fix(ci): match `?` and `[...]` in scope globs per character by sileht · Pull Request #1883 · Mergifyio/mergify-cli

sileht · 2026-10-05T14:37:30Z

globset compiles a glob to a (?-u) byte regex, so in mergify ci scopes a ? matched one byte and [é] was a class of é's two UTF-8
bytes. The engine matches str, so on any non-ASCII path a pattern
like critical/caf?.txt put critical/café.txt in the scope on the
engine side and not in ci scopes, which then skipped the scope-gated
job.

globset still parses the glob (braces, escapes, literal_separator
all unchanged); its regex is recompiled in Unicode mode, with the
\xNN byte escapes folded back into the characters they spell.

Fixes MRGFY-10066

Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com

mergify · 2026-10-05T14:37:41Z

Merge Protections

🔴 2 of 6 protections blocking · waiting on 👀 reviews

	Protection	Waiting on
🔴	👀 Review Requirements	👀 reviews
🔴	🔎 Reviews	👀 reviews
🟢	🤖 Continuous Integration	—
🟢	Enforce conventional commit	—
🟢	📕 PR description	—
🟢	🚦 Auto-queue	—

🔴 👀 Review Requirements

Waiting for

#approved-reviews-by>=2

This rule is failing.

🔴 🔎 Reviews

Waiting for

#review-requested = 0

This rule is failing.

#review-requested = 0
#changes-requested-reviews-by = 0
#review-threads-unresolved = 0

Show 4 satisfied protections

🟢 🤖 Continuous Integration

all of:
- check-success=ci-gate

🟢 Enforce conventional commit

Make sure that we follow https://www.conventionalcommits.org/en/v1.0.0/

title ~= ^(fix|feat|internal|docs|style|refactor|perf|test|build|ci|chore|revert|ui)(?:\(.+\))?!?:

🟢 📕 PR description

body ~= (?ms:.{48,})

🟢 🚦 Auto-queue

When all merge protections are satisfied, this pull request will be queued automatically.

Copilot

Copilot review overview

🟢 Approval recommended

The focused Unicode correction includes regression coverage, with no unresolved review findings.

Review effort: Balanced
Findings: None

What changed in this PR

Fixes Unicode scope matching in mergify ci scopes so ? and character classes match characters rather than UTF-8 bytes.

Changes:

Recompiles globset patterns as Unicode-aware regular expressions.
Adds regression tests and documents known engine differences.
Adds the workspace regex dependency.

File	Description
`crates/mergify-ci/src/scopes_detect/matching.rs`	Implements Unicode matching and adds regression coverage.
`crates/mergify-ci/Cargo.toml`	Declares the shared `regex` dependency.
`Cargo.lock`	Records the dependency for `mergify-ci`.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

JulianMaurin

日本語 !

mergify · 2026-10-05T15:01:57Z

Merge Queue Status

✅ Entered queue — 2026-10-05 15:01 UTC · Rule: default · triggered by merge protections
🚫 Left the queue — 2026-10-05 15:13 UTC · at 91ce6c76968a2c9c16032be8504ab696e51bd329

This pull request spent 11 minutes 46 seconds in the queue, with no time running CI.

Reason

The pull request conflicts with the base branch

The pull request was held in the queue, conflicting with at least one of 2 pull requests ahead of it. The conflict became a base branch one after #1879, #1881 merged.

Hint

You should update or rebase your pull request.

If you want to requeue this pull request, you can post a @mergifyio queue comment.

Tick the box to put this pull request back in the merge queue (same as @mergifyio queue).

Requeue this pull request

mergify · 2026-10-05T15:15:25Z

@sileht this pull request is now in conflict 😩

globset compiles a glob to a `(?-u)` byte regex, so in `mergify ci scopes` a `?` matched one byte and `[é]` was a class of é's two UTF-8 bytes. The engine matches `str`, so on any non-ASCII path a pattern like `critical/caf?.txt` put `critical/café.txt` in the scope on the engine side and not in `ci scopes`, which then skipped the scope-gated job. globset still parses the glob (braces, escapes, `literal_separator` all unchanged); its regex is recompiled in Unicode mode, with the `\xNN` byte escapes folded back into the characters they spell. Fixes MRGFY-10066 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Change-Id: I030bec464bbd2a85413524f6e652ea850265a905

sileht · 2026-10-05T15:24:11Z

Revision history

#	Type	Changes	Reason	Date
1	initial	`91ce6c7`		2026-10-05 15:24 UTC
2	content	`91ce6c7 → 5ec5782`		2026-10-05 15:24 UTC

Pull request has been modified.

sileht · 2026-10-06T05:49:33Z

Superseded by #1888 (MRGFY-10075), which includes this fix and replaces globset entirely. MRGFY-10066 closes with #1888.

`mergify ci scopes` matched scope globs with globset, the engine with its own brace expansion plus Python's `glob.translate`. Even with `*` bounded and `?` matching characters, the two dialects still disagreed on ASCII patterns, so one config could select different files in CI and in the merge queue: - `[^x]` negated in globset, it is a class holding `^` and `x` for the engine; - a class spanning a separator (`x[a/b]y`) was a class in globset, the engine splits the pattern on separators first, so it is literal text; - `\` was an escape in globset, it is a path separator for the engine. Replace globset with `engine_glob`, a port of the engine's `globs.expand_braces` and of `glob.translate(recursive=True, include_hidden=True, seps=["/", "\\"])`, emitting a `regex` crate pattern. `engine_glob_cases.json` holds patterns and paths with the verdicts the engine itself gives, generated by `engine_glob_cases.py` from the engine's `globs.py` on Python 3.14 with the `regex` module, and a test checks this side reproduces every one. A local run of the same comparison over ~190,000 random pattern/path pairs found no other difference. The one deliberate divergence: the `regex` module reads `[:name:]` inside a set as a POSIX class, which swallows the glob class's closing bracket. The engine then rejects the pattern or matches something meaningless; the CLI refuses it as a configuration error. A regex that fails to compile is a bug in the port, so it is reported as such rather than as the user's config. This also matches `?` and `[...]` per character on non-ASCII paths, which #1883 (MRGFY-10066) did on top of globset; that pull request is closed in favour of this one. Fixes MRGFY-10075 Fixes MRGFY-10066 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Change-Id: I67724907949c50cfc3f10f37dd968cc2e68dab2e

`mergify ci scopes` matched scope globs with globset, the engine with its own brace expansion plus Python's `glob.translate`. Even with `*` bounded and `?` matching characters, the two dialects still disagreed on ASCII patterns, so one config could select different files in CI and in the merge queue: - `[^x]` negated in globset, it is a class holding `^` and `x` for the engine; - a class spanning a separator (`x[a/b]y`) was a class in globset, the engine splits the pattern on separators first, so it is literal text; - `\` was an escape in globset, it is a path separator for the engine. Replace globset with `engine_glob`, a port of the engine's `globs.expand_braces` and of `glob.translate(recursive=True, include_hidden=True, seps=["/", "\\"])`, emitting a `regex` crate pattern. `engine_glob_cases.json` holds patterns and paths with the verdicts the engine itself gives, generated by `engine_glob_cases.py` from the engine's `globs.py` on Python 3.14 with the `regex` module, and a test checks this side reproduces every one. A local run of the same comparison over ~190,000 random pattern/path pairs found no other difference. The one deliberate divergence: the `regex` module reads `[:name:]` inside a set as a POSIX class, which swallows the glob class's closing bracket. The engine then rejects the pattern or matches something meaningless; the CLI refuses it as a configuration error. A regex that fails to compile is a bug in the port, so it is reported as such rather than as the user's config. This also matches `?` and `[...]` per character on non-ASCII paths, which #1883 (MRGFY-10066) did on top of globset; that pull request is closed in favour of this one. Fixes MRGFY-10075 Fixes MRGFY-10066 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Copilot AI balanced review requested due to automatic review settings October 5, 2026 14:37

mergify Bot deployed to Mergify Merge Protections October 5, 2026 14:37 Active

Copilot started reviewing on behalf of sileht October 5, 2026 14:38

View session

sileht deployed to func-tests-live October 5, 2026 14:38 — with GitHub Actions Active

Copilot AI reviewed Oct 5, 2026

View reviewed changes

mergify Bot requested a review from a team October 5, 2026 14:46

sileht marked this pull request as ready for review October 5, 2026 14:53

JulianMaurin previously approved these changes Oct 5, 2026

View reviewed changes

mergify Bot requested a review from a team October 5, 2026 14:58

remyduthu previously approved these changes Oct 5, 2026

View reviewed changes

mergify Bot added queued dequeued and removed queued labels Oct 5, 2026

mergify Bot added the conflict label Oct 5, 2026

sileht force-pushed the devs/sileht/mrgfy-10066-scopes-glob-char-semantics/match-scope-globs-per-character--030bec46 branch from 91ce6c7 to 5ec5782 Compare October 5, 2026 15:24

sileht deployed to func-tests-live October 5, 2026 15:24 — with GitHub Actions Active

mergify Bot had a problem deploying to Mergify Merge Protections October 5, 2026 15:24 Failure

mergify Bot removed dequeued conflict labels Oct 5, 2026

mergify Bot requested a review from a team October 5, 2026 15:30

sileht mentioned this pull request Oct 5, 2026

fix(ci): compile scope globs exactly as the engine does #1888

Merged

sileht closed this Oct 6, 2026

Navigation Menu

Repository navigation

fix(ci): match `?` and `[...]` in scope globs per character - #1883

fix(ci): match `?` and `[...]` in scope globs per character#1883
sileht wants to merge 1 commit into
mainfrom
devs/sileht/mrgfy-10066-scopes-glob-char-semantics/match-scope-globs-per-character--030bec46

sileht commented Oct 5, 2026

Uh oh!

mergify Bot commented Oct 5, 2026 •

edited

Loading

🟢 🤖 Continuous Integration

🟢 Enforce conventional commit

🟢 📕 PR description

🟢 🚦 Auto-queue

Uh oh!

Copilot AI left a comment

Uh oh!

JulianMaurin left a comment

Uh oh!

mergify Bot commented Oct 5, 2026 •

edited

Loading

Uh oh!

mergify Bot commented Oct 5, 2026

Uh oh!

sileht commented Oct 5, 2026

Uh oh!

sileht commented Oct 6, 2026

Uh oh!

This branch had an error being deployed

Reviewers

Uh oh!

Assignees

Labels

Milestone

Development

Uh oh!

4 participants

Conversation

sileht commented Oct 5, 2026

Uh oh!

mergify Bot commented Oct 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merge Protections

🔴 👀 Review Requirements

🔴 🔎 Reviews

🟢 🤖 Continuous Integration

🟢 Enforce conventional commit

🟢 📕 PR description

🟢 🚦 Auto-queue

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Copilot review overview

🟢 Approval recommended

Uh oh!

JulianMaurin left a comment

Choose a reason for hiding this comment

Uh oh!

mergify Bot commented Oct 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merge Queue Status

Reason

Hint

Uh oh!

mergify Bot commented Oct 5, 2026

Uh oh!

sileht commented Oct 5, 2026

Revision history

Uh oh!

sileht commented Oct 6, 2026

Uh oh!

This branch had an error being deployed

Reviewers

Uh oh!

Assignees

Labels

Milestone

Development

Uh oh!

4 participants

mergify Bot commented Oct 5, 2026 •

edited

Loading

mergify Bot commented Oct 5, 2026 •

edited

Loading