Thoughts on file ids

[vila]

>>>>> Jelmer Vernooij <jelmer at samba.org> writes:

<snip/>

    > This would help tremendously with parallel imports. 

It seems to me that the issue is in our data model. The per-file graph
is defined as being the set of all revisions texts where the key
includes the file-id.

If a per-file graph was allowed to include revision texts from several
file-ids, that would directly address the parallel import problem.

    > The main problem I see with something like this is that it doesn't help
    > with the problem of texts with the same content having a different file
    > id/revision and being stored multiple times in the repository.

The above will partially address that. Now, we mainly store deltas and
we don't necessarily store the deltas corresponding to the user changes
(we sometimes store whatever delta we feel appropriate (against the left
parent for example) IIUC).

       Vincent