Cascade version title changes to linker chunks

[yonadavGit]

Description

A brief description of the PR

Code Changes

The following changes were made to the files below

Notes

Any additional notes go here

[gitvelocity-reviewer]

📊 Code Quality Score: 15/100

Base Score 37 × ESF 0.4 = 14.8, rounded to 15

Category	Score	Factors
🔭 Scope	7/20	Three files modified: dependencies.py gains one subscribe call, marked_up_text_chunk.py gains a 16-line function and a docstring update, and the test file gains one standalone test function and a test isolation fix. All changes are within the marked_up_text_chunk subsystem.
🏗️ Architecture	5/20	The new handler follows the existing pattern established by process_index_delete and process_category_path_change in the same module. No new module boundary, no new dependency, and no new abstraction is introduced.
⚙️ Implementation	7/20	process_version_title_change builds a MongoDB $and/$or query using Ref(ver.title).regex(as_list=True) to scope updates to the correct book's refs, then calls update_many on both db.marked_up_text_chunks and db.linker_output with a $set on versionTitle. The language filter is applied via the shared query dict.
⚠️ Risk	6/20	update_many runs against two production collections without a feature flag or dry-run mode. A malformed ver.title passed to Ref() could raise at cascade time. The regex pattern list from Ref.regex(as_list=True) determines the blast radius; an overly broad pattern could update unintended documents. No rollback mechanism is documented.
✅ Quality	10/15	The test in sefaria/model/tests/marked_up_text_chunk.py uses SimpleNamespace to drive process_version_title_change and asserts three cases across both collections: matching ref+language updated, matching ref but wrong language not updated, out-of-scope ref not updated. Cleanup runs in finally blocks. The test isolation fix for test_incorrect_text_span uses deepcopy and finally to restore original_spans. No test covers the subscription wiring in dependencies.py or the multi-pattern regex path.
🔒 Perf / Security	2/5	update_many is used for bulk efficiency rather than per-document updates. No index analysis is provided for the versionTitle, language, and ref fields used in the query.

Was this score accurate? 👍 Yes · 👎 No

How this was scored →

_{Scored by GitVelocity · How are scores calculated?}