Re: Modif timestamps of files in CVS repository
Larry Jones wrote: Laurent CHARLES writes: In our case, we only had such 2-digit dates for year 02, and we did not have any date prior to 2002. No recent version of CVS could have done that, only an ancient version that uses external RCS commands rather than the integrated RCS library along with an antique version of RCS. Yes. Actually we faced this problem after migrating a "cvsroot" from and old infrastructure to a new one, with new tools. The problem seems now solved for us. Thank you for your help. -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Re: Modif timestamps of files in CVS repository
Larry Jones wrote: (...) I'm sorry, I don't understand what you mean. Do you mean that your RCS file contains 2-digit dates for 2000 and later? Yes. Apparently set by some old version of cvs. Let me make sure we're talking about the same thing: Are you saying that your RCS files contain dates that have 04 to mean 2004, or is it just that recent RCS files still have dates with 99 to mean 1999? Yes. In our case, we only had such 2-digit dates for year 02, and we did not have any date prior to 2002. -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Re: Modif timestamps of files in CVS repository
Jim.Hyslop wrote: I'm sorry, I don't understand what you mean. Do you mean that your RCS file contains 2-digit dates for 2000 and later? Yes. Apparently set by some old version of cvs. -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Re: Modif timestamps of files in CVS repository
Hello, Thank you all for your replies. Laurent wrote: I have some dirty admin hacks to do on CVS repository files (*,v). I wonder if there is any risk or incidence to change the file timestamps. We want to rewrite the revision time information within the *,v files, but not to change the date, only to change the way the date is displayed. This is because the old date format with 2 digits for the year crashes viewCVS, while it is ok if the year is written with 4 digits. We take the risk to hack the files. The tests we have done so far show it seems to work as expected. And about this: That is an exceedingly bad idea. It's not a question of "old" format versus "new" format -- the RCS file format explicitly requires 2 digits for years 1900 through 1999 and all digits for subsequent years (it does not support years prior to 1900). I agree, but we checked that not 2 digit dates are prior 2000. So it is ok _in our case_. Thank you again for your help. -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Re: Modif timestamps of files in CVS repository
Laurent CHARLES wrote: I have some dirty admin hacks to do on CVS repository files (*,v). I wonder if there is any risk or incidence to change the file timestamps. NB: I accept the risk to badly hack the files themselves. Jim.Hyslop wrote: So, let me get this straight - you want to change the timestamp of the checkin. Yes, but it was not the sense of my question. First of all, why? Is there a specific reason you want to rewrite history? I'd be concerned about what would happen if you changed the timestamps such that an earlier revision number has a later date than a later revision number, i.e.: rev 1.1: January 6, 2005 rev 1.2: December 31, 2004 We want to rewrite the revision time information within the *,v files, but not to change the date, only to change the way the date is displayed. This is because the old date format with 2 digits for the year crashes viewCVS, while it is ok if the year is written with 4 digits. My question was about the timestamp of the file itself, not the one within the file. Because doing the hack above changes the modification date of all the files, even those that has not changed since ages. I wondered if CVS uses this information (file timestamp), or if it only relies on the date information present within the files. Sorry not to have been clear enough. Thank you -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Modif timestamps of files in CVS repository
Hello, I have some dirty admin hacks to do on CVS repository files (*,v). I wonder if there is any risk or incidence to change the file timestamps. NB: I accept the risk to badly hack the files themselves. Thank you for your support Regards -- -- Laurent ___ Info-cvs mailing list Info-cvs@gnu.org http://lists.gnu.org/mailman/listinfo/info-cvs
Re: CVS in web development environment
Hi, Ian Michell wrote: We are currently thinking of migrating to CVS for all our development. It is fine for java development with multiple devlopers as we can compile the code in our own sandboxes. But on the web development side it is a little trickier with multiple developers as we have one development server that is used for testing, and everyone currently tests and develops out of the same directory. One way of sandboxes this, is to allow the developers to checkout into thier own home directory (on the development server). Our main problem is we have over 100 virtual hosts in the apache conf file pointing to our development server, therefore we need to find a way to allow each developer to make use of these virtual hosts, and develop and test out of their own sandboxes. Are there any ideas, or any common ways to solve this problem? I don't know if there is a common way. We manage some similar issues with web applications, not with as many virtual servers though. Here is a attempt to give you some ideas of simple principles we try to respect. To simplify, we can consider we have 3 types of files to manage in an application; - Sources - Configuration - Data The sources are simply managed with a classical repository. Checkouts, updates, merges, etc. are the everyday life of the developpers in their sandboxes. The configuration files are managed the same way if they apply to all machines and services. If conf files can not be applied to all machines (like 'httpd.conf' for ex), we either use one conf file per machine (eg 'httpd.conf.machine'), and soft links it(*) to the ad hoc file on every machine, or we have a process to generate it from a template. (*) the developers can also start the service specifying the conf file. We may have a Makefile in a "main" directory with a rules such as 'update'. Thus, 'make update' runs for example, a 'cvs update' in every directories (eg: /etc, /var/www, etc.), and may run some processes as well (e.g create links, generate config files). Managing data is more complicated. I mean data that can be considered as application sources (initialisations for ex), not users' or application data. We try to manage them with scripts under control (for example sql scripts to run to initialise a database). A similar 'make update' process may be used if relevant. On the server side, developers have usually their own server running one instance of the application, based on their current sandbox sources. It is also possible to run several instances of the application on the same server, each based on its own set of sources and configuration files. This might be interesting when tests require the usage of a single database. With such structure of our works, we have a pretty good control of the whole application, and can setup very quickly a new developping environment based on any version of the application. Hope it helps. I'm interested in any feedback people can give. Another comment... Sorry to post it here... Why CVS? Why not starting with Subversion or Arch? -- -- Laurent ___ Info-cvs mailing list [EMAIL PROTECTED] http://lists.gnu.org/mailman/listinfo/info-cvs
CVS Layout problem
I trying to setup a new CVS repository. I would like my various modules to share the same src/ tree so I will not have a seperate subdirectory for each project. Here's my first approach: 0. cvs -d $CVSROOT init 1. cd $home/project1 2. cvs import -m "Imported Source" . project1 start This works fine. I can look in my repository and I see: ls -l $CVSROOT CVSROOT/ src/ project1,v* project1.xml,v And $CVSROOT/src contains my java sources: src/com/ox/db/*.java src/com/ox/utils/*.java etc. I than modified the CVSROOT/modules file to create new module: % cat modules project1./ ./project1 ./project1.xml \ &src/com/ox/db \ &src/com/ox/esi \ &src/com/ox/rv \ &src/com/ox/security \ &src/com/ox/trade \ &src/com/ox/tools \ &src/com/ox/utils src/com src/com src/com/ox src/com/ox src/com/ox/db src/com/ox/db src/com/ox/esi src/com/ox/esi src/com/ox/rv src/com/ox/rv src/com/ox/security src/com/ox/security src/com/ox/toolssrc/com/ox/tools src/com/ox/tradesrc/com/ox/trade src/com/ox/utilssrc/com/ox/utils I can now checkout my project just fine: % cvs checkout project1 Of course, now everything shows up in $home/project1/ but I can build the project and run it just fine. But I *CANNOT* check it back in and I have not idea why. I retried from scratch a couple of times and got the same resutls: % cvs checkin project1 cvs commit: Examining project1 cvs commit: Up-to-date check failed for `project1/.' cvs commit: Examining project1/src cvs commit: Examining project1/src/com cvs commit: Examining project1/src/com/ox cvs commit: Examining project1/src/com/ox/db cvs commit: Examining project1/src/com/ox/esi cvs commit: Examining project1/src/com/ox/esi/command cvs commit: Examining project1/src/com/ox/esi/event cvs commit: Examining project1/src/com/ox/esi/record cvs commit: Examining project1/src/com/ox/rv cvs commit: Examining project1/src/com/ox/security cvs commit: Examining project1/src/com/ox/security/rv cvs commit: Examining project1/src/com/ox/tools cvs commit: Examining project1/src/com/ox/trade cvs commit: Examining project1/src/com/ox/utils cvs [commit aborted]: correct above errors first! >From the I tried "cd project1" and than "cvs commit ." or "cvs commit" and I get the same results. I also tried various combinations with "cvs update" but I alway get: cvs update: conflict: project1/. is modified but no longer in the repository I'm not sure I'm following what the problem is. -- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com| but me. To err is Human, to forgive is against Departmental Policy! ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
How to move trunk to brance?
Hello, I checked in v2.1 sources yesterday. Now I want to make them branch and use Mar 1 data as the trunk. Can you tell me how to do it? Thanks, Charles Sun ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
RE: The recursive behavior of CVS
Actually, there are TWO -d options, one to the left and one to the right of your checkout command. The one on the left sets the CVS root. The one on the right points out to CVS where the work area is. It seems, from a little experimentation, that CVS needs to find A root, and will take it from one of the work area's CVS directories if it isn't set on the command line or in the environment. If your current working directory isn't a work area, and you don't use that righthand-side -d option to find the working directory, CVS declines to make assumptions about where the work area is and thus you need to set the CVSROOT yourself. As to what happens when you checkout something you already have and you supply a different CVSROOT than expected, you'll have to find out on your own. Also, the rules update follows are its own. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
How to admin deep hierachy directories
I have a tool that generates many new files and folders in my working area. These files shall be CVS controlled. For example, my working area looks like: Project_folder --- folderA --- subfolderA1 --- fileA1.1 -- folderB --- subfolderB1 --- fileB1.1 After running my tool, it becomes: Project_folder --- folderA --- subfolderA1 --- fileA1.1(updated) -- subfolderA2 --- fileA2.1 -- folderB --- subfolderB1 --- fileB1.1 -- fileB1.2 -- folderC --- fileC.1 I don't want to find and add each new files and folders one by one. What is the easiest way to maintain my working area? I want to add a symbolic link in my working area. Can CVS keep this and distribute it to other co-developers? Charles ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: User variable problem
There's a shell setting that makes it complain when you use an undefined value.
(But I don't recall what it is). If you are using ksh, you can use
${parameter:-word} to substitute "word" in place of "parameter" if parameter
isn't defined.
DEFAULT /serv/CVSROOT/commitcheck ${COMMITCODE:-JUNK}
Then, in your script, if $1 == Junk, then you know the user left it off.
-CTH
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
Re: restricting concurrent editing
> > can any body tell me how to restrict users editing the same file at a time > That's against the philosophy behind CVS. CVS was designed to explicitly > _allow_ people from editing a file at the same time. In case of no Not quite. CVS is designed to make concurrent editing feasable. That doesn't mean you need to do it, and if you do it, you don't necessarily have to do it on the same branch. There are two ways to deal with your problem. The complex one is to make everybody work on their own branch. This will have "overhead" in your process. More simply, and probably what you want is the command "cvs admin -l". It will lock a revision until you checkin or somebody else uses "cvs admin -u". -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Multiple tags
> But as Larry said to the original poster: > > No. You're expected to tag all the files in a module, > > not just some of them. > Indeed. In CVS, tags are meant to label the state of the module No. That is untrue. Otherwise you wouldn't be able to place a tag on a single file or checkout a single file. What you both ment to say, is that if you checkout a module, any tag referenced is assumed to exist over the whole module (or else the file is excluded WRT that tag). That is why, cvs "cleans up" the work are to give you what you asked for (which in this case is not what you want). The net result is the same however; use a script and work by file. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
fetching multiple tags
I found this question interesting, since I'm about to address it in a CVS wrapper I'm writing... Marc Tessier wrote: > I am wandering if there is any options to extract the most recent > version of a list of TAG? I have tag A, B, C and the I want to get the > most recent files of all three in one cvs checkout or cvs export so I > don't have to check the files version one by one. Larry replied: > No. You're expected to tag all the files in a module, > not just some of them. Perhaps I'm answering a different question. Here's my first solution: $ cvs checkout -r C f1 f2 f3 ... $ cvs checkout -r B f1 f2 f3 ... $ cvs checkout -r A f1 f2 f3 ... Where the first checkout gets the files with tag C. The second checkout replaces any file that has a B tag (and leaves behind any file that should be deleted as-of B I think). The third command overlays all the files tagged with A. While this approach is "wasteful" in that it can fetch 3 different versions of the same file, it meets your criteria for not having to think very hard to make it work right. A scripted approach, which would be more efficient is to collect the list of files and revisions first, and then get only the "good" ones. $ cvs -n checkout -r C<<-- returns the list that has tag C. $ cvs -n checkout -r B<<-- returns the list that has tag B. $ cvs -n checkout -r A<<-- returns the list that has tag A. Armed with the 3 lists, you can only fetch each file once. Though I don't have an example database handy with deleted files, so I can't say what to do about discarding them from the list. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
cvs checkout -d -N (more)
Also, this sequence works too... $ cvs checkout -d WorkArea4 -N mod1/file3 $ cvs checkout -d WorkArea4 -N file1 $ cvs checkout -d WorkArea4 file2 Work Area State: File: Module/file: new -d-d (-N) old -d (-N) -d -N Here are each of the cases laid out (oh, I'm on solaris using 1.11.2 $ cvs checkout -d WorkArea1 -N file1 cvs checkout: existing repository /BBROOT0/CVSROOT/Emptydir does not match /BBROOT0 cvs checkout: ignoring module file1 $ cvs checkout -d WorkArea2 file1 U /home/cthart/work/WorkArea2/file1 $ cvs checkout -d WorkArea3 -N mod1/file3 U /home/cthart/work/WorkArea3/mod1/file3 $ cvs checkout -d WorkArea4 mod1/file3 U /home/cthart/work/WorkArea4/file3 $ cvs checkout -d WorkArea5 file2 U /home/cthart/work/WorkArea5/file2 $ cvs checkout -d WorkArea5 -N file1 U /home/cthart/work/WorkArea5/./file1 $ cvs checkout -d WorkArea6 file2 U /home/cthart/work/WorkArea6/file2 $ cvs checkout -d WorkArea6 file1 U /home/cthart/work/WorkArea6/file1 $ cvs checkout -d WorkArea7 file2 U /home/cthart/work/WorkArea7/file2 $ cvs checkout -d WorkArea7 -N mod1/file3 U /home/cthart/work/WorkArea7/mod1/file3 $ cvs checkout -d WorkArea8 file2 U /home/cthart/work/WorkArea8/file2 $ cvs checkout -d WorkArea8 mod1/file3 cvs checkout: existing repository /BBROOT0 does not match /BBROOT0/mod1 cvs checkout: ignoring module mod1/file3 -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
cvs checkout -N -d xxx
This works flawlessly: $ mkdir WorkArea1 $ cd WorkArea1 $ cvs checkout file1 file2 mod1/file3 As does this (note the missing -N at the beginning): $ cvs checkout -d WorkArea2 file1 $ cvs checkout -d WorkArea2 -N file2 $ cvs checkout -d WorkArea2 -N mod1/file3 But, the first and third commands of this sequence fail: $ cvs checkout -d WorkArea3 -N file1 $ cvs checkout -d WorkArea3 file2 $ cvs checkout -d WorkArea3 mod1/file3 Can somebody clarify the purpose of the -N switch and how it interacts with -d, since the Cederqvist text doesn't seem to relate to this behavior at all. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
starting with an empty repository (more)
I made some progress... $ cvs checkout -l . $ touch file0 $ mkdir module $ cvs add module $ touch module/file2 $ cvs add module/file2 The module itself needs to be added to then get the file in. Further, if I don't want ALL of the files in the repository root, (caused by the initial check out), the solution seems to be name one file rather than ".". -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
starting with an empty repository
Maybe it's me and not cvs 1.11.2 ... I think the following command sequence should work... $ mkdir workarea $ cd workarea $ cvs checkout -l . $ touch file0 $ mkdir module $ touch module/file2 $ cvs add module/file2 ... I am looking for a command sequence to get file2 into CVS without first populating my local area with all of the CVS's world or getting any other junk (file0) sucked in. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
checkout -r dst -j src fyle
I'm writing a script to merge files from one branch (src) to another (dst). If a file has never been merged between the two, then the command "checkout -r dst -j src fyle" should checkout the file on the dst branch and then merge the changes from the tip of the src branch into it. If this merge has to be repeated, because of new work on the src branch, the command "checkout -r dst -j OLDM -j src fyle", where OLDM was src branch tip when the previous merge was done, would do the subsequent merge. But if I don't want to do this per file and want to work on a whole repository, I have to use the -j -j format for the files that don't have a label, yet. If I put the label on the "common ancestor" version, will it work? If there is no common ancestor, should I use verison 1.1.1.1? -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Security, audits and pserver
um, I'm a newbie at CVS, so I've read more of the documentation than anything else, but the answers I've seen so far for the security question seem to have missed one vital point. People have write access to spots in the repository, therefore they, just like CVS, can write as they please to the ,v files. That means that there is no guarentee that the history that is present in the repository reflects the actual events in history. This hole, even if it is not exploited in an organization because everybody is "good", is still a hole, and won't pass an audit. What I'll be testing next week, is a CVS server where no "users" exist in /etc/passwd, and all access rights are granted through pserver mapping CVS user IDs to security accounts. The admin can still bypass audit controls, but that's better than having my 1,000 users being enabled to. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
anomoly merging from one branch to another
I presume that the following commit message is a benign case of version 1.1.1.1 being renamed on output to the user. But I'm a bit more worried about the message from checkout. What I did was to create a new file on a branch (so there is no 1.1 yet), and then I used the branch copy, v1.1.2.1, to make yet another branch copy, v1.1.4.1. $ cvs checkout -r bb -j 1.1.2.1 wrk/fyle.txt cvs server: warning: new-born wrk/fyle.txt has disappeared U wrk/fyle.txt $ cvs commit -m Deliver Task 1.1.2.1 work to stage wrk/fyle.txt Checking in wrk/fyle.txt; /home/cvsroot1/wrk/Attic/fyle.txt,v <-- fyle.txt new revision: 1.1.4.1; previous revision: 1.1 done -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
repeated merges from a branch
I have reached the point where I need to merge from a branch to main a second time with the same file. The documentation I have found dedicates just enough words to say use "cvs checkout -j once -j twice file" and it's the human's responsibility to guess the right value for "once". This strikes me as being a real nasty limitation. There is no meta data preserved to figure this out with? -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
to release or not to release
If I am reading the documentation correctly, I can forgo the release command if I change history logging in config to ignore checkout/release events. All I need to do after that seems to be to use rm -rf * in the work directory. to get the same impact as release. -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
update -l -r brnch dir doesn't seem to do anything
I've used CVS for a bit over 24 hours, having used ClearCase for the past 16 years, and I am trying to get a handle on how NOT to get a whole directory dumped into my work area. We have a directory with approximately 1,500 source files in it. (No unfortunat ely, I can't make a smaller directory, since this is only one of 475 directorie s I have to worry about). I have figured out that I can start with an empty work area and say $ cvs update -r brnch bmf/foo.x and just get one file moved into the directory. But, if I start with an empty work area and I want to create a new file in the big directory, I can't get CVS to give me an empty place to work. (At least the cvs add command refuses to work in anything other than a work directory blessed by CVS). I figured out that I can do an mkdir and cvs add, and cvs correctly notices that the new directory is redundant, but shouldn't "update -l -r brnch dir" work too? -CTH ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
cvs does not add hierachy files
Hello, I am a newbie. I added many files and directories in the original check out. Is there a simple way to add all of them? ¡°cvs add¡± does not add hierarchy files and directories. ¡°cvs import¡± will create unnecessary tags on other cvs files. I wish ¡°cvs add¡± has a ¡°-R¡± option. Charles
FW: How do I run cvs as server on Win2K?
Please ignore last post! I was looking at the wrong cvs product - as Anders indicated, cvsnt is what I'm looking for. Unfortunately, due to security I cannot download files with ".exe" extensions - is there an ftp site where I can download the below mentioned version from, or somewhere with a zip version so I don't have the security problem? Thanks again, Charles -Original Message- From: Anders Truelsen [mailto:[EMAIL PROTECTED]] Sent: Wednesday, August 14, 2002 1:59 AM To: Cummings, Charles (Debt Markets Technology); [EMAIL PROTECTED] Subject: RE: How do I run cvs as server on Win2K? Hi Charles If you want to run a cvs server on Win2K you should pick up the latest build (1.11.1.3 Build 57g) of CVSNT at http://www.cvsnt.org Regards, anders -Original Message----- From: Cummings, Charles (Debt Markets Technology) [mailto:[EMAIL PROTECTED]] Sent: 13. august 2002 23:46 To: '[EMAIL PROTECTED]' Subject: How do I run cvs as server on Win2K? Greetings - I've installed v. 1.10 on win2k successfully in that I can define modules, check in and out, etc., all from the command line. I'm trying to access my repository on the same computer using jcvs, but get the following: Opening server '2401@localhost' Failed to open '2401@localhost' 1. Is this possible (i.e. use jcvs to access cvs on a win2k computer locally and remotely)? 2. If it is possible, do I need to do anything specific to getting cvs to run as a server? The documentation discusses unix configuration, but I didn't notice anything specific to running a windows service? 3. How are settings such as port configured (default port implies to me that it can be changed if desired). I've tried the ip address as well as the computer's network name, all to no avail (same basic error message). Any suggestions greatly appreciated - will move everything to Linux or Solaris as soon as possible, but at moment am stuck with above mentioned environment. Charles Cummings ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
FW: How do I run cvs as server on Win2K?
Thanks, Anders - I've updated cvs.exe to 1.11.2 - I believe that's the latest. At any rate, I still get the behavior described below. Is there any configuration above and beyond what I did to get it to work from the command line? It seems to me that since there is a port and hostname, there should be a process running for the jcvs app to communicate. I haven't found anything in the documentation about that for Windows. Thanks, Charles -Original Message- From: Anders Truelsen [mailto:[EMAIL PROTECTED]] Sent: Wednesday, August 14, 2002 1:59 AM To: Cummings, Charles (Debt Markets Technology); [EMAIL PROTECTED] Subject: RE: How do I run cvs as server on Win2K? Hi Charles If you want to run a cvs server on Win2K you should pick up the latest build (1.11.1.3 Build 57g) of CVSNT at http://www.cvsnt.org Regards, anders -Original Message- From: Cummings, Charles (Debt Markets Technology) [mailto:[EMAIL PROTECTED]] Sent: 13. august 2002 23:46 To: '[EMAIL PROTECTED]' Subject: How do I run cvs as server on Win2K? Greetings - I've installed v. 1.10 on win2k successfully in that I can define modules, check in and out, etc., all from the command line. I'm trying to access my repository on the same computer using jcvs, but get the following: Opening server '2401@localhost' Failed to open '2401@localhost' 1. Is this possible (i.e. use jcvs to access cvs on a win2k computer locally and remotely)? 2. If it is possible, do I need to do anything specific to getting cvs to run as a server? The documentation discusses unix configuration, but I didn't notice anything specific to running a windows service? 3. How are settings such as port configured (default port implies to me that it can be changed if desired). I've tried the ip address as well as the computer's network name, all to no avail (same basic error message). Any suggestions greatly appreciated - will move everything to Linux or Solaris as soon as possible, but at moment am stuck with above mentioned environment. Charles Cummings ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
How do I run cvs as server on Win2K?
Greetings - I've installed v. 1.10 on win2k successfully in that I can define modules, check in and out, etc., all from the command line. I'm trying to access my repository on the same computer using jcvs, but get the following: Opening server '2401@localhost' Failed to open '2401@localhost' 1. Is this possible (i.e. use jcvs to access cvs on a win2k computer locally and remotely)? 2. If it is possible, do I need to do anything specific to getting cvs to run as a server? The documentation discusses unix configuration, but I didn't notice anything specific to running a windows service? 3. How are settings such as port configured (default port implies to me that it can be changed if desired). I've tried the ip address as well as the computer's network name, all to no avail (same basic error message). Any suggestions greatly appreciated - will move everything to Linux or Solaris as soon as possible, but at moment am stuck with above mentioned environment. Charles Cummings ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS vs. Perforce
Hello, I was hoping to get some opinions from people on this list. Currently, I'm looking into getting a new version control system. I have 23 users on Windows and Macintosh and a few users are on Linux. Microsoft Visual SourceSafe 5.0 has almost become unusable for us. It becomes corrupted almost once every week or two! Because we need a Macintosh CodeWarrior IDE plug-in I think that CVS and Perforce are really the only ones for me to look at. The Windows clients need a Visual C++ 6.0 plug-in. It would be great to have stand-along GUIs on both. Right now I'm leaning a little bit towards Perforce. Perforce has a number of features that CVS doesn't seem to have. I might be wrong. Some of these features include atomic check-ins (related changes are checked in as a single atomic transaction), change lists (checkouts/checkins are grouped and can be tracked), better branching and merging (so I've been told), and official tech support. I have also heard a lot of good things about CVS. Also, several people that I have talked with have recommended the Subversion Project (http://subversion.tigris.org/). A big difference between CVS and Perforce is cost. CVS is free and Perforce will cost more $12,000 to license. Does anyone have opinions to share regarding this decision? Some Perforce users have told me that it's worth the money, but I really don't know how much experience they have with CVS. Thanks for any help, Chuck ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS problem
I am having problems connecting to a CVS server I am attempting to configure. I believe I have setup the server per all the Documentation I can find but I recieve the following message when I attempt to connect with WinCVS. cvs [login aborted]: connect to xxx.xxx.xxx.xxx:2401 failed: Connection refused I am using Redhat 7.2 with the default install version of CVS. 2401 is listed in my services file. I have defined CVSROOT in WinCVS as: :pserver:[EMAIL PROTECTED]:/home/cvsroot Anyone have any suggestions? __ Do You Yahoo!? Yahoo! Movies - coverage of the 74th Academy Awards® http://movies.yahoo.com/ ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Newbe CVS install question with WINCVS
I have installed CVS on a suse 7.1 Linux configuration. CVS seems to run fine in native mode. I am trying to use WINCVS 1.2 for connection from an Win2000 machine. I have the cvspserver line in inetd.conf as per the documentation. I have set up a passwd file and used htpasswd to create encrypted passwords. No matter what I try I get *CVS exited normally with code 1* And nothing happens It seems like I've tried all I can think of. Any help ? Any thing else ? Thanks in advance. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re: Why is revision on 1st import 1.1.1.1 instead of 1.1?
On Sun, 24 Jun 2001 [EMAIL PROTECTED] wrote: > > I have a question about a CVS behaviour that makes me wonder for a > long time. When I import some files with the following command: > > $ cvs import -m "1st import" some/path/in/repository mygroup start > > their revision numbers are all 1.1.1.1 - after a change+commit they > become 1.2, but why aren't they they 1.1 before? Is there a way to > make them 1.1 immediately on import? > > Any hints appreciated! > Hi, Really short answer: This is normal behavior. Short answer: In general, revision numbers are internally used by cvs. Don't stress about them. In particular, revision 1.1.1.1 (and everything on the 1.1.1 branch) has a special meaning to cvs. A revision 1.1 does exist in your files, and it's the same as 1.1.1.1. You do not want to try to change this. Long answer: The 1.1.1 branch is used for the vendor branch feature of cvs. (Ever wondered why you need the mygroup and start args to import?) Please see the documentation on vendor branches for more information. (Hmmm... why is the short answer longer? :-) HTH, Charles ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS and AFS
[Sorry my previous send of this message got switch to base64 encdoing becuase he body of the message contained a character. Here it is again... CK] We use CVS in a mixed Windows and Linux environment. Recently we switched from accessing the CVS repository vis ssh to a Linux machine on which CVSROOT was a local disk to having the CVS repository in AFS and having all clients access this as a local file system. The motivation for this change was * to simplify user authentication; * to allow finer grain control over access to the repository. While things are basically working OK, we are running into some problems as I'd be interested in hearing other peoples' experiences. (1) Windows does loose matching on the case of filenames. On one occasion this has led to the filenames in the repository been "spontaneously" changed. The way this happened was (I think) * On a Linux system, I did 'cvs remove NOTES.DOC', followed by 'cvs add Notes.doc', i.e., a 'safe' rename. * On an NT system (using WinCVS 1.06), a user modifies NOTES.DOC (which he still has checked out) and CVS apparently matches this with Notes.doc in the repository and alters the repository name to NOTES.doc when he checks this in. * I do a 'cvs update' and am surprised to see NOTES.DOC resurrected. It would seem that the Windows versions of CVS should try REALLY hard not to do this. (2) If the CVS repository is an AFS and if access to the repository is granted by virtue of an AFS token in the name of NAME@CELL, this name should be used in the log messages instead of the possibly random Windows (or Unix) user name. Lots of the Windows checkins are by 'Administrator' which is not very informative. (3) If an NT user checked out administrative files in $CVSROOT/CVSROOT, modifies one of them and checks it back in the latest release in CVSROOT carries DOS conventions. When accessing the repository from Linux, I then get cvs log: syntax error in /afs/sarnoff.com/common/cvs/CVSROOT/config: line '^M' is missing '=' Either the Unix clients should care about foreign newline conventions in $CVSROOT/CVSROOT/config, or the Windows clients should ensure that these files have Unix newline conventions. (4) It's too bad that AFS's notion of machine-independent pathname doesn't carry through to Windows. Having to set up drive letter e.g., Z: pointing to /afs/CELL/ breaks this notion and is so retro. I'd like to be able to use :local:\\afs\CELL\common\cvs instead of :local:z:\common\cvs (5) Finally, we love to have network-level encryption of the data in AFS files. Is this on anyones radar screen? -- Charles Karney Email: [EMAIL PROTECTED] Sarnoff Corporation Phone: +1 609 734 2312 Princeton, NJ 08543-5300Fax:+1 609 734 2586 ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS and AFS
We use CVS in a mixed Windows and Linux environment. Recently we switched from accessing the CVS repository vis ssh to a Linux machine on which CVSROOT was a local disk to having the CVS repository in AFS and having all clients access this as a local file system. The motivation for this change was * to simplify user authentication; * to allow finer grain control over access to the repository. While things are basically working OK, we are running into some problems as I'd be interested in hearing other peoples' experiences. (1) Windows does loose matching on the case of filenames. On one occasion this has led to the filenames in the repository been "spontaneously" changed. The way this happened was (I think) * On a Linux system, I did 'cvs remove NOTES.DOC', followed by 'cvs add Notes.doc', i.e., a 'safe' rename. * On an NT system (using WinCVS 1.06), a user modifies NOTES.DOC (which he still has checked out) and CVS apparently matches this with Notes.doc in the repository and alters the repository name to NOTES.doc when he checks this in. * I do a 'cvs update' and am surprised to see NOTES.DOC resurrected. It would seem that the Windows versions of CVS should try REALLY hard not to do this. (2) If the CVS repository is an AFS and if access to the repository is granted by virtue of an AFS token in the name of NAME@CELL, this name should be used in the log messages instead of the possibly random Windows (or Unix) user name. Lots of the Windows checkins are by 'Administrator' which is not very informative. (3) If an NT user checked out administrative files in $CVSROOT/CVSROOT, modifies one of them and checks it back in the latest release in CVSROOT carries DOS conventions. When accessing the repository from Linux, I then get cvs log: syntax error in /afs/sarnoff.com/common/cvs/CVSROOT/config: line ' ' is missing '=' Either the Unix clients should care about foreign newline conventions in $CVSROOT/CVSROOT/config, or the Windows clients should ensure that these files have Unix newline conventions. (4) It's too bad that AFS's notion of machine-independent pathname doesn't carry through to Windows. Having to set up drive letter e.g., Z: pointing to /afs/CELL/ breaks this notion and is so retro. I'd like to be able to use :local:\\afs\CELL\common\cvs instead of :local:z:\common\cvs (5) Finally, we love to have network-level encryption of the data in AFS files. Is this on anyones radar screen? -- Charles Karney Email: [EMAIL PROTECTED] Sarnoff Corporation Phone: +1 609 734 2312 Princeton, NJ 08543-5300Fax:+1 609 734 2586"wèrû&j)b b²Ò'~/² !¶Úþf¢ ?¨¥©ÿ+-wèþ)ß¡Ëì
Putting Gif Files into CVS
Hello
I am adminstrating my first CVS repository and I need help with gif
files.
My environment is a CVS Repository on a Sun machine where NT clients are
accessing the Repository.
The procedure I have been using is:
1. import the files
2. find . -name "*.gif" -exec cvs admin -kb {} \;
3. find . -name "*.gif" -exec cvs update {} \;
This procedure has worked for MSWord documents, Power Point Files and
Jar Files; however it has not worked for gif files.
How do I put a gif file into CVS ???
Thank you for any help you provide.
___
Info-cvs mailing list
[EMAIL PROTECTED]
http://mail.gnu.org/mailman/listinfo/info-cvs
Remove me
Remove me Please
How to browse the CVS repository?
dear list, A silly question: Is there any way to browse the files of CVS repository without checking them out? My team members want to see what's there in the repository before checking them out. Thanks! Charles ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Is it possible to merge together two files that CVS does not recoginize as the same?
Hey all, I would appreciate any input you might have on how best to accomplish the following: Our project is in a transitional state and a file has been copied from one module to another with some concurrent development taking place on both revisions of that file. Now, the changes that were made to one copy need to be joined with those made in the other copy in a different module. So: module1/fileX needs to be merged with module2/folder/fileX The only way I can think of to attempt this is: cp module1/fileX module2/folder/fileX cvs ci fileX and then merging the differences between the two revisions so created. Any idea if this will work? Any suggestions for how else to accomplish this? I would imagine that having a file editor that can diff and merge two files would be the best way to do it, and we are going to try that approach, but I am curious now as to whether this can be accomplished using CVS. Ch. ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Checking out a "subset" of a CVS module
Greetings. We are setting up CVS to track revisions of scripts. Nothing really complicated. I understand how CVS imports modules that can be collections of directories and files, and that when you check out a module CVS sends a working copy of everything in the module. Just the same, we were wondering if it is possible to check out a "subset" of the module. For example, if a module contains 200 scripts, can we select just one of the scripts to checkout? Can we check out just one script that may be contained in a directory that is several directories deep in a module? Thanks, Charles ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
CVS pserver
I'm new to this and am attempting to set up a pserver on a linux box which will service multiple windows machines running wincvs. I have thus far installed cvs and created a repository. I'm not sure how to actually get pserver listening on the port I'm using. Any help on what command line needs to be added to what file would be greatly appreciated. From reading I found that the line goes to the inetd.conf file but I can't find it on RH7.0. Thanks Charles
Windows GUI fro CVS
I'm looking for a good GUI for windows machines to connect to a Linux box with CVS, any recommendations? Thanks Charles
tag limits?
Hello, I've recently read Martin Fowler's paper on continouous integration (http://www.martinfowler.com/articles/continuousIntegration.html). This paper seems to advocate tagging or "labeling" each build with a build number regardless of whether is is any kind of milestone, release or whatever. The paper also indicates that builds may happen as often as every 10-15 minutes or so. This could result in many, many tags or "labels" being assigned to a given version of a file; could be dozens, or hundreds. Can CVS support this? Is there any limit on tags? Any comments on the whole concept in the context of CVS? Regards, Chuck ___ Info-cvs mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/info-cvs
Re:CVS/ANT/Win2K
Oops - I didn't put in that "file:" stuff, my mail client did The actual paths I tried are "//Beta/Projects@C/CVS" and ":local://Beta/Projects@C/CVS" Original message follows: >Hi,>I am trying to execute CVS from ANT under Win2000. My problem is to make it>recognize the repository.>If I specify "-d file://Beta/Projects@C/CVS" (a UNC path) the message is>"file://Beta/Projects@C/CVS must be an absolute pathname">Putting :local: in front of the path doesn't help either. ANT/CVS complains>"cvs.exe checkout: missing hostname in CVSROOT:>:local://Beta/Projects@C/CVS>cvs.exe [checkout aborted]: Bad CVSROOT.">I have tried various combinations of forward & backslashes as well as some>bad language, but nothing seems to work.>Can someone help me please? Charles de VilliersSalesBid Ltd. Charles de VilliersSalesBid Ltd.
CVS/ANT/Win2K
Hi,I am trying to execute CVS from ANT under Win2000. My problem is to make itrecognize the repository.If I specify "-d file://Beta/Projects@C/CVS" (a UNC path) the message is"file://Beta/Projects@C/CVS must be an absolute pathname"Putting :local: in front of the path doesn't help either. ANT/CVS complains"cvs.exe checkout: missing hostname in CVSROOT::local://Beta/Projects@C/CVScvs.exe [checkout aborted]: Bad CVSROOT."I have tried various combinations of forward & backslashes as well as somebad language, but nothing seems to work.Can someone help me please? Charles de VilliersSalesBid Ltd.
unsubscribe
cause the huge amount of mails, i consider to substitute for another email account. thanks. _ һ·ÉÏÓÐÄ㣬¿àÒ»µãÒ²Ô¸Ò⣡ --ΪÄãϲ°®ÓëÖ§³ÖµÄ263Ê׶¼ÔÚÏߣ¨http://www.263.net£©Í¶Ò»Æ±£¡ ÎÒҪͶƱ£¡£¨http://fsurvey.cnnic.net.cn/survey/index.html£©
