Re: Merge errors from HEAD to branch
On Tue, Mar 16, 2004 at 11:08:27AM -0500, Jim.Hyslop wrote: Now let's say I want to merge in the referred changes. The way I see it (as per book and info reading) is: cvs up -j HEAD:2 days ago -j HEAD ChangeLog HEAD has a specific meaning - it means the latest revision on the trunk [footnote]. Because you have specified an exact revision, the date component will be ignored (the date component can only be used with a _branch_ tag), so your command boils down to: cvs up -j HEAD -j HEAD which tells CVS Take the difference between revision 'HEAD' and revision 'HEAD', and apply that difference to my local file. Since the two revisions are identical, there is no difference and nothing to apply. If this is the first time you are merging (which I strongly suspect, given your question), then CVS will use the ancestor revision (in the specific Hmm. My situation was a bit peculiar: I wanted to merge in *only* the changes that were done in the last 2 days, ignoring changes done before that (which is why I specified the date modifier to HEAD); I still don't see how I can do that easily (without specifying tags to delimit the merge?). Perhaps CVS doesn't allow you to merge arbitrary timeslices of your tree without having specific tags to merge against; perhaps it's silly anyway to assume it would let you merge only a subset of the changes done to HEAD since the branch (given the chance that you're going to mess up is large since you're not respecting the historical evolution of the tree). I ended up just generating a diff of everything changes in the past two days and merging it in manually, but I would love to know if there's a more correct solution. Take care, -- Christian Robottom Reis | http://async.com.br/~kiko/ | [+55 16] 261 2331 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Merge errors from HEAD to branch
Hello there, I'm trying to merge a set of changes from HEAD to a branch opened a while back. The changes I would like to merge were committed to HEAD between 2 days ago and today; I'll use ChangeLog as an example because it's a file being actively changed on both HEAD and branch. I have the file on a branch checkout: [EMAIL PROTECTED]:~/devel/IC-0.6$ cvs status ChangeLog === File: ChangeLog Status: Up-to-date Working revision:1.222.2.5 Repository revision: 1.222.2.5 /cvs/IndexedCatalog/ChangeLog,v Sticky Tag: INDEXED_CATALOG_0_6_0_branch (branch: 1.222.2) Sticky Date: (none) Sticky Options: (none) Now let's say I want to merge in the referred changes. The way I see it (as per book and info reading) is: cvs up -j HEAD:2 days ago -j HEAD ChangeLog But what I get back is: cvs update: file ChangeLog exists, but has been added in revision HEAD and no changes are merged onto the ChangeLog. I've googled up and down for the error. It *seems* as though the error results when CVS tries to create the file and sees it already exists in the branch; however in this case it should definitely be possible to track the fact that the files in branch and HEAD have a common ancestor (version 1.222 in fact), and that therefore they are the same file. Is this a gotcha or am I doing something silly? Is there a way to do this type of merge, or must I do it manually? Take care, -- Christian Robottom Reis | http://async.com.br/~kiko/ | [+55 16] 261 2331 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Problem with cvslogin and GETPASS on HP/UX
Hi there, I've just noticed that logins on my HP/UX box here didnt work to the repository. Since we have a number of different OS' here I found it very odd. It turns out the problem is GETPASS (which is a macro to native getpass() ) truncating passwords at 8 chars, which makes it useless for reading most passwords for cvs. Has anyone run into this problem before? And does anybody know of a decent getpass() replacement that we could use for this purpose to let cvs login properly over the multipls OS' without depending on a native (potentially broken) version. I've tried using Bugzilla at cvshome.org but since it denies creating new bugs (why? would be very nice) perhaps someone in the development team could create one with my proposal? I'll try and find time to dig up a replacement for getpass to provide a patch. Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 272 3330 | NMFL ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Problem with cvslogin and GETPASS on HP/UX
On Mon, 17 Dec 2001, Christian Robottom Reis wrote: I'll try and find time to dig up a replacement for getpass to provide a patch. An alternative I see is using the getline library by Chris Thewalt, which is included in a number of packages including ncftp. It provides a getpass function with out the 8-char limitation, and it supposedly works cross-platform. Anybodyhave any comments? Take care, -- Christian Reis, Senior Engineer, Async Open Source, Brazil. http://async.com.br/~kiko/ | [+55 16] 272 3330 | NMFL ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: list modules
On Sun, 20 May 2001, Andy Nuss wrote: 1. list modules on server, 2. choose the module I want, AFAICS this _is_ obtainable using 'cvs history', but it seems like a hack. I don't know if anybody has a better solution, though. Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS bookmarker or addressbook
Us people that delve in the maze of Free Software repositories have one task that's dull -- keeping track of the CVSROOT strings for every project we want to pull. Is there something shell-oriented that provides something like that? A means to add and use root strings by a key or an index? This would be quite useful, and I'm sure I'm not the only one looking. :-) Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS bookmarker or addressbook
On Tue, 22 May 2001, Larry Jones wrote: Once you've checked something out, CVS automatically remembers the root for it (in CVS/Root), so there's no need for you to keep track of them. Yes, unless it's time I did a new checkout, or am on a different box than originally, in which case moving CVS/Root around is an alternative, albeit inconvenient. Are there more convenient solutions, however? Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS LOGIN
On Thu, 17 May 2001, Larry Jones wrote: Christian Robottom Reis writes: you could also use the new password syntax that pserver offers now What new password syntax?!? Well, from http://www.cvshome.org/dev/NEWS-1.11.1p1.txt * A password and a port number may now be specified in CVSROOT for pserver connections. The new format is: :pserver:[[user][:password]@]host[:[port]]/path Am I mistaken? Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS: Unknown Error 523
I'm running CVS on a Linux-ix86 box, running glibc-2.1.2 and cvs-1.10.8. I've been using CVS for years now, without issue. However, my recent upgrade to linux-2.4.4 with reiserfs has caused me to get some peculiar errors. It only happens on checkouts from remote pservers (my local pserver works fine) onto NFS-mounted partitions. I understand it's probably a filesystem error, but since it only happens to CVS, perhaps somebody has run into this before: blackjesus:~ cvs -d:pserver:[EMAIL PROTECTED]:/cvs-repositor co StandaloneZODB [snip] U bsddb3Storage/docs/custom_zodb.py U bsddb3Storage/docs/notes cvs server: Updating bsddb3Storage/test cvs server: Updating bsddb3Storage/tests cvs checkout: cannot remove StandaloneZODB/bsddb3Storage/tests: Unknown error 523 cvs checkout: cannot remove StandaloneZODB/bsddb3Storage/test: Unknown error 523 cvs checkout: cannot remove StandaloneZODB/BTrees/test: Unknown error 523 cvs checkout: cannot remove StandaloneZODB/ZEO/test: Unknown error 523 Anyone seen this before? Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: handling of MsWord-Documents in CVS ?
On Wed, 16 May 2001, Mike Castle wrote: Dump word. Use TeX or roff. That's _sane advice_. Although do note that MS' shift to XML-based formats could make this change. I'm not sure if the change information would make much sense, but it's a start. Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS: Unknown Error 523
On Wed, 16 May 2001, Mike Castle wrote: On Wed, May 16, 2001 at 05:41:52PM -0300, Christian Robottom Reis wrote: I've been using CVS for years now, without issue. However, my recent upgrade to linux-2.4.4 with reiserfs has caused me to get some peculiar errors. It only happens on checkouts from remote pservers (my local pserver works fine) onto NFS-mounted partitions. You're not using a reiserfs partition via NFS are you? There are known issues with that just plain not working. Mike, I understand there are issues; I was wondering if somebody has an idea of the fix! :-) It didn't happen with 2.4.3, and I _have_ applied the reiserfs-nfs patches. I've written to reiserfs and nfs to work with them for a solution (it's failing a getdents() call). Thanks for letting me know it's not just me, anyway. Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: CVS LOGIN
On Wed, 16 May 2001, G. Shi wrote: However, it seems that cvs login is interactive. Is there any way to include login, update logout in a script? You could use expect(1) to automate that, and you could also use the new password syntax that pserver offers now -- I believe only 1.11 has it, however. Take care, -- /\/\ Christian Reis, Senior Engineer, Async Open Source, Brazil ~\/~ http://async.com.br/~kiko/ | [+55 16] 274 4311 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
