Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to sales@march-hare.com.
Tony, Tony, I have merged the passwd.c file from HEAD into the release branch to check up on the empty password handling. It now works OK but to get there I had to do something I did not expect was necessary. Can you please explain where what is done? On my local machine I had the Alpha version build 59 running as a service. The path points to the CVSNT directory of course. On the NT4 test server I replaced the 57a version with the one with the fix included. Then on a command prompt in my own machine I did the following: C:\test\castor>set cvsroot=:ntserver:castor:/test C:\test\castor>cvs passwd -r bosse -a tom Adding user tom at castor New password: cvs [passwd aborted]: A password must be supplied A bit of a surprise to me. So I replaced my local cvsnt from alpha build 59 to beta 57b (just compiled). And now the expected result was there: C:\test\castor>cvs passwd -r bosse -a tom Adding user tom at castor New password: Verify password: The passwd file now contains this line: tom::bosse Question: I always thought that the local cvs program just would pass the command over to the remote server and show the output from that to the user, not involving itself in the handling of the command. But this seems not to be the case since the local cvs.exe was not updated with the passwd fix and it barfed at the empty password. The server would not have complained. So where is CVS processing actually done? On the client or on the server? I am a bit confused now. /Bo (Bo Berglund, developer in Sweden) _______________________________________________ Cvsnt mailing list Cvsnt at cvsnt.org http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt https://www.march-hare.com/cvspro/en.asp#downcvs