Currently we have a single extended regular expression defining
files that clean-includes should ignore. This is now very long
and awkward to read and edit.

Switch to having a list of newline-separated EREs that we write
to a file for grep's -f option, so we can express them more
legibly in the shell script. We allow for comments in the
regex list, which lets us document why we have put the
exclusions in.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20260116124005.925382-4-peter.maydell@linaro.org
---
 scripts/clean-includes | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/scripts/clean-includes b/scripts/clean-includes
index 568033ee9c..5ab3b07196 100755
--- a/scripts/clean-includes
+++ b/scripts/clean-includes
@@ -42,8 +42,6 @@
 GIT=no
 DUPHEAD=no
 
-# Extended regular expression defining files to ignore when using --all
-XDIRREGEX='^(tests/tcg|tests/multiboot|tests/fp|tests/plugin|tests/uefi-test-tools|pc-bios|subprojects|contrib/plugins|tools/ebpf|ebpf/rss.bpf.skeleton.h|linux-user/(mips64|x86_64)/(cpu_loop|signal).c)'
 
 while true
 do
@@ -83,15 +81,33 @@ if [ "$1" = "--all" ]; then
     set -- '.'
 fi
 
-# We assume there are no files in the tree with spaces in their name
-set -- $(git ls-files "$@" | grep '\.[ch]$' | grep -E -v "$XDIRREGEX")
-
 # Annoyingly coccinelle won't read a scriptfile unless its
 # name ends '.cocci', so write it out to a tempfile with the
 # right kind of name.
 COCCIFILE="$(mktemp --suffix=.cocci)"
+REGEXFILE="$(mktemp --suffix=.regex)"
+
+trap 'rm -f -- "$COCCIFILE" "$REGEXFILE"' INT TERM HUP EXIT
+
+# List of extended regular expressions defining files to ignore
+# Comments starting with '#' are permitted
+grep -v '^#' >"$REGEXFILE" <<EOT
+# These tests are generally standalone binaries
+^tests/(tcg|multiboot|fp|plugin|uefi-test-tools)
+# BIOS sources and third-party subprojects don't follow our rules
+^pc-bios
+^subprojects
+# plugin binaries are standalone
+^contrib/plugins
+# the ebpf tool is standalone, and the skeleton header is autogenerated
+^tools/ebpf
+^ebpf/rss.bpf.skeleton.h
+# These files just include some other .c file and have no content themselves
+^linux-user/(mips64|x86_64)/(cpu_loop|signal).c
+EOT
 
-trap 'rm -f -- "$COCCIFILE"' INT TERM HUP EXIT
+# We assume there are no files in the tree with spaces in their name
+set -- $(git ls-files "$@" | grep '\.[ch]$' | grep -E -v -f "$REGEXFILE")
 
 cat >"$COCCIFILE" <<EOT
 @@
-- 
2.47.3