[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn update error !!

From: Himanshu Raina <raina_himanshu_at_yahoo.com>
Date: Thu, 16 Jul 2009 12:16:01 +0530 (IST)

Guess I wasn't clear with the requirement. Anyways I got it working with svn update --force. The requirement is explained  below

I have a working copy of my repo on my webserver . This working copy contains both versioned and unversioned projects (I can't commit all, as some of these projects aren't active basis subscription but are still needed). Tomorrow if I want to make changes to one of these unversioned projects (say if the customer wants to get some content changed) I'll copy/move this code to my SVN and commit it. Now when I used to run svn update on my working copy it used to fail as it already had a version copy. In order to fix this I used to remove the unversioned copy from my wc and ideally on running svn update it should have checked out the versioned copy of that project. The problem started here as on running svn update for the second time the update used to hang or used to give error code 400 etc..

Hope this clarifies everything. Initially I thought it could be because I am using capistrano for release management but even without it (i.e by manually running svn update on working copy I got this error).

Regards
Himanshu Raina

--- On Thu, 16/7/09, Ryan Schmidt <subversion-2009b_at_ryandesign.com> wrote:

From: Ryan Schmidt <subversion-2009b_at_ryandesign.com>
Subject: Re: svn update error !!
To: "Himanshu Raina" <raina_himanshu_at_yahoo.com>
Cc: users_at_subversion.tigris.org
Date: Thursday, 16 July, 2009, 5:49 AM

On Jul 15, 2009, at 03:22, Himanshu Raina wrote:

> Since the data size is couple of GB's we are using checkout as 
> opposed to svn export. The issue I'm facing is given below
>
> When I run svn update on my working copy in case a project of the 
> same name already exists the update stops. Later when I remove the 
> conflicting project (i.e rm project) and run svn update keeps 
> running i.e. it doesn't return to the prompt. I tried cleanup and 
> ran svn update again but no result.
>
> Morever what I have observed is usually svn update on working copy 
> gives the following output
>
> [designer_at_myserver current]$ svn status
> ?      himanshu
>
> Now suppose I commit project1 to my repo and run svn update on my wc
>
> svn: Failed to add directory 'himanshu': object of the same name 
> already exists
> command finished
>
> On working copy the svn status gives
> [designer_at_myserver current]$ svn status
> ?      himanshu
> !      .    (!  means item is missing (removed by non-svn command) 
> or incomplete).
>
>
> Now how do I get rid of this??

Like Jochen, I'm unclear on exactly what commands you ran, what 
happened, what you expected, or what you were actually trying to do. 
But taking a guess: is it perhaps that you have a working copy of 
your entire repository, and to create a new project, you create (with 
your OS, not with Subversion) a new directory and start adding files 
to it, and when you want to import it into Subversion you use "svn 
import", and then you find when you "svn update" your working copy 
that it complains that the project's directory already exists?

If so, there are a number of issues going on. Foremost is that "svn 
import" imports an unversioned directory into the repository. It does 
not transform that unversioned directory into a working copy, and the 
intended workflow is that you then remove the directory and check it 
out from the repository, or in your case, update the existing whole-
repository working copy that you already have. You may find it easier 
to do a so-called "in-place import" instead of using "svn import". 
The "in-place import" does transform an unversioned directory into a 
working copy and you just continue using it, rather than having to 
throw it away and get a new one, and is described here:

http://subversion.tigris.org/faq.html#in-place-import

Finally, it is unusual to have a working copy of the entire 
repository, because most repositories contain not only the trunk of a 
project (or of multiple projects), but also branches and tags, and 
there's almost no reason why you would want to check out the trunk 
and all branches and all tags. Instead, it's usual to check out a 
working copy of just the trunk or branch or tag of just the project 
you're interested in.

If you want to continue using the single monolithic working copy of 
everything, then you don't need "svn import" or the "in-place import" 
to add a new directory; all you need to do is "svn add" and "svn 
commit" it.

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2371648

To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].

      Love Cricket? Check out live scores, photos, video highlights and more. Click here http://cricket.yahoo.com

------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=1065&dsMessageId=2371678

To unsubscribe from this discussion, e-mail: [users-unsubscribe_at_subversion.tigris.org].
Received on 2009-07-16 08:47:31 CEST

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.