http://nagoya.apache.org/bugzilla/show_bug.cgi?id=961
*** shadow/961 Thu Mar 15 01:55:26 2001 --- shadow/961.tmp.5024 Mon Mar 19 05:51:49 2001 *************** *** 30,33 **** ------- Additional Comments From [EMAIL PROTECTED] 2001-03-15 01:55 ------- This is what the fixcrlf task is supposed to do. Any reason, why you cannot ! use it? --- 30,48 ---- ------- Additional Comments From [EMAIL PROTECTED] 2001-03-15 01:55 ------- This is what the fixcrlf task is supposed to do. Any reason, why you cannot ! use it? ! ! ------- Additional Comments From [EMAIL PROTECTED] 2001-03-19 05:51 ------- ! fixcrlf did not work correctly with UTF-16 files. From the carriage return word ! 0x000d, it removed only the 0x0d byte, and left the 0x00 byte in the file. ! Therefore all the rest unicode chars went broken. ! ! P.S. At the moment I'm calling Perl script from Ant and do the cr removal ! there. So at the moment this UTF-16 support in fixcrlf is not so critical for ! me. The Perl script of mine should be ANSI, UTF-8 and UTF-16 combatible, i.e. ! even as it does not know the file encoding, it should not break the files. Here ! is the main part of the script if someone's interested how to remove crs: ! $file =~ s/\x00\x0d\x00\x0a/\x00\x0a/g; ! $file =~ s/\x00\x0a\x00\x0d/\x00\x0a/g; ! $file =~ s/\x0d\x0a/\x0a/g; ! $file =~ s/\x0a\x0d/\x0a/g;
