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 email@example.com.
Ok. That makes sense. I was wrong that "U" files came over with the old timestamp every time. It's just the "U" files that are new that come over with the older timestamp. Thanks Paul! __________________________________ Reed Lawson IGT Firmware Engineering (775) 448-0755 -----Original Message----- From: Paul Russell [mailto:paul_newsgroups at yahoo.ca] Sent: Monday, September 08, 2003 9:58 PM To: Lawson.Reed; cvsnt at cvsnt.org cvsnt downloads at march-hare.com @CVSNT on Twitter CVSNT on Facebook Cc: Mason.Michael Subject: Re: [cvsnt] Timestamp problem I'm sure I'll be corrected if I'm wrong, but I think its like this: Fresh Checkout - Get files with timestamp from last commit. Update - Use current time. The reason is that if you: -compile/build using a make program -do an update that causes changes (regressed or another's work) -do another compile/build, then you want the changes compiled. If the updated files had the timestamp from the repository, then they might not get rebuilt by many make programs that rely on timestamps. You may end up with object/binary files that are out of sync with your source (murder to debug). So this kind of update always gets the latest time stamp. If you want the original time stamps for say a distribution, then check out to a new location (temporary folder). -Paul R. "Lawson.Reed" <Reed.Lawson at IGT.com> wrote: Hi, One of our users just showed me something that shocked me. I was in disbelief at first, but he is right.... If you run update and the file comes over as a patch "P" then the time stamp on the file is the current time. If its a "U", then it gets the time stamp of when that version was created in the archive. Is this normal behavior? (please say no) This does not seem right to me. I would expect if I check out a particular version of a file that the time stamp on the file (in Windows Explorer) would represent the time it was created in the archive on the server regardless of how cvs chooses to transport it. The $Header$ keyword displays the time the file was created in the archive. The Entries file and the File System timestamp agree with each other, but not the time/date text in the $Header$. Can someone try this and tell me if they see the same thing? We are using the latest beta from the WinCVS GUI site 1.3.b13 4/17/03 The cvs version on our server is slightly newer. cvs version Client: Concurrent Versions System (CVSNT) 2.0.2 (client/server) Server: Concurrent Versions System (CVSNT) 2.0.5 (client/server) Is this my problem? If so, is there a way to update the "cvs" that WinCVS is using without breaking WinCVS? Or do I need to downgrade the server to 2.0.2? Thanks, __________________________________ Reed Lawson IGT Firmware Engineering (775) 448-0755 _______________________________________________ cvsnt mailing list cvsnt at cvsnt.org cvsnt downloads at march-hare.com @CVSNT on Twitter CVSNT on Facebook http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt https://www.march-hare.com/cvspro/en.asp#downcvs _____ Post your free ad now! <http://ca.personals.yahoo.com/> Yahoo! Canada Personals