Mercurial > hg > nginx-vendor-0-7

view contrib/unicode2nginx/unicode-to-nginx.pl @ 268:a0c9f21ee120 NGINX_0_5_4

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
nginx 0.5.4 *) Feature: the "perl" directive may be used inside the "limit_except" block. *) Bugfix: the ngx_http_dav_module required the "Date" request header line for the DELETE method. *) Bugfix: if one only parameter was used in the "dav_access" directive, then nginx might report about configuration error. *) Bugfix: a segmentation fault might occur if the $host variable was used; bug appeared in 0.4.14.
author Igor Sysoev <http://sysoev.ru>
date Fri, 15 Dec 2006 00:00:00 +0300
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: '$_'";
	}
}

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