Are you sure you are specifying a REPOSITORY and not a working copy???

BOb


From: David Goldsmith [mailto:eulergaussriem...@gmail.com]
Sent: Thursday, October 31, 2013 4:17 PM
To: Bob Archer
Cc: users@subversion.apache.org
Subject: Re: Windows file:/// URL format for svnsync

OK, now we're going in circles: what you cut and paste is what I've been 
trying, over and over again, to no avail (as I said in my OP, it didn't like 
the way I'm specifying my source folder.)  I'm sure there's some subtlety 
that--as a user, not an admin--I'm missing, that those overly-simplified 
instructions aren't including, but I can't seem to communicate what it is.  
I'll try posting a transcript of my efforts.
DG

YOU!...are Big Data<http://en.wikipedia.org/wiki/Big_data>.

On Thu, Oct 31, 2013 at 1:08 PM, Bob Archer 
<bob.arc...@amsi.com<mailto:bob.arc...@amsi.com>> wrote:
Not sure how we got off list. But, I just used svnsync to move a repo to google 
code.. it was pretty easy. Just follow the instructions on their wiki page:

How do I import an existing Subversion repository?
To upload the history of an existing Subversion repository, use the svnsync 
tool that ships with Subversion 1.4 or later. Run svnsync help to read more 
about this tool.
Note that your Google Code repository must be reset to revision 
0<https://code.google.com/p/support/source/detail?r=0> for this to work. Your 
project's Source tab will display instructions on how to reset the repository 
yourself. (Note: you must be a project owner to reset your own repository, and 
also to push code up with svnsync.)
Here's a sample transcript that demonstrates how you can push history from an 
existing repository (located at file:///my/repos<file:///\\my\repos>) to your 
repository on Google Code:
  $ svnsync init --username YOURUSERNAME 
https://YOURPROJECT.googlecode.com/svn<http://YOURPROJECT.googlecode.com/svn> 
file:///path/to/localrepos<file:///\\path\to\localrepos>
  Copied properties for revision 0.
  $ svnsync sync --username YOURUSERNAME 
https://YOURPROJECT.googlecode.com/svn<http://YOURPROJECT.googlecode.com/svn>
  Committed revision 1.
  Copied properties for revision 1.
  Committed revision 2.
  Copied properties for revision 2.
  [...]
When prompted for your password, use your googlecode.com<http://googlecode.com> 
password, which can be found on the settings 
tab<http://code.google.com/hosting/settings> of your profile page.
Running svnsync on a large repository will take a significant amount of time. 
If you are disconnected during the process, you may see the error message 
"svnsync: Couldn't get lock on destination repos after 10 attempts". If this 
happens, you can remove the lock yourself; see the "Locks" section of 
svnsync.txt<http://svn.collab.net/repos/svn/trunk/notes/svnsync.txt>.

BOb


From: David Goldsmith 
[mailto:eulergaussriem...@gmail.com<mailto:eulergaussriem...@gmail.com>]
Sent: Thursday, October 31, 2013 3:25 PM

To: Bob Archer
Subject: Re: Windows file:///<file:///\\> URL format for svnsync

They used to have restrictions--for security--on what kind of executable code 
you could upload, but I think that was (is?) just for Google docs; AFA Google 
Code is concerned, seeing as how it's intended as an open source code hub, they 
allow just about anything (at least, anything text-based)--if there are 
restrictions, they don't feature them prominently, so I'm not aware of them.

YOU!...are Big Data<http://en.wikipedia.org/wiki/Big_data>.

On Thu, Oct 31, 2013 at 11:23 AM, Bob Archer 
<bob.arc...@amsi.com<mailto:bob.arc...@amsi.com>> wrote:
I’ve never used google reports. Do they let you upload and import a dump file?

From: David Goldsmith 
[mailto:eulergaussriem...@gmail.com<mailto:eulergaussriem...@gmail.com>]
Sent: Thursday, October 31, 2013 2:05 PM
To: Bob Archer

Subject: Re: Windows file:///<file:///\\> URL format for svnsync

Why wouldn't I just upload it to Google?  I guess what I'm missing is: how does 
SVN "recognize" history?  By path and filename?  So, if I preserve that, 
shouldn't that be enough to preserve the history?  Thanks again!
DG

YOU!...are Big Data<http://en.wikipedia.org/wiki/Big_data>.

On Thu, Oct 31, 2013 at 6:49 AM, Bob Archer 
<bob.arc...@amsi.com<mailto:bob.arc...@amsi.com>> wrote:
Good point… our just use svnrdump to dump the repo… then you can take that 
file, load it into a repo on another machine and then sync that to google.

From: Bert Huijben [mailto:b...@qqmail.nl<mailto:b...@qqmail.nl>]
Sent: Wednesday, October 30, 2013 7:02 PM
To: 'David Goldsmith'
Cc: users@subversion.apache.org<mailto:users@subversion.apache.org>
Subject: RE: Windows file:/// URL format for svnsync

‘svn info WORKINGCOPY’ will tell you the url in the repository and the 
repository root.

Assuming that you have access to the entire repository you probably want to 
sync from the repository root to an empty repository to have a local backup. 
(Look in http://svnbook.red-bean.com for details on how to setup the right 
hooks, etc)

And with a bit less time pressure you can then sync that to google code.

                Bert

From: David Goldsmith 
[mailto:eulergaussriem...@gmail.com<mailto:eulergaussriem...@gmail.com>]
Sent: woensdag 30 oktober 2013 22:59
Cc: users@subversion.apache.org<mailto:users@subversion.apache.org>
Subject: Re: Windows file:/// URL format for svnsync

Hi, Ben, and thanks for your reply.  Unfortunately, I think that was the first 
(or perhaps the second) thing I tried, to no avail (also unfortunately, I'm 
away from my work computer for the rest of the day so I can't check my command 
history or the error message it generated).

As I think about my sitch, I realize that the folder I thought was the root of 
the repository probably isn't, because it's the root of the directory tree in 
which reside all the files that I edit day-to-day, and that's supposed to be a 
working copy, not the repository itself, correct? Assuming that's correct, my 
ultimate goal is to "relocate" my project, with history, to a new, empty Google 
code project (already created and reset)--how should I proceed: should I 
continue to try to svnsync my new Google project to my existing repository (to 
which I'll never again have access after tomorrow), and if so, how do I find my 
repository from knowing where a working copy is ('cause, clearly, I've 
forgotten)?  Or should I just upload my working copy from its root, and then 
check that out to any place else I want to be able to work on it--would such an 
upload include the history, and would Google Code recognize it?  Please 
advise/help!

Thanks,

DG

YOU!...are Big Data<http://en.wikipedia.org/wiki/Big_data>.

On Wed, Oct 30, 2013 at 1:29 PM, Ben Reser 
<b...@reser.org<mailto:b...@reser.org>> wrote:
On 10/30/13 1:08 PM, David Goldsmith wrote:
> Hi!  I can't seem to get the formatting for my source repository name--which 
> is
> a Windows directory--correct for svnsync: I've tried forward slashes and
> backslashes, quotes and no quotes, relative path and absolute path--nada.  My
> repository, in Windows syntax, is C:\MWDM--how do I specify this as part of 
> the
> source argument to svnsync?  Thanks,
file:///C:/MWDM<file:///C:\MWDM>

Note that there are 3 forward slashes before the path because you want a blank
host entry.

http://svnbook.red-bean.com/en/1.7/svn.basic.in-action.html#svn.advanced.reposurls




Reply via email to