[Bug build/34410] New: [pre-commit] Don't call identify for each file
"vries at gcc dot gnu.org via Gdb-prs" <[email protected]>
| Newsgroups | gmane.comp.gdb.bugs.discuss |
|---|---|
| Message-ID | <[email protected]/bugzilla/> |
https://sourceware.org/bugzilla/show_bug.cgi?id=34410
Bug ID: 34410
Summary: [pre-commit] Don't call identify for each file
Product: gdb
Version: HEAD
Status: NEW
Severity: enhancement
Priority: P2
Component: build
Assignee: unassigned at sourceware dot org
Reporter: vries at gcc dot gnu.org
Target Milestone: ---
Pre-commit hooks run on files that match both the files and types filters.
The files filter uses fields files and exclude.
The types filter uses fields types, types_or and types_exclude.
The default values for the types filter fields are:
- types: [file]
- types_or: []
- types_exclude: []
If we'd have types == [] and types_or == [], there would be no need for
pre-commit to run identify on each matching file. Currently pre-commit doesn't
have this optimization, but it doesn't seem difficult to implement in
Classifier.by_types.
The file type classification done by identify is file/symlink/directory/socket,
so the question is whether dropping the file requirement introduces any real
regressions.
--
You are receiving this mail because:
You are on the CC list for the bug.