Docserver: Make ObjectStore set calls return Futures

Docserver: Make ObjectStore set calls return Futures

This allows (but does not require) consumers to wait on
the completion of Set/SetMulti operations.

BUG=None
R=kalman@chromium.org
NOTRY=true

Committed: https://crrev.com/9cc03f8a9321155e7da37ca5860cc92b15cc187b
Cr-Commit-Position: refs/heads/master@{#300938}

[Ken Rockot(use gerrit already), 2014-10-23 19:46:04]

Patchset #1 (id:1) has been deleted

[Ken Rockot(use gerrit already), 2014-10-23 19:46:36]

Ben could you take a look? This is extracted from the other large (and soon
less-large) CL.

[not at google - send to devlin, 2014-10-23 20:00:19]

lgtm

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/memcache_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/memcache_object_store.py:26: return
Future(callback=lambda: None)
Any reason why this isn't Future(value=None)?

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/persistent_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/persistent_object_store.py:21: for key,
value in mapping.items()]
iteritems

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/persistent_object_store.py:27: return
All(futures)
this is another case where you should be able to put the generator inside the
All method without constructing an intermediate list.

[Ken Rockot(use gerrit already), 2014-10-23 20:16:18]

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/cache_chain_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/cache_chain_object_store.py:32: return
All(object_store.SetMulti(mapping)
Actually we don't want to force callers to Get the returned future in order for
the set 
operation to complete. So I'm re-listifying this to force evaluation of the
generator (and thus force the RPCs to be started).

(I started running the long test cycle after uploading and just discovered this
bug)

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/memcache_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/memcache_object_store.py:26: return
Future(callback=lambda: None)
On 2014/10/23 20:00:19, kalman wrote:
> Any reason why this isn't Future(value=None)?

Yes. The reason is that I'm bad at Future. Fixed.

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/persistent_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/persistent_object_store.py:27: return
All(futures)
On 2014/10/23 20:00:19, kalman wrote:
> this is another case where you should be able to put the generator inside the
> All method without constructing an intermediate list.

Done.

Given my previous comment, note that unlike CacheChainObjectStore we've already
started all the RPCs, so lazy generator evaluation is really OK here.

[Ken Rockot(use gerrit already), 2014-10-23 20:17:34]

The CQ bit was checked by rockot@chromium.org

[not at google - send to devlin, 2014-10-23 20:18:08]

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
File chrome/common/extensions/docs/server2/cache_chain_object_store.py (right):

https://codereview.chromium.org/658733004/diff/20001/chrome/common/extensions...
chrome/common/extensions/docs/server2/cache_chain_object_store.py:32: return
All(object_store.SetMulti(mapping)
On 2014/10/23 20:16:18, Ken Rockot wrote:
> Actually we don't want to force callers to Get the returned future in order
for
> the set 
> operation to complete. So I'm re-listifying this to force evaluation of the
> generator (and thus force the RPCs to be started).
> 
> (I started running the long test cycle after uploading and just discovered
this
> bug)

ah good point. You might want to add a comment for future adventurers.

[commit-bot: I haz the power, 2014-10-23 20:18:33]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/658733004/40001

[commit-bot: I haz the power, 2014-10-23 20:20:01]

Committed patchset #2 (id:40001)

[commit-bot: I haz the power, 2014-10-23 20:20:41]

Patchset 2 (id:??) landed as
https://crrev.com/9cc03f8a9321155e7da37ca5860cc92b15cc187b
Cr-Commit-Position: refs/heads/master@{#300938}

[Ken Rockot(use gerrit already), 2014-10-23 20:20:52]

On 2014/10/23 20:20:01, I haz the power (commit-bot) wrote:
> Committed patchset #2 (id:40001)

Doh. I'll add a comment soon.