You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@subversion.apache.org by David Bartmess <di...@edingo.net> on 2010/09/21 19:08:54 UTC

Problem with first revision of file being checked in, then deployed

 I'm automating the deployment of our software, and came across a problem.

In order to only pull the files changed for a specific revision, I do a 
"diff -r<PriorRev>:<CurrRev> --summarize". I then parse the output 
through sed to get just the filename.

All this works normally, but if the PriorRev of the file doesn't exist 
(it was added in the CurrRev revision), it gives me an error (below) and 
doesn't return anything. My question is, is there anything I can do 
besides having to parse the output from the command for that error 
message? I don't want to have to pull ALL the files in revision CurrRev, 
because some of these are SQL scripts and can't be re-run again, if they 
were from a prior revision.

svn: Unable to find repository location for 
'https://subversion.assemb<snip>' in revision 40851

-- 
"Dingo" Dave Bartmess
Broomfield, CO. USA
http://edingo.net


Re: Problem with first revision of file being checked in, then deployed

Posted by David Bartmess <di...@edingo.net>.
 On 9/21/2010 1:22 PM, Bostjan Skufca wrote:
> For database up/downgrades you should look into dbdeploy.
>
> Also, my svn client (latest) nicely shows all files modified AND
> added. The letter M or A in front of filename signifies it.
>
> b.
>
>
>
> On 21 September 2010 21:08, David Bartmess<di...@edingo.net>  wrote:
>>   I'm automating the deployment of our software, and came across a problem.
>>
>> In order to only pull the files changed for a specific revision, I do a
>> "diff -r<PriorRev>:<CurrRev>  --summarize". I then parse the output through
>> sed to get just the filename.
>>
>> All this works normally, but if the PriorRev of the file doesn't exist (it
>> was added in the CurrRev revision), it gives me an error (below) and doesn't
>> return anything. My question is, is there anything I can do besides having
>> to parse the output from the command for that error message? I don't want to
>> have to pull ALL the files in revision CurrRev, because some of these are
>> SQL scripts and can't be re-run again, if they were from a prior revision.
>>
>> svn: Unable to find repository location for
>> 'https://subversion.assemb<snip>' in revision 40851
>>
I think the problem comes from the fact that the directory holding the 
file was added at the same time as the file itself, so SVN doesn't see 
that directory until after an update or export is done.

-- 
"Dingo" Dave Bartmess
Broomfield, CO. USA
http://edingo.net


Re: Problem with first revision of file being checked in, then deployed

Posted by Bostjan Skufca <bo...@a2o.si>.
For database up/downgrades you should look into dbdeploy.

Also, my svn client (latest) nicely shows all files modified AND
added. The letter M or A in front of filename signifies it.

b.



On 21 September 2010 21:08, David Bartmess <di...@edingo.net> wrote:
>  I'm automating the deployment of our software, and came across a problem.
>
> In order to only pull the files changed for a specific revision, I do a
> "diff -r<PriorRev>:<CurrRev> --summarize". I then parse the output through
> sed to get just the filename.
>
> All this works normally, but if the PriorRev of the file doesn't exist (it
> was added in the CurrRev revision), it gives me an error (below) and doesn't
> return anything. My question is, is there anything I can do besides having
> to parse the output from the command for that error message? I don't want to
> have to pull ALL the files in revision CurrRev, because some of these are
> SQL scripts and can't be re-run again, if they were from a prior revision.
>
> svn: Unable to find repository location for
> 'https://subversion.assemb<snip>' in revision 40851
>
> --
> "Dingo" Dave Bartmess
> Broomfield, CO. USA
> http://edingo.net
>
>
>

Re: Problem with first revision of file being checked in, then deployed

Posted by Ryan Schmidt <su...@ryandesign.com>.
On Sep 21, 2010, at 14:08, David Bartmess wrote:

> I'm automating the deployment of our software, and came across a problem.
> 
> In order to only pull the files changed for a specific revision, I do a "diff -r<PriorRev>:<CurrRev> --summarize". I then parse the output through sed to get just the filename.
> 
> All this works normally, but if the PriorRev of the file doesn't exist (it was added in the CurrRev revision), it gives me an error (below) and doesn't return anything. My question is, is there anything I can do besides having to parse the output from the command for that error message? I don't want to have to pull ALL the files in revision CurrRev, because some of these are SQL scripts and can't be re-run again, if they were from a prior revision.
> 
> svn: Unable to find repository location for 'https://subversion.assemb<snip>' in revision 40851

diff summarize works fine for me in Subversion 1.6.12 even when the file doesn't exist in the prior rev. Are you using the latest Subversion?

Another possibility would be to use "svn log -q -v -r<PriorRev>:<CurrRev>" instead of diff summarize.