streaming push alpha-testing
John Arbash Meinel
john at arbash-meinel.com
Tue Feb 24 18:05:47 GMT 2009
-----BEGIN PGP SIGNED MESSAGE-----
Robert Collins wrote:
> On Mon, 2009-02-23 at 10:55 -0600, John Arbash Meinel wrote:
>> I realize this may seem trivial, but having no progress information
>> ~10min is a non trivial change.
> For sure. You should still have been getting bytecounting progress, and
> Andrew and I do want to have progress bars doing their thing.
Well, this was local branching, so no byte counting. It was also copying
into a GC repository, so recompressing all of the texts was taking a
while, and it was nice to know how far along it was.
>> Should we be putting "get_nested_progress_bar()" into the
>> get_record_stream code?
> Possibly. Certainly we do need to get reasonable progress bars back.
> OTOH progress bars are a bit of a sign of a slow program :P.
True enough. Though downloading XX MB and compressing 10,000 texts will
be slow for any program.
> Andrew and I don't have a canned answer yet - we deferred it until we
> have the actual fetching polished. Some possibilities are - bars on the
> sender, bars in the receiver, embedded progress in the stream, stream
> metadata to allow receiver bars, a simple counter of objects received
> I'm kindof partial to simply showing objects handled, for pull 'objects
> received', for push 'objects sent', and where/when we know giving a
> total counter.
I'm fine with that.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
-----END PGP SIGNATURE-----
More information about the bazaar