Add text element metadata to protocol, app server, and core (#9331)

The second part of breaking up PR
https://github.com/openai/codex/pull/9116

Summary:

- Add `TextElement` / `ByteRange` to protocol user inputs and user
message events with defaults.
- Thread `text_elements` through app-server v1/v2 request handling and
history rebuild.
- Preserve UI metadata only in user input/events (not `ContentItem`)
while keeping local image attachments in user events for rehydration.

Details:

- Protocol: `UserInput::Text` carries `text_elements`;
`UserMessageEvent` carries `text_elements` + `local_images`.
Serialization includes empty vectors for backward compatibility.
- app-server-protocol: v1 defines `V1TextElement` / `V1ByteRange` in
camelCase with conversions; v2 uses its own camelCase wrapper.
- app-server: v1/v2 input mapping includes `text_elements`; thread
history rebuilds include them.
- Core: user event emission preserves UI metadata while model history
stays clean; history replay round-trips the metadata.
This commit is contained in:
charley-oai
2026-01-15 17:26:41 -08:00
committed by GitHub
parent 004a74940a
commit 1fa8350ae7
18 changed files with 416 additions and 46 deletions

View File

@@ -6,6 +6,8 @@ use codex_core::protocol::ItemCompletedEvent;
use codex_core::protocol::ItemStartedEvent;
use codex_core::protocol::Op;
use codex_protocol::items::TurnItem;
use codex_protocol::user_input::ByteRange;
use codex_protocol::user_input::TextElement;
use codex_protocol::user_input::UserInput;
use core_test_support::responses::ev_assistant_message;
use core_test_support::responses::ev_completed;
@@ -38,12 +40,18 @@ async fn user_message_item_is_emitted() -> anyhow::Result<()> {
let first_response = sse(vec![ev_response_created("resp-1"), ev_completed("resp-1")]);
mount_sse_once(&server, first_response).await;
let text_elements = vec![TextElement {
byte_range: ByteRange { start: 0, end: 6 },
placeholder: Some("<file>".into()),
}];
let expected_input = UserInput::Text {
text: "please inspect sample.txt".into(),
text_elements: text_elements.clone(),
};
codex
.submit(Op::UserInput {
items: (vec![UserInput::Text {
text: "please inspect sample.txt".into(),
text_elements: Vec::new(),
}]),
items: vec![expected_input.clone()],
final_output_json_schema: None,
})
.await?;
@@ -66,20 +74,16 @@ async fn user_message_item_is_emitted() -> anyhow::Result<()> {
.await;
assert_eq!(started_item.id, completed_item.id);
assert_eq!(
started_item.content,
vec![UserInput::Text {
text: "please inspect sample.txt".into(),
text_elements: Vec::new(),
}]
);
assert_eq!(
completed_item.content,
vec![UserInput::Text {
text: "please inspect sample.txt".into(),
text_elements: Vec::new(),
}]
);
assert_eq!(started_item.content, vec![expected_input.clone()]);
assert_eq!(completed_item.content, vec![expected_input]);
let legacy_message = wait_for_event_match(&codex, |ev| match ev {
EventMsg::UserMessage(event) => Some(event.clone()),
_ => None,
})
.await;
assert_eq!(legacy_message.message, "please inspect sample.txt");
assert_eq!(legacy_message.text_elements, text_elements);
Ok(())
}