How to merge from HEAD into another branch?

I have (by mistake) imported (or checked in, I'm not sure) some new files/folders onto my main trunk (HEAD). I want to also have the new files/folders on one of my branches (DEV) but can't seem to figure out how to do this. I have tried from a DEV sandbox: $ cvs up -jHEAD

merging from HEAD into another branch?

Firstly please excuse the silly disclaimer at the end of my email. I have tried posting to this group using Google groups but for some reason the post don't seem to make it to the mailing list. If anyone can let me know how to post to the mailing list using Google groups (or any other public

merging from HEAD into a branch?

I've added a new directory (with files) to my HEAD branch and I'd like to include it into my sandbox which is for a branch called TEST. I've tried 'cvs update -jHEAD' but it doesn't do anything ... What is the correct way to merge changes from the HEAD into a branch? Thanks, Jc Please ignore

Re: merging from HEAD into a branch?

Rohan Nandode wrote: Try cvs update -d -jHEAD Tried, it didn't change anything. $ cvs -nq update -d -jHEAD RCS file: /repos/MRW/mizuhoApp/src/com/mizuho/rws/build.xml,v retrieving revision 1.1 retrieving revision 1.2 Merging differences between 1.1 and 1.2 into build.xml cvs update: build.xml: No

Protocol error: uncounted data discarded

I am trying to merge one branch into HEAD but I get the following strange error. $ cvs -nq update -j DEV M dao/EQRes_rwsConn.java RCS file: /repos/MRW/mizuhoApp/src/com/mizuho/rws/common/business/dao/EQRes_rwsConn.java,v retrieving revision 1.1 retrieving revision 1.1.2.1 Merging differences

Re: Protocol error: uncounted data discarded

to get the version numbers of the client and server you are using. If either of them is less than cvs 1.11.17 for the stable release $ cvs version Client: Concurrent Versions System (CVS) 1.11.15 (client/server) Server: Concurrent Versions System (CVS) 1.11.17 (client/server) In the general case,

Re: Protocol error: uncounted data discarded

Mark D. Baushke wrote: Yup, cvs 1.11.x has this problem both of those releases: It is fixed in cvs 1.11.17.1 (will be in cvs 1.11.18) with this patch: Thanks for replying, even though I still that nasty legal disclaimer. I'll ask the sysadmins to upgrade. As long as 1.11.17.1 is not a beta or

diff'ing a branch against HEAD doesn't work

@@ -A file in the TEST branch but not in HEAD +A file in HEAD and on the TEST branch What is the reason why I can diff from HEAD to another branch but not from a branch to HEAD? Thanks! -- Jean-Christian Imbeault Note: This e-mail contains privileged and confidential information

Re: diff'ing a branch against HEAD doesn't work

Spiro Trikaliotis wrote: Try cvs diff -u -N -r HEAD .test3 Thanks for trying but that didn't help. Even if the file is not new I still can't get diff to when I try to diff a branch to HEAD, but I can diff HEAD to a branch. == When in HEAD, diff works == $ cvs diff -r TEST .test3 Index: .test3

Re: diff'ing a branch against HEAD doesn't work

I found this from an old post and wonder if it could be what is causing my problem? From http://lists.gnu.org/archive/html/info-cvs/2004-03/msg00181.html [Footnote] There is one exception - for the 'diff' command, HEAD means the most recent revision in the currently-checked-out branch. If this

Why does CVS remember the last login user?

login but it doesn't seem to have any effect. Is there a simple way to share an account for CVS but have have the ability for people to login to cvs under different user names? Thanks! -- Jean-Christian Imbeault Assistant Manager Technology Department _ Mizuho

RE: How to keep developers separate from committers?

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Your sysadmin is mistaken. The commitinfo trigger is able to run cvs -Qn status on the list of files and determine the 'Sticky Tag:' line that relates to the file being committed. I don't know much about cvs

RE: How to keep developers separate from committers?

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of This issue has been raised multiple times in the info-cvs mailing list. You may find review of the list useful. For an example of one of the threads, follow this link:

RE: How to keep developers separate from committers?

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of For an example of one of the threads, follow this link: http://mail.gnu.org/archive/html/info-cvs/2002-05/msg00149.html http://mail.gnu.org/archive/html/info-cvs/2004-01/msg00024.html This looks exactly like the kind of

Merging from one module into another

At my company we have a repository with two projects (I hope I have the terminology right). One project is A and the other is A_DEV. Our developers work and commit only into A_DEV. When everything is ready for release we move the changes into A and release. My question is what is the easiest (or

Undoing renaming a directory

One of our developers just checked in new code (more than once) and by mistake renamed a directory from src to source. Is there a simple way to undo this? I've been reading the archives and it seems that there isn't ... Thanks, Jean-Christian Imbeault

RE: Merging from one module into another

From: [EMAIL PROTECTED] Using your terminology, cvs co A cvs update -j A -j A_DEV (Resolve conflicts, if any) cvs commit That doesn't seem to work. I get: $ cvs co A [snip] $ cvs -n update -j A -j A_DEV cvs server: WARNING: global `-l' option ignored. cvs

How to keep developers separate from committers?

merging easier but he said that going that route would make it impossible to restrict the developer's access to just one branch. What is a better approach to what we want to achieve, i.e. restricting access but making merging simple? Thanks, Jean-Christian Imbeault