Matthew Honnibal
26446aa728
Avoid loading all French exceptions on import
...
Move exceptions loading behind a get_tokenizer_exceptions() function
for French, instead of loading into the top-level namespace. This
cuts import times from 0.6s to 0.2s, at the expense of making the
French data a little different from the others (there's no top-level
TOKENIZER_EXCEPTIONS variable.) The current solution feels somewhat
unsatisfying.
2017-02-25 11:55:00 +01:00
ines
0e2e331b58
Convert exceptions to Python list
2017-02-24 18:22:40 +01:00
ines
f08e180a47
Make groups non-capturing
...
Prevents hitting the 100 named groups limit in Python
2017-02-10 13:35:02 +01:00
ines
fa3b8512da
Use consistent imports and exports
...
Bundle everything in language_data to keep it consistent with other
languages and make TOKENIZER_EXCEPTIONS importable from there.
2017-02-10 13:34:09 +01:00
ines
21f09d10d7
Revert "Revert "Merge pull request #818 from raphael0202/tokenizer_exceptions""
...
This reverts commit f02a2f9322
2017-02-10 13:17:05 +01:00
ines
f02a2f9322
Revert "Merge pull request #818 from raphael0202/tokenizer_exceptions"
...
This reverts commit b95afdf39cb0ccf32378
2017-02-09 17:07:21 +01:00
Raphaël Bournhonesque
5d706ab95d
Merge tokenizer exceptions from PR #802
2017-02-09 16:30:28 +01:00
Raphaël Bournhonesque
85f951ca99
Add tokenizer exceptions for French
2017-02-02 08:36:16 +01:00
