[MERGE] Remove the basis_tree parameter to record_iter_changes.

Robert Collins robert.collins at canonical.com
Sat Mar 21 11:06:31 GMT 2009


On Sat, 2009-03-21 at 20:43 +1000, James Westby wrote:

> > >   * The current commit code expends some effort to go from 
> > >     specific_files to specific_file_ids, that doesn't seem totally 
> > >     redundant:
> > 
> > It should be - in what way isn't it? 
> > 
> > > > # If provided, ensure the specified files are versioned
> 
>  ^ that's the comment from the file. There is a bit more discussion
> (and an "XXX" IIRC), so it seems that this could be cleaned up at the
> same time.
> 
> To me it reads like specific_files=["non-existent"] currently errors
> in commit.py, but wouldn't error if passed to iter_changes. Not too
> difficult to fix I feel.

specific_files=['missing'], require_versioned=True :)

> > >   * Tree.add(['a/b']) where 'a' is not added leads to an iter_changes
> > >     that just adds 'a/b' without adding 'a', and the current code fails
> > >     to commit this as it never adds the parent. Something needs to 
> > >     change here.
> > 
> > Tree.add(['a/b']) should error loudly if a is not added - please file a
> > critical bug on this.
> > For tests, either add a and b, or use tree.smart_add which will add 'a'
> > if needed.
> 
> I was remembering wrong, the failing test is
> bzrlib.tests.test_commit.TestCommit.test_commit_new_subdir_child_selective
> which adds both, but only specifies the child.

Hmm, perhaps iter_changes there should include the parent? 

-Rob
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20090321/9666f55f/attachment.pgp 


More information about the bazaar mailing list