[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Feature Wish

From: Ralf Wiebicke <ralf_at_rw7.de>
Date: 2000-10-11 17:32:47 CEST

Hi there!

I recently made a disturbing experience with CVS. I hope SVN will do better
once it's released, so I'll describe the problem.

I was a newbie on CVS. I already dealt with the issue of different line
breaks on different systems. So I was glad to read, that CVS will handle this
automagically. However, I managed to get DOS line breaks into the repository.

All I did was using a UNIX CVS client to import text files with DOS line
breaks. The UNIX client doesn't check for UNIX line breaks, it just assumes
that they are there.

For the UNIX client, the 0D is just a character on the line. If another unix
client checks out, it gets correct DOS text files. However, I a DOS client
checks out these files, the 0D0A becomes 0D0D0A. Some editors treat this as
two line breaks, which messes up the whole text file.

To make the trouble complete, CVS somehow managed to add a single UNIX line
break at the end of each file. Thus, the files where mixed line break
flavour, which causes GNU recode to refuse the dos2unix translation (no
--force switch there). I finally used zip/unzip to do the conversion.

Please !!!, make subversion better.

Ralf Wiebicke.

PS: I'm not subscribed to this list. If there are any comments, which I
should know, please send me a copy.

Ralf Wiebicke <ralf@rw7.de>
Received on Sat Oct 21 14:36:10 2006

This is an archived mail posted to the Subversion Dev mailing list.