[OSM-dev] Minute Diffs Broken

[Matt Amos]

On Tue, May 5, 2009 at 1:45 PM, Greg Troxel <gdt at ir.bbn.com> wrote:
> I think we have
>
>  uploads == db transactions (perhaps "microchangesets" of "changeset fragments"??)
>
>  changesets == (some group of uploads, with a common id and comment)
>
>  minute diffs == (some collection of uploads)

your summary is correct. we do have a terminology problem with
"changesets", as there's repeated confusion about their lack of
transactionality.

> I don't have a clue about how the uploads are numbered

they aren't explicitly numbered, but might be implicitly numbered by
txid or commit-time.

> and how easy it is to extract all of them,

that would appear to be difficult ;-)

> but given that the main DB can have committed
> transactions with uploads that are not part of a closed changeset, I
> think the minute diff and replicated dbs should have that too.

indeed. i don't think changesets were intended as units of
replication, especially since elements used in a closed changeset
might have previous versions in still-open changesets.

while frederik is right that knowing a changeset is closed is very
useful, it creates difficulties when used for replication.

cheers,

matt