curl/.github/scripts/cleancmd.pl
Daniel Stenberg cc16211a30
cleancmd.pl: strip out backticked words
To make sure they are not spellchecked. Also, leaving two backticks is
not good because they cause the spellchecker to misinterpret the
markdown file so they have to be removed as well.
2025-01-03 18:36:50 +01:00

58 lines
1.1 KiB
Perl
Executable File

#!/usr/bin/env perl
# Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
#
# SPDX-License-Identifier: curl
#
# Input: a cmdline docs markdown, it gets modified *in place*
#
# The main purpose is to strip off the leading meta-data part, but also to
# clean up whatever else the spell checker might have a problem with that we
# still deem is fine.
my $header = 1;
while(1) {
# set this if the markdown has no meta-data header to skip
if($ARGV[0] eq "--no-header") {
shift @ARGV;
$header = 0;
}
else {
last;
}
}
my $f = $ARGV[0];
open(F, "<$f") or die;
my $ignore = $header;
my $sepcount = 0;
my @out;
while(<F>) {
if(/^---/ && $header) {
if(++$sepcount == 2) {
$ignore = 0;
}
next;
}
next if($ignore);
# strip out backticked words
$_ =~ s/`[^`]+`//g;
# strip out all long command line options
$_ =~ s/--[a-z0-9-]+//g;
# strip out https URLs, we don't want them spellchecked
$_ =~ s!https://[a-z0-9\#_/.-]+!!gi;
push @out, $_;
}
close(F);
if(!$ignore) {
open(O, ">$f") or die;
print O @out;
close(O);
}