[Logo] JCVSForum - Community Support For JCVS Users and Developers
  [Search] Search   [Recent Topics] Recent Topics   [Hottest Topics] Hottest Topics   [Members]  Member Listing   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
Messages posted by: time
Forum Index » Profile for time » Messages posted by time
Author Message
The chacl command is not currently specifically implemented in jCVS II, the application. However, it should be straightfoward to implement the command if you are simply using the cvs package for your custom client.
jCVS is currently under the GPL. The jCVS library is LGPL. The source should be properly labeled.

I believe the problem here is the use of backslashes. Please try replacing the '\\' backslashes with '/' forward slashes.

Well, in theory, this request is only different from normal requests in that is does not involve a working directory. So you would just build up the CVSRequest in all the same ways, except you would not have to send any "CVSEntrys" to the server, and you would not expect to have to process anything related to CVSEntry's in the response.

So, I suppose that I would start with the "checkout" command, which has the perfect little "universe" defined in the CheckOutPanel code in JCVS II. The code is completely self-contained in that class I believe. Since the checkout command does not have a working directory to begin with, it does not require one to send the request, and there are not CVSEntry's to send either. Of course, the command will in turn process the entries being sent down as it builds the working directory, which is different than your case.

You will probably need to define a CVSUserInterface subclass to provide to capture the output, since you are not looking to get files, but stdout and stderr I think, but I am not sure about this since I have not looked at the command in question.

Between those two classes, and the examples (check out the JavaScripts in the latest download, which are insightful), I think you can put together enough code to get started, then we can deal with what you discover.
If I get a chance (which is a serious longshot), I will take a stab at this, since it appears to be desired by several, and I need to add a few patches. Ping me in a week if you remember.
Thank you very much for your followup on this problem!
When you want to display the "raw content" of a document, you need to click on the little icon to the left of the document's name in the index page. This will deliver the document's data "as-is", and should include the ".doc" suffix needed to get the browser to properly handle the download.
From what I can remember off the top of my head, there is currently no "implementation" of the rlog command. However, to implement such a command would seem a trivial effort, since it appears to be a "non-working-directory" command. In other words, it does not require you to checkout a working directory, and so does not require setting up all of that information to run the command.

Sadly, at the moment my job is demanding all of my time, so I do not have the time to make the mods. If you are interested in trying to write the code, I would be very happy to help you and answer any questions you might have. Otherwise, if you ping me in a month or so, I may have the time to knock this one out.
Must apologize one more time for the downtime. Host move was botched by myself, but I think we are now fully recovered.

I was unaware that CVS had provided "creating users" as a feature. In the past, all CVS users were actual logins. Thus, CVS was not able to create users. I realize that CVS has since added its own concept of users in CVSROOT files, but I was unaware that a command has been created for the automatic management of those files.
JCVSWeb, and the jCVS package in general, does not directly access CVS repositories. JCVSWeb requires client/server protocol access to a CVS server that serves up the repository in question.

There is still inadequate documentation for jCVSWeb. My apologies.
This appears to be strictly an issue with file/directory permissions on the server. Since jCVS does nothing more than log-in using the information you provide, I would first ask if there is any difference between the logins? If not, then the other other possible difference is that jCVS is performing an operation (such as watches) that tortoise is not performing. In either case, this is definitely a server issue and one of permissions. I would look at the CVS server file in question.
The problem is not immediately obvious to me.
I will look some more when I get a chance.

jCVS is a Java-based package that allows one to communicate with a CVS server. CVS is a "Concurrent Versioning System", which allows developers to manage the versions of their source or other files. jCVS provides a User Application that implements a CVS "client" with a simple GUI interface. jCVSWeb provides a webapp that allows the presentation of an CVS repository from anywhere on the internet. The com.ice.cvsc package provides an API for developers to implements other CVS client implementations.

Thank you very much for this report.

CVS would not leave itself with an "unresolvable" condition such as the one you describe. My point is that this must be a bug in the assumptions that jCVS is making, not in the protocol itself.

I will look into this as a bug in jCVS.


Forum Index » Profile for time » Messages posted by time
Go to:   
Powered by JForum 2.1.9 © JForum Team