pytest-benchmark (and its dependency "statistics") needs py 3.3+.
we do not want to run benchmarks on travis anyway, but even its presence
makes trouble due to this, so we just remove it from there again.
as --benchmark-skip is also removed from tox configuration, this means benchmarks
are run by default when pytest-benchmark is installed.
Instead of "realistic data", I chose the test data to be either all-zero (all-ascii-zero to be precise)
or all-random and benchmark them separately.
So we can better determine the cause (deduplication or storage) in case we see some performance regression.
"help" is benchmarked to see the minimum runtime when it basically does nothing.
also:
- refactor archiver execution core functionality into exec_cmd() so it can be used more flexibly
- tox: usually we want to skip benchmarks, only run them if requested manually
- install pytest-benchmark - run tox with "-r" to have it installed into your .tox envs
note that we do not depend on attic to build borg, just to do those
tests. if attic goes away, we could eventually do this another way or
just stop testing this altogether.
README.rst (shown on github and also at the start of the html docs) shall
be like an elevator speech - convince readers in a very short time.
this is most important, everything else can come after we got the reader's interest.
include README into docs to avoid duplication.
also include CHANGES into docs.
add developer docs, move examples from tox.ini there
add separate support docs
remove glossary, most of what was there can be understood by an admin from context
move attic and compatibility note to the end
the usage of some py.test features in the previous locking changesets somehow broke the test environment.
it seems to help to point py.test to some specific package or module (and not just invoke "py.test" without arguments).
