>It doesn't really make sense that cvs would work where cvs-checkout would >fail, 
>since I assume cvs-checkout is just cvs with a defaulted command >attribute--but 
>that's what's happening here.

No it doesn't.  I sent an email asking for more information on the <exec/> run and the 
<cvs-checkout/> run.  I switched isp's so I am not sure if the email go through.  What 
I would appreciate is if you could confirm that both tasks are using the same version 
of cvs.  You can do this by passing in the debug switch to NAnt and looking through 
the output (you may need to pipe to a file, or record the output).

>This worked, but cvs.exe does not allow the password to be specified as an >argument. 
> However, since CVS was already setup on my machine, it was finding >the password 
>somewhere.

The password is found in a file called .cvspass, the cvsbook variable section is a 
good recourse on how this (should be) resolved: 
http://cvsbook.red-bean.com/cvsbook.html#Environment%20Variables.  You can also 
specify the location using the passfile attribute in the cvs tasks.



Cheers,


Clayton


-----Original Message-----
From:   Merrill Cornish [mailto:[EMAIL PROTECTED]
Sent:   Mon 7/19/2004 10:50 AM
To:     [EMAIL PROTECTED]
Cc:     
Subject:        [Nant-users] cvs-checkout fails; cvs works
Last week I reported that the cvs-checkout task (using CVSNT in place of the Sharp 
CVS) failed in that while the checkout itself worked fine, the top-level CVS\Entries 
file was missing some of the subdirectory names.  As a result, the cvs tree checked 
out by cvs-checkout could not be updated.

I initially worked around the problem by using the exec task to directly execute 
cvs.exe.  This worked, but cvs.exe does not allow the password to be specified as an 
argument.  However, since CVS was already setup on my machine, it was finding the 
password somewhere.  Nevertheless, I wanted the NAnt script to be self-contained.

So--this morning I tried using the basic cvs task itself.  THIS time, not only does 
the checkout work, but CVS\Entries is correct, so the checked out tree is updateable.

It doesn't really make sense that cvs would work where cvs-checkout would fail, since 
I assume cvs-checkout is just cvs with a defaulted command attribute--but that's 
what's happening here.


Merrill


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=4721&alloc_id=10040&op=click
_______________________________________________
Nant-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/nant-users





-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idG21&alloc_id040&op=click
_______________________________________________
Nant-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/nant-users

Reply via email to