git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* git cvsimport error
@ 2008-05-21 19:14 Adam Mercer
  2008-05-21 19:39 ` Adam Mercer
  0 siblings, 1 reply; 7+ messages in thread
From: Adam Mercer @ 2008-05-21 19:14 UTC (permalink / raw)
  To: git

Hi

I'm trying to convert a CVS repository to GIT (1.5.5.1 on Intel Mac OS
X 10.5.2) using the following:

[ram@mimir ~]$ git cvsimport -v -d :local:/Users/ram/test_cvs -C glue.git glue
Initialized empty Git repository in /Users/ram/glue.git/.git/
Running cvsps...
cvs_direct initialized to CVSROOT /Users/ram/test_cvs
cvs rlog: Logging glue
cvs rlog: Logging glue/bin
cvs rlog: Logging glue/debian
cvs rlog: Logging glue/doc
cvs rlog: Logging glue/etc
cvs rlog: Logging glue/glue
cvs rlog: Logging glue/glue/ligolw
cvs rlog: Logging glue/glue/ligolw/utils
cvs rlog: Logging glue/glue/segfindserver
cvs rlog: Logging glue/glue/segfindserver/segments_1_7
cvs rlog: Logging glue/man
cvs rlog: Logging glue/man/man1
cvs rlog: Logging glue/sbin
cvs rlog: Logging glue/src
cvs rlog: Logging glue/src/conf
cvs rlog: Logging glue/src/conf/db2
cvs rlog: Logging glue/src/conf/mysql
cvs rlog: Logging glue/src/conf/q_replication
cvs rlog: Logging glue/src/conf/q_replication/cit
cvs rlog: Logging glue/src/conf/q_replication/lho
cvs rlog: Logging glue/src/conf/q_replication/llo
cvs rlog: Logging glue/src/segments
cvs rlog: Logging glue/test
Fetching glue/ldbd.py   v 1.1
Unknown: error
[ram@mimir ~]$

How can I get a more verbose error message to try and fix what's going
wrong here?

Cheers

Adam

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: git cvsimport error
  2008-05-21 19:14 git cvsimport error Adam Mercer
@ 2008-05-21 19:39 ` Adam Mercer
  2008-05-21 19:51   ` Craig L. Ching
  0 siblings, 1 reply; 7+ messages in thread
From: Adam Mercer @ 2008-05-21 19:39 UTC (permalink / raw)
  To: git

On Wed, May 21, 2008 at 2:14 PM, Adam Mercer <ramercer@gmail.com> wrote:

> Fetching glue/ldbd.py   v 1.1
> Unknown: error
> [ram@mimir ~]$

Done some digging with cvsps itself and it looks like the lbdb.py,v
file has become corrupted from a very early version...

Is there a way that I can tell git to ignore the early revisions?

Cheers

Adam

^ permalink raw reply	[flat|nested] 7+ messages in thread

* RE: git cvsimport error
  2008-05-21 19:39 ` Adam Mercer
@ 2008-05-21 19:51   ` Craig L. Ching
  2008-05-23 20:01     ` Adam Mercer
       [not found]     ` <799406d60805231301l1ff158b7k73bb193c472a8211@mail.gmail.com>
  0 siblings, 2 replies; 7+ messages in thread
From: Craig L. Ching @ 2008-05-21 19:51 UTC (permalink / raw)
  To: Adam Mercer, git


> On Wed, May 21, 2008 at 2:14 PM, Adam Mercer 
> <ramercer@gmail.com> wrote:
> 
> > Fetching glue/ldbd.py   v 1.1
> > Unknown: error
> > [ram@mimir ~]$
> 
> Done some digging with cvsps itself and it looks like the 
> lbdb.py,v file has become corrupted from a very early version...
> 
> Is there a way that I can tell git to ignore the early revisions?
> 
Not a cvsps expert, but I'm at the same point as you in converting a
repo ;-)  Would the -d option on cvsps do what you want?

> Cheers
> 
> Adam
> --
> 
Cheers,
Craig

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: git cvsimport error
  2008-05-21 19:51   ` Craig L. Ching
@ 2008-05-23 20:01     ` Adam Mercer
       [not found]     ` <799406d60805231301l1ff158b7k73bb193c472a8211@mail.gmail.com>
  1 sibling, 0 replies; 7+ messages in thread
From: Adam Mercer @ 2008-05-23 20:01 UTC (permalink / raw)
  To: git

On Wed, May 21, 2008 at 2:51 PM, Craig L. Ching <cching@mqsoftware.com> wrote:

> Not a cvsps expert, but I'm at the same point as you in converting a
> repo ;-)  Would the -d option on cvsps do what you want?

It should but I can't get it too work, the ldbd.py,v file is OK from
2003/05/06 07:39:52, so after consulting the man pages for both
git-cvsimport and cvsps I ran the following

$ git cvsimport -v  -p "-d" -p "2003/05/06 07:40:00"
-d:local:/Users/ram/test_cvs -C glue.git glue

yet git/cvsps still tries to retrieve the bad revisions, prior to the
above date. I'm clearly missing something?

Cheers

Adam

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: git cvsimport error
       [not found]       ` <63BEA5E623E09F4D92233FB12A9F794301FC8BCA@emailmn.mqsoftware.com>
@ 2008-05-23 20:24         ` Adam Mercer
  2008-05-23 20:30           ` Shawn O. Pearce
  0 siblings, 1 reply; 7+ messages in thread
From: Adam Mercer @ 2008-05-23 20:24 UTC (permalink / raw)
  To: Craig L. Ching; +Cc: git

On Fri, May 23, 2008 at 3:05 PM, Craig L. Ching <cching@mqsoftware.com> wrote:
> Yeah, I don't know much about it at all, but my advice would be to run
> cvsps by itself and generate a cache that git-fast-import can import.
> That's the approach we're taking because it gives us a chance to figure
> out where things have gone wrong as we have some bad, buggy commits in
> our archive (CVS created the bad commits).

I thought of using git-fast-import but the problem is that the CVS
repository is still in use so I need to keep the git and CVS repos
synchronised and it seems like git-fast-import can't be used in this
case.

Cheers

Adam

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: git cvsimport error
  2008-05-23 20:24         ` Adam Mercer
@ 2008-05-23 20:30           ` Shawn O. Pearce
  2008-05-24  5:23             ` Adam Mercer
  0 siblings, 1 reply; 7+ messages in thread
From: Shawn O. Pearce @ 2008-05-23 20:30 UTC (permalink / raw)
  To: Adam Mercer; +Cc: Craig L. Ching, git

Adam Mercer <ramercer@gmail.com> wrote:
> On Fri, May 23, 2008 at 3:05 PM, Craig L. Ching <cching@mqsoftware.com> wrote:
> > Yeah, I don't know much about it at all, but my advice would be to run
> > cvsps by itself and generate a cache that git-fast-import can import.
> > That's the approach we're taking because it gives us a chance to figure
> > out where things have gone wrong as we have some bad, buggy commits in
> > our archive (CVS created the bad commits).
> 
> I thought of using git-fast-import but the problem is that the CVS
> repository is still in use so I need to keep the git and CVS repos
> synchronised and it seems like git-fast-import can't be used in this
> case.

You can run fast-import in incremental behavior.  You just have to
restart each branch with a "from refs/heads/master^0" or whatever.
There's some discussion of this in the (rather large) fast-import
man page.

-- 
Shawn.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: git cvsimport error
  2008-05-23 20:30           ` Shawn O. Pearce
@ 2008-05-24  5:23             ` Adam Mercer
  0 siblings, 0 replies; 7+ messages in thread
From: Adam Mercer @ 2008-05-24  5:23 UTC (permalink / raw)
  To: Shawn O. Pearce; +Cc: Craig L. Ching, git

On Fri, May 23, 2008 at 3:30 PM, Shawn O. Pearce <spearce@spearce.org> wrote:

> You can run fast-import in incremental behavior.  You just have to
> restart each branch with a "from refs/heads/master^0" or whatever.
> There's some discussion of this in the (rather large) fast-import
> man page.

Thanks I'll take a look at that as I've found that I can pass the -d
option to cvsps which _should_ tell it to ignore revisions before a
certain date, but even doing this cvsimport tries to import the broken
revision prior date!

Cheers

Adam

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2008-05-24  5:24 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-05-21 19:14 git cvsimport error Adam Mercer
2008-05-21 19:39 ` Adam Mercer
2008-05-21 19:51   ` Craig L. Ching
2008-05-23 20:01     ` Adam Mercer
     [not found]     ` <799406d60805231301l1ff158b7k73bb193c472a8211@mail.gmail.com>
     [not found]       ` <63BEA5E623E09F4D92233FB12A9F794301FC8BCA@emailmn.mqsoftware.com>
2008-05-23 20:24         ` Adam Mercer
2008-05-23 20:30           ` Shawn O. Pearce
2008-05-24  5:23             ` Adam Mercer

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).