Looking through the documentation I think we are hitting this problem:

"This file should then be selected for the "Topic text upload" field. The 
"Start tag", "End tag" and "Module" fields have no relevance in this situation. 
Because of an issue in Subversion, running the svn diff command outside of the 
repository root will not allow Codestriker to download and review the full file 
(although the patch segments can still be reviewed). If you commit your code 
first, then let Codestriker make the diff, as explained in the next section, 
then you can avoid this limitation."

I'll check and see how the problematic Topics were created.

Chris

-----Original Message-----
From: Chris Welch 
Sent: Wednesday, February 29, 2012 11:34 AM
To: 'codestriker-user@lists.sourceforge.net'
Subject: RE: [Codestriker-user] Parallel view bug using SVN, get uninitialized 
value error

I tried the code change and the Parallel display still comes up empty.  Do we 
have to recreate the Topic for this code change to have effect (I.e. the change 
is used during Topic creation, not dynamically during Topic display)?

Tx,

Chris

From: Chris Welch 
Sent: Wednesday, February 29, 2012 9:44 AM
To: codestriker-user@lists.sourceforge.net
Subject: RE: [Codestriker-user] Parallel view bug using SVN, get uninitialized 
value error

Great, tx gentlemen I'll look into both of these suggestions.  I suspect the 
different branch is the most likely cause as many of our projects make heavy 
use of externals.

Chris

From: David Carson [mailto:dccar...@gmail.com] 
Sent: Wednesday, February 29, 2012 8:29 AM
To: David Sitsky
Cc: Chris Welch; codestriker-user@lists.sourceforge.net
Subject: Re: [Codestriker-user] Parallel view bug using SVN, get uninitialized 
value error

This couldn't have to do with 'highlight' being defined, but no highlight 
utility being installed, could it?  Seems like when I hit that particular 
error, it showed up as data not being retrieved as expected.
On Wed, Feb 29, 2012 at 12:42 AM, David Sitsky <david.sit...@gmail.com> wrote:
Hi Chris,

That error message in the logs basically indicates that no data was
retrieved from the subversion server for some reason.  Normally the
logs would contain some information about it.

I've sometimes seen this issue happen in the past where the revisions
were created on different branches that what the repository is based
off.

In Subversion.pm:
http://codestriker.cvs.sourceforge.net/viewvc/codestriker/codestriker/lib/Codestriker/Repository/Subversion.pm?revision=1.29&view=markup
around line 77, you could try replacing:

push @args, @{ $self->{userCmdLine} };
push @args, $self->{repository_url} . '/' . $filename . '@' . $revision;

with

push @args, '-r';
push @args, $revision;
push @args, @{ $self->{userCmdLine} };
push @args, $self->{repository_url} . '/' . $filename;

to see if it makes any difference.

There was some semantic differences with how subversion treats @r
retrievals vs -r retrievals.  A long time ago, somebody made the
change to use @r retrievals

Cheers,
David

On Wed, Feb 29, 2012 at 4:10 AM, Chris Welch <chris.we...@jdsu.com> wrote:
> Using Codestriker 1.9.10 with SVN 1.6.  Patches are typically created using 
> TortoiseSVN 1.6.2 (right click, generate patch).
>
> Most of the time this works but we have some Topics that the full file 
> content is blank when you do a Parallel view.  This failure occurs regardless 
> of browser.
>
> The Apache log shows the following error for this request:
>
> [Tue Feb 28 11:27:37 2012] [error] [client 10.13.36.124] [Tue Feb 28 11:27:37 
> 2012] codestriker.pl: Use of uninitialized value in concatenation (.) or 
> string at /var/www/codestriker/lib/Codestriker/Action/ViewTopicFile.pm line 
> 113., referer: 
> http://aontools/codestriker/codestriker.pl?action=view&topic=2687633&mode=2&fview=0
>
> The code in question is:
>
>        # Output those lines leading up to the start of the next delta.
>        # Build up a delta with no changes, and render it.
>        my $next_delta_linenumber = $delta->{old_linenumber};
>        for (my $i = $old_linenumber; $i < $next_delta_linenumber; $i++) {
>            $delta_text .= " $filedata[$i]\n";
>            $old_linenumber++;
>
> with the failure occurring at the delta_text concatenation (I.e. 
> $filedata[$i] is uninitialized).
>
> Any ideas on how to fix or work around this?
>
> Tx,
>
> Chris Welch
>
>
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> Codestriker-user mailing list
> Codestriker-user@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/codestriker-user
------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Codestriker-user mailing list
Codestriker-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/codestriker-user


------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Codestriker-user mailing list
Codestriker-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/codestriker-user

Reply via email to