view contrib/unicode2nginx/unicode-to-nginx.pl @ 372:6639b93e81b2 NGINX_0_6_30

nginx 0.6.30 *) Change: now if an "include" directive pattern does not match any file, then nginx does not issue an error. *) Feature: now the time in directives may be specified without spaces, for example, "1h50m". *) Bugfix: memory leaks if the "ssl_verify_client" directive was on. Thanks to Chavelle Vincent. *) Bugfix: the "sub_filter" directive might set text to change into output. *) Bugfix: the "error_page" directive did not take into account arguments in redirected URI. *) Bugfix: now nginx always opens files in binary mode under Cygwin. *) Bugfix: nginx could not be built on OpenBSD; bug appeared in 0.6.15.
author Igor Sysoev <http://sysoev.ru>
date Tue, 29 Apr 2008 00:00:00 +0400
parents fa32d59d9a15
children
line wrap: on
line source

#!/usr/bin/perl -w

# Convert unicode mappings to nginx configuration file format.

# You may find useful mappings in various places, including
# unicode.org official site:
#
# http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP1251.TXT
# http://www.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8-R.TXT

# Needs perl 5.6 or later.

# Written by Maxim Dounin, mdounin@rambler-co.ru

###############################################################################

require 5.006;

while (<>) {
	# Skip comments and empty lines

	next if /^#/;
	next if /^\s*$/;
	chomp;

	# Convert mappings

	if (/^\s*0x(..)\s*0x(....)\s*(#.*)/) {
		# Mapping <from-code> <unicode-code> "#" <unicode-name>
		my $cs_code = $1;
		my $un_code = $2;
		my $un_name = $3;

		# Produce UTF-8 sequence from character code;

		my $un_utf8 = join('', map { sprintf("%02X", $_) } unpack("C*", pack("U", hex($un_code))));

		print "    $cs_code  $un_utf8 ; $un_name\n";

	} else {
		warn "Unrecognized line: '$_'";
	}
}

###############################################################################