On Thu, May 9, 2019 at 10:03 AM Joe Perches <j...@perches.com> wrote: > > On Thu, 2019-05-09 at 09:21 +0200, Antonio Borneo wrote: > > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl > [] > > @@ -2224,7 +2229,7 @@ sub string_find_replace { > > sub tabify { > > my ($leading) = @_; > > > > - my $source_indent = 8; > > + my $source_indent = $tabsize; > > my $max_spaces_before_tab = $source_indent - 1; > > I didn't test this. > > Does this work properly if --tab-size=1 ? > Maybe die if ($tabsize < 2); is necessary?
I have tested it and it works fine, but now that you point it, I rechecked the code. There is already this in checkpatch sub tabify { ... my $source_indent = $tabsize; my $max_spaces_before_tab = $source_indent - 1; ... #Remove spaces before a tab 1 while $leading =~ s@^([\t]*)( {1,$max_spaces_before_tab})\t@$1\t@g; that works fine. But we could have in the future some other test in checkpatch that uses "$tabsize - 1" and does not get the proper validation for --tab-size=1 Maybe it's safer to put die if ($tabsize < 2) right now, and avoid future headaches. With a comment to explains this choice. Antonio