* fix(lint): make worker/frontend separation lint work on Windows
* chore: update cljs:electron-watch script to include test flag
* chore: remove dead root dependencies
* chore(deps): converge better-sqlite3 to 12.6.2 across deps packages
* chore(deps): converge fs-extra to ^11.3.0 across package roots
* fix(test): correct parameters for create-if-not-exists function
* chore(deps): converge cljs-bean to 1.9.0 across deps roots
* fix(tests): escape regex in cljs:run-test script
* chore: pin root packageManager to yarn 1.22.22
* chore(build): replace del with fs.rmSync in gulp clean
* chore(build): replace npm-run-all with npm-run-all2
* chore(security): upgrade dompurify and unify sanitizer path
* chore(observability): upgrade web sentry to 8.x
* chore: remove unused react-draggable dependencies
* chore(ci): fix windows release artifact collection
* fix(build): create static dir before gulp clean scans it
* fix: update nbb-logseq dependency to version feat-db-v33
* fix(test): move start-time initialization after clone repo
* fix(deps): update nbb dependencies and adjust test script paths to compatible with windows path delimiter
* chore(deps): remove dead meander dependency
---------
Co-authored-by: Tienson Qin <tiensonqin@gmail.com>
When new built-in properties are supported in exporter, all-built-in-names
and file-built-in-property-names should be updated. No need
to handle block/properties-order and block/properties-text-values
as those are ignored
Template blocks were being copied instead of updated in place. Would
feel buggy to user as Template blocks would show up twice in search and
would not be where they had defined them
* Fix import Zotero not recognized as asset
* Fix the inconsistency between imported Zotero label and actual filename
* Import support for zotero linked file like using zotfile
* Fix zotero highlighting block error after importing from file
* compatible with old configuration
* fix zotero picture highlight block does not consider as asset after importing
* Reconstructed to meet bb requirements
* empty datascript to avoid Conflicting upsert error after importing file to db graph
* use external-file-name instead of adding new property
* fix zotero link file can't open in asset page
* compatible with commit 33db791
* compatible with windows path & support zotero path can be relative for ci testing
* add zotero importing test
* remove prn log
* remove useless line
* Revert commit 45ebb9e
Future imports will be performed in the worker, will no longer encounter these issues
* refactor: update today page check to use async block retrieval
* Revert "refactor: update today page check to use async block retrieval"
This reverts commit 6750333df1.
* fix multiline block containing attributes will be truncated and lose data after importing
* fix the issues with block attribute name contains "/" after importing
* standardize eol, all files except .bat files use lf
* compatible with windows path
* fix multiple line importing issues
Also remove the following related file schema
attributes and their dependent code:
:block/marker, :block/priority, :block/scheduled,
:block/deadline,:block/repeated?, :block/pre-block?,
:block/properties-order, :block/properties-text-values, :block/macros,
:block/invalid-properties
Also update validation in DB scripts:
* replace alternative validate-db impl with standard CLI one
* Add validation to import script
* Choose to decouple other DB scripts from validate-db as it may go away
later
Asset files that are edn or md were being treated like assets and
extracted as doc files. This was noticed because pdf annotation edn was being
needlessly extracted
