DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18476>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=18476

copy with filtering modifies lineendings





------- Additional Comments From [EMAIL PROTECTED]  2003-04-22 00:07 -------
I have not seen that problem, but remember that I made the
patch on linux.
I have now tested on windows but have run out of time today to
make a patch on windows (needed to download cvsnt instead of cygwin
cvs as the latter treated the files as unix, and it takes a long
time to cvs ant on a modem).

On doing this I found another bug in the tests.
in testHeadTail, the expected outout is now incorrect (
it should be "Line  3\r\nLine  4\r\n").

All the tests thus passed, except the last test.

I could not understand why the unix tests passed so I rebooted
into linux and found the problem - the file names used in
the test where of a different case to the files on the filesystem
(head-tail.headtail.test as against head-tail.headTail.test). So
on unix the test was comparing two non-existent files!

If you could apply the patches to all the files except the head-tail.test
file, and add a new line using an editor to the head-tail.test, the
line ending problem (\n instead of \r\n) may be resolved.

Also I see that I included the wrong patched file (CommandlineJavaTest.java)
in patch 5850. The file should have been
src/etc/testcases/taskdefs/copy.filterset, and the change was to add a
new line to the end of the file. Sorry..., I will make a new patch tomorrow
from windows.

Peter

Reply via email to