Seed when computing hashes now a parameter for all C-level

[lgautier]

Initial changes to expose hashing seed used at C-level as a parameter (see issue #76).

If deemed reasonable this can be propagated up to command-line.

• Is it mergeable?
• make test Did it pass the tests?
• make coverage Is the new code covered?
• Did it change the command-line interface? Only additions are allowed
  without a major version increment. Changing file formats also requires a
  major version number increment.
• Was a spellchecker run on the source code and documentation after
  changes were made?

[codecov-io]

Current coverage is 76.47% (diff: 75.00%)

Merging #79 into master will increase coverage by 2.04%

@@             master        #79   diff @@
==========================================
  Files            15         17     +2   
  Lines          2010       2274   +264   
  Methods          46         48     +2   
  Messages          0          0          
  Branches         84         84          
==========================================
+ Hits           1496       1739   +243   
- Misses          489        510    +21   
  Partials         25         25          

Powered by Codecov. Last update 3138a76...487f8e3

[ctb]

Is this ready to be reviewed @lgautier ?

[ctb]

On a quick skim, the only problem I see is that the seed code isn't executed by any tests, nor is the seed saved in any signature/estimator location. I'm happy to address both of those post-merge over in #91, where I'm mucking about with that stuff.

[lgautier]

It is in the sense that pull request is mostly consistent and self-sufficient. The only possible inconsistency is that until the seed is saved into JSON/YAML the consistency between save/load cycles can be broken, also this is not an issue for the casual user since the command line is not letting one play with the seed.

I am seeing the following missing:

• bubbling the seed up to the command line interface
• save the seed in the JSON / YAML (for the time it must be supported)
• read the seed from the JSON (assume default seed with a warning if missing seed ?)

I'd say review and merge now, than quickly add the save/read seed to/from JSON/YAML ASAP.

[ctb]

Agreed! I can do the add/save etc in #91.

[ctb]

LGTM.

[lgautier]

On a quick skim, the only problem I see is that the seed code isn't executed by any tests, nor is the seed saved in any signature/estimator location. I'm happy to address both of those post-merge over in #91, where I'm mucking about with that stuff.

Pickling / unplicking should not be problem since this seed is a default type... but I realizing while writing this that seed is not stored at the Python level. I am happy to give an hand in #91 about this if needed.