;
}
search_start = rest_start;
}
normalized.push_str(&replaced[search_start..]);
replaced = normalized;
}
if !replaced.ends_with('\n') {
replaced.push('\n');
}
replaced
}
#[test]
fn build_undo_patch_returns_none_without_baseline() {
let mut tracker = TurnDiffTracker::new();
assert_eq!(tracker.build_undo_patch().unwrap(), None);
}
#[test]
fn build_undo_patch_restores_updated_file() {
let dir = tempdir().unwrap();
let path = dir.path().join("undo.txt");
fs::write(&path, "before\n").unwrap();
let mut tracker = TurnDiffTracker::new();
let update_changes = HashMap::from([(
path.clone(),
FileChange::Update {
unified_diff: String::new(),
move_path: None,
},
)]);
tracker.on_patch_begin(&update_changes);
fs::write(&path, "after\n").unwrap();
let patch = tracker
.build_undo_patch()
.expect("undo patch")
.expect("some undo patch");
let normalized = normalize_patch_for_test(&patch, dir.path());
let expected = concat!(
"*** Begin Patch\n",
"*** Delete File: <TMP>/undo.txt\n",
"*** Add File: <TMP>/undo.txt\n",
"+before\n",
"*** End Patch\n",
);
assert_eq!(normalized, expected);
}
#[test]
fn build_undo_patch_restores_deleted_file() {
let dir = tempdir().unwrap();
let path = dir.path().join("gone.txt");
fs::write(&path, "gone\n").unwrap();
let mut tracker = TurnDiffTracker::new();
let delete_changes = HashMap::from([(
path.clone(),
FileChange::Delete {
content: "gone\n".to_string(),
},
)]);
tracker.on_patch_begin(&delete_changes);
fs::remove_file(&path).unwrap();
let patch = tracker
.build_undo_patch()
.expect("undo patch")
.expect("some undo patch");
let normalized = normalize_patch_for_test(&patch, dir.path());
let expected = concat!(
"*** Begin Patch\n",
"*** Add File: <TMP>/gone.txt\n",
"+gone\n",
"*** End Patch\n",
);
assert_eq!(normalized, expected);
}
#[test]
fn build_undo_patch_rejects_non_utf8_content() {
let dir = tempdir().unwrap();
let path = dir.path().join("binary.bin");
fs::write(&path, [0xff, 0xfe, 0x00]).unwrap();
let mut tracker = TurnDiffTracker::new();
let update_changes = HashMap::from([(
path.clone(),
FileChange::Update {
unified_diff: String::new(),
move_path: None,
},
)]);
tracker.on_patch_begin(&update_changes);
let err = tracker.build_undo_patch().unwrap_err();
let message = format!("{err:#}");
assert!(
message.contains("undo is not supported for non-UTF8 baseline file"),
"unexpected error message: {message}"
);
}
#[test]
fn accumulates_add_and_update() {
let mut acc = TurnDiffTracker::new();

View File

@@ -0,0 +1,22 @@
use thiserror::Error;
#[derive(Debug, Error)]
pub(crate) enum UnifiedExecError {
#[error("Failed to create unified exec session: {pty_error}")]
CreateSession {
#[source]
pty_error: anyhow::Error,
},
#[error("Unknown session id {session_id}")]
UnknownSessionId { session_id: i32 },
#[error("failed to write to stdin")]
WriteToStdin,
#[error("missing command line for unified exec request")]
MissingCommandLine,
}
impl UnifiedExecError {
pub(crate) fn create_session(error: anyhow::Error) -> Self {
Self::CreateSession { pty_error: error }
}
}

View File

@@ -0,0 +1,653 @@
use portable_pty::CommandBuilder;
use portable_pty::PtySize;
use portable_pty::native_pty_system;
use std::collections::HashMap;
use std::collections::VecDeque;
use std::io::ErrorKind;
use std::io::Read;
use std::sync::Arc;
use std::sync::Mutex as StdMutex;
use std::sync::atomic::AtomicBool;
use std::sync::atomic::AtomicI32;
use std::sync::atomic::Ordering;
use tokio::sync::Mutex;
use tokio::sync::Notify;
use tokio::sync::mpsc;
use tokio::task::JoinHandle;
use tokio::time::Duration;
use tokio::time::Instant;
use crate::exec_command::ExecCommandSession;
use crate::truncate::truncate_middle;
mod errors;
pub(crate) use errors::UnifiedExecError;
const DEFAULT_TIMEOUT_MS: u64 = 1_000;
const MAX_TIMEOUT_MS: u64 = 60_000;
const UNIFIED_EXEC_OUTPUT_MAX_BYTES: usize = 128 * 1024; // 128 KiB
#[derive(Debug)]
pub(crate) struct UnifiedExecRequest<'a> {
pub session_id: Option<i32>,
pub input_chunks: &'a [String],
pub timeout_ms: Option<u64>,
}
#[derive(Debug, Clone, PartialEq)]
pub(crate) struct UnifiedExecResult {
pub session_id: Option<i32>,
pub output: String,
}
#[derive(Debug, Default)]
pub(crate) struct UnifiedExecSessionManager {
next_session_id: AtomicI32,
sessions: Mutex<HashMap<i32, ManagedUnifiedExecSession>>,
}
#[derive(Debug)]
struct ManagedUnifiedExecSession {
session: ExecCommandSession,
output_buffer: OutputBuffer,
/// Notifies waiters whenever new output has been appended to
/// `output_buffer`, allowing clients to poll for fresh data.
output_notify: Arc<Notify>,
output_task: JoinHandle<()>,
}
#[derive(Debug, Default)]
struct OutputBufferState {
chunks: VecDeque<Vec<u8>>,
total_bytes: usize,
}
impl OutputBufferState {
fn push_chunk(&mut self, chunk: Vec<u8>) {
self.total_bytes = self.total_bytes.saturating_add(chunk.len());
self.chunks.push_back(chunk);
let mut excess = self
.total_bytes
.saturating_sub(UNIFIED_EXEC_OUTPUT_MAX_BYTES);
while excess > 0 {
match self.chunks.front_mut() {
Some(front) if excess >= front.len() => {
excess -= front.len();
self.total_bytes = self.total_bytes.saturating_sub(front.len());
self.chunks.pop_front();
}
Some(front) => {
front.drain(..excess);
self.total_bytes = self.total_bytes.saturating_sub(excess);
break;
}
None => break,
}
}
}
fn drain(&mut self) -> Vec<Vec<u8>> {
let drained: Vec<Vec<u8>> = self.chunks.drain(..).collect();
self.total_bytes = 0;
drained
}
}
type OutputBuffer = Arc<Mutex<OutputBufferState>>;
type OutputHandles = (OutputBuffer, Arc<Notify>);
impl ManagedUnifiedExecSession {
fn new(session: ExecCommandSession) -> Self {
let output_buffer = Arc::new(Mutex::new(OutputBufferState::default()));
let output_notify = Arc::new(Notify::new());
let mut receiver = session.output_receiver();
let buffer_clone = Arc::clone(&output_buffer);
let notify_clone = Arc::clone(&output_notify);
let output_task = tokio::spawn(async move {
while let Ok(chunk) = receiver.recv().await {
let mut guard = buffer_clone.lock().await;
guard.push_chunk(chunk);
drop(guard);
notify_clone.notify_waiters();
}
});
Self {
session,
output_buffer,
output_notify,
output_task,
}
}
fn writer_sender(&self) -> mpsc::Sender<Vec<u8>> {
self.session.writer_sender()
}
fn output_handles(&self) -> OutputHandles {
(
Arc::clone(&self.output_buffer),
Arc::clone(&self.output_notify),
)
}
fn has_exited(&self) -> bool {
self.session.has_exited()
}
}
impl Drop for ManagedUnifiedExecSession {
fn drop(&mut self) {
self.output_task.abort();
}
}
impl UnifiedExecSessionManager {
pub async fn handle_request(
&self,
request: UnifiedExecRequest<'_>,
) -> Result<UnifiedExecResult, UnifiedExecError> {
let (timeout_ms, timeout_warning) = match request.timeout_ms {
Some(requested) if requested > MAX_TIMEOUT_MS => (
MAX_TIMEOUT_MS,
Some(format!(
"Warning: requested timeout {requested}ms exceeds maximum of {MAX_TIMEOUT_MS}ms; clamping to {MAX_TIMEOUT_MS}ms.\n"
)),
),
Some(requested) => (requested, None),
None => (DEFAULT_TIMEOUT_MS, None),
};
let mut new_session: Option<ManagedUnifiedExecSession> = None;
let session_id;
let writer_tx;
let output_buffer;
let output_notify;
if let Some(existing_id) = request.session_id {
let mut sessions = self.sessions.lock().await;
match sessions.get(&existing_id) {
Some(session) => {
if session.has_exited() {
sessions.remove(&existing_id);
return Err(UnifiedExecError::UnknownSessionId {
session_id: existing_id,
});
}
let (buffer, notify) = session.output_handles();
session_id = existing_id;
writer_tx = session.writer_sender();
output_buffer = buffer;
output_notify = notify;
}
None => {
return Err(UnifiedExecError::UnknownSessionId {
session_id: existing_id,
});
}
}
drop(sessions);
} else {
let command = request.input_chunks.to_vec();
let new_id = self.next_session_id.fetch_add(1, Ordering::SeqCst);
let session = create_unified_exec_session(&command).await?;
let managed_session = ManagedUnifiedExecSession::new(session);
let (buffer, notify) = managed_session.output_handles();
writer_tx = managed_session.writer_sender();
output_buffer = buffer;
output_notify = notify;
session_id = new_id;
new_session = Some(managed_session);
};
if request.session_id.is_some() {
let joined_input = request.input_chunks.join(" ");
if !joined_input.is_empty() && writer_tx.send(joined_input.into_bytes()).await.is_err()
{
return Err(UnifiedExecError::WriteToStdin);
}
}
let mut collected: Vec<u8> = Vec::with_capacity(4096);
let start = Instant::now();
let deadline = start + Duration::from_millis(timeout_ms);
loop {
let drained_chunks;
let mut wait_for_output = None;
{
let mut guard = output_buffer.lock().await;
drained_chunks = guard.drain();
if drained_chunks.is_empty() {
wait_for_output = Some(output_notify.notified());
}
}
if drained_chunks.is_empty() {
let remaining = deadline.saturating_duration_since(Instant::now());
if remaining == Duration::ZERO {
break;
}
let notified = wait_for_output.unwrap_or_else(|| output_notify.notified());
tokio::pin!(notified);
tokio::select! {
_ = &mut notified => {}
_ = tokio::time::sleep(remaining) => break,
}
continue;
}
for chunk in drained_chunks {
collected.extend_from_slice(&chunk);
}
if Instant::now() >= deadline {
break;
}
}
let (output, _maybe_tokens) = truncate_middle(
&String::from_utf8_lossy(&collected),
UNIFIED_EXEC_OUTPUT_MAX_BYTES,
);
let output = if let Some(warning) = timeout_warning {
format!("{warning}{output}")
} else {
output
};
let should_store_session = if let Some(session) = new_session.as_ref() {
!session.has_exited()
} else if request.session_id.is_some() {
let mut sessions = self.sessions.lock().await;
if let Some(existing) = sessions.get(&session_id) {
if existing.has_exited() {
sessions.remove(&session_id);
false
} else {
true
}
} else {
false
}
} else {
true
};
if should_store_session {
if let Some(session) = new_session {
self.sessions.lock().await.insert(session_id, session);
}
Ok(UnifiedExecResult {
session_id: Some(session_id),
output,
})
} else {
Ok(UnifiedExecResult {
session_id: None,
output,
})
}
}
}
async fn create_unified_exec_session(
command: &[String],
) -> Result<ExecCommandSession, UnifiedExecError> {
if command.is_empty() {
return Err(UnifiedExecError::MissingCommandLine);
}
let pty_system = native_pty_system();
let pair = pty_system
.openpty(PtySize {
rows: 24,
cols: 80,
pixel_width: 0,
pixel_height: 0,
})
.map_err(UnifiedExecError::create_session)?;
// Safe thanks to the check at the top of the function.
let mut command_builder = CommandBuilder::new(command[0].clone());
for arg in &command[1..] {
command_builder.arg(arg);
}
let mut child = pair
.slave
.spawn_command(command_builder)
.map_err(UnifiedExecError::create_session)?;
let killer = child.clone_killer();
let (writer_tx, mut writer_rx) = mpsc::channel::<Vec<u8>>(128);
let (output_tx, _) = tokio::sync::broadcast::channel::<Vec<u8>>(256);
let mut reader = pair
.master
.try_clone_reader()
.map_err(UnifiedExecError::create_session)?;
let output_tx_clone = output_tx.clone();
let reader_handle = tokio::task::spawn_blocking(move || {
let mut buf = [0u8; 8192];
loop {
match reader.read(&mut buf) {
Ok(0) => break,
Ok(n) => {
let _ = output_tx_clone.send(buf[..n].to_vec());
}
Err(ref e) if e.kind() == ErrorKind::Interrupted => continue,
Err(ref e) if e.kind() == ErrorKind::WouldBlock => {
std::thread::sleep(Duration::from_millis(5));
continue;
}
Err(_) => break,
}
}
});
let writer = pair
.master
.take_writer()
.map_err(UnifiedExecError::create_session)?;
let writer = Arc::new(StdMutex::new(writer));
let writer_handle = tokio::spawn({
let writer = writer.clone();
async move {
while let Some(bytes) = writer_rx.recv().await {
let writer = writer.clone();
let _ = tokio::task::spawn_blocking(move || {
if let Ok(mut guard) = writer.lock() {
use std::io::Write;
let _ = guard.write_all(&bytes);
let _ = guard.flush();
}
})
.await;
}
}
});
let exit_status = Arc::new(AtomicBool::new(false));
let wait_exit_status = Arc::clone(&exit_status);
let wait_handle = tokio::task::spawn_blocking(move || {
let _ = child.wait();
wait_exit_status.store(true, Ordering::SeqCst);
});
Ok(ExecCommandSession::new(
writer_tx,
output_tx,
killer,
reader_handle,
writer_handle,
wait_handle,
exit_status,
))
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn push_chunk_trims_only_excess_bytes() {
let mut buffer = OutputBufferState::default();
buffer.push_chunk(vec![b'a'; UNIFIED_EXEC_OUTPUT_MAX_BYTES]);
buffer.push_chunk(vec![b'b']);
buffer.push_chunk(vec![b'c']);
assert_eq!(buffer.total_bytes, UNIFIED_EXEC_OUTPUT_MAX_BYTES);
assert_eq!(buffer.chunks.len(), 3);
assert_eq!(
buffer.chunks.front().unwrap().len(),
UNIFIED_EXEC_OUTPUT_MAX_BYTES - 2
);
assert_eq!(buffer.chunks.pop_back().unwrap(), vec![b'c']);
assert_eq!(buffer.chunks.pop_back().unwrap(), vec![b'b']);
}
#[cfg(unix)]
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn unified_exec_persists_across_requests_jif() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let open_shell = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["bash".to_string(), "-i".to_string()],
timeout_ms: Some(1_500),
})
.await?;
let session_id = open_shell.session_id.expect("expected session_id");
manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &[
"export".to_string(),
"CODEX_INTERACTIVE_SHELL_VAR=codex\n".to_string(),
],
timeout_ms: Some(2_500),
})
.await?;
let out_2 = manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &["echo $CODEX_INTERACTIVE_SHELL_VAR\n".to_string()],
timeout_ms: Some(1_500),
})
.await?;
assert!(out_2.output.contains("codex"));
Ok(())
}
#[cfg(unix)]
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn multi_unified_exec_sessions() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let shell_a = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["/bin/bash".to_string(), "-i".to_string()],
timeout_ms: Some(1_500),
})
.await?;
let session_a = shell_a.session_id.expect("expected session id");
manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_a),
input_chunks: &["export CODEX_INTERACTIVE_SHELL_VAR=codex\n".to_string()],
timeout_ms: Some(1_500),
})
.await?;
let out_2 = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &[
"echo".to_string(),
"$CODEX_INTERACTIVE_SHELL_VAR\n".to_string(),
],
timeout_ms: Some(1_500),
})
.await?;
assert!(!out_2.output.contains("codex"));
let out_3 = manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_a),
input_chunks: &["echo $CODEX_INTERACTIVE_SHELL_VAR\n".to_string()],
timeout_ms: Some(1_500),
})
.await?;
assert!(out_3.output.contains("codex"));
Ok(())
}
#[cfg(unix)]
#[tokio::test]
async fn unified_exec_timeouts() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let open_shell = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["bash".to_string(), "-i".to_string()],
timeout_ms: Some(1_500),
})
.await?;
let session_id = open_shell.session_id.expect("expected session id");
manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &[
"export".to_string(),
"CODEX_INTERACTIVE_SHELL_VAR=codex\n".to_string(),
],
timeout_ms: Some(1_500),
})
.await?;
let out_2 = manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &["sleep 5 && echo $CODEX_INTERACTIVE_SHELL_VAR\n".to_string()],
timeout_ms: Some(10),
})
.await?;
assert!(!out_2.output.contains("codex"));
tokio::time::sleep(Duration::from_secs(7)).await;
let empty = Vec::new();
let out_3 = manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &empty,
timeout_ms: Some(100),
})
.await?;
assert!(out_3.output.contains("codex"));
Ok(())
}
#[cfg(unix)]
#[tokio::test]
async fn requests_with_large_timeout_are_capped() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let result = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["echo".to_string(), "codex".to_string()],
timeout_ms: Some(120_000),
})
.await?;
assert!(result.output.starts_with(
"Warning: requested timeout 120000ms exceeds maximum of 60000ms; clamping to 60000ms.\n"
));
assert!(result.output.contains("codex"));
Ok(())
}
#[cfg(unix)]
#[tokio::test]
async fn completed_commands_do_not_persist_sessions() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let result = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["/bin/echo".to_string(), "codex".to_string()],
timeout_ms: Some(1_500),
})
.await?;
assert!(result.session_id.is_none());
assert!(result.output.contains("codex"));
assert!(manager.sessions.lock().await.is_empty());
Ok(())
}
#[cfg(unix)]
#[tokio::test]
async fn correct_path_resolution() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let result = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["echo".to_string(), "codex".to_string()],
timeout_ms: Some(1_500),
})
.await?;
assert!(result.session_id.is_none());
assert!(result.output.contains("codex"));
assert!(manager.sessions.lock().await.is_empty());
Ok(())
}
#[cfg(unix)]
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn reusing_completed_session_returns_unknown_session() -> Result<(), UnifiedExecError> {
let manager = UnifiedExecSessionManager::default();
let open_shell = manager
.handle_request(UnifiedExecRequest {
session_id: None,
input_chunks: &["/bin/bash".to_string(), "-i".to_string()],
timeout_ms: Some(1_500),
})
.await?;
let session_id = open_shell.session_id.expect("expected session id");
manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &["exit\n".to_string()],
timeout_ms: Some(1_500),
})
.await?;
tokio::time::sleep(Duration::from_millis(200)).await;
let err = manager
.handle_request(UnifiedExecRequest {
session_id: Some(session_id),
input_chunks: &[],
timeout_ms: Some(100),
})
.await
.expect_err("expected unknown session error");
match err {
UnifiedExecError::UnknownSessionId { session_id: err_id } => {
assert_eq!(err_id, session_id);
}
other => panic!("expected UnknownSessionId, got {other:?}"),
}
assert!(!manager.sessions.lock().await.contains_key(&session_id));
Ok(())
}
}

View File

@@ -1,4 +1,6 @@
use assert_cmd::Command as AssertCommand;
use codex_core::RolloutRecorder;
use codex_core::protocol::GitInfo;
use codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR;
use std::time::Duration;
use std::time::Instant;
@@ -77,6 +79,22 @@ async fn chat_mode_stream_cli() {
assert_eq!(hi_lines, 1, "Expected exactly one line with 'hi'");
server.verify().await;
// Verify a new session rollout was created and is discoverable via list_conversations
let page = RolloutRecorder::list_conversations(home.path(), 10, None)
.await
.expect("list conversations");
assert!(
!page.items.is_empty(),
"expected at least one session to be listed"
);
// First line of head must be the SessionMeta payload (id/timestamp)
let head0 = page.items[0].head.first().expect("missing head record");
assert!(head0.get("id").is_some(), "head[0] missing id");
assert!(
head0.get("timestamp").is_some(),
"head[0] missing timestamp"
);
}
/// Verify that passing `-c experimental_instructions_file=...` to the CLI
@@ -297,8 +315,10 @@ async fn integration_creates_and_checks_session_file() {
Ok(v) => v,
Err(_) => continue,
};
if item.get("type").and_then(|t| t.as_str()) == Some("message")
&& let Some(c) = item.get("content")
if item.get("type").and_then(|t| t.as_str()) == Some("response_item")
&& let Some(payload) = item.get("payload")
&& payload.get("type").and_then(|t| t.as_str()) == Some("message")
&& let Some(c) = payload.get("content")
&& c.to_string().contains(&marker)
{
matching_path = Some(path.to_path_buf());
@@ -361,9 +381,16 @@ async fn integration_creates_and_checks_session_file() {
.unwrap_or_else(|_| panic!("missing session meta line"));
let meta: serde_json::Value = serde_json::from_str(meta_line)
.unwrap_or_else(|_| panic!("Failed to parse session meta line as JSON"));
assert!(meta.get("id").is_some(), "SessionMeta missing id");
assert_eq!(
meta.get("type").and_then(|v| v.as_str()),
Some("session_meta")
);
let payload = meta
.get("payload")
.unwrap_or_else(|| panic!("Missing payload in meta line"));
assert!(payload.get("id").is_some(), "SessionMeta missing id");
assert!(
meta.get("timestamp").is_some(),
payload.get("timestamp").is_some(),
"SessionMeta missing timestamp"
);
@@ -375,8 +402,10 @@ async fn integration_creates_and_checks_session_file() {
let Ok(item) = serde_json::from_str::<serde_json::Value>(line) else {
continue;
};
if item.get("type").and_then(|t| t.as_str()) == Some("message")
&& let Some(c) = item.get("content")
if item.get("type").and_then(|t| t.as_str()) == Some("response_item")
&& let Some(payload) = item.get("payload")
&& payload.get("type").and_then(|t| t.as_str()) == Some("message")
&& let Some(c) = payload.get("content")
&& c.to_string().contains(&marker)
{
found_message = true;
@@ -388,8 +417,7 @@ async fn integration_creates_and_checks_session_file() {
"No message found in session file containing the marker"
);
// Second run: resume should create a NEW session file that contains both old and new history.
let orig_len = content.lines().count();
// Second run: resume should update the existing file.
let marker2 = format!("integration-resume-{}", Uuid::new_v4());
let prompt2 = format!("echo {marker2}");
// Crossplatform safe resume override. On Windows, backslashes in a TOML string must be escaped
@@ -449,8 +477,8 @@ async fn integration_creates_and_checks_session_file() {
}
let resumed_path = resumed_path.expect("No resumed session file found containing the marker2");
// Resume should have written to a new file, not the original one.
assert_ne!(
// Resume should write to the existing log file.
assert_eq!(
resumed_path, path,
"resume should create a new session file"
);
@@ -464,14 +492,6 @@ async fn integration_creates_and_checks_session_file() {
resumed_content.contains(&marker2),
"resumed file missing resumed marker"
);
// Original file should remain unchanged.
let content_after = std::fs::read_to_string(&path).unwrap();
assert_eq!(
content_after.lines().count(),
orig_len,
"original rollout file should not change on resume"
);
}
/// Integration test to verify git info is collected and recorded in session files.
@@ -598,7 +618,7 @@ async fn integration_git_info_unit_test() {
// 5. Test serialization to ensure it works in SessionMeta
let serialized = serde_json::to_string(&git_info).unwrap();
let deserialized: codex_core::git_info::GitInfo = serde_json::from_str(&serialized).unwrap();
let deserialized: GitInfo = serde_json::from_str(&serialized).unwrap();
assert_eq!(git_info.commit_hash, deserialized.commit_hash);
assert_eq!(git_info.branch, deserialized.branch);

View File

@@ -15,6 +15,7 @@ use core_test_support::wait_for_event;
use serde_json::json;
use std::io::Write;
use tempfile::TempDir;
use uuid::Uuid;
use wiremock::Mock;
use wiremock::MockServer;
use wiremock::ResponseTemplate;
@@ -122,11 +123,22 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
let tmpdir = TempDir::new().unwrap();
let session_path = tmpdir.path().join("resume-session.jsonl");
let mut f = std::fs::File::create(&session_path).unwrap();
// First line: meta (content not used by reader other than non-empty)
let convo_id = Uuid::new_v4();
writeln!(
f,
"{}",
serde_json::json!({"meta":"test","instructions":"be nice"})
json!({
"timestamp": "2024-01-01T00:00:00.000Z",
"type": "session_meta",
"payload": {
"id": convo_id,
"timestamp": "2024-01-01T00:00:00Z",
"instructions": "be nice",
"cwd": ".",
"originator": "test_originator",
"cli_version": "test_version"
}
})
)
.unwrap();
@@ -138,7 +150,17 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
text: "resumed user message".to_string(),
}],
};
writeln!(f, "{}", serde_json::to_string(&prior_user).unwrap()).unwrap();
let prior_user_json = serde_json::to_value(&prior_user).unwrap();
writeln!(
f,
"{}",
json!({
"timestamp": "2024-01-01T00:00:01.000Z",
"type": "response_item",
"payload": prior_user_json
})
)
.unwrap();
// Prior item: system message (excluded from API history)
let prior_system = codex_protocol::models::ResponseItem::Message {
@@ -148,7 +170,17 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
text: "resumed system instruction".to_string(),
}],
};
writeln!(f, "{}", serde_json::to_string(&prior_system).unwrap()).unwrap();
let prior_system_json = serde_json::to_value(&prior_system).unwrap();
writeln!(
f,
"{}",
json!({
"timestamp": "2024-01-01T00:00:02.000Z",
"type": "response_item",
"payload": prior_system_json
})
)
.unwrap();
// Prior item: assistant message
let prior_item = codex_protocol::models::ResponseItem::Message {
@@ -158,7 +190,17 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
text: "resumed assistant message".to_string(),
}],
};
writeln!(f, "{}", serde_json::to_string(&prior_item).unwrap()).unwrap();
let prior_item_json = serde_json::to_value(&prior_item).unwrap();
writeln!(
f,
"{}",
json!({
"timestamp": "2024-01-01T00:00:03.000Z",
"type": "response_item",
"payload": prior_item_json
})
)
.unwrap();
drop(f);
// Mock server that will receive the resumed request
@@ -196,16 +238,13 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
.await
.expect("create new conversation");
// 1) Assert initial_messages contains the prior user + assistant messages as EventMsg entries
// 1) Assert initial_messages only includes existing EventMsg entries; response items are not converted
let initial_msgs = session_configured
.initial_messages
.clone()
.expect("expected initial messages for resumed session");
.expect("expected initial messages option for resumed session");
let initial_json = serde_json::to_value(&initial_msgs).unwrap();
let expected_initial_json = serde_json::json!([
{ "type": "user_message", "message": "resumed user message", "kind": "plain" },
{ "type": "agent_message", "message": "resumed assistant message" }
]);
let expected_initial_json = json!([]);
assert_eq!(initial_json, expected_initial_json);
// 2) Submit new input; the request body must include the prior item followed by the new user input.
@@ -221,7 +260,7 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
let request = &server.received_requests().await.unwrap()[0];
let request_body = request.body_json::<serde_json::Value>().unwrap();
let expected_input = serde_json::json!([
let expected_input = json!([
{
"type": "message",
"role": "user",
@@ -238,7 +277,25 @@ async fn resume_includes_initial_messages_and_sends_prior_items() {
"content": [{ "type": "input_text", "text": "hello" }]
}
]);
assert_eq!(request_body["input"], expected_input);
let input_array = request_body
.get("input")
.and_then(|v| v.as_array())
.cloned()
.expect("input array in request body");
let filtered: Vec<serde_json::Value> = input_array
.into_iter()
.filter(|item| {
let text = item
.get("content")
.and_then(|c| c.as_array())
.and_then(|a| a.first())
.and_then(|o| o.get("text"))
.and_then(|t| t.as_str())
.unwrap_or("");
!text.contains("<environment_context>")
})
.collect();
assert_eq!(serde_json::json!(filtered), expected_input);
}
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
@@ -371,56 +428,6 @@ async fn includes_base_instructions_override_in_request() {
);
}
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn originator_config_override_is_used() {
// Mock server
let server = MockServer::start().await;
let first = ResponseTemplate::new(200)
.insert_header("content-type", "text/event-stream")
.set_body_raw(sse_completed("resp1"), "text/event-stream");
Mock::given(method("POST"))
.and(path("/v1/responses"))
.respond_with(first)
.expect(1)
.mount(&server)
.await;
let model_provider = ModelProviderInfo {
base_url: Some(format!("{}/v1", server.uri())),
..built_in_model_providers()["openai"].clone()
};
let codex_home = TempDir::new().unwrap();
let mut config = load_default_config_for_test(&codex_home);
config.model_provider = model_provider;
config.responses_originator_header = "my_override".to_owned();
let conversation_manager =
ConversationManager::with_auth(CodexAuth::from_api_key("Test API Key"));
let codex = conversation_manager
.new_conversation(config)
.await
.expect("create new conversation")
.conversation;
codex
.submit(Op::UserInput {
items: vec![InputItem::Text {
text: "hello".into(),
}],
})
.await
.unwrap();
wait_for_event(&codex, |ev| matches!(ev, EventMsg::TaskComplete(_))).await;
let request = &server.received_requests().await.unwrap()[0];
let request_originator = request.headers.get("originator").unwrap();
assert_eq!(request_originator.to_str().unwrap(), "my_override");
}
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn chatgpt_auth_sends_correct_request() {
if std::env::var(CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
@@ -546,15 +553,12 @@ async fn prefers_chatgpt_token_when_config_prefers_chatgpt() {
config.model_provider = model_provider;
config.preferred_auth_method = AuthMode::ChatGPT;
let auth_manager = match CodexAuth::from_codex_home(
codex_home.path(),
config.preferred_auth_method,
&config.responses_originator_header,
) {
Ok(Some(auth)) => codex_core::AuthManager::from_auth_for_testing(auth),
Ok(None) => panic!("No CodexAuth found in codex_home"),
Err(e) => panic!("Failed to load CodexAuth: {e}"),
};
let auth_manager =
match CodexAuth::from_codex_home(codex_home.path(), config.preferred_auth_method) {
Ok(Some(auth)) => codex_core::AuthManager::from_auth_for_testing(auth),
Ok(None) => panic!("No CodexAuth found in codex_home"),
Err(e) => panic!("Failed to load CodexAuth: {e}"),
};
let conversation_manager = ConversationManager::new(auth_manager);
let NewConversation {
conversation: codex,
@@ -622,15 +626,12 @@ async fn prefers_apikey_when_config_prefers_apikey_even_with_chatgpt_tokens() {
config.model_provider = model_provider;
config.preferred_auth_method = AuthMode::ApiKey;
let auth_manager = match CodexAuth::from_codex_home(
codex_home.path(),
config.preferred_auth_method,
&config.responses_originator_header,
) {
Ok(Some(auth)) => codex_core::AuthManager::from_auth_for_testing(auth),
Ok(None) => panic!("No CodexAuth found in codex_home"),
Err(e) => panic!("Failed to load CodexAuth: {e}"),
};
let auth_manager =
match CodexAuth::from_codex_home(codex_home.path(), config.preferred_auth_method) {
Ok(Some(auth)) => codex_core::AuthManager::from_auth_for_testing(auth),
Ok(None) => panic!("No CodexAuth found in codex_home"),
Err(e) => panic!("Failed to load CodexAuth: {e}"),
};
let conversation_manager = ConversationManager::new(auth_manager);
let NewConversation {
conversation: codex,
@@ -967,34 +968,6 @@ async fn history_dedupes_streamed_and_final_messages_across_turns() {
assert_eq!(requests.len(), 3, "expected 3 requests (one per turn)");
// Replace full-array compare with tail-only raw JSON compare using a single hard-coded value.
let r3_tail_expected = serde_json::json!([
{
"type": "message",
"role": "user",
"content": [{"type":"input_text","text":"U1"}]
},
{
"type": "message",
"role": "assistant",
"content": [{"type":"output_text","text":"Hey there!\n"}]
},
{
"type": "message",
"role": "user",
"content": [{"type":"input_text","text":"U2"}]
},
{
"type": "message",
"role": "assistant",
"content": [{"type":"output_text","text":"Hey there!\n"}]
},
{
"type": "message",
"role": "user",
"content": [{"type":"input_text","text":"U3"}]
}
]);
let r3_input_array = requests[2]
.body_json::<serde_json::Value>()
.unwrap()
@@ -1002,12 +975,60 @@ async fn history_dedupes_streamed_and_final_messages_across_turns() {
.and_then(|v| v.as_array())
.cloned()
.expect("r3 missing input array");
// skipping earlier context and developer messages
let tail_len = r3_tail_expected.as_array().unwrap().len();
let actual_tail = &r3_input_array[r3_input_array.len() - tail_len..];
// We only assert on the last 5 items of the input history for request 3.
// With per-turn environment context injected, the last 5 should be:
// [env_ctx, U2, assistant("Hey there!\n"), env_ctx, U3]
let actual_tail = &r3_input_array[r3_input_array.len() - 5..];
// env_ctx 1
assert_eq!(actual_tail[0]["type"], serde_json::json!("message"));
assert_eq!(actual_tail[0]["role"], serde_json::json!("user"));
let env_text_1 = &actual_tail[0]["content"][0]["text"];
assert!(
env_text_1
.as_str()
.expect("env text should be string")
.contains("<environment_context>")
);
// U2
assert_eq!(
serde_json::Value::Array(actual_tail.to_vec()),
r3_tail_expected,
"request 3 tail mismatch",
actual_tail[1],
serde_json::json!({
"type": "message",
"role": "user",
"content": [ { "type": "input_text", "text": "U2" } ]
})
);
// assistant response
assert_eq!(
actual_tail[2],
serde_json::json!({
"type": "message",
"role": "assistant",
"content": [ { "type": "output_text", "text": "Hey there!\n" } ]
})
);
// env_ctx 2
assert_eq!(actual_tail[3]["type"], serde_json::json!("message"));
assert_eq!(actual_tail[3]["role"], serde_json::json!("user"));
let env_text_2 = &actual_tail[3]["content"][0]["text"];
assert!(
env_text_2
.as_str()
.expect("env text should be string")
.contains("<environment_context>")
);
// U3
assert_eq!(
actual_tail[4],
serde_json::json!({
"type": "message",
"role": "user",
"content": [ { "type": "input_text", "text": "U3" } ]
})
);
}

View File

@@ -1,12 +1,16 @@
use codex_core::CodexAuth;
use codex_core::ContentItem;
use codex_core::ConversationManager;
use codex_core::ModelProviderInfo;
use codex_core::NewConversation;
use codex_core::ResponseItem;
use codex_core::built_in_model_providers;
use codex_core::protocol::ConversationHistoryResponseEvent;
use codex_core::protocol::ConversationPathResponseEvent;
use codex_core::protocol::EventMsg;
use codex_core::protocol::InputItem;
use codex_core::protocol::Op;
use codex_core::protocol::RolloutItem;
use codex_core::protocol::RolloutLine;
use core_test_support::load_default_config_for_test;
use core_test_support::wait_for_event;
use tempfile::TempDir;
@@ -71,84 +75,121 @@ async fn fork_conversation_twice_drops_to_first_message() {
let _ = wait_for_event(&codex, |ev| matches!(ev, EventMsg::TaskComplete(_))).await;
}
// Request history from the base conversation.
codex.submit(Op::GetHistory).await.unwrap();
// Request history from the base conversation to obtain rollout path.
codex.submit(Op::GetPath).await.unwrap();
let base_history =
wait_for_event(&codex, |ev| matches!(ev, EventMsg::ConversationHistory(_))).await;
// Capture entries from the base history and compute expected prefixes after each fork.
let entries_after_three = match &base_history {
EventMsg::ConversationHistory(ConversationHistoryResponseEvent { entries, .. }) => {
entries.clone()
}
wait_for_event(&codex, |ev| matches!(ev, EventMsg::ConversationPath(_))).await;
let base_path = match &base_history {
EventMsg::ConversationPath(ConversationPathResponseEvent { path, .. }) => path.clone(),
_ => panic!("expected ConversationHistory event"),
};
// History layout for this test:
// [0] user instructions,
// [1] environment context,
// [2] "first" user message,
// [3] "second" user message,
// [4] "third" user message.
// Fork 1: drops the last user message and everything after.
let expected_after_first = vec![
entries_after_three[0].clone(),
entries_after_three[1].clone(),
entries_after_three[2].clone(),
entries_after_three[3].clone(),
];
// GetHistory flushes before returning the path; no wait needed.
// Fork 2: drops the last user message and everything after.
// [0] user instructions,
// [1] environment context,
// [2] "first" user message,
let expected_after_second = vec![
entries_after_three[0].clone(),
entries_after_three[1].clone(),
entries_after_three[2].clone(),
];
// Helper: read rollout items (excluding SessionMeta) from a JSONL path.
let read_items = |p: &std::path::Path| -> Vec<RolloutItem> {
let text = std::fs::read_to_string(p).expect("read rollout file");
let mut items: Vec<RolloutItem> = Vec::new();
for line in text.lines() {
if line.trim().is_empty() {
continue;
}
let v: serde_json::Value = serde_json::from_str(line).expect("jsonl line");
let rl: RolloutLine = serde_json::from_value(v).expect("rollout line");
match rl.item {
RolloutItem::SessionMeta(_) => {}
other => items.push(other),
}
}
items
};
// Fork once with n=1 → drops the last user message and everything after.
// Compute expected prefixes after each fork by truncating base rollout at nth-from-last user input.
let base_items = read_items(&base_path);
let find_user_input_positions = |items: &[RolloutItem]| -> Vec<usize> {
let mut pos = Vec::new();
for (i, it) in items.iter().enumerate() {
if let RolloutItem::ResponseItem(ResponseItem::Message { role, content, .. }) = it
&& role == "user"
{
// Consider any user message as an input boundary; recorder stores both EventMsg and ResponseItem.
// We specifically look for input items, which are represented as ContentItem::InputText.
if content
.iter()
.any(|c| matches!(c, ContentItem::InputText { .. }))
{
pos.push(i);
}
}
}
pos
};
let user_inputs = find_user_input_positions(&base_items);
// After dropping last user input (n=1), cut strictly before that input if present, else empty.
let cut1 = user_inputs
.get(user_inputs.len().saturating_sub(1))
.copied()
.unwrap_or(0);
let expected_after_first: Vec<RolloutItem> = base_items[..cut1].to_vec();
// After dropping again (n=1 on fork1), compute expected relative to fork1's rollout.
// Fork once with n=1 → drops the last user input and everything after.
let NewConversation {
conversation: codex_fork1,
..
} = conversation_manager
.fork_conversation(entries_after_three.clone(), 1, config_for_fork.clone())
.fork_conversation(1, config_for_fork.clone(), base_path.clone())
.await
.expect("fork 1");
codex_fork1.submit(Op::GetHistory).await.unwrap();
codex_fork1.submit(Op::GetPath).await.unwrap();
let fork1_history = wait_for_event(&codex_fork1, |ev| {
matches!(ev, EventMsg::ConversationHistory(_))
matches!(ev, EventMsg::ConversationPath(_))
})
.await;
let entries_after_first_fork = match &fork1_history {
EventMsg::ConversationHistory(ConversationHistoryResponseEvent { entries, .. }) => {
assert!(matches!(
fork1_history,
EventMsg::ConversationHistory(ConversationHistoryResponseEvent { ref entries, .. }) if *entries == expected_after_first
));
entries.clone()
}
let fork1_path = match &fork1_history {
EventMsg::ConversationPath(ConversationPathResponseEvent { path, .. }) => path.clone(),
_ => panic!("expected ConversationHistory event after first fork"),
};
// GetHistory on fork1 flushed; the file is ready.
let fork1_items = read_items(&fork1_path);
pretty_assertions::assert_eq!(
serde_json::to_value(&fork1_items).unwrap(),
serde_json::to_value(&expected_after_first).unwrap()
);
// Fork again with n=1 → drops the (new) last user message, leaving only the first.
let NewConversation {
conversation: codex_fork2,
..
} = conversation_manager
.fork_conversation(entries_after_first_fork.clone(), 1, config_for_fork.clone())
.fork_conversation(1, config_for_fork.clone(), fork1_path.clone())
.await
.expect("fork 2");
codex_fork2.submit(Op::GetHistory).await.unwrap();
codex_fork2.submit(Op::GetPath).await.unwrap();
let fork2_history = wait_for_event(&codex_fork2, |ev| {
matches!(ev, EventMsg::ConversationHistory(_))
matches!(ev, EventMsg::ConversationPath(_))
})
.await;
assert!(matches!(
fork2_history,
EventMsg::ConversationHistory(ConversationHistoryResponseEvent { ref entries, .. }) if *entries == expected_after_second
));
let fork2_path = match &fork2_history {
EventMsg::ConversationPath(ConversationPathResponseEvent { path, .. }) => path.clone(),
_ => panic!("expected ConversationHistory event after second fork"),
};
// GetHistory on fork2 flushed; the file is ready.
let fork1_items = read_items(&fork1_path);
let fork1_user_inputs = find_user_input_positions(&fork1_items);
let cut_last_on_fork1 = fork1_user_inputs
.get(fork1_user_inputs.len().saturating_sub(1))
.copied()
.unwrap_or(0);
let expected_after_second: Vec<RolloutItem> = fork1_items[..cut_last_on_fork1].to_vec();
let fork2_items = read_items(&fork2_path);
pretty_assertions::assert_eq!(
serde_json::to_value(&fork2_items).unwrap(),
serde_json::to_value(&expected_after_second).unwrap()
);
}

View File

@@ -191,7 +191,8 @@ async fn prompt_tools_are_consistent_across_requests() {
let expected_instructions: &str = include_str!("../../prompt.md");
// our internal implementation is responsible for keeping tools in sync
// with the OpenAI schema, so we just verify the tool presence here
let expected_tools_names: &[&str] = &["shell", "update_plan", "apply_patch", "view_image"];
let expected_tools_names: &[&str] =
&["unified_exec", "update_plan", "apply_patch", "view_image"];
let body0 = requests[0].body_json::<serde_json::Value>().unwrap();
assert_eq!(
body0["instructions"],
@@ -271,7 +272,7 @@ async fn prefixes_context_and_instructions_once_and_consistently_across_requests
let shell = default_user_shell().await;
let expected_env_text = format!(
let expected_env_text_init = format!(
r#"<environment_context>
<cwd>{}</cwd>
<approval_policy>on-request</approval_policy>
@@ -284,13 +285,28 @@ async fn prefixes_context_and_instructions_once_and_consistently_across_requests
None => String::new(),
}
);
// Per-turn environment context omits the shell tag.
let expected_env_text_turn = format!(
r#"<environment_context>
<cwd>{}</cwd>
<approval_policy>on-request</approval_policy>
<sandbox_mode>read-only</sandbox_mode>
<network_access>restricted</network_access>
</environment_context>"#,
cwd.path().to_string_lossy(),
);
let expected_ui_text =
"<user_instructions>\n\nbe consistent and helpful\n\n</user_instructions>";
let expected_env_msg = serde_json::json!({
let expected_env_msg_init = serde_json::json!({
"type": "message",
"role": "user",
"content": [ { "type": "input_text", "text": expected_env_text } ]
"content": [ { "type": "input_text", "text": expected_env_text_init } ]
});
let expected_env_msg_turn = serde_json::json!({
"type": "message",
"role": "user",
"content": [ { "type": "input_text", "text": expected_env_text_turn } ]
});
let expected_ui_msg = serde_json::json!({
"type": "message",
@@ -306,7 +322,12 @@ async fn prefixes_context_and_instructions_once_and_consistently_across_requests
let body1 = requests[0].body_json::<serde_json::Value>().unwrap();
assert_eq!(
body1["input"],
serde_json::json!([expected_ui_msg, expected_env_msg, expected_user_message_1])
serde_json::json!([
expected_ui_msg,
expected_env_msg_init,
expected_env_msg_turn,
expected_user_message_1
])
);
let expected_user_message_2 = serde_json::json!({
@@ -318,7 +339,7 @@ async fn prefixes_context_and_instructions_once_and_consistently_across_requests
let expected_body2 = serde_json::json!(
[
body1["input"].as_array().unwrap().as_slice(),
[expected_user_message_2].as_slice(),
[expected_env_msg_turn, expected_user_message_2].as_slice(),
]
.concat()
);
@@ -426,11 +447,17 @@ async fn overrides_turn_context_but_keeps_cached_prefix_and_key_constant() {
// After overriding the turn context, the environment context should be emitted again
// reflecting the new approval policy and sandbox settings. Omit cwd because it did
// not change.
let expected_env_text_2 = r#"<environment_context>
let expected_env_text_2 = format!(
r#"<environment_context>
<approval_policy>never</approval_policy>
<sandbox_mode>workspace-write</sandbox_mode>
<network_access>enabled</network_access>
</environment_context>"#;
<writable_roots>
<root>{}</root>
</writable_roots>
</environment_context>"#,
writable.path().to_string_lossy()
);
let expected_env_msg_2 = serde_json::json!({
"type": "message",
"role": "user",
@@ -540,10 +567,24 @@ async fn per_turn_overrides_keep_cached_prefix_and_key_constant() {
"role": "user",
"content": [ { "type": "input_text", "text": "hello 2" } ]
});
let expected_env_text_2 = format!(
r#"<environment_context>
<cwd>{}</cwd>
<approval_policy>never</approval_policy>
<sandbox_mode>workspace-write</sandbox_mode>
<network_access>enabled</network_access>
</environment_context>"#,
new_cwd.path().to_string_lossy()
);
let expected_env_msg_2 = serde_json::json!({
"type": "message",
"role": "user",
"content": [ { "type": "input_text", "text": expected_env_text_2 } ]
});
let expected_body2 = serde_json::json!(
[
body1["input"].as_array().unwrap().as_slice(),
[expected_user_message_2].as_slice(),
[expected_env_msg_2, expected_user_message_2].as_slice(),
]
.concat()
);

View File

@@ -159,6 +159,41 @@ async fn read_only_forbids_all_writes() {
.await;
}
/// Verify that user lookups via `pwd.getpwuid(os.getuid())` work under the
/// seatbelt sandbox. Prior to allowing the necessary machlookup for
/// OpenDirectory libinfo, this would fail with `KeyError: getpwuid(): uid not found`.
#[tokio::test]
async fn python_getpwuid_works_under_seatbelt() {
if std::env::var(CODEX_SANDBOX_ENV_VAR) == Ok("seatbelt".to_string()) {
eprintln!("{CODEX_SANDBOX_ENV_VAR} is set to 'seatbelt', skipping test.");
return;
}
// ReadOnly is sufficient here since we are only exercising user lookup.
let policy = SandboxPolicy::ReadOnly;
let mut child = spawn_command_under_seatbelt(
vec![
"python3".to_string(),
"-c".to_string(),
// Print the passwd struct; success implies lookup worked.
"import pwd, os; print(pwd.getpwuid(os.getuid()))".to_string(),
],
&policy,
std::env::current_dir().expect("should be able to get current dir"),
StdioPolicy::RedirectForShellTool,
HashMap::new(),
)
.await
.expect("should be able to spawn python under seatbelt");
let status = child
.wait()
.await
.expect("should be able to wait for child process");
assert!(status.success(), "python exited with {status:?}");
}
#[expect(clippy::expect_used)]
fn create_test_scenario(tmp: &TempDir) -> TestScenario {
let repo_parent = tmp.path().to_path_buf();

View File

@@ -25,7 +25,6 @@ codex-common = { path = "../common", features = [
"sandbox_summary",
] }
codex-core = { path = "../core" }
codex-login = { path = "../login" }
codex-ollama = { path = "../ollama" }
codex-protocol = { path = "../protocol" }
owo-colors = "4.2.0"

View File

@@ -26,6 +26,7 @@ use codex_core::protocol::TurnAbortReason;
use codex_core::protocol::TurnDiffEvent;
use codex_core::protocol::WebSearchBeginEvent;
use codex_core::protocol::WebSearchEndEvent;
use codex_protocol::num_format::format_with_separators;
use owo_colors::OwoColorize;
use owo_colors::Style;
use shlex::try_join;
@@ -194,7 +195,7 @@ impl EventProcessor for EventProcessorWithHumanOutput {
ts_println!(
self,
"tokens used: {}",
usage_info.total_token_usage.blended_total()
format_with_separators(usage_info.total_token_usage.blended_total())
);
}
}
@@ -522,6 +523,7 @@ impl EventProcessor for EventProcessorWithHumanOutput {
history_log_id: _,
history_entry_count: _,
initial_messages: _,
rollout_path: _,
} = session_configured_event;
ts_println!(
@@ -557,7 +559,7 @@ impl EventProcessor for EventProcessorWithHumanOutput {
}
},
EventMsg::ShutdownComplete => return CodexStatus::Shutdown,
EventMsg::ConversationHistory(_) => {}
EventMsg::ConversationPath(_) => {}
EventMsg::UserMessage(_) => {}
}
CodexStatus::Running

View File

@@ -190,7 +190,6 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
let conversation_manager = ConversationManager::new(AuthManager::shared(
config.codex_home.clone(),
config.preferred_auth_method,
config.responses_originator_header.clone(),
));
let NewConversation {
conversation_id: _,

View File

@@ -15,9 +15,7 @@ path = "src/lib.rs"
workspace = true
[target.'cfg(target_os = "linux")'.dependencies]
anyhow = "1"
clap = { version = "4", features = ["derive"] }
codex-common = { path = "../common", features = ["cli"] }
codex-core = { path = "../core" }
landlock = "0.4.1"
libc = "0.2.175"

View File

@@ -17,7 +17,6 @@ serde = { version = "1", features = ["derive"] }
serde_json = "1"
sha2 = "0.10"
tempfile = "3"
thiserror = "2.0.16"
tiny_http = "0.12"
tokio = { version = "1", features = [
"io-std",
@@ -31,5 +30,4 @@ urlencoding = "2.1"
webbrowser = "1.0"
[dev-dependencies]
pretty_assertions = "1.4.1"
tempfile = "3"

View File

@@ -16,6 +16,7 @@ use base64::Engine;
use chrono::Utc;
use codex_core::auth::AuthDotJson;
use codex_core::auth::get_auth_file;
use codex_core::default_client::ORIGINATOR;
use codex_core::token_data::TokenData;
use codex_core::token_data::parse_id_token;
use rand::RngCore;
@@ -35,11 +36,10 @@ pub struct ServerOptions {
pub port: u16,
pub open_browser: bool,
pub force_state: Option<String>,
pub originator: String,
}
impl ServerOptions {
pub fn new(codex_home: PathBuf, client_id: String, originator: String) -> Self {
pub fn new(codex_home: PathBuf, client_id: String) -> Self {
Self {
codex_home,
client_id: client_id.to_string(),
@@ -47,7 +47,6 @@ impl ServerOptions {
port: DEFAULT_PORT,
open_browser: true,
force_state: None,
originator,
}
}
}
@@ -103,14 +102,7 @@ pub fn run_login_server(opts: ServerOptions) -> io::Result<LoginServer> {
let server = Arc::new(server);
let redirect_uri = format!("http://localhost:{actual_port}/auth/callback");
let auth_url = build_authorize_url(
&opts.issuer,
&opts.client_id,
&redirect_uri,
&pkce,
&state,
&opts.originator,
);
let auth_url = build_authorize_url(&opts.issuer, &opts.client_id, &redirect_uri, &pkce, &state);
if opts.open_browser {
let _ = webbrowser::open(&auth_url);
@@ -311,7 +303,6 @@ fn build_authorize_url(
redirect_uri: &str,
pkce: &PkceCodes,
state: &str,
originator: &str,
) -> String {
let query = vec![
("response_type", "code"),
@@ -323,7 +314,7 @@ fn build_authorize_url(
("id_token_add_organizations", "true"),
("codex_cli_simplified_flow", "true"),
("state", state),
("originator", originator),
("originator", ORIGINATOR.value.as_str()),
];
let qs = query
.into_iter()

View File

@@ -102,7 +102,6 @@ async fn end_to_end_login_flow_persists_auth_json() {
port: 0,
open_browser: false,
force_state: Some(state),
originator: "test_originator".to_string(),
};
let server = run_login_server(opts).unwrap();
let login_port = server.actual_port;
@@ -161,7 +160,6 @@ async fn creates_missing_codex_home_dir() {
port: 0,
open_browser: false,
force_state: Some(state),
originator: "test_originator".to_string(),
};
let server = run_login_server(opts).unwrap();
let login_port = server.actual_port;
@@ -202,7 +200,6 @@ async fn cancels_previous_login_server_when_port_is_in_use() {
port: 0,
open_browser: false,
force_state: Some("cancel_state".to_string()),
originator: "test_originator".to_string(),
};
let first_server = run_login_server(first_opts).unwrap();
@@ -221,7 +218,6 @@ async fn cancels_previous_login_server_when_port_is_in_use() {
port: login_port,
open_browser: false,
force_state: Some("cancel_state_2".to_string()),
originator: "test_originator".to_string(),
};
let second_server = run_login_server(second_opts).unwrap();

View File

@@ -64,6 +64,9 @@ async fn main() -> Result<()> {
name: "codex-mcp-client".to_owned(),
version: env!("CARGO_PKG_VERSION").to_owned(),
title: Some("Codex".to_string()),
// This field is used by Codex when it is an MCP server: it should
// not be used when Codex is an MCP client.
user_agent: None,
},
protocol_version: MCP_SCHEMA_VERSION.to_owned(),
};

View File

@@ -26,7 +26,6 @@ schemars = "0.8.22"
serde = { version = "1", features = ["derive"] }
serde_json = "1"
shlex = "1.3.0"
strum_macros = "0.27.2"
tokio = { version = "1", features = [
"io-std",
"macros",
@@ -41,8 +40,9 @@ uuid = { version = "1", features = ["serde", "v4"] }
[dev-dependencies]
assert_cmd = "2"
base64 = "0.22"
mcp_test_support = { path = "tests/common" }
os_info = "3.12.0"
pretty_assertions = "1.4.1"
tempfile = "3"
tokio-test = "0.4"
wiremock = "0.6"

View File

@@ -1,8 +1,3 @@
use std::collections::HashMap;
use std::path::PathBuf;
use std::sync::Arc;
use std::time::Duration;
use crate::error_code::INTERNAL_ERROR_CODE;
use crate::error_code::INVALID_REQUEST_ERROR_CODE;
use crate::json_to_toml::json_to_toml;
@@ -14,11 +9,15 @@ use codex_core::ConversationManager;
use codex_core::Cursor as RolloutCursor;
use codex_core::NewConversation;
use codex_core::RolloutRecorder;
use codex_core::SessionMeta;
use codex_core::auth::CLIENT_ID;
use codex_core::auth::get_auth_file;
use codex_core::auth::try_read_auth_json;
use codex_core::config::Config;
use codex_core::config::ConfigOverrides;
use codex_core::config::ConfigToml;
use codex_core::config::load_config_as_toml;
use codex_core::default_client::get_codex_user_agent;
use codex_core::exec::ExecParams;
use codex_core::exec_env::create_env;
use codex_core::get_platform_sandbox;
@@ -38,6 +37,8 @@ use codex_protocol::mcp_protocol::AddConversationListenerParams;
use codex_protocol::mcp_protocol::AddConversationSubscriptionResponse;
use codex_protocol::mcp_protocol::ApplyPatchApprovalParams;
use codex_protocol::mcp_protocol::ApplyPatchApprovalResponse;
use codex_protocol::mcp_protocol::ArchiveConversationParams;
use codex_protocol::mcp_protocol::ArchiveConversationResponse;
use codex_protocol::mcp_protocol::AuthMode;
use codex_protocol::mcp_protocol::AuthStatusChangeNotification;
use codex_protocol::mcp_protocol::ClientRequest;
@@ -48,6 +49,7 @@ use codex_protocol::mcp_protocol::ExecArbitraryCommandResponse;
use codex_protocol::mcp_protocol::ExecCommandApprovalParams;
use codex_protocol::mcp_protocol::ExecCommandApprovalResponse;
use codex_protocol::mcp_protocol::ExecOneOffCommandParams;
use codex_protocol::mcp_protocol::GetUserAgentResponse;
use codex_protocol::mcp_protocol::GetUserSavedConfigResponse;
use codex_protocol::mcp_protocol::GitDiffToRemoteResponse;
use codex_protocol::mcp_protocol::InputItem as WireInputItem;
@@ -67,12 +69,25 @@ use codex_protocol::mcp_protocol::SendUserMessageResponse;
use codex_protocol::mcp_protocol::SendUserTurnParams;
use codex_protocol::mcp_protocol::SendUserTurnResponse;
use codex_protocol::mcp_protocol::ServerNotification;
use codex_protocol::mcp_protocol::UserInfoResponse;
use codex_protocol::mcp_protocol::UserSavedConfig;
use codex_protocol::models::ContentItem;
use codex_protocol::models::ResponseItem;
use codex_protocol::protocol::InputMessageKind;
use codex_protocol::protocol::USER_MESSAGE_BEGIN;
use mcp_types::JSONRPCErrorError;
use mcp_types::RequestId;
use std::collections::HashMap;
use std::ffi::OsStr;
use std::path::PathBuf;
use std::sync::Arc;
use std::time::Duration;
use tokio::select;
use tokio::sync::Mutex;
use tokio::sync::oneshot;
use tracing::error;
use tracing::info;
use tracing::warn;
use uuid::Uuid;
// Duration before a ChatGPT login attempt is abandoned.
@@ -136,6 +151,9 @@ impl CodexMessageProcessor {
ClientRequest::ResumeConversation { request_id, params } => {
self.handle_resume_conversation(request_id, params).await;
}
ClientRequest::ArchiveConversation { request_id, params } => {
self.archive_conversation(request_id, params).await;
}
ClientRequest::SendUserMessage { request_id, params } => {
self.send_user_message(request_id, params).await;
}
@@ -169,6 +187,12 @@ impl CodexMessageProcessor {
ClientRequest::GetUserSavedConfig { request_id } => {
self.get_user_saved_config(request_id).await;
}
ClientRequest::GetUserAgent { request_id } => {
self.get_user_agent(request_id).await;
}
ClientRequest::UserInfo { request_id } => {
self.get_user_info(request_id).await;
}
ClientRequest::ExecOneOffCommand { request_id, params } => {
self.exec_one_off_command(request_id, params).await;
}
@@ -180,11 +204,7 @@ impl CodexMessageProcessor {
let opts = LoginServerOptions {
open_browser: false,
..LoginServerOptions::new(
config.codex_home.clone(),
CLIENT_ID.to_string(),
config.responses_originator_header.clone(),
)
..LoginServerOptions::new(config.codex_home.clone(), CLIENT_ID.to_string())
};
enum LoginChatGptReply {
@@ -384,6 +404,12 @@ impl CodexMessageProcessor {
self.outgoing.send_response(request_id, response).await;
}
async fn get_user_agent(&self, request_id: RequestId) {
let user_agent = get_codex_user_agent();
let response = GetUserAgentResponse { user_agent };
self.outgoing.send_response(request_id, response).await;
}
async fn get_user_saved_config(&self, request_id: RequestId) {
let toml_value = match load_config_as_toml(&self.config.codex_home) {
Ok(val) => val,
@@ -419,6 +445,18 @@ impl CodexMessageProcessor {
self.outgoing.send_response(request_id, response).await;
}
async fn get_user_info(&self, request_id: RequestId) {
// Read alleged user email from auth.json (best-effort; not verified).
let auth_path = get_auth_file(&self.config.codex_home);
let alleged_user_email = match try_read_auth_json(&auth_path) {
Ok(auth) => auth.tokens.and_then(|t| t.id_token.email),
Err(_) => None,
};
let response = UserInfoResponse { alleged_user_email };
self.outgoing.send_response(request_id, response).await;
}
async fn exec_one_off_command(&self, request_id: RequestId, params: ExecOneOffCommandParams) {
tracing::debug!("ExecOneOffCommand params: {params:?}");
@@ -513,6 +551,7 @@ impl CodexMessageProcessor {
let response = NewConversationResponse {
conversation_id,
model: session_configured.model,
rollout_path: session_configured.rollout_path,
};
self.outgoing.send_response(request_id, response).await;
}
@@ -559,16 +598,11 @@ impl CodexMessageProcessor {
}
};
// Build summaries
let mut items: Vec<ConversationSummary> = Vec::new();
for it in page.items.into_iter() {
let (timestamp, preview) = extract_ts_and_preview(&it.head);
items.push(ConversationSummary {
path: it.path,
preview,
timestamp,
});
}
let items = page
.items
.into_iter()
.filter_map(|it| extract_conversation_summary(it.path, &it.head))
.collect();
// Encode next_cursor as a plain string
let next_cursor = match page.next_cursor {
@@ -622,19 +656,29 @@ impl CodexMessageProcessor {
session_configured,
..
}) => {
let event = codex_core::protocol::Event {
let event = Event {
id: "".to_string(),
msg: codex_core::protocol::EventMsg::SessionConfigured(
session_configured.clone(),
),
msg: EventMsg::SessionConfigured(session_configured.clone()),
};
self.outgoing.send_event_as_notification(&event, None).await;
let initial_messages = session_configured.initial_messages.map(|msgs| {
msgs.into_iter()
.filter(|event| {
// Don't send non-plain user messages (like user instructions
// or environment context) back so they don't get rendered.
if let EventMsg::UserMessage(user_message) = event {
return matches!(user_message.kind, Some(InputMessageKind::Plain));
}
true
})
.collect()
});
// Reply with conversation id + model and initial messages (when present)
let response = codex_protocol::mcp_protocol::ResumeConversationResponse {
conversation_id,
model: session_configured.model.clone(),
initial_messages: session_configured.initial_messages.clone(),
initial_messages,
};
self.outgoing.send_response(request_id, response).await;
}
@@ -649,6 +693,141 @@ impl CodexMessageProcessor {
}
}
async fn archive_conversation(&self, request_id: RequestId, params: ArchiveConversationParams) {
let ArchiveConversationParams {
conversation_id,
rollout_path,
} = params;
// Verify that the rollout path is in the sessions directory or else
// a malicious client could specify an arbitrary path.
let rollout_folder = self.config.codex_home.join(codex_core::SESSIONS_SUBDIR);
let canonical_rollout_path = tokio::fs::canonicalize(&rollout_path).await;
let canonical_rollout_path = if let Ok(path) = canonical_rollout_path
&& path.starts_with(&rollout_folder)
{
path
} else {
let error = JSONRPCErrorError {
code: INVALID_REQUEST_ERROR_CODE,
message: format!(
"rollout path `{}` must be in sessions directory",
rollout_path.display()
),
data: None,
};
self.outgoing.send_error(request_id, error).await;
return;
};
let required_suffix = format!("{}.jsonl", conversation_id.0);
let Some(file_name) = canonical_rollout_path.file_name().map(OsStr::to_owned) else {
let error = JSONRPCErrorError {
code: INVALID_REQUEST_ERROR_CODE,
message: format!(
"rollout path `{}` missing file name",
rollout_path.display()
),
data: None,
};
self.outgoing.send_error(request_id, error).await;
return;
};
if !file_name
.to_string_lossy()
.ends_with(required_suffix.as_str())
{
let error = JSONRPCErrorError {
code: INVALID_REQUEST_ERROR_CODE,
message: format!(
"rollout path `{}` does not match conversation id {conversation_id}",
rollout_path.display()
),
data: None,
};
self.outgoing.send_error(request_id, error).await;
return;
}
let removed_conversation = self
.conversation_manager
.remove_conversation(&conversation_id)
.await;
if let Some(conversation) = removed_conversation {
info!("conversation {conversation_id} was active; shutting down");
let conversation_clone = conversation.clone();
let notify = Arc::new(tokio::sync::Notify::new());
let notify_clone = notify.clone();
// Establish the listener for ShutdownComplete before submitting
// Shutdown so it is not missed.
let is_shutdown = tokio::spawn(async move {
loop {
select! {
_ = notify_clone.notified() => {
break;
}
event = conversation_clone.next_event() => {
if let Ok(event) = event && matches!(event.msg, EventMsg::ShutdownComplete) {
break;
}
}
}
}
});
// Request shutdown.
match conversation.submit(Op::Shutdown).await {
Ok(_) => {
// Successfully submitted Shutdown; wait before proceeding.
select! {
_ = is_shutdown => {
// Normal shutdown: proceed with archive.
}
_ = tokio::time::sleep(Duration::from_secs(10)) => {
warn!("conversation {conversation_id} shutdown timed out; proceeding with archive");
notify.notify_one();
}
}
}
Err(err) => {
error!("failed to submit Shutdown to conversation {conversation_id}: {err}");
notify.notify_one();
// Perhaps we lost a shutdown race, so let's continue to
// clean up the .jsonl file.
}
}
}
// Move the .jsonl file to the archived sessions subdir.
let result: std::io::Result<()> = async {
let archive_folder = self
.config
.codex_home
.join(codex_core::ARCHIVED_SESSIONS_SUBDIR);
tokio::fs::create_dir_all(&archive_folder).await?;
tokio::fs::rename(&canonical_rollout_path, &archive_folder.join(&file_name)).await?;
Ok(())
}
.await;
match result {
Ok(()) => {
let response = ArchiveConversationResponse {};
self.outgoing.send_response(request_id, response).await;
}
Err(err) => {
let error = JSONRPCErrorError {
code: INTERNAL_ERROR_CODE,
message: format!("failed to archive conversation: {err}"),
data: None,
};
self.outgoing.send_error(request_id, error).await;
}
}
}
async fn send_user_message(&self, request_id: RequestId, params: SendUserMessageParams) {
let SendUserMessageParams {
conversation_id,
@@ -822,11 +1001,11 @@ impl CodexMessageProcessor {
let mut params = match serde_json::to_value(event.clone()) {
Ok(serde_json::Value::Object(map)) => map,
Ok(_) => {
tracing::error!("event did not serialize to an object");
error!("event did not serialize to an object");
continue;
}
Err(err) => {
tracing::error!("failed to serialize event: {err}");
error!("failed to serialize event: {err}");
continue;
}
};
@@ -1009,7 +1188,7 @@ fn derive_config_from_params(
async fn on_patch_approval_response(
event_id: String,
receiver: tokio::sync::oneshot::Receiver<mcp_types::Result>,
receiver: oneshot::Receiver<mcp_types::Result>,
codex: Arc<CodexConversation>,
) {
let response = receiver.await;
@@ -1051,14 +1230,14 @@ async fn on_patch_approval_response(
async fn on_exec_approval_response(
event_id: String,
receiver: tokio::sync::oneshot::Receiver<mcp_types::Result>,
receiver: oneshot::Receiver<mcp_types::Result>,
conversation: Arc<CodexConversation>,
) {
let response = receiver.await;
let value = match response {
Ok(value) => value,
Err(err) => {
tracing::error!("request failed: {err:?}");
error!("request failed: {err:?}");
return;
}
};
@@ -1085,37 +1264,100 @@ async fn on_exec_approval_response(
}
}
fn extract_ts_and_preview(head: &[serde_json::Value]) -> (Option<String>, String) {
let ts = head
.first()
.and_then(|v| v.get("timestamp"))
.and_then(|v| v.as_str())
.map(|s| s.to_string());
let preview = find_first_user_text(head).unwrap_or_default();
(ts, preview)
fn extract_conversation_summary(
path: PathBuf,
head: &[serde_json::Value],
) -> Option<ConversationSummary> {
let session_meta = match head.first() {
Some(first_line) => serde_json::from_value::<SessionMeta>(first_line.clone()).ok()?,
None => return None,
};
let preview = head
.iter()
.filter_map(|value| serde_json::from_value::<ResponseItem>(value.clone()).ok())
.find_map(|item| match item {
ResponseItem::Message { content, .. } => {
content.into_iter().find_map(|content| match content {
ContentItem::InputText { text } => {
match InputMessageKind::from(("user", &text)) {
InputMessageKind::Plain => Some(text),
_ => None,
}
}
_ => None,
})
}
_ => None,
})?;
let preview = match preview.find(USER_MESSAGE_BEGIN) {
Some(idx) => preview[idx + USER_MESSAGE_BEGIN.len()..].trim(),
None => preview.as_str(),
};
let timestamp = if session_meta.timestamp.is_empty() {
None
} else {
Some(session_meta.timestamp.clone())
};
Some(ConversationSummary {
conversation_id: session_meta.id,
timestamp,
path,
preview: preview.to_string(),
})
}
fn find_first_user_text(head: &[serde_json::Value]) -> Option<String> {
use codex_core::protocol::InputMessageKind;
for v in head.iter() {
let t = v.get("type").and_then(|x| x.as_str()).unwrap_or("");
if t != "message" {
continue;
}
if v.get("role").and_then(|x| x.as_str()) != Some("user") {
continue;
}
if let Some(arr) = v.get("content").and_then(|c| c.as_array()) {
for c in arr.iter() {
if let (Some("input_text"), Some(txt)) =
(c.get("type").and_then(|t| t.as_str()), c.get("text"))
&& let Some(s) = txt.as_str()
&& matches!(InputMessageKind::from(("user", s)), InputMessageKind::Plain)
{
return Some(s.to_string());
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use pretty_assertions::assert_eq;
use serde_json::json;
#[test]
fn extract_conversation_summary_prefers_plain_user_messages() {
let conversation_id =
ConversationId(Uuid::parse_str("3f941c35-29b3-493b-b0a4-e25800d9aeb0").unwrap());
let timestamp = Some("2025-09-05T16:53:11.850Z".to_string());
let path = PathBuf::from("rollout.jsonl");
let head = vec![
json!({
"id": conversation_id.0,
"timestamp": timestamp,
"cwd": "/",
"originator": "codex",
"cli_version": "0.0.0",
"instructions": null
}),
json!({
"type": "message",
"role": "user",
"content": [{
"type": "input_text",
"text": "<user_instructions>\n<AGENTS.md contents>\n</user_instructions>".to_string(),
}],
}),
json!({
"type": "message",
"role": "user",
"content": [{
"type": "input_text",
"text": format!("<prior context> {USER_MESSAGE_BEGIN}Count to 5"),
}],
}),
];
let summary = extract_conversation_summary(path.clone(), &head).expect("summary");
assert_eq!(summary.conversation_id, conversation_id);
assert_eq!(
summary.timestamp,
Some("2025-09-05T16:53:11.850Z".to_string())
);
assert_eq!(summary.path, path);
assert_eq!(summary.preview, "Count to 5");
}
None
}

View File

@@ -5,6 +5,10 @@
use std::collections::HashMap;
use std::sync::Arc;
use crate::exec_approval::handle_exec_approval_request;
use crate::outgoing_message::OutgoingMessageSender;
use crate::outgoing_message::OutgoingNotificationMeta;
use crate::patch_approval::handle_patch_approval_request;
use codex_core::CodexConversation;
use codex_core::ConversationManager;
use codex_core::NewConversation;
@@ -26,11 +30,6 @@ use mcp_types::TextContent;
use serde_json::json;
use tokio::sync::Mutex;
use crate::exec_approval::handle_exec_approval_request;
use crate::outgoing_message::OutgoingMessageSender;
use crate::outgoing_message::OutgoingNotificationMeta;
use crate::patch_approval::handle_patch_approval_request;
pub(crate) const INVALID_PARAMS_ERROR_CODE: i64 = -32602;
/// Run a complete Codex session and stream events back to the client.
@@ -278,7 +277,7 @@ async fn run_codex_tool_session_inner(
| EventMsg::GetHistoryEntryResponse(_)
| EventMsg::PlanUpdate(_)
| EventMsg::TurnAborted(_)
| EventMsg::ConversationHistory(_)
| EventMsg::ConversationPath(_)
| EventMsg::UserMessage(_)
| EventMsg::ShutdownComplete => {
// For now, we do not do anything extra for these

View File

@@ -14,6 +14,8 @@ use codex_protocol::mcp_protocol::ConversationId;
use codex_core::AuthManager;
use codex_core::ConversationManager;
use codex_core::config::Config;
use codex_core::default_client::USER_AGENT_SUFFIX;
use codex_core::default_client::get_codex_user_agent;
use codex_core::protocol::Submission;
use mcp_types::CallToolRequestParams;
use mcp_types::CallToolResult;
@@ -54,11 +56,8 @@ impl MessageProcessor {
config: Arc<Config>,
) -> Self {
let outgoing = Arc::new(outgoing);
let auth_manager = AuthManager::shared(
config.codex_home.clone(),
config.preferred_auth_method,
config.responses_originator_header.clone(),
);
let auth_manager =
AuthManager::shared(config.codex_home.clone(), config.preferred_auth_method);
let conversation_manager = Arc::new(ConversationManager::new(auth_manager.clone()));
let codex_message_processor = CodexMessageProcessor::new(
auth_manager,
@@ -211,6 +210,14 @@ impl MessageProcessor {
return;
}
let client_info = params.client_info;
let name = client_info.name;
let version = client_info.version;
let user_agent_suffix = format!("{name}; {version}");
if let Ok(mut suffix) = USER_AGENT_SUFFIX.lock() {
*suffix = Some(user_agent_suffix);
}
self.initialized = true;
// Build a minimal InitializeResult. Fill with placeholders.
@@ -231,6 +238,7 @@ impl MessageProcessor {
name: "codex-mcp-server".to_string(),
version: env!("CARGO_PKG_VERSION").to_string(),
title: Some("Codex".to_string()),
user_agent: Some(get_codex_user_agent()),
},
};

View File

@@ -262,6 +262,7 @@ mod tests {
use codex_protocol::mcp_protocol::LoginChatGptCompleteNotification;
use pretty_assertions::assert_eq;
use serde_json::json;
use tempfile::NamedTempFile;
use uuid::Uuid;
use super::*;
@@ -272,6 +273,7 @@ mod tests {
let outgoing_message_sender = OutgoingMessageSender::new(outgoing_tx);
let conversation_id = ConversationId::new();
let rollout_file = NamedTempFile::new().unwrap();
let event = Event {
id: "1".to_string(),
msg: EventMsg::SessionConfigured(SessionConfiguredEvent {
@@ -280,6 +282,7 @@ mod tests {
history_log_id: 1,
history_entry_count: 1000,
initial_messages: None,
rollout_path: rollout_file.path().to_path_buf(),
}),
};
@@ -305,12 +308,14 @@ mod tests {
let outgoing_message_sender = OutgoingMessageSender::new(outgoing_tx);
let conversation_id = ConversationId::new();
let rollout_file = NamedTempFile::new().unwrap();
let session_configured_event = SessionConfiguredEvent {
session_id: conversation_id,
model: "gpt-4o".to_string(),
history_log_id: 1,
history_entry_count: 1000,
initial_messages: None,
rollout_path: rollout_file.path().to_path_buf(),
};
let event = Event {
id: "1".to_string(),
@@ -340,6 +345,7 @@ mod tests {
"history_log_id": session_configured_event.history_log_id,
"history_entry_count": session_configured_event.history_entry_count,
"type": "session_configured",
"rollout_path": rollout_file.path().to_path_buf(),
}
});
assert_eq!(params.unwrap(), expected_params);

View File

@@ -13,16 +13,14 @@ codex-core = { path = "../../../core" }
codex-mcp-server = { path = "../.." }
codex-protocol = { path = "../../../protocol" }
mcp-types = { path = "../../../mcp-types" }
os_info = "3.12.0"
pretty_assertions = "1.4.1"
serde = { version = "1" }
serde_json = "1"
shlex = "1.3.0"
tempfile = "3"
tokio = { version = "1", features = [
"io-std",
"macros",
"process",
"rt-multi-thread",
] }
uuid = { version = "1", features = ["serde", "v4"] }
wiremock = "0.6"

View File

@@ -13,6 +13,7 @@ use anyhow::Context;
use assert_cmd::prelude::*;
use codex_mcp_server::CodexToolCallParam;
use codex_protocol::mcp_protocol::AddConversationListenerParams;
use codex_protocol::mcp_protocol::ArchiveConversationParams;
use codex_protocol::mcp_protocol::CancelLoginChatGptParams;
use codex_protocol::mcp_protocol::GetAuthStatusParams;
use codex_protocol::mcp_protocol::InterruptConversationParams;
@@ -53,6 +54,18 @@ pub struct McpProcess {
impl McpProcess {
pub async fn new(codex_home: &Path) -> anyhow::Result<Self> {
Self::new_with_env(codex_home, &[]).await
}
/// Creates a new MCP process, allowing tests to override or remove
/// specific environment variables for the child process only.
///
/// Pass a tuple of (key, Some(value)) to set/override, or (key, None) to
/// remove a variable from the child's environment.
pub async fn new_with_env(
codex_home: &Path,
env_overrides: &[(&str, Option<&str>)],
) -> anyhow::Result<Self> {
// Use assert_cmd to locate the binary path and then switch to tokio::process::Command
let std_cmd = StdCommand::cargo_bin("codex-mcp-server")
.context("should find binary for codex-mcp-server")?;
@@ -67,6 +80,17 @@ impl McpProcess {
cmd.env("CODEX_HOME", codex_home);
cmd.env("RUST_LOG", "debug");
for (k, v) in env_overrides {
match v {
Some(val) => {
cmd.env(k, val);
}
None => {
cmd.env_remove(k);
}
}
}
let mut process = cmd
.kill_on_drop(true)
.spawn()
@@ -114,6 +138,7 @@ impl McpProcess {
name: "elicitation test".into(),
title: Some("Elicitation Test".into()),
version: "0.0.0".into(),
user_agent: None,
},
protocol_version: mcp_types::MCP_SCHEMA_VERSION.into(),
};
@@ -128,6 +153,14 @@ impl McpProcess {
.await?;
let initialized = self.read_jsonrpc_message().await?;
let os_info = os_info::get();
let user_agent = format!(
"codex_cli_rs/0.0.0 ({} {}; {}) {} (elicitation test; 0.0.0)",
os_info.os_type(),
os_info.version(),
os_info.architecture().unwrap_or("unknown"),
codex_core::terminal::user_agent()
);
assert_eq!(
JSONRPCMessage::Response(JSONRPCResponse {
jsonrpc: JSONRPC_VERSION.into(),
@@ -141,7 +174,8 @@ impl McpProcess {
"serverInfo": {
"name": "codex-mcp-server",
"title": "Codex",
"version": "0.0.0"
"version": "0.0.0",
"user_agent": user_agent
},
"protocolVersion": mcp_types::MCP_SCHEMA_VERSION
})
@@ -186,6 +220,15 @@ impl McpProcess {
self.send_request("newConversation", params).await
}
/// Send an `archiveConversation` JSON-RPC request.
pub async fn send_archive_conversation_request(
&mut self,
params: ArchiveConversationParams,
) -> anyhow::Result<i64> {
let params = Some(serde_json::to_value(params)?);
self.send_request("archiveConversation", params).await
}
/// Send an `addConversationListener` JSON-RPC request.
pub async fn send_add_conversation_listener_request(
&mut self,
@@ -247,6 +290,16 @@ impl McpProcess {
self.send_request("getUserSavedConfig", None).await
}
/// Send a `getUserAgent` JSON-RPC request.
pub async fn send_get_user_agent_request(&mut self) -> anyhow::Result<i64> {
self.send_request("getUserAgent", None).await
}
/// Send a `userInfo` JSON-RPC request.
pub async fn send_user_info_request(&mut self) -> anyhow::Result<i64> {
self.send_request("userInfo", None).await
}
/// Send a `listConversations` JSON-RPC request.
pub async fn send_list_conversations_request(
&mut self,

View File

@@ -0,0 +1,105 @@
use std::path::Path;
use codex_core::ARCHIVED_SESSIONS_SUBDIR;
use codex_protocol::mcp_protocol::ArchiveConversationParams;
use codex_protocol::mcp_protocol::ArchiveConversationResponse;
use codex_protocol::mcp_protocol::NewConversationParams;
use codex_protocol::mcp_protocol::NewConversationResponse;
use mcp_test_support::McpProcess;
use mcp_test_support::to_response;
use mcp_types::JSONRPCResponse;
use mcp_types::RequestId;
use tempfile::TempDir;
use tokio::time::timeout;
const DEFAULT_READ_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(10);
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn archive_conversation_moves_rollout_into_archived_directory() {
let codex_home = TempDir::new().expect("create temp dir");
create_config_toml(codex_home.path()).expect("write config.toml");
let mut mcp = McpProcess::new(codex_home.path())
.await
.expect("spawn mcp process");
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize())
.await
.expect("initialize timeout")
.expect("initialize request");
let new_request_id = mcp
.send_new_conversation_request(NewConversationParams {
model: Some("mock-model".to_string()),
..Default::default()
})
.await
.expect("send newConversation");
let new_response: JSONRPCResponse = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_response_message(RequestId::Integer(new_request_id)),
)
.await
.expect("newConversation timeout")
.expect("newConversation response");
let NewConversationResponse {
conversation_id,
rollout_path,
..
} = to_response::<NewConversationResponse>(new_response)
.expect("deserialize newConversation response");
assert!(
rollout_path.exists(),
"expected rollout path {} to exist",
rollout_path.display()
);
let archive_request_id = mcp
.send_archive_conversation_request(ArchiveConversationParams {
conversation_id,
rollout_path: rollout_path.clone(),
})
.await
.expect("send archiveConversation");
let archive_response: JSONRPCResponse = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_response_message(RequestId::Integer(archive_request_id)),
)
.await
.expect("archiveConversation timeout")
.expect("archiveConversation response");
let _: ArchiveConversationResponse =
to_response::<ArchiveConversationResponse>(archive_response)
.expect("deserialize archiveConversation response");
let archived_directory = codex_home.path().join(ARCHIVED_SESSIONS_SUBDIR);
let archived_rollout_path =
archived_directory.join(rollout_path.file_name().unwrap_or_else(|| {
panic!("rollout path {} missing file name", rollout_path.display())
}));
assert!(
!rollout_path.exists(),
"expected rollout path {} to be moved",
rollout_path.display()
);
assert!(
archived_rollout_path.exists(),
"expected archived rollout path {} to exist",
archived_rollout_path.display()
);
}
fn create_config_toml(codex_home: &Path) -> std::io::Result<()> {
let config_toml = codex_home.join("config.toml");
std::fs::write(config_toml, config_contents())
}
fn config_contents() -> &'static str {
r#"model = "mock-model"
approval_policy = "never"
sandbox_mode = "read-only"
"#
}

View File

@@ -41,7 +41,7 @@ async fn get_auth_status_no_auth() {
let codex_home = TempDir::new().unwrap_or_else(|e| panic!("create tempdir: {e}"));
create_config_toml(codex_home.path()).expect("write config.toml");
let mut mcp = McpProcess::new(codex_home.path())
let mut mcp = McpProcess::new_with_env(codex_home.path(), &[("OPENAI_API_KEY", None)])
.await
.expect("spawn mcp process");
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize())

View File

@@ -90,6 +90,7 @@ async fn test_codex_jsonrpc_conversation_flow() {
let NewConversationResponse {
conversation_id,
model,
rollout_path: _,
} = new_conv_resp;
assert_eq!(model, "mock-model");

View File

@@ -59,6 +59,7 @@ async fn test_conversation_create_and_send_message_ok() {
let NewConversationResponse {
conversation_id,
model,
rollout_path: _,
} = to_response::<NewConversationResponse>(new_conv_resp)
.expect("deserialize newConversation response");
assert_eq!(model, "o3");

View File

@@ -156,14 +156,45 @@ fn create_fake_rollout(codex_home: &Path, filename_ts: &str, meta_rfc3339: &str,
let file_path = dir.join(format!("rollout-{filename_ts}-{uuid}.jsonl"));
let mut lines = Vec::new();
// Meta line with timestamp
lines.push(json!({"timestamp": meta_rfc3339}).to_string());
// Minimal user message entry as a persisted response item
// Meta line with timestamp (flattened meta in payload for new schema)
lines.push(
json!({
"type":"message",
"role":"user",
"content":[{"type":"input_text","text": preview}]
"timestamp": meta_rfc3339,
"type": "session_meta",
"payload": {
"id": uuid,
"timestamp": meta_rfc3339,
"cwd": "/",
"originator": "codex",
"cli_version": "0.0.0",
"instructions": null
}
})
.to_string(),
);
// Minimal user message entry as a persisted response item (with envelope timestamp)
lines.push(
json!({
"timestamp": meta_rfc3339,
"type":"response_item",
"payload": {
"type":"message",
"role":"user",
"content":[{"type":"input_text","text": preview}]
}
})
.to_string(),
);
// Add a matching user message event line to satisfy filters
lines.push(
json!({
"timestamp": meta_rfc3339,
"type":"event_msg",
"payload": {
"type":"user_message",
"message": preview,
"kind": "plain"
}
})
.to_string(),
);

View File

@@ -46,7 +46,7 @@ async fn logout_chatgpt_removes_auth() {
login_with_api_key(codex_home.path(), "sk-test-key").expect("seed api key");
assert!(codex_home.path().join("auth.json").exists());
let mut mcp = McpProcess::new(codex_home.path())
let mut mcp = McpProcess::new_with_env(codex_home.path(), &[("OPENAI_API_KEY", None)])
.await
.expect("spawn mcp process");
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize())

View File

@@ -1,4 +1,5 @@
// Aggregates all former standalone integration tests as modules.
mod archive_conversation;
mod auth;
mod codex_message_processor_flow;
mod codex_tool;
@@ -8,3 +9,5 @@ mod interrupt;
mod list_resume;
mod login;
mod send_message;
mod user_agent;
mod user_info;

View File

@@ -0,0 +1,50 @@
use codex_protocol::mcp_protocol::GetUserAgentResponse;
use mcp_test_support::McpProcess;
use mcp_test_support::to_response;
use mcp_types::JSONRPCResponse;
use mcp_types::RequestId;
use pretty_assertions::assert_eq;
use tempfile::TempDir;
use tokio::time::timeout;
const DEFAULT_READ_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(10);
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn get_user_agent_returns_current_codex_user_agent() {
let codex_home = TempDir::new().unwrap_or_else(|err| panic!("create tempdir: {err}"));
let mut mcp = McpProcess::new(codex_home.path())
.await
.expect("spawn mcp process");
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize())
.await
.expect("initialize timeout")
.expect("initialize request");
let request_id = mcp
.send_get_user_agent_request()
.await
.expect("send getUserAgent");
let response: JSONRPCResponse = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_response_message(RequestId::Integer(request_id)),
)
.await
.expect("getUserAgent timeout")
.expect("getUserAgent response");
let os_info = os_info::get();
let user_agent = format!(
"codex_cli_rs/0.0.0 ({} {}; {}) {} (elicitation test; 0.0.0)",
os_info.os_type(),
os_info.version(),
os_info.architecture().unwrap_or("unknown"),
codex_core::terminal::user_agent()
);
let received: GetUserAgentResponse =
to_response(response).expect("deserialize getUserAgent response");
let expected = GetUserAgentResponse { user_agent };
assert_eq!(received, expected);
}

View File

@@ -0,0 +1,78 @@
use std::time::Duration;
use anyhow::Context;
use base64::Engine;
use base64::engine::general_purpose::URL_SAFE_NO_PAD;
use codex_core::auth::AuthDotJson;
use codex_core::auth::get_auth_file;
use codex_core::auth::write_auth_json;
use codex_core::token_data::IdTokenInfo;
use codex_core::token_data::TokenData;
use codex_protocol::mcp_protocol::UserInfoResponse;
use mcp_test_support::McpProcess;
use mcp_test_support::to_response;
use mcp_types::JSONRPCResponse;
use mcp_types::RequestId;
use pretty_assertions::assert_eq;
use serde_json::json;
use tempfile::TempDir;
use tokio::time::timeout;
const DEFAULT_READ_TIMEOUT: Duration = Duration::from_secs(10);
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn user_info_returns_email_from_auth_json() {
let codex_home = TempDir::new().expect("create tempdir");
let auth_path = get_auth_file(codex_home.path());
let mut id_token = IdTokenInfo::default();
id_token.email = Some("user@example.com".to_string());
id_token.raw_jwt = encode_id_token_with_email("user@example.com").expect("encode id token");
let auth = AuthDotJson {
openai_api_key: None,
tokens: Some(TokenData {
id_token,
access_token: "access".to_string(),
refresh_token: "refresh".to_string(),
account_id: None,
}),
last_refresh: None,
};
write_auth_json(&auth_path, &auth).expect("write auth.json");
let mut mcp = McpProcess::new(codex_home.path())
.await
.expect("spawn mcp process");
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize())
.await
.expect("initialize timeout")
.expect("initialize request");
let request_id = mcp.send_user_info_request().await.expect("send userInfo");
let response: JSONRPCResponse = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_response_message(RequestId::Integer(request_id)),
)
.await
.expect("userInfo timeout")
.expect("userInfo response");
let received: UserInfoResponse = to_response(response).expect("deserialize userInfo response");
let expected = UserInfoResponse {
alleged_user_email: Some("user@example.com".to_string()),
};
assert_eq!(received, expected);
}
fn encode_id_token_with_email(email: &str) -> anyhow::Result<String> {
let header_b64 = URL_SAFE_NO_PAD.encode(
serde_json::to_vec(&json!({ "alg": "none", "typ": "JWT" }))
.context("serialize jwt header")?,
);
let payload =
serde_json::to_vec(&json!({ "email": email })).context("serialize jwt payload")?;
let payload_b64 = URL_SAFE_NO_PAD.encode(payload);
Ok(format!("{header_b64}.{payload_b64}.signature"))
}

View File

@@ -9,4 +9,4 @@ workspace = true
[dependencies]
serde = { version = "1", features = ["derive"] }
serde_json = "1"
ts-rs = { version = "11", features = ["serde-json-impl"] }
ts-rs = { version = "11", features = ["serde-json-impl", "no-serde-warnings"] }

View File

@@ -265,8 +265,11 @@ class StructField:
name: str
type_name: str
serde: str | None = None
comment: str | None = None
def append(self, out: list[str], supports_const: bool) -> None:
if self.comment:
out.append(f" // {self.comment}\n")
if self.serde:
out.append(f" {self.serde}\n")
if self.viz == "const":
@@ -312,6 +315,18 @@ def define_struct(
else:
fields.append(StructField("pub", rs_prop.name, prop_type, rs_prop.serde))
# Special-case: add Codex-specific user_agent to Implementation
if name == "Implementation":
fields.append(
StructField(
"pub",
"user_agent",
"Option<String>",
'#[serde(default, skip_serializing_if = "Option::is_none")]',
"This is an extra field that the Codex MCP server sends as part of InitializeResult.",
)
)
if implements_request_trait(name):
add_trait_impl(name, "ModelContextProtocolRequest", fields, out)
elif implements_notification_trait(name):

View File

@@ -487,6 +487,9 @@ pub struct Implementation {
#[serde(default, skip_serializing_if = "Option::is_none")]
pub title: Option<String>,
pub version: String,
// This is an extra field that the Codex MCP server sends as part of InitializeResult.
#[serde(default, skip_serializing_if = "Option::is_none")]
pub user_agent: Option<String>,
}
#[derive(Debug, Clone, PartialEq, Deserialize, Serialize, TS)]

View File

@@ -62,6 +62,7 @@ fn deserialize_initialize_request() {
name: "acme-client".into(),
title: Some("Acme".to_string()),
version: "1.2.3".into(),
user_agent: None,
},
protocol_version: "2025-06-18".into(),
}

View File

@@ -24,9 +24,7 @@ tokio = { version = "1", features = [
"rt-multi-thread",
"signal",
] }
toml = "0.9.5"
tracing = { version = "0.1.41", features = ["log"] }
wiremock = "0.6"
[dev-dependencies]
tempfile = "3"

View File

@@ -16,6 +16,7 @@ path = "src/main.rs"
[dependencies]
anyhow = "1"
mcp-types = { path = "../mcp-types" }
codex-protocol = { path = "../protocol" }
ts-rs = "11"
clap = { version = "4", features = ["derive"] }

View File

@@ -16,37 +16,33 @@ pub fn generate_ts(out_dir: &Path, prettier: Option<&Path>) -> Result<()> {
ensure_dir(out_dir)?;
// Generate TS bindings
mcp_types::InitializeResult::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ConversationId::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::InputItem::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ClientRequest::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ServerRequest::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::NewConversationParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::NewConversationResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::AddConversationListenerParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ListConversationsResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ResumeConversationResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ArchiveConversationResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::AddConversationSubscriptionResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::RemoveConversationListenerParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::RemoveConversationSubscriptionResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::SendUserMessageParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::SendUserMessageResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::SendUserTurnParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::SendUserTurnResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::InterruptConversationParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::InterruptConversationResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GitDiffToRemoteParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GitDiffToRemoteResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::LoginChatGptResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::LoginChatGptCompleteNotification::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::CancelLoginChatGptParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::CancelLoginChatGptResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::LogoutChatGptParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::LogoutChatGptResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GetAuthStatusParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GetAuthStatusResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ApplyPatchApprovalParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ApplyPatchApprovalResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ExecCommandApprovalParams::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::ExecCommandApprovalResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GetUserSavedConfigResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::GetUserAgentResponse::export_all_to(out_dir)?;
codex_protocol::mcp_protocol::UserInfoResponse::export_all_to(out_dir)?;
// All notification types reachable from this enum will be generated by
// induction, so they do not need to be listed individually.
codex_protocol::mcp_protocol::ServerNotification::export_all_to(out_dir)?;
generate_index_ts(out_dir)?;

View File

@@ -12,17 +12,24 @@ workspace = true
[dependencies]
base64 = "0.22.1"
icu_decimal = "2.0.0"
icu_locale_core = "2.0.0"
mcp-types = { path = "../mcp-types" }
mime_guess = "2.0.5"
serde = { version = "1", features = ["derive"] }
serde_bytes = "0.11"
serde_json = "1"
serde_with = { version = "3.14.0", features = ["macros", "base64"] }
strum = "0.27.2"
strum_macros = "0.27.2"
sys-locale = "0.3.2"
tracing = "0.1.41"
ts-rs = { version = "11", features = ["uuid-impl", "serde-json-impl"] }
ts-rs = { version = "11", features = ["uuid-impl", "serde-json-impl", "no-serde-warnings"] }
uuid = { version = "1", features = ["serde", "v4"] }
[dev-dependencies]
pretty_assertions = "1.4.1"
tempfile = "3"
[package.metadata.cargo-shear]
# Required because the not imported as strum_macros in non-nightly builds.
ignored = ["strum"]

View File

@@ -1,8 +1,9 @@
use serde::Deserialize;
use serde::Serialize;
use std::path::PathBuf;
use ts_rs::TS;
#[derive(Serialize, Deserialize, Debug, Clone)]
#[derive(Serialize, Deserialize, Debug, Clone, TS)]
pub struct CustomPrompt {
pub name: String,
pub path: PathBuf,

View File

@@ -3,6 +3,7 @@ pub mod custom_prompts;
pub mod mcp_protocol;
pub mod message_history;
pub mod models;
pub mod num_format;
pub mod parse_command;
pub mod plan_tool;
pub mod protocol;

View File

@@ -19,7 +19,7 @@ use strum_macros::Display;
use ts_rs::TS;
use uuid::Uuid;
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize, TS, Hash, Default)]
#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize, TS, Hash)]
#[ts(type = "string")]
pub struct ConversationId(pub Uuid);
@@ -29,6 +29,12 @@ impl ConversationId {
}
}
impl Default for ConversationId {
fn default() -> Self {
Self::new()
}
}
impl Display for ConversationId {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
write!(f, "{}", self.0)
@@ -85,6 +91,11 @@ pub enum ClientRequest {
request_id: RequestId,
params: ResumeConversationParams,
},
ArchiveConversation {
#[serde(rename = "id")]
request_id: RequestId,
params: ArchiveConversationParams,
},
SendUserMessage {
#[serde(rename = "id")]
request_id: RequestId,
@@ -137,6 +148,14 @@ pub enum ClientRequest {
#[serde(rename = "id")]
request_id: RequestId,
},
GetUserAgent {
#[serde(rename = "id")]
request_id: RequestId,
},
UserInfo {
#[serde(rename = "id")]
request_id: RequestId,
},
/// Execute a command (argv vector) under the server's sandbox.
ExecOneOffCommand {
#[serde(rename = "id")]
@@ -193,9 +212,10 @@ pub struct NewConversationParams {
pub struct NewConversationResponse {
pub conversation_id: ConversationId,
pub model: String,
pub rollout_path: PathBuf,
}
#[derive(Serialize, Deserialize, Debug, Clone)]
#[derive(Serialize, Deserialize, Debug, Clone, TS)]
#[serde(rename_all = "camelCase")]
pub struct ResumeConversationResponse {
pub conversation_id: ConversationId,
@@ -218,6 +238,7 @@ pub struct ListConversationsParams {
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct ConversationSummary {
pub conversation_id: ConversationId,
pub path: PathBuf,
pub preview: String,
/// RFC3339 timestamp string for the session start, if available.
@@ -251,6 +272,18 @@ pub struct AddConversationSubscriptionResponse {
pub subscription_id: Uuid,
}
/// The [`ConversationId`] must match the `rollout_path`.
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct ArchiveConversationParams {
pub conversation_id: ConversationId,
pub rollout_path: PathBuf,
}
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct ArchiveConversationResponse {}
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct RemoveConversationSubscriptionResponse {}
@@ -339,6 +372,22 @@ pub struct GetAuthStatusResponse {
pub auth_token: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct GetUserAgentResponse {
pub user_agent: String,
}
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct UserInfoResponse {
/// Note: `alleged_user_email` is not currently verified. We read it from
/// the local auth.json, which the user could theoretically modify. In the
/// future, we may add logic to verify the email against the server before
/// returning it.
pub alleged_user_email: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, TS)]
#[serde(rename_all = "camelCase")]
pub struct GetUserSavedConfigResponse {
@@ -621,4 +670,10 @@ mod tests {
serde_json::to_value(&request).unwrap(),
);
}
#[test]
fn test_conversation_id_default_is_not_zeroes() {
let id = ConversationId::default();
assert_ne!(id.0, Uuid::nil());
}
}

View File

@@ -1,7 +1,8 @@
use serde::Deserialize;
use serde::Serialize;
use ts_rs::TS;
#[derive(Serialize, Deserialize, Debug, Clone)]
#[derive(Serialize, Deserialize, Debug, Clone, TS)]
pub struct HistoryEntry {
pub conversation_id: String,
pub ts: u64,

View File

@@ -6,10 +6,11 @@ use serde::Deserialize;
use serde::Deserializer;
use serde::Serialize;
use serde::ser::Serializer;
use ts_rs::TS;
use crate::protocol::InputItem;
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ResponseInputItem {
Message {
@@ -30,7 +31,7 @@ pub enum ResponseInputItem {
},
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ContentItem {
InputText { text: String },
@@ -38,7 +39,7 @@ pub enum ContentItem {
OutputText { text: String },
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ResponseItem {
Message {
@@ -48,7 +49,7 @@ pub enum ResponseItem {
content: Vec<ContentItem>,
},
Reasoning {
#[serde(default)]
#[serde(default, skip_serializing)]
id: String,
summary: Vec<ReasoningItemReasoningSummary>,
#[serde(default, skip_serializing_if = "should_serialize_reasoning_content")]
@@ -114,7 +115,6 @@ pub enum ResponseItem {
status: Option<String>,
action: WebSearchAction,
},
#[serde(other)]
Other,
}
@@ -159,7 +159,7 @@ impl From<ResponseInputItem> for ResponseItem {
}
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(rename_all = "snake_case")]
pub enum LocalShellStatus {
Completed,
@@ -167,13 +167,13 @@ pub enum LocalShellStatus {
Incomplete,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum LocalShellAction {
Exec(LocalShellExecAction),
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
pub struct LocalShellExecAction {
pub command: Vec<String>,
pub timeout_ms: Option<u64>,
@@ -182,7 +182,7 @@ pub struct LocalShellExecAction {
pub user: Option<String>,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum WebSearchAction {
Search {
@@ -192,13 +192,13 @@ pub enum WebSearchAction {
Other,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ReasoningItemReasoningSummary {
SummaryText { text: String },
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ReasoningItemContent {
ReasoningText { text: String },
@@ -242,7 +242,7 @@ impl From<Vec<InputItem>> for ResponseInputItem {
/// If the `name` of a `ResponseItem::FunctionCall` is either `container.exec`
/// or shell`, the `arguments` field should deserialize to this struct.
#[derive(Deserialize, Debug, Clone, PartialEq)]
#[derive(Deserialize, Debug, Clone, PartialEq, TS)]
pub struct ShellToolCallParams {
pub command: Vec<String>,
pub workdir: Option<String>,
@@ -256,7 +256,7 @@ pub struct ShellToolCallParams {
pub justification: Option<String>,
}
#[derive(Debug, Clone, PartialEq)]
#[derive(Debug, Clone, PartialEq, TS)]
pub struct FunctionCallOutputPayload {
pub content: String,
pub success: Option<bool>,

View File

@@ -0,0 +1,98 @@
use std::sync::OnceLock;
use icu_decimal::DecimalFormatter;
use icu_decimal::input::Decimal;
use icu_decimal::options::DecimalFormatterOptions;
use icu_locale_core::Locale;
fn make_local_formatter() -> Option<DecimalFormatter> {
let loc: Locale = sys_locale::get_locale()?.parse().ok()?;
DecimalFormatter::try_new(loc.into(), DecimalFormatterOptions::default()).ok()
}
fn make_en_us_formatter() -> DecimalFormatter {
#![allow(clippy::expect_used)]
let loc: Locale = "en-US".parse().expect("en-US wasn't a valid locale");
DecimalFormatter::try_new(loc.into(), DecimalFormatterOptions::default())
.expect("en-US wasn't a valid locale")
}
fn formatter() -> &'static DecimalFormatter {
static FORMATTER: OnceLock<DecimalFormatter> = OnceLock::new();
FORMATTER.get_or_init(|| make_local_formatter().unwrap_or_else(make_en_us_formatter))
}
/// Format a u64 with locale-aware digit separators (e.g. "12345" -> "12,345"
/// for en-US).
pub fn format_with_separators(n: u64) -> String {
formatter().format(&Decimal::from(n)).to_string()
}
fn format_si_suffix_with_formatter(n: u64, formatter: &DecimalFormatter) -> String {
if n < 1000 {
return formatter.format(&Decimal::from(n)).to_string();
}
// Format `n / scale` with the requested number of fractional digits.
let format_scaled = |n: u64, scale: u64, frac_digits: u32| -> String {
let value = n as f64 / scale as f64;
let scaled: u64 = (value * 10f64.powi(frac_digits as i32)).round() as u64;
let mut dec = Decimal::from(scaled);
dec.multiply_pow10(-(frac_digits as i16));
formatter.format(&dec).to_string()
};
const UNITS: [(u64, &str); 3] = [(1_000, "K"), (1_000_000, "M"), (1_000_000_000, "G")];
let f = n as f64;
for &(scale, suffix) in &UNITS {
if (100.0 * f / scale as f64).round() < 1000.0 {
return format!("{}{}", format_scaled(n, scale, 2), suffix);
} else if (10.0 * f / scale as f64).round() < 1000.0 {
return format!("{}{}", format_scaled(n, scale, 1), suffix);
} else if (f / scale as f64).round() < 1000.0 {
return format!("{}{}", format_scaled(n, scale, 0), suffix);
}
}
// Above 1000G, keep wholeG precision.
format!(
"{}G",
format_with_separators(((n as f64) / 1e9).round() as u64)
)
}
/// Format token counts to 3 significant figures, using base-10 SI suffixes.
///
/// Examples (en-US):
/// - 999 -> "999"
/// - 1200 -> "1.20K"
/// - 123456789 -> "123M"
pub fn format_si_suffix(n: u64) -> String {
format_si_suffix_with_formatter(n, formatter())
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn kmg() {
let formatter = make_en_us_formatter();
let fmt = |n: u64| format_si_suffix_with_formatter(n, &formatter);
assert_eq!(fmt(0), "0");
assert_eq!(fmt(999), "999");
assert_eq!(fmt(1_000), "1.00K");
assert_eq!(fmt(1_200), "1.20K");
assert_eq!(fmt(10_000), "10.0K");
assert_eq!(fmt(100_000), "100K");
assert_eq!(fmt(999_500), "1.00M");
assert_eq!(fmt(1_000_000), "1.00M");
assert_eq!(fmt(1_234_000), "1.23M");
assert_eq!(fmt(12_345_678), "12.3M");
assert_eq!(fmt(999_950_000), "1.00G");
assert_eq!(fmt(1_000_000_000), "1.00G");
assert_eq!(fmt(1_234_000_000), "1.23G");
// Above 1000G we keep wholeG precision (no higher unit supported here).
assert_eq!(fmt(1_234_000_000_000), "1,234G");
}
}

View File

@@ -1,7 +1,8 @@
use serde::Deserialize;
use serde::Serialize;
use ts_rs::TS;
#[derive(Debug, Clone, PartialEq, Eq, Deserialize, Serialize)]
#[derive(Debug, Clone, PartialEq, Eq, Deserialize, Serialize, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
pub enum ParsedCommand {
Read {

View File

@@ -1,8 +1,9 @@
use serde::Deserialize;
use serde::Serialize;
use ts_rs::TS;
// Types for the TODO tool arguments matching codex-vscode/todo-mcp/src/main.rs
#[derive(Debug, Clone, Serialize, Deserialize)]
#[derive(Debug, Clone, Serialize, Deserialize, TS)]
#[serde(rename_all = "snake_case")]
pub enum StepStatus {
Pending,
@@ -10,14 +11,14 @@ pub enum StepStatus {
Completed,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[derive(Debug, Clone, Serialize, Deserialize, TS)]
#[serde(deny_unknown_fields)]
pub struct PlanItemArg {
pub step: String,
pub status: StepStatus,
}
#[derive(Debug, Clone, Serialize, Deserialize)]
#[derive(Debug, Clone, Serialize, Deserialize, TS)]
#[serde(deny_unknown_fields)]
pub struct UpdatePlanArgs {
#[serde(default)]

View File

@@ -16,6 +16,7 @@ use crate::custom_prompts::CustomPrompt;
use crate::mcp_protocol::ConversationId;
use crate::message_history::HistoryEntry;
use crate::models::ResponseItem;
use crate::num_format::format_with_separators;
use crate::parse_command::ParsedCommand;
use crate::plan_tool::UpdatePlanArgs;
use mcp_types::CallToolResult;
@@ -32,6 +33,7 @@ pub const USER_INSTRUCTIONS_OPEN_TAG: &str = "<user_instructions>";
pub const USER_INSTRUCTIONS_CLOSE_TAG: &str = "</user_instructions>";
pub const ENVIRONMENT_CONTEXT_OPEN_TAG: &str = "<environment_context>";
pub const ENVIRONMENT_CONTEXT_CLOSE_TAG: &str = "</environment_context>";
pub const USER_MESSAGE_BEGIN: &str = "## My request for Codex:";
/// Submission Queue Entry - requests from user
#[derive(Debug, Clone, Deserialize, Serialize)]
@@ -58,9 +60,6 @@ pub enum Op {
items: Vec<InputItem>,
},
/// Undo the most recently applied turn diff using the local git repo.
UndoLastTurnDiff,
/// Similar to [`Op::UserInput`], but contains additional context required
/// for a turn of a [`crate::codex_conversation::CodexConversation`].
UserTurn {
@@ -150,7 +149,7 @@ pub enum Op {
/// Request the full in-memory conversation transcript for the current session.
/// Reply is delivered via `EventMsg::ConversationHistory`.
GetHistory,
GetPath,
/// Request the list of MCP tools available across all configured servers.
/// Reply is delivered via `EventMsg::McpListToolsResponse`.
@@ -406,7 +405,7 @@ pub struct Event {
}
/// Response event from the agent
#[derive(Debug, Clone, Deserialize, Serialize, Display)]
#[derive(Debug, Clone, Deserialize, Serialize, Display, TS)]
#[serde(tag = "type", rename_all = "snake_case")]
#[strum(serialize_all = "snake_case")]
pub enum EventMsg {
@@ -500,27 +499,27 @@ pub enum EventMsg {
/// Notification that the agent is shutting down.
ShutdownComplete,
ConversationHistory(ConversationHistoryResponseEvent),
ConversationPath(ConversationPathResponseEvent),
}
// Individual event payload types matching each `EventMsg` variant.
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ErrorEvent {
pub message: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TaskCompleteEvent {
pub last_agent_message: Option<String>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TaskStartedEvent {
pub model_context_window: Option<u64>,
}
#[derive(Debug, Clone, Deserialize, Serialize, Default)]
#[derive(Debug, Clone, Deserialize, Serialize, Default, TS)]
pub struct TokenUsage {
pub input_tokens: u64,
pub cached_input_tokens: u64,
@@ -529,7 +528,7 @@ pub struct TokenUsage {
pub total_tokens: u64,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TokenUsageInfo {
pub total_token_usage: TokenUsage,
pub last_token_usage: TokenUsage,
@@ -566,7 +565,7 @@ impl TokenUsageInfo {
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TokenCountEvent {
pub info: Option<TokenUsageInfo>,
}
@@ -648,19 +647,26 @@ impl From<TokenUsage> for FinalOutput {
impl fmt::Display for FinalOutput {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
let token_usage = &self.token_usage;
write!(
f,
"Token usage: total={} input={}{} output={}{}",
token_usage.blended_total(),
token_usage.non_cached_input(),
format_with_separators(token_usage.blended_total()),
format_with_separators(token_usage.non_cached_input()),
if token_usage.cached_input() > 0 {
format!(" (+ {} cached)", token_usage.cached_input())
format!(
" (+ {} cached)",
format_with_separators(token_usage.cached_input())
)
} else {
String::new()
},
token_usage.output_tokens,
format_with_separators(token_usage.output_tokens),
if token_usage.reasoning_output_tokens > 0 {
format!(" (reasoning {})", token_usage.reasoning_output_tokens)
format!(
" (reasoning {})",
format_with_separators(token_usage.reasoning_output_tokens)
)
} else {
String::new()
}
@@ -668,12 +674,12 @@ impl fmt::Display for FinalOutput {
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentMessageEvent {
pub message: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
#[serde(rename_all = "snake_case")]
pub enum InputMessageKind {
/// Plain user text (default)
@@ -684,11 +690,13 @@ pub enum InputMessageKind {
EnvironmentContext,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct UserMessageEvent {
pub message: String,
#[serde(skip_serializing_if = "Option::is_none")]
pub kind: Option<InputMessageKind>,
#[serde(skip_serializing_if = "Option::is_none")]
pub images: Option<Vec<String>>,
}
impl<T, U> From<(T, U)> for InputMessageKind
@@ -714,35 +722,35 @@ where
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentMessageDeltaEvent {
pub delta: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentReasoningEvent {
pub text: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentReasoningRawContentEvent {
pub text: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentReasoningRawContentDeltaEvent {
pub delta: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentReasoningSectionBreakEvent {}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct AgentReasoningDeltaEvent {
pub delta: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct McpInvocation {
/// Name of the MCP server as defined in the config.
pub server: String,
@@ -752,18 +760,19 @@ pub struct McpInvocation {
pub arguments: Option<serde_json::Value>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct McpToolCallBeginEvent {
/// Identifier so this can be paired with the McpToolCallEnd event.
pub call_id: String,
pub invocation: McpInvocation,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct McpToolCallEndEvent {
/// Identifier for the corresponding McpToolCallBegin that finished.
pub call_id: String,
pub invocation: McpInvocation,
#[ts(type = "string")]
pub duration: Duration,
/// Result of the tool call. Note this could be an error.
pub result: Result<CallToolResult, String>,
@@ -778,12 +787,12 @@ impl McpToolCallEndEvent {
}
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct WebSearchBeginEvent {
pub call_id: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct WebSearchEndEvent {
pub call_id: String,
pub query: String,
@@ -791,13 +800,127 @@ pub struct WebSearchEndEvent {
/// Response payload for `Op::GetHistory` containing the current session's
/// in-memory transcript.
#[derive(Debug, Clone, Deserialize, Serialize)]
pub struct ConversationHistoryResponseEvent {
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ConversationPathResponseEvent {
pub conversation_id: ConversationId,
pub entries: Vec<ResponseItem>,
pub path: PathBuf,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ResumedHistory {
pub conversation_id: ConversationId,
pub history: Vec<RolloutItem>,
pub rollout_path: PathBuf,
}
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub enum InitialHistory {
New,
Resumed(ResumedHistory),
Forked(Vec<RolloutItem>),
}
impl InitialHistory {
pub fn get_rollout_items(&self) -> Vec<RolloutItem> {
match self {
InitialHistory::New => Vec::new(),
InitialHistory::Resumed(resumed) => resumed.history.clone(),
InitialHistory::Forked(items) => items.clone(),
}
}
pub fn get_response_items(&self) -> Vec<ResponseItem> {
match self {
InitialHistory::New => Vec::new(),
InitialHistory::Resumed(resumed) => resumed
.history
.iter()
.filter_map(|ri| match ri {
RolloutItem::ResponseItem(item) => Some(item.clone()),
_ => None,
})
.collect(),
InitialHistory::Forked(items) => items
.iter()
.filter_map(|ri| match ri {
RolloutItem::ResponseItem(item) => Some(item.clone()),
_ => None,
})
.collect(),
}
}
pub fn get_event_msgs(&self) -> Option<Vec<EventMsg>> {
match self {
InitialHistory::New => None,
InitialHistory::Resumed(resumed) => Some(
resumed
.history
.iter()
.filter_map(|ri| match ri {
RolloutItem::EventMsg(ev) => Some(ev.clone()),
_ => None,
})
.collect(),
),
InitialHistory::Forked(items) => Some(
items
.iter()
.filter_map(|ri| match ri {
RolloutItem::EventMsg(ev) => Some(ev.clone()),
_ => None,
})
.collect(),
),
}
}
}
#[derive(Serialize, Deserialize, Clone, Default, Debug, TS)]
pub struct SessionMeta {
pub id: ConversationId,
pub timestamp: String,
pub cwd: PathBuf,
pub originator: String,
pub cli_version: String,
pub instructions: Option<String>,
}
#[derive(Serialize, Deserialize, Debug, Clone, TS)]
pub struct SessionMetaLine {
#[serde(flatten)]
pub meta: SessionMeta,
#[serde(skip_serializing_if = "Option::is_none")]
pub git: Option<GitInfo>,
}
#[derive(Serialize, Deserialize, Debug, Clone, TS)]
#[serde(tag = "type", content = "payload", rename_all = "snake_case")]
pub enum RolloutItem {
SessionMeta(SessionMetaLine),
ResponseItem(ResponseItem),
EventMsg(EventMsg),
}
#[derive(Serialize, Deserialize, Clone)]
pub struct RolloutLine {
pub timestamp: String,
#[serde(flatten)]
pub item: RolloutItem,
}
#[derive(Serialize, Deserialize, Clone, Debug, TS)]
pub struct GitInfo {
/// Current commit hash (SHA)
#[serde(skip_serializing_if = "Option::is_none")]
pub commit_hash: Option<String>,
/// Current branch name
#[serde(skip_serializing_if = "Option::is_none")]
pub branch: Option<String>,
/// Repository URL (if available from remote)
#[serde(skip_serializing_if = "Option::is_none")]
pub repository_url: Option<String>,
}
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ExecCommandBeginEvent {
/// Identifier so this can be paired with the ExecCommandEnd event.
pub call_id: String,
@@ -808,7 +931,7 @@ pub struct ExecCommandBeginEvent {
pub parsed_cmd: Vec<ParsedCommand>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ExecCommandEndEvent {
/// Identifier for the ExecCommandBegin that finished.
pub call_id: String,
@@ -822,12 +945,13 @@ pub struct ExecCommandEndEvent {
/// The command's exit code.
pub exit_code: i32,
/// The duration of the command execution.
#[ts(type = "string")]
pub duration: Duration,
/// Formatted output from the command, as seen by the model.
pub formatted_output: String,
}
#[derive(Debug, Clone, Deserialize, Serialize, PartialEq)]
#[derive(Debug, Clone, Deserialize, Serialize, PartialEq, TS)]
#[serde(rename_all = "snake_case")]
pub enum ExecOutputStream {
Stdout,
@@ -835,7 +959,7 @@ pub enum ExecOutputStream {
}
#[serde_as]
#[derive(Debug, Clone, Deserialize, Serialize, PartialEq)]
#[derive(Debug, Clone, Deserialize, Serialize, PartialEq, TS)]
pub struct ExecCommandOutputDeltaEvent {
/// Identifier for the ExecCommandBegin that produced this chunk.
pub call_id: String,
@@ -843,10 +967,11 @@ pub struct ExecCommandOutputDeltaEvent {
pub stream: ExecOutputStream,
/// Raw bytes from the stream (may not be valid UTF-8).
#[serde_as(as = "serde_with::base64::Base64")]
#[ts(type = "string")]
pub chunk: Vec<u8>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ExecApprovalRequestEvent {
/// Identifier for the associated exec call, if available.
pub call_id: String,
@@ -859,7 +984,7 @@ pub struct ExecApprovalRequestEvent {
pub reason: Option<String>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ApplyPatchApprovalRequestEvent {
/// Responses API call id for the associated patch apply call, if available.
pub call_id: String,
@@ -872,17 +997,17 @@ pub struct ApplyPatchApprovalRequestEvent {
pub grant_root: Option<PathBuf>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct BackgroundEventEvent {
pub message: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct StreamErrorEvent {
pub message: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct PatchApplyBeginEvent {
/// Identifier so this can be paired with the PatchApplyEnd event.
pub call_id: String,
@@ -892,7 +1017,7 @@ pub struct PatchApplyBeginEvent {
pub changes: HashMap<PathBuf, FileChange>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct PatchApplyEndEvent {
/// Identifier for the PatchApplyBegin that finished.
pub call_id: String,
@@ -904,12 +1029,12 @@ pub struct PatchApplyEndEvent {
pub success: bool,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TurnDiffEvent {
pub unified_diff: String,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct GetHistoryEntryResponseEvent {
pub offset: usize,
pub log_id: u64,
@@ -919,19 +1044,19 @@ pub struct GetHistoryEntryResponseEvent {
}
/// Response payload for `Op::ListMcpTools`.
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct McpListToolsResponseEvent {
/// Fully qualified tool name -> tool definition.
pub tools: std::collections::HashMap<String, McpTool>,
}
/// Response payload for `Op::ListCustomPrompts`.
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct ListCustomPromptsResponseEvent {
pub custom_prompts: Vec<CustomPrompt>,
}
#[derive(Debug, Default, Clone, Deserialize, Serialize)]
#[derive(Debug, Default, Clone, Deserialize, Serialize, TS)]
pub struct SessionConfiguredEvent {
/// Name left as session_id instead of conversation_id for backwards compatibility.
pub session_id: ConversationId,
@@ -949,6 +1074,8 @@ pub struct SessionConfiguredEvent {
/// When present, UIs can use these to seed the history.
#[serde(skip_serializing_if = "Option::is_none")]
pub initial_messages: Option<Vec<EventMsg>>,
pub rollout_path: PathBuf,
}
/// User's decision in response to an ExecApprovalRequest.
@@ -988,7 +1115,7 @@ pub enum FileChange {
},
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct Chunk {
/// 1-based line index of the first line in the original file
pub orig_index: u32,
@@ -996,7 +1123,7 @@ pub struct Chunk {
pub inserted_lines: Vec<String>,
}
#[derive(Debug, Clone, Deserialize, Serialize)]
#[derive(Debug, Clone, Deserialize, Serialize, TS)]
pub struct TurnAbortedEvent {
pub reason: TurnAbortReason,
}
@@ -1011,13 +1138,15 @@ pub enum TurnAbortReason {
#[cfg(test)]
mod tests {
use super::*;
use pretty_assertions::assert_eq;
use serde_json::json;
use tempfile::NamedTempFile;
/// Serialize Event to verify that its JSON representation has the expected
/// amount of nesting.
#[test]
fn serialize_event() {
let conversation_id = ConversationId(uuid::uuid!("67e55044-10b1-426f-9247-bb680e5fe0c8"));
let rollout_file = NamedTempFile::new().unwrap();
let event = Event {
id: "1234".to_string(),
msg: EventMsg::SessionConfigured(SessionConfiguredEvent {
@@ -1026,13 +1155,22 @@ mod tests {
history_log_id: 0,
history_entry_count: 0,
initial_messages: None,
rollout_path: rollout_file.path().to_path_buf(),
}),
};
let serialized = serde_json::to_string(&event).unwrap();
assert_eq!(
serialized,
r#"{"id":"1234","msg":{"type":"session_configured","session_id":"67e55044-10b1-426f-9247-bb680e5fe0c8","model":"codex-mini-latest","history_log_id":0,"history_entry_count":0}}"#
);
let expected = json!({
"id": "1234",
"msg": {
"type": "session_configured",
"session_id": "67e55044-10b1-426f-9247-bb680e5fe0c8",
"model": "codex-mini-latest",
"history_log_id": 0,
"history_entry_count": 0,
"rollout_path": format!("{}", rollout_file.path().display()),
}
});
assert_eq!(expected, serde_json::to_value(&event).unwrap());
}
#[test]

View File

@@ -59,9 +59,7 @@ ratatui = { version = "0.29.0", features = [
"unstable-rendered-line-info",
"unstable-widget-ref",
] }
ratatui-image = "8.0.0"
regex-lite = "0.1"
reqwest = { version = "0.12", features = ["json"] }
serde = { version = "1", features = ["derive"] }
serde_json = { version = "1", features = ["preserve_order"] }
shlex = "1.3.0"
@@ -81,12 +79,10 @@ tokio-stream = "0.1.17"
tracing = { version = "0.1.41", features = ["log"] }
tracing-appender = "0.2.3"
tracing-subscriber = { version = "0.3.19", features = ["env-filter"] }
tui-input = "0.14.0"
tui-markdown = "0.3.3"
pulldown-cmark = "0.10"
unicode-segmentation = "1.12.0"
unicode-width = "0.1"
url = "2"
uuid = "1"
pathdiff = "0.2"
[target.'cfg(unix)'.dependencies]

View File

@@ -1,14 +1,17 @@
use std::path::PathBuf;
use crate::app::App;
use crate::backtrack_helpers;
use crate::pager_overlay::Overlay;
use crate::tui;
use crate::tui::TuiEvent;
use codex_core::protocol::ConversationHistoryResponseEvent;
use codex_core::protocol::ConversationPathResponseEvent;
use codex_protocol::mcp_protocol::ConversationId;
use color_eyre::eyre::Result;
use crossterm::event::KeyCode;
use crossterm::event::KeyEvent;
use crossterm::event::KeyEventKind;
/// Aggregates all backtrack-related state used by the App.
#[derive(Default)]
pub(crate) struct BacktrackState {
@@ -97,7 +100,7 @@ impl App {
) {
self.backtrack.pending = Some((base_id, drop_last_messages, prefill));
self.app_event_tx.send(crate::app_event::AppEvent::CodexOp(
codex_core::protocol::Op::GetHistory,
codex_core::protocol::Op::GetPath,
));
}
@@ -264,7 +267,7 @@ impl App {
pub(crate) async fn on_conversation_history_for_backtrack(
&mut self,
tui: &mut tui::Tui,
ev: ConversationHistoryResponseEvent,
ev: ConversationPathResponseEvent,
) -> Result<()> {
if let Some((base_id, _, _)) = self.backtrack.pending.as_ref()
&& ev.conversation_id == *base_id
@@ -280,14 +283,14 @@ impl App {
async fn fork_and_switch_to_new_conversation(
&mut self,
tui: &mut tui::Tui,
ev: ConversationHistoryResponseEvent,
ev: ConversationPathResponseEvent,
drop_count: usize,
prefill: String,
) {
let cfg = self.chat_widget.config_ref().clone();
// Perform the fork via a thin wrapper for clarity/testability.
let result = self
.perform_fork(ev.entries.clone(), drop_count, cfg.clone())
.perform_fork(ev.path.clone(), drop_count, cfg.clone())
.await;
match result {
Ok(new_conv) => {
@@ -300,13 +303,11 @@ impl App {
/// Thin wrapper around ConversationManager::fork_conversation.
async fn perform_fork(
&self,
entries: Vec<codex_protocol::models::ResponseItem>,
path: PathBuf,
drop_count: usize,
cfg: codex_core::config::Config,
) -> codex_core::error::Result<codex_core::NewConversation> {
self.server
.fork_conversation(entries, drop_count, cfg)
.await
self.server.fork_conversation(drop_count, cfg, path).await
}
/// Install a forked conversation into the ChatWidget and update UI to reflect selection.

View File

@@ -1,4 +1,4 @@
use codex_core::protocol::ConversationHistoryResponseEvent;
use codex_core::protocol::ConversationPathResponseEvent;
use codex_core::protocol::Event;
use codex_file_search::FileMatch;
@@ -58,5 +58,5 @@ pub(crate) enum AppEvent {
UpdateSandboxPolicy(SandboxPolicy),
/// Forwarded conversation history snapshot from the current conversation.
ConversationHistory(ConversationHistoryResponseEvent),
ConversationHistory(ConversationPathResponseEvent),
}

View File

@@ -0,0 +1,15 @@
use std::io::Read;
use std::io::{self};
fn main() {
let mut input = String::new();
if let Err(err) = io::stdin().read_to_string(&mut input) {
eprintln!("failed to read stdin: {err}");
std::process::exit(1);
}
let parser = pulldown_cmark::Parser::new(&input);
for event in parser {
println!("{event:?}");
}
}

View File

@@ -1,4 +1,5 @@
use codex_core::protocol::TokenUsageInfo;
use codex_protocol::num_format::format_si_suffix;
use crossterm::event::KeyCode;
use crossterm::event::KeyEvent;
use crossterm::event::KeyEventKind;
@@ -1276,8 +1277,11 @@ impl WidgetRef for ChatComposer {
let token_usage = &token_usage_info.total_token_usage;
hint.push(" ".into());
hint.push(
Span::from(format!("{} tokens used", token_usage.blended_total()))
.style(Style::default().add_modifier(Modifier::DIM)),
Span::from(format!(
"{} tokens used",
format_si_suffix(token_usage.blended_total())
))
.style(Style::default().add_modifier(Modifier::DIM)),
);
let last_token_usage = &token_usage_info.last_token_usage;
if let Some(context_window) = token_usage_info.model_context_window {

View File

@@ -252,6 +252,11 @@ impl TextArea {
modifiers: KeyModifiers::CONTROL,
..
} => self.delete_backward(1),
KeyEvent {
code: KeyCode::Delete,
modifiers: KeyModifiers::ALT,
..
} => self.delete_forward_word(),
KeyEvent {
code: KeyCode::Delete,
..
@@ -435,6 +440,18 @@ impl TextArea {
self.replace_range(start..self.cursor_pos, "");
}
/// Delete text to the right of the cursor using "word" semantics.
///
/// Deletes from the current cursor position through the end of the next word as determined
/// by `end_of_next_word()`. Any whitespace (including newlines) between the cursor and that
/// word is included in the deletion.
pub fn delete_forward_word(&mut self) {
let end = self.end_of_next_word();
if end > self.cursor_pos {
self.replace_range(self.cursor_pos..end, "");
}
}
pub fn kill_to_end_of_line(&mut self) {
let eol = self.end_of_current_line();
if self.cursor_pos == eol {
@@ -1104,6 +1121,79 @@ mod tests {
assert_eq!(t.cursor(), 3);
}
#[test]
fn delete_forward_word_variants() {
let mut t = ta_with("hello world ");
t.set_cursor(0);
t.delete_forward_word();
assert_eq!(t.text(), " world ");
assert_eq!(t.cursor(), 0);
let mut t = ta_with("hello world ");
t.set_cursor(1);
t.delete_forward_word();
assert_eq!(t.text(), "h world ");
assert_eq!(t.cursor(), 1);
let mut t = ta_with("hello world");
t.set_cursor(t.text().len());
t.delete_forward_word();
assert_eq!(t.text(), "hello world");
assert_eq!(t.cursor(), t.text().len());
let mut t = ta_with("foo \nbar");
t.set_cursor(3);
t.delete_forward_word();
assert_eq!(t.text(), "foo");
assert_eq!(t.cursor(), 3);
let mut t = ta_with("foo\nbar");
t.set_cursor(3);
t.delete_forward_word();
assert_eq!(t.text(), "foo");
assert_eq!(t.cursor(), 3);
let mut t = ta_with("hello world ");
t.set_cursor(t.text().len() + 10);
t.delete_forward_word();
assert_eq!(t.text(), "hello world ");
assert_eq!(t.cursor(), t.text().len());
}
#[test]
fn delete_forward_word_handles_atomic_elements() {
let mut t = TextArea::new();
t.insert_element("<element>");
t.insert_str(" tail");
t.set_cursor(0);
t.delete_forward_word();
assert_eq!(t.text(), " tail");
assert_eq!(t.cursor(), 0);
let mut t = TextArea::new();
t.insert_str(" ");
t.insert_element("<element>");
t.insert_str(" tail");
t.set_cursor(0);
t.delete_forward_word();
assert_eq!(t.text(), " tail");
assert_eq!(t.cursor(), 0);
let mut t = TextArea::new();
t.insert_str("prefix ");
t.insert_element("<element>");
t.insert_str(" tail");
// cursor in the middle of the element, delete_forward_word deletes the element
let elem_range = t.elements[0].range.clone();
t.cursor_pos = elem_range.start + (elem_range.len() / 2);
t.delete_forward_word();
assert_eq!(t.text(), "prefix tail");
assert_eq!(t.cursor(), elem_range.start);
}
#[test]
fn cursor_left_and_right_handle_graphemes() {
let mut t = ta_with("a👍b");
@@ -1174,6 +1264,21 @@ mod tests {
assert_eq!(t.cursor(), 6);
}
#[test]
fn delete_forward_word_with_without_alt_modifier() {
let mut t = ta_with("hello world");
t.set_cursor(0);
t.input(KeyEvent::new(KeyCode::Delete, KeyModifiers::ALT));
assert_eq!(t.text(), " world");
assert_eq!(t.cursor(), 0);
let mut t = ta_with("hello");
t.set_cursor(0);
t.input(KeyEvent::new(KeyCode::Delete, KeyModifiers::NONE));
assert_eq!(t.text(), "ello");
assert_eq!(t.cursor(), 0);
}
#[test]
fn control_h_backspace() {
// Test Ctrl+H as backspace

View File

@@ -411,8 +411,6 @@ impl ChatWidget {
fn on_background_event(&mut self, message: String) {
debug!("BackgroundEvent: {message}");
self.add_to_history(history_cell::new_background_event(message));
self.request_redraw();
}
fn on_stream_error(&mut self, message: String) {
@@ -864,9 +862,6 @@ impl ChatWidget {
tx.send(AppEvent::DiffResult(text));
});
}
SlashCommand::Undo => {
self.open_undo_confirmation_popup();
}
SlashCommand::Mention => {
self.insert_str("@");
}
@@ -1088,7 +1083,7 @@ impl ChatWidget {
self.on_user_message_event(ev);
}
}
EventMsg::ConversationHistory(ev) => {
EventMsg::ConversationPath(ev) => {
self.app_event_tx
.send(crate::app_event::AppEvent::ConversationHistory(ev));
}
@@ -1258,43 +1253,6 @@ impl ChatWidget {
);
}
fn open_undo_confirmation_popup(&mut self) {
let confirm_message = "Undoing the last Codex turn diff.".to_string();
let undo_actions: Vec<SelectionAction> = vec![Box::new(move |tx| {
tx.send(AppEvent::InsertHistoryCell(Box::new(
history_cell::new_background_event(confirm_message.clone()),
)));
tx.send(AppEvent::CodexOp(Op::UndoLastTurnDiff));
})];
let items = vec![
SelectionItem {
name: "Undo last turn diff".to_string(),
description: Some(
"Revert files that Codex changed during the most recent turn.".to_string(),
),
is_current: false,
actions: undo_actions,
},
SelectionItem {
name: "Cancel".to_string(),
description: Some("Close without undoing any files.".to_string()),
is_current: false,
actions: Vec::new(),
},
];
self.bottom_pane.show_selection_view(
"Undo last Codex turn?".to_string(),
Some(
"Codex will apply a patch to restore files from before the previous turn."
.to_string(),
),
Some("Press Enter to confirm or Esc to cancel".to_string()),
items,
);
}
/// Set the approval policy in the widget's config copy.
pub(crate) fn set_approval_policy(&mut self, policy: AskForApproval) {
self.config.approval_policy = policy;

View File

@@ -0,0 +1,18 @@
---
source: tui/src/chatwidget/tests.rs
expression: visual
---
> -- Indented code block (4 spaces)
SELECT *
FROM "users"
WHERE "email" LIKE '%@example.com';
```sh
printf 'fenced within fenced\n'
```
{
// comment allowed in jsonc
"path": "C:\\Program Files\\App",
"regex": "^foo.*(bar)?$"
}

View File

@@ -13,7 +13,6 @@ use codex_core::protocol::AgentMessageEvent;
use codex_core::protocol::AgentReasoningDeltaEvent;
use codex_core::protocol::AgentReasoningEvent;
use codex_core::protocol::ApplyPatchApprovalRequestEvent;
use codex_core::protocol::BackgroundEventEvent;
use codex_core::protocol::Event;
use codex_core::protocol::EventMsg;
use codex_core::protocol::ExecApprovalRequestEvent;
@@ -36,6 +35,7 @@ use std::fs::File;
use std::io::BufRead;
use std::io::BufReader;
use std::path::PathBuf;
use tempfile::NamedTempFile;
use tokio::sync::mpsc::unbounded_channel;
fn test_config() -> Config {
@@ -134,7 +134,7 @@ fn resumed_initial_messages_render_history() {
let (mut chat, mut rx, _ops) = make_chatwidget_manual();
let conversation_id = ConversationId::new();
let rollout_file = NamedTempFile::new().unwrap();
let configured = codex_core::protocol::SessionConfiguredEvent {
session_id: conversation_id,
model: "test-model".to_string(),
@@ -144,11 +144,13 @@ fn resumed_initial_messages_render_history() {
EventMsg::UserMessage(UserMessageEvent {
message: "hello from user".to_string(),
kind: Some(InputMessageKind::Plain),
images: None,
}),
EventMsg::AgentMessage(AgentMessageEvent {
message: "assistant reply".to_string(),
}),
]),
rollout_path: rollout_file.path().to_path_buf(),
};
chat.handle_codex_event(Event {
@@ -178,6 +180,10 @@ fn resumed_initial_messages_render_history() {
);
}
#[cfg_attr(
target_os = "macos",
ignore = "system configuration APIs are blocked under macOS seatbelt"
)]
#[tokio::test(flavor = "current_thread")]
async fn helpers_are_available_and_do_not_panic() {
let (tx_raw, _rx) = unbounded_channel::<AppEvent>();
@@ -615,58 +621,6 @@ fn disabled_slash_command_while_task_running_snapshot() {
assert_snapshot!(blob);
}
#[test]
fn undo_command_requires_confirmation() {
let (mut chat, mut rx, _op_rx) = make_chatwidget_manual();
chat.dispatch_command(SlashCommand::Undo);
assert!(rx.try_recv().is_err(), "undo should require confirmation");
chat.handle_key_event(KeyEvent::new(KeyCode::Enter, KeyModifiers::NONE));
let mut undo_requested = false;
let mut history_lines = Vec::new();
while let Ok(event) = rx.try_recv() {
match event {
AppEvent::InsertHistoryCell(cell) => {
history_lines.push(cell.display_lines(80));
}
AppEvent::CodexOp(Op::UndoLastTurnDiff) => {
undo_requested = true;
}
_ => {}
}
}
assert!(undo_requested, "expected undo op after confirmation");
let combined = history_lines
.iter()
.map(|lines| lines_to_single_string(lines))
.collect::<String>();
assert!(combined.contains("Undoing the last Codex turn diff."));
}
#[test]
fn background_events_are_rendered_in_history() {
let (mut chat, mut rx, _op_rx) = make_chatwidget_manual();
chat.handle_codex_event(Event {
id: "undo".to_string(),
msg: EventMsg::BackgroundEvent(BackgroundEventEvent {
message: "Reverted last turn diff.".to_string(),
}),
});
let history = drain_insert_history(&mut rx);
let combined = history
.iter()
.map(|lines| lines_to_single_string(lines))
.collect::<String>();
assert!(combined.contains("Reverted last turn diff."));
}
#[tokio::test(flavor = "current_thread")]
async fn binary_size_transcript_snapshot() {
let (mut chat, mut rx, _op_rx) = make_chatwidget_manual();
@@ -1802,3 +1756,123 @@ fn chatwidget_exec_and_status_layout_vt100_snapshot() {
let visual = vt_lines.join("\n");
assert_snapshot!(visual);
}
// E2E vt100 snapshot for complex markdown with indented and nested fenced code blocks
#[test]
fn chatwidget_markdown_code_blocks_vt100_snapshot() {
let (mut chat, mut rx, _op_rx) = make_chatwidget_manual();
// Simulate a final agent message via streaming deltas instead of a single message
chat.handle_codex_event(Event {
id: "t1".into(),
msg: EventMsg::TaskStarted(TaskStartedEvent {
model_context_window: None,
}),
});
// Build a vt100 visual from the history insertions only (no UI overlay)
let width: u16 = 80;
let height: u16 = 50;
let backend = ratatui::backend::TestBackend::new(width, height);
let mut term = crate::custom_terminal::Terminal::with_options(backend).expect("terminal");
// Place viewport at the last line so that history lines insert above it
term.set_viewport_area(Rect::new(0, height - 1, width, 1));
let mut ansi: Vec<u8> = Vec::new();
// Simulate streaming via AgentMessageDelta in 2-character chunks (no final AgentMessage).
let source: &str = r#"
-- Indented code block (4 spaces)
SELECT *
FROM "users"
WHERE "email" LIKE '%@example.com';
````markdown
```sh
printf 'fenced within fenced\n'
```
````
```jsonc
{
// comment allowed in jsonc
"path": "C:\\Program Files\\App",
"regex": "^foo.*(bar)?$"
}
```
"#;
let mut it = source.chars();
loop {
let mut delta = String::new();
match it.next() {
Some(c) => delta.push(c),
None => break,
}
if let Some(c2) = it.next() {
delta.push(c2);
}
chat.handle_codex_event(Event {
id: "t1".into(),
msg: EventMsg::AgentMessageDelta(AgentMessageDeltaEvent { delta }),
});
// Drive commit ticks and drain emitted history lines into the vt100 buffer.
loop {
chat.on_commit_tick();
let mut inserted_any = false;
while let Ok(app_ev) = rx.try_recv() {
if let AppEvent::InsertHistoryCell(cell) = app_ev {
let lines = cell.display_lines(width);
crate::insert_history::insert_history_lines_to_writer(
&mut term, &mut ansi, lines,
);
inserted_any = true;
}
}
if !inserted_any {
break;
}
}
}
// Finalize the stream without sending a final AgentMessage, to flush any tail.
chat.handle_codex_event(Event {
id: "t1".into(),
msg: EventMsg::TaskComplete(TaskCompleteEvent {
last_agent_message: None,
}),
});
for lines in drain_insert_history(&mut rx) {
crate::insert_history::insert_history_lines_to_writer(&mut term, &mut ansi, lines);
}
let mut parser = vt100::Parser::new(height, width, 0);
parser.process(&ansi);
let mut vt_lines: Vec<String> = (0..height)
.map(|row| {
let mut s = String::with_capacity(width as usize);
for col in 0..width {
if let Some(cell) = parser.screen().cell(row, col) {
if let Some(ch) = cell.contents().chars().next() {
s.push(ch);
} else {
s.push(' ');
}
} else {
s.push(' ');
}
}
s.trim_end().to_string()
})
.collect();
// Compact trailing blank rows for a stable snapshot
while matches!(vt_lines.last(), Some(l) if l.trim().is_empty()) {
vt_lines.pop();
}
let visual = vt_lines.join("\n");
assert_snapshot!(visual);
}

View File

@@ -1 +0,0 @@

View File

@@ -49,34 +49,60 @@ pub struct PastedImageInfo {
/// Capture image from system clipboard, encode to PNG, and return bytes + info.
#[cfg(not(target_os = "android"))]
pub fn paste_image_as_png() -> Result<(Vec<u8>, PastedImageInfo), PasteImageError> {
let _span = tracing::debug_span!("paste_image_as_png").entered();
tracing::debug!("attempting clipboard image read");
let mut cb = arboard::Clipboard::new()
.map_err(|e| PasteImageError::ClipboardUnavailable(e.to_string()))?;
let img = cb
.get_image()
.map_err(|e| PasteImageError::NoImage(e.to_string()))?;
let w = img.width as u32;
let h = img.height as u32;
// Sometimes images on the clipboard come as files (e.g. when copy/pasting from
// Finder), sometimes they come as image data (e.g. when pasting from Chrome).
// Accept both, and prefer files if both are present.
let files = cb
.get()
.file_list()
.map_err(|e| PasteImageError::ClipboardUnavailable(e.to_string()));
let dyn_img = if let Some(img) = files
.unwrap_or_default()
.into_iter()
.find_map(|f| image::open(f).ok())
{
tracing::debug!(
"clipboard image opened from file: {}x{}",
img.width(),
img.height()
);
img
} else {
let _span = tracing::debug_span!("get_image").entered();
let img = cb
.get_image()
.map_err(|e| PasteImageError::NoImage(e.to_string()))?;
let w = img.width as u32;
let h = img.height as u32;
tracing::debug!("clipboard image opened from image: {}x{}", w, h);
let Some(rgba_img) = image::RgbaImage::from_raw(w, h, img.bytes.into_owned()) else {
return Err(PasteImageError::EncodeFailed("invalid RGBA buffer".into()));
};
image::DynamicImage::ImageRgba8(rgba_img)
};
let mut png: Vec<u8> = Vec::new();
let Some(rgba_img) = image::RgbaImage::from_raw(w, h, img.bytes.into_owned()) else {
return Err(PasteImageError::EncodeFailed("invalid RGBA buffer".into()));
};
let dyn_img = image::DynamicImage::ImageRgba8(rgba_img);
tracing::debug!("clipboard image decoded RGBA {w}x{h}");
{
let span =
tracing::debug_span!("encode_image", byte_length = tracing::field::Empty).entered();
let mut cursor = std::io::Cursor::new(&mut png);
dyn_img
.write_to(&mut cursor, image::ImageFormat::Png)
.map_err(|e| PasteImageError::EncodeFailed(e.to_string()))?;
span.record("byte_length", png.len());
}
tracing::debug!("clipboard image encoded to PNG ({}) bytes", png.len());
Ok((
png,
PastedImageInfo {
width: w,
height: h,
width: dyn_img.width(),
height: dyn_img.height(),
encoded_format: EncodedImageFormat::Png,
},
))

View File

@@ -28,6 +28,7 @@ use codex_core::protocol::SandboxPolicy;
use codex_core::protocol::SessionConfiguredEvent;
use codex_core::protocol::TokenUsage;
use codex_protocol::mcp_protocol::ConversationId;
use codex_protocol::num_format::format_with_separators;
use codex_protocol::parse_command::ParsedCommand;
use image::DynamicImage;
use image::ImageReader;
@@ -604,6 +605,7 @@ pub(crate) fn new_session_info(
history_log_id: _,
history_entry_count: _,
initial_messages: _,
rollout_path: _,
} = event;
if is_first_event {
let cwd_str = match relativize_to_home(&config.cwd) {
@@ -964,7 +966,7 @@ pub(crate) fn new_status_output(
// Input: <input> [+ <cached> cached]
let mut input_line_spans: Vec<Span<'static>> = vec![
" • Input: ".into(),
usage.non_cached_input().to_string().into(),
format_with_separators(usage.non_cached_input()).into(),
];
if usage.cached_input_tokens > 0 {
let cached = usage.cached_input_tokens;
@@ -974,12 +976,12 @@ pub(crate) fn new_status_output(
// Output: <output>
lines.push(Line::from(vec![
" • Output: ".into(),
usage.output_tokens.to_string().into(),
format_with_separators(usage.output_tokens).into(),
]));
// Total: <total>
lines.push(Line::from(vec![
" • Total: ".into(),
usage.blended_total().to_string().into(),
format_with_separators(usage.blended_total()).into(),
]));
PlainHistoryCell { lines }
@@ -1064,11 +1066,6 @@ pub(crate) fn new_stream_error_event(message: String) -> PlainHistoryCell {
PlainHistoryCell { lines }
}
pub(crate) fn new_background_event(message: String) -> PlainHistoryCell {
let lines: Vec<Line<'static>> = vec![vec![padded_emoji("").into(), message.into()].into()];
PlainHistoryCell { lines }
}
/// Render a userfriendly plan update styled like a checkbox todo list.
pub(crate) fn new_plan_update(update: UpdatePlanArgs) -> PlanUpdateCell {
let UpdatePlanArgs { explanation, plan } = update;
@@ -1182,13 +1179,14 @@ pub(crate) fn new_proposed_command(command: &[String]) -> PlainHistoryCell {
let mut lines: Vec<Line<'static>> = Vec::new();
lines.push(Line::from(vec!["".into(), "Proposed Command".bold()]));
let cmd_lines: Vec<Line<'static>> = cmd
.lines()
.map(|part| Line::from(part.to_string()))
.collect();
let highlighted_lines = crate::render::highlight::highlight_bash_to_lines(&cmd);
let initial_prefix: Span<'static> = "".dim();
let subsequent_prefix: Span<'static> = " ".into();
lines.extend(prefix_lines(cmd_lines, initial_prefix, subsequent_prefix));
lines.extend(prefix_lines(
highlighted_lines,
initial_prefix,
subsequent_prefix,
));
PlainHistoryCell { lines }
}

View File

@@ -97,7 +97,17 @@ pub fn insert_history_lines_to_writer<B, W>(
for line in wrapped {
queue!(writer, Print("\r\n")).ok();
write_spans(writer, &line).ok();
// Merge line-level style into each span so that ANSI colors reflect
// line styles (e.g., blockquotes with green fg).
let merged_spans: Vec<Span> = line
.spans
.iter()
.map(|s| Span {
style: s.style.patch(line.style),
content: s.content.clone(),
})
.collect();
write_spans(writer, merged_spans.iter()).ok();
}
queue!(writer, ResetScrollRegion).ok();
@@ -264,6 +274,10 @@ where
#[cfg(test)]
mod tests {
use super::*;
use crate::markdown_render::render_markdown_text;
use ratatui::layout::Rect;
use ratatui::style::Color;
use vt100::Parser;
#[test]
fn writes_bold_then_regular_spans() {
@@ -292,4 +306,240 @@ mod tests {
String::from_utf8(expected).unwrap()
);
}
#[test]
fn vt100_blockquote_line_emits_green_fg() {
// Set up a small off-screen terminal
let width: u16 = 40;
let height: u16 = 10;
let backend = ratatui::backend::TestBackend::new(width, height);
let mut term = crate::custom_terminal::Terminal::with_options(backend).expect("terminal");
// Place viewport on the last line so history inserts scroll upward
let viewport = Rect::new(0, height - 1, width, 1);
term.set_viewport_area(viewport);
// Build a blockquote-like line: apply line-level green style and prefix "> "
let mut line: Line<'static> = Line::from(vec!["> ".into(), "Hello world".into()]);
line = line.style(Color::Green);
let mut ansi: Vec<u8> = Vec::new();
insert_history_lines_to_writer(&mut term, &mut ansi, vec![line]);
// Parse ANSI using vt100 and assert at least one non-default fg color appears
let mut parser = Parser::new(height, width, 0);
parser.process(&ansi);
let mut saw_colored = false;
'outer: for row in 0..height {
for col in 0..width {
if let Some(cell) = parser.screen().cell(row, col)
&& cell.has_contents()
&& cell.fgcolor() != vt100::Color::Default
{
saw_colored = true;
break 'outer;
}
}
}
assert!(
saw_colored,
"expected at least one colored cell in vt100 output"
);
}
#[test]
fn vt100_blockquote_wrap_preserves_color_on_all_wrapped_lines() {
// Force wrapping by using a narrow viewport width and a long blockquote line.
let width: u16 = 20;
let height: u16 = 8;
let backend = ratatui::backend::TestBackend::new(width, height);
let mut term = crate::custom_terminal::Terminal::with_options(backend).expect("terminal");
// Viewport is the last line so history goes directly above it.
let viewport = Rect::new(0, height - 1, width, 1);
term.set_viewport_area(viewport);
// Create a long blockquote with a distinct prefix and enough text to wrap.
let mut line: Line<'static> = Line::from(vec![
"> ".into(),
"This is a long quoted line that should wrap".into(),
]);
line = line.style(Color::Green);
let mut ansi: Vec<u8> = Vec::new();
insert_history_lines_to_writer(&mut term, &mut ansi, vec![line]);
// Parse and inspect the final screen buffer.
let mut parser = Parser::new(height, width, 0);
parser.process(&ansi);
let screen = parser.screen();
// Collect rows that are non-empty; these should correspond to our wrapped lines.
let mut non_empty_rows: Vec<u16> = Vec::new();
for row in 0..height {
let mut any = false;
for col in 0..width {
if let Some(cell) = screen.cell(row, col)
&& cell.has_contents()
&& cell.contents() != "\0"
&& cell.contents() != " "
{
any = true;
break;
}
}
if any {
non_empty_rows.push(row);
}
}
// Expect at least two rows due to wrapping.
assert!(
non_empty_rows.len() >= 2,
"expected wrapped output to span >=2 rows, got {non_empty_rows:?}",
);
// For each non-empty row, ensure all non-space cells are using a non-default fg color.
for row in non_empty_rows {
for col in 0..width {
if let Some(cell) = screen.cell(row, col) {
let contents = cell.contents();
if !contents.is_empty() && contents != " " {
assert!(
cell.fgcolor() != vt100::Color::Default,
"expected non-default fg on row {row} col {col}, got {:?}",
cell.fgcolor()
);
}
}
}
}
}
#[test]
fn vt100_colored_prefix_then_plain_text_resets_color() {
let width: u16 = 40;
let height: u16 = 6;
let backend = ratatui::backend::TestBackend::new(width, height);
let mut term = crate::custom_terminal::Terminal::with_options(backend).expect("terminal");
let viewport = Rect::new(0, height - 1, width, 1);
term.set_viewport_area(viewport);
// First span colored, rest plain.
let line: Line<'static> = Line::from(vec![
Span::styled("1. ", ratatui::style::Style::default().fg(Color::LightBlue)),
Span::raw("Hello world"),
]);
let mut ansi: Vec<u8> = Vec::new();
insert_history_lines_to_writer(&mut term, &mut ansi, vec![line]);
let mut parser = Parser::new(height, width, 0);
parser.process(&ansi);
let screen = parser.screen();
// Find the first non-empty row; verify first three cells are colored, following cells default.
'rows: for row in 0..height {
let mut has_text = false;
for col in 0..width {
if let Some(cell) = screen.cell(row, col)
&& cell.has_contents()
&& cell.contents() != " "
{
has_text = true;
break;
}
}
if !has_text {
continue;
}
// Expect "1. Hello world" starting at col 0.
for col in 0..3 {
let cell = screen.cell(row, col).unwrap();
assert!(
cell.fgcolor() != vt100::Color::Default,
"expected colored prefix at col {col}, got {:?}",
cell.fgcolor()
);
}
for col in 3..(3 + "Hello world".len() as u16) {
let cell = screen.cell(row, col).unwrap();
assert_eq!(
cell.fgcolor(),
vt100::Color::Default,
"expected default color for plain text at col {col}, got {:?}",
cell.fgcolor()
);
}
break 'rows;
}
}
#[test]
fn vt100_deep_nested_mixed_list_third_level_marker_is_colored() {
// Markdown with five levels (ordered → unordered → ordered → unordered → unordered).
let md = "1. First\n - Second level\n 1. Third level (ordered)\n - Fourth level (bullet)\n - Fifth level to test indent consistency\n";
let text = render_markdown_text(md);
let lines: Vec<Line<'static>> = text.lines.clone();
let width: u16 = 60;
let height: u16 = 12;
let backend = ratatui::backend::TestBackend::new(width, height);
let mut term = crate::custom_terminal::Terminal::with_options(backend).expect("terminal");
let viewport = ratatui::layout::Rect::new(0, height - 1, width, 1);
term.set_viewport_area(viewport);
let mut ansi: Vec<u8> = Vec::new();
insert_history_lines_to_writer(&mut term, &mut ansi, lines);
let mut parser = Parser::new(height, width, 0);
parser.process(&ansi);
let screen = parser.screen();
// Reconstruct screen rows as strings to locate the 3rd level line.
let mut rows: Vec<String> = Vec::with_capacity(height as usize);
for row in 0..height {
let mut s = String::with_capacity(width as usize);
for col in 0..width {
if let Some(cell) = screen.cell(row, col) {
if let Some(ch) = cell.contents().chars().next() {
s.push(ch);
} else {
s.push(' ');
}
} else {
s.push(' ');
}
}
rows.push(s.trim_end().to_string());
}
let needle = "1. Third level (ordered)";
let row_idx = rows
.iter()
.position(|r| r.contains(needle))
.unwrap_or_else(|| {
panic!("expected to find row containing {needle:?}, have rows: {rows:?}")
});
let col_start = rows[row_idx].find(needle).unwrap() as u16; // column where '1' starts
// Verify that the numeric marker ("1.") at the third level is colored
// (non-default fg) and the content after the following space resets to default.
for c in [col_start, col_start + 1] {
let cell = screen.cell(row_idx as u16, c).unwrap();
assert!(
cell.fgcolor() != vt100::Color::Default,
"expected colored 3rd-level marker at row {row_idx} col {c}, got {:?}",
cell.fgcolor()
);
}
let content_col = col_start + 3; // skip '1', '.', and the space
if let Some(cell) = screen.cell(row_idx as u16, content_col) {
assert_eq!(
cell.fgcolor(),
vt100::Color::Default,
"expected default color for 3rd-level content at row {row_idx} col {content_col}, got {:?}",
cell.fgcolor()
);
}
}
}

View File

@@ -45,6 +45,7 @@ pub mod insert_history;
mod key_hint;
pub mod live_wrap;
mod markdown;
mod markdown_render;
mod markdown_stream;
pub mod onboarding;
mod pager_overlay;
@@ -61,11 +62,6 @@ mod user_approval_widget;
mod version;
mod wrapping;
// Internal vt100-based replay tests live as a separate source file to keep them
// close to the widget code. Include them in unit tests.
#[cfg(test)]
mod chatwidget_stream_tests;
#[cfg(not(debug_assertions))]
mod updates;
@@ -217,6 +213,7 @@ pub async fn run_main(
let file_layer = tracing_subscriber::fmt::layer()
.with_writer(non_blocking)
.with_target(false)
.with_span_events(tracing_subscriber::fmt::format::FmtSpan::CLOSE)
.with_filter(env_filter());
if cli.oss {
@@ -311,11 +308,7 @@ async fn run_ratatui_app(
..
} = cli;
let auth_manager = AuthManager::shared(
config.codex_home.clone(),
config.preferred_auth_method,
config.responses_originator_header.clone(),
);
let auth_manager = AuthManager::shared(config.codex_home.clone(), config.preferred_auth_method);
let login_status = get_login_status(&config);
let should_show_onboarding =
should_show_onboarding(login_status, &config, should_show_trust_screen);
@@ -399,11 +392,7 @@ fn get_login_status(config: &Config) -> LoginStatus {
// Reading the OpenAI API key is an async operation because it may need
// to refresh the token. Block on it.
let codex_home = config.codex_home.clone();
match CodexAuth::from_codex_home(
&codex_home,
config.preferred_auth_method,
&config.responses_originator_header,
) {
match CodexAuth::from_codex_home(&codex_home, config.preferred_auth_method) {
Ok(Some(auth)) => LoginStatus::AuthMode(auth.mode),
Ok(None) => LoginStatus::NotAuthenticated,
Err(err) => {

Some files were not shown because too many files have changed in this diff Show More