zodb / relstorage / 357

96%

master: 92%

push

travis-ci

Major optimizations for persistent cache files.

Typical operations for large caches will be 3x faster or more. Python 2
and <3.4 need more memory to write, though.

The benchmark is much better and more realistic which is what drove
these changes.

Introduced an option to not compress the local cache files; they aren't
compressed by default anymore because that was a big part of the
slowness.

32 of 32 new or added lines in 2 files covered. (100.0%)

4664 of 4871 relevant lines covered (95.75%)

7.74 hits per line