Re: Can't identify parent of branch

John McNamee <[email protected]>
Newsgroups gmane.comp.version-control.subversion.cvs2svn.user
Message-ID <[email protected]>
That makes sense, but the same pattern (file created on branch) exists in
thousands of other files, and cvs2svn selects the trunk as the parent.

If this were merely cosmetic, I wouldn't care.  The operational problem I'm
having has to do with symbol hints.  My conversion project involves taking
a single large CVS repository and breaking it up into ~50 separate SVN
repositories.  I've got a "master" symbol hints file that I want to use
for these separate conversions.  The hints file was created based on a
run through the entire CVS repository, and then some hand editing was
applied.  The entry for the "symbol00904" branch lists the parent as the
trunk.  cvs2svn generates an error if the symbol hints declare a parent
that isn't one of the "possible parents".


-------- Original Message --------
Subject: Re: Can't identify parent of branch
From: Michael Haggerty <[email protected]>
To: John McNamee <[email protected]>
Date: 11/8/2010 10:41 PM

> On 11/08/2010 09:55 PM, John McNamee wrote:
>>  I've got a small number of files that confuse cvs2svn, and I can't
>>  figure out why.  The failing files seem more-or-less the same as tens
>>  of thousands of other files in the repository that convert correctly.
>>  The problem is the cvs2svn can't determine the parent of a branch.
>>
>>  I found a relatively simple file that demonstrates the problem,
>>  which I'm attaching to this message (file0001,v).  cvs2svn can't
>>  identify a parent for symbol00904 (the actual parent is the trunk).
>>
>>  I'm using cvs2svn version 2.3.0.
>>
>>  cvs2svn --write-symbol-info=symbol-info.txt -v -v --dry-run /path/to/cvsroot
>>
>>  Any ideas?
>
> Please notice that the 1.1 revision of the file is "dead".  This means
> that the file wasn't present on trunk when symbol00904 was originally
> created, then was later added on the branch.  Later it was added to
> trunk (with version 1.2).
>
> Since there *was* no revision 1.1, it is not possible for trunk to be
> the parent of symbol00904.  The very first appearance of the file is on
> symbol00904.
>
> Do you see something incorrect in the repository output by cvs2svn, or
> are you just concerned that no "possible parents" are listed for
> symbol00904 in symbol-info.txt?
>
> Michael

------------------------------------------------------
http://cvs2svn.tigris.org/ds/viewMessage.do?dsForumId=1670&dsMessageId=2680444

To unsubscribe from this discussion, e-mail: [[email protected]].
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.