On Monday, February 25, 2013 2:17:07 PM UTC-5, Matthew Woehlke wrote:
> On 2013-02-25 11:54, age...@themactionfaction.com <javascript:> wrote: 
> > I am migrating from RB 1.6.3 on SQLite and local git repos to RB 1.7.5 
> on 
> > PostgreSQL and remote git repos and cgit. I have successfully migrated 
> the 
> > data and reviewboard 1.7.5 is running, however, the diff viewer is 
> broken. 
> > It seems that RB is passing non-existent SHA hashes to cgit so cgit 
> always 
> > returns the latest version of the file. The resultant error is: 
> This isn't a commit ID, but a blob ID. 
> If you use 'git show <sha>', you should get a dump of the file contents. 
> It sounds like your 'raw file URL' for your repository is not set 
> correctly. Make sure it is set to a template from which cgit will accept 
> a *blob ID* (not a commit ID) to retrieve a specific version of a file. 

Thanks for the pointer, Matthew. We are using cgit It seems that 
if I use http://cgit/cgit/repo/blob/<filename>?id2=<revision>, I always get 
the latest version of the file on the master. cgit seems to ignore id2 if 
the blob hash is not found in the repository. However, if I switch to 
?id=<revision>, then all the diffs seem to work. 

I found that you had answered a similar question here: 

I looked at the cgit source and the difference is id=>sha1 and id2=>sha2. 
In cmd.c:

cgit_print_blob(ctx->qry.sha1, ctx->qry.path, ctx->qry.head);

clearly indicates that only sha1 is used to output blobs which implies use 
of "id" and not "id2". 

Why do the reviewboard docs indicate use of "id2", then?

Want to help the Review Board project? Donate today at 
Happy user? Let us know at http://www.reviewboard.org/users/
To unsubscribe from this group, send email to 
For more options, visit this group at 
You received this message because you are subscribed to the Google Groups 
"reviewboard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to reviewboard+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to