From 4558c8f9ff1e3855a72e647380d18d7a07b70eb9 Mon Sep 17 00:00:00 2001
From: Alex Turbov <i.zaufi@gmail.com>
Date: Mon, 10 Feb 2025 00:56:39 +0400
Subject: ci: Use `typos` to check commit messages

Commit messages in the CMake repo may have Git hashes that can trigger
false positives in the `typos` checker.

Revert commit ddebf4653d (ci: Do not check commit messages with 'typos'
due to false positives, 2025-02-07).  Instead, copy the top-level config
file to a temporary location, add a configuration option to mark
10-hex-digit identifiers as always valid, and pass it to `typos` at the
commit checking phase.
---
 .gitlab/ci/typos.bash | 15 ++++++++++++---
 .typos.toml           |  5 +++++
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/.gitlab/ci/typos.bash b/.gitlab/ci/typos.bash
index 7293a03..4c92383 100755
--- a/.gitlab/ci/typos.bash
+++ b/.gitlab/ci/typos.bash
@@ -7,8 +7,17 @@ result=0
 echo "Running 'typos' on source code..."
 typos || result=1
 
-# FIXME(typos): checking commit messages hits false positives
-# on "words" inside commit hashes.  We'd need a way to disable
-# checking of combined identifiers to avoid this.
+cfg='.typos.toml'
+tmp_cfg="${TEMP:-/tmp}/$cfg"
+# Uncomment `extend-ignore-identifiers-re` in the top-level config file
+# to make Git hashes (possibly used in commit messages) valid "identifiers".
+sed 's/^#\s*\(extend-ignore-identifiers-re\)/\1/' "$cfg" >"$tmp_cfg"
+
+if [ -n "$CI_MERGE_REQUEST_DIFF_BASE_SHA" ]; then
+  for COMMIT in $(git rev-list "^$CI_MERGE_REQUEST_DIFF_BASE_SHA" "$CI_COMMIT_SHA"); do
+    echo "Running 'typos' on commit message of $COMMIT..."
+    git show --format=%B -s "$COMMIT" | typos -c "$tmp_cfg" - || result=1
+  done
+fi
 
 exit $result
diff --git a/.typos.toml b/.typos.toml
index c494769..c06745a 100644
--- a/.typos.toml
+++ b/.typos.toml
@@ -11,6 +11,11 @@ extend-ignore-re = [
   , "(?Rm)^.*(#|/(/|\\*)|\\.\\.)\\s*(NOQA|noqa):? spellcheck(: *|=| +)disable-line$"
   ]
 locale = "en-us"
+# ATTENTION If, for any reason, you want to add the
+# `extend-ignore-identifiers-re` to this section,
+# please also modify the `.gitlab/ci/typos.bash`
+# script accordingly.
+#extend-ignore-identifiers-re=["\\b[0-9a-f]{10}\\b"]
 
 # Add repo-wide false positives here in the form of `word = "word"`.
 # Check the manual for details.
-- 
cgit v0.12