summaryrefslogtreecommitdiff
path: root/scripts
diff options
context:
space:
mode:
authorMauro Carvalho Chehab <mchehab@kernel.org>2022-10-17 13:04:46 +0200
committerMauro Carvalho Chehab <mchehab@kernel.org>2024-02-15 21:04:46 +0100
commit64bf05b11415d74d945841c73a281f1d635ce862 (patch)
tree2cc671cb370062ceae9d4630d435068d4eeb1790 /scripts
parent9a6c8fdfa98f3dcacd005b4bb76743c6989290ed (diff)
scripts/code_cov_parse_info: better handle include regexes
When there are both exclude and include regexes, the include ones may be overriding the more generic excluded ones. So, handle them first. Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org> Reviewed-by: Kamil Konieczny <kamil.konieczny@linux.intel.com>
Diffstat (limited to 'scripts')
-rwxr-xr-xscripts/code_cov_parse_info38
1 files changed, 21 insertions, 17 deletions
diff --git a/scripts/code_cov_parse_info b/scripts/code_cov_parse_info
index d3739211b..ceea67a13 100755
--- a/scripts/code_cov_parse_info
+++ b/scripts/code_cov_parse_info
@@ -34,17 +34,24 @@ sub is_function_excluded($)
my $func = shift;
+ # Handle includes first, as, when there are both include and exclude
+ # includes should take preference, as they can be overriding exclude
+ # rules
+ foreach my $r (@func_include_regexes) {
+ return 0 if ($func =~ m/$r/);
+ }
+
foreach my $r (@func_exclude_regexes) {
return 1 if ($func =~ m/$r/);
}
- return 0 if (!@func_include_regexes);
-
- foreach my $r (@func_include_regexes) {
- return 0 if ($func =~ m/$r/);
+ # If there are no exclude regexes, only include functions that are
+ # explicitly included.
+ if ($#func_exclude_regexes == 0) {
+ return 1;
}
- return 1;
+ return 0;
}
sub is_file_excluded($)
@@ -1178,9 +1185,10 @@ Each line at B<[filter's file]> may contain a new regex:
=back
-When both include and exclude regexes are found, exclude regexes are
-applied first and any functions that don't match the include regular
-expressions from the B<[filter's file]> will be ignored.
+Include regexes are handled first, as they can override an exclude regex.
+
+When just include regexes are used, any functions that don't match the
+include regular expressions from the B<[filter's file]> will be ignored.
Please notice that, when this filter is used, B<--ignore-unused> will be
automaticaly enabled, as the final goal is to report per-function usage.
@@ -1189,7 +1197,8 @@ automaticaly enabled, as the final goal is to report per-function usage.
Include B<regex> to the function filter. Can be used multiple times.
-When used together with B<--func-filters>, regexes here are handled first.
+When used together with B<--func-filters> or B<--exclude-func>, regexes
+here are handled first.
Please notice that, when this filter is used, B<--ignore-unused> will be
automaticaly enabled, as the final goal is to report per-function usage.
@@ -1198,8 +1207,6 @@ automaticaly enabled, as the final goal is to report per-function usage.
Include B<regex> to the function filter. Can be used multiple times.
-When used together with B<--func-filters>, regexes here are handled first.
-
Please notice that, when this filter is used, B<--ignore-unused> will be
automaticaly enabled, as the final goal is to report per-function usage.
@@ -1229,22 +1236,19 @@ Each line at B<[filter's file]> may contain a new regex:
=back
-When both include and exclude regexes are found, exclude regexes are
-applied first and any functions that don't match the include regular
-expressions from the B<[filter's file]> will be ignored.
+Include regexes are handled first, as they can override an exclude regex.
=item B<--include-src> B<regex>
Include B<regex> to the sources filter. Can be used multiple times.
-When used together with B<--src-filters>, regexes here are handled first.
+When used together with B<--src-filters> and B<--exclude-src>, regexes
+here are handled first.
=item B<--exclude-src> B<regex>
Include B<regex> to the sources filter. Can be used multiple times.
-When used together with B<--src-filters>, regexes here are handled first.
-
=item B<--ignore-unused> or B<--ignore_unused>
Filters out unused C files and headers from the code coverage results.