Storm

lp:~jamesh/storm/sqlite-sync

Created by James Henstridge and last modified
Get this branch:
bzr branch lp:~jamesh/storm/sqlite-sync
Only James Henstridge can upload to this branch. If you are James Henstridge please log in for upload directions.

Branch merges

Related bugs

Related blueprints

Branch information

Owner:
James Henstridge
Project:
Storm
Status:
Merged

Recent revisions

249. By James Henstridge

Move non-timeout related SQLite tests to use unsynchronised connections
to remove fsync() delays.

248. By James Henstridge

Add synchronous parameter to sqlite connections.

247. By James Henstridge

Add __slots__ definitions to Variable and Expr classes and subclasses.
This reduces the memory usage of the most common Storm objects by
removing their class dictionaries. [r=niemeyer,radix]

246. By Gustavo Niemeyer

Updated NEWS file.

245. By Gustavo Niemeyer

Merging need-for-speed-revenge branch [a=niemeyer,jamesh,radix]
[r=jamesh,therve]

This branch introduces a number of fixes, optimizations, and extensions
on the cextensions module, with speedup purposes. The module is now
built by default, but still disabled unless the STORM_CEXTENSIONS=1
environment variable is defined.

Besides these, Chris also provided a mechanism to cache ClassAliases,
to prevent the cost of rebuilding the ClassInfo for each alias.

244. By Gustavo Niemeyer

Use items() instead of iteritems() when iterating the class dictionary
because the class is shared between all threads, and it may actually
be rightfully modified by one of them while the iteration is going on.

[f=242930] [trivial]

243. By Gustavo Niemeyer

Added note about the advisory ordering to the NEWS file.

242. By Gustavo Niemeyer

Merge default-ordering branch [r=jkakar,radix]

This introduces advisory flush ordering. Objects will be flushed in the
order they became dirty if possible (IOW, if there are no other rules
enforcing a different order).

241. By Thomas Herve

Merge zstorm-synch-240801 [f=240801] [r=jkakar,jamesh]

Fix management of existing named stores in ZStorm.create: before the store was
created before checking if it already exists, so was the synchronizer. The
main visible problem was an intermittent failure in the zstorm tests.

240. By James Henstridge

[trivial] Extend the pg_sleep() timeout in the PostgreSQL timeout tracer
test to 0.5s. With the 5ms timeout being used before, the query
sometimes didn't get cancelled. Note that the query will be canelled
long before the 0.5s elapses.

This branch contains Public information 
Everyone can see this information.

Subscribers

No subscribers.