Add strategy index test to travis.

[drvinceknight]

Closes #957. I am expecting this to fail on travis: two modules are not currently indexed (hmm and memorytwo). Using that as a test that this script works (once travis fails, will push the fix).

[drvinceknight]

4738a96 has failed as expected (travis build here: https://travis-ci.org/Axelrod-Python/Axelrod/builds/223638310):

 0.04s$ python run_strategy_indexer.py
memorytwo not in index
The command "python run_strategy_indexer.py" exited with 1.

Just pushing the fix.

[marcharper]

Can we improve the docstring? Technically these aren't modules. Also let's use the standard style for parameters (just module_path is ok?) and the return value.

[marcharper]

could use a single if (either way is ok)

[marcharper]

"*.py" might be slightly more future proof

[drvinceknight]

I think I've addressed your comments @marcharper. I've also added type hinting :)

[meatballs]

I'm happy with this but I think an improvement might be to use the Pathlib module for file paths.

I don't think there's a problem here, but I've have had problems with file paths on windows before and Pathlib handles all that perfectly.

[drvinceknight]

That's a good shout. I'll do that and push today 👍

…

On Thu, 20 Apr 2017, 08:20 Owen Campbell, ***@***.***> wrote: I'm happy with this but I think an improvement might be to use the Pathlib module for file paths. I don't think there's a problem here, but I've have had problems with file paths on windows before and Pathlib handles all that perfectly. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#980 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACCGWgPFUEipTyorDVLn58_7Y_ZNfONKks5rxwchgaJpZM4NB8PY> .

[drvinceknight]

but I've have had problems with file paths on windows before and Pathlib handles all that perfectly.

Great shout on Pathlib, it cleared up a lot of the code (no real point in 2 of the 3 functions as they're just simple method calls of the Path class). I'd seen it used before but never played around with it myself: it's lovely!.

Note: I removed the type specification in the docstring. This is mainly because I noticed just before pushing that it had become a lie (I changed the type hint to be a Path and didn't change the docstring). Seems sensible to me to remove the types from the docstrings when we have type hints in place but I'm happy to throw them back in for this function if we want them.

[meatballs]

Great shout on Pathlib, it cleared up a lot of the code (no real point in 2 of the 3 functions as they're just simple method calls of the Path class). I'd seen it used before but never played around with it myself: it's lovely!.

Yep. I'm a big fan!