diff --git a/examples/pipeline/multi_processing.py b/examples/pipeline/multi_processing.py index 07d3165f9..4c4d29c13 100644 --- a/examples/pipeline/multi_processing.py +++ b/examples/pipeline/multi_processing.py @@ -10,12 +10,13 @@ Compatible with: spaCy v2.0.0+ """ from __future__ import print_function, unicode_literals -from toolz import partition_all from pathlib import Path from joblib import Parallel, delayed +from functools import partial import thinc.extra.datasets import plac import spacy +from spacy.util import minibatch @plac.annotations( @@ -35,10 +36,10 @@ def main(output_dir, model="en_core_web_sm", n_jobs=4, batch_size=1000, limit=10 data, _ = thinc.extra.datasets.imdb() texts, _ = zip(*data[-limit:]) print("Processing texts...") - partitions = partition_all(batch_size, texts) - executor = Parallel(n_jobs=n_jobs) - do = delayed(transform_texts) - tasks = (do(nlp, i, batch, output_dir) for i, batch in enumerate(partitions)) + partitions = minibatch(texts, size=batch_size) + executor = Parallel(n_jobs=n_jobs, backend="multiprocessing", prefer="processes") + do = delayed(partial(transform_texts, nlp)) + tasks = (do(i, batch, output_dir) for i, batch in enumerate(partitions)) executor(tasks) diff --git a/examples/training/train_tagger.py b/examples/training/train_tagger.py index 0673d06d2..db0627270 100644 --- a/examples/training/train_tagger.py +++ b/examples/training/train_tagger.py @@ -31,7 +31,7 @@ TAG_MAP = {"N": {"pos": "NOUN"}, "V": {"pos": "VERB"}, "J": {"pos": "ADJ"}} # strings are unicode and that the number of tags assigned matches spaCy's # tokenization. If not, you can always add a 'words' key to the annotations # that specifies the gold-standard tokenization, e.g.: -# ("Eatblueham", {'words': ['Eat', 'blue', 'ham'] 'tags': ['V', 'J', 'N']}) +# ("Eatblueham", {'words': ['Eat', 'blue', 'ham'], 'tags': ['V', 'J', 'N']}) TRAIN_DATA = [ ("I like green eggs", {"tags": ["N", "V", "J", "N"]}), ("Eat blue ham", {"tags": ["V", "J", "N"]}), diff --git a/fabfile.py b/fabfile.py index 268ebb6d7..0e69551c3 100644 --- a/fabfile.py +++ b/fabfile.py @@ -59,6 +59,7 @@ def make(): def sdist(): with virtualenv(VENV_DIR) as venv_local: with lcd(path.dirname(__file__)): + local('python -m pip install -U setuptools') local('python setup.py sdist') def wheel(): diff --git a/spacy/about.py b/spacy/about.py index b11b2c81e..9947bf71a 100644 --- a/spacy/about.py +++ b/spacy/about.py @@ -4,7 +4,7 @@ # fmt: off __title__ = "spacy-nightly" -__version__ = "2.1.0a7" +__version__ = "2.1.0a8" __summary__ = "Industrial-strength Natural Language Processing (NLP) with Python and Cython" __uri__ = "https://spacy.io" __author__ = "Explosion AI" diff --git a/spacy/lang/char_classes.py b/spacy/lang/char_classes.py index 892de6374..4906436e3 100644 --- a/spacy/lang/char_classes.py +++ b/spacy/lang/char_classes.py @@ -5,41 +5,143 @@ split_chars = lambda char: list(char.strip().split(" ")) merge_chars = lambda char: char.strip().replace(" ", "|") group_chars = lambda char: char.strip().replace(" ", "") -# used https://unicode.org/cldr/utility/list-unicodeset.jsp to convert categories into code points -# \p{L}&&\p{Bengali} -# https://en.wikipedia.org/wiki/Bengali_(Unicode_block) _bengali = r"\u0980-\u09FF" -# \p{L}&&\p{Hebrew} -# https://en.wikipedia.org/wiki/Unicode_and_HTML_for_the_Hebrew_alphabet _hebrew = r"\u0591-\u05F4\uFB1D-\uFB4F" -# \p{Ll}&&\p{Latin} -_latin_lower = r"a-z\u00DF-\u00F6\u00F8-\u00FF\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175\u0177\u017A\u017C\u017E-\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242\u0247\u0249\u024B\u024D\u024F-\u0293\u0295-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFF\u214E\u2184\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\uFB00-\uFB06\uFF41-\uFF5A" -# \p{Lu}&&\p{Latin} -_latin_upper = r"A-Z\u00C0-\u00D6\u00D8-\u00DE\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176\u0178\u0179\u017B\u017D\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFE\u212A\u212B\u2132\u2183\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8\uFF21-\uFF3A" -# [\p{Ll}||\p{Lu}]&&\p{Latin} -latin = r"A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u01BA\u01BC-\u01BF\u01C4\u01C6\u01C7\u01C9\u01CA\u01CC-\u01F1\u01F3-\u0293\u0295-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A\u1E00-\u1EFF\u212A\u212B\u2132\u214E\u2183\u2184\u2C60-\u2C7B\u2C7E\u2C7F\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA\uAB30-\uAB5A\uAB60-\uAB64\uFB00-\uFB06\uFF21-\uFF3A\uFF41-\uFF5A" +# Latin standard +_latin_u_standard = r"A-Z" +_latin_l_standard = r"a-z" +_latin_standard = _latin_u_standard + _latin_l_standard -# \p{L}&&\p{Arabic} -_persian = r"\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB" +_latin_u_standard_fullwidth = r"\uFF21-\uFF3A" +_latin_l_standard_fullwidth = r"\uFF41-\uFF5A" +_latin_standard_fullwidth = _latin_u_standard_fullwidth + _latin_l_standard_fullwidth + +# letters with diacritics - French, German, Icelandic, Spanish +_latin_u_supplement = r"\u00C0-\u00D6\u00D8-\u00DE" +_latin_l_supplement = r"\u00DF-\u00F6\u00F8-\u00FF" +_latin_supplement = r"\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u00FF" + +# letters with diacritics - Catalan, Czech, Latin, Latvian, Lithuanian, Polish, Slovak, Turkish, Welsh +_latin_u_extendedA = r"\u0100\u0102\u0104\u0106\u0108\u010A\u010C\u010E\u0110\u0112\u0114\u0116\u0118\u011A\u011C" \ + r"\u011E\u0120\u0122\u0124\u0126\u0128\u012A\u012C\u012E\u0130\u0132\u0134\u0136\u0139\u013B" \ + r"\u013D\u013F\u0141\u0143\u0145\u0147\u014A\u014C\u014E\u0150\u0152\u0154\u0156\u0158" \ + r"\u015A\u015C\u015E\u0160\u0162\u0164\u0166\u0168\u016A\u016C\u016E\u0170\u0172\u0174\u0176" \ + r"\u0178\u0179\u017B\u017D" +_latin_l_extendedA = r"\u0101\u0103\u0105\u0107\u0109\u010B\u010D\u010F\u0111\u0113\u0115\u0117\u0119\u011B\u011D" \ + r"\u011F\u0121\u0123\u0125\u0127\u0129\u012B\u012D\u012F\u0131\u0133\u0135\u0137\u0138\u013A" \ + r"\u013C\u013E\u0140\u0142\u0144\u0146\u0148\u0149\u014B\u014D\u014F\u0151\u0153\u0155\u0157" \ + r"\u0159\u015B\u015D\u015F\u0161\u0163\u0165\u0167\u0169\u016B\u016D\u016F\u0171\u0173\u0175" \ + r"\u0177\u017A\u017C\u017E\u017F" +_latin_extendedA = r"\u0100-\u017F" + +# special characters - Khoisan, Pan-Nigerian, Pinyin, Romanian +# those that are a combination of both upper and lower letters are only included in the group _latin_extendedB +_latin_u_extendedB = r"\u0181\u0182\u0184\u0186\u0187\u0189-\u018B\u018E-\u0191\u0193\u0194\u0196-\u0198\u019C" \ + r"\u019D\u019F\u01A0\u01A2\u01A4\u01A6\u01A7\u01A9\u01AC\u01AE\u01AF\u01B1-\u01B3\u01B5" \ + r"\u01B7\u01B8\u01BC\u01C4\u01C7\u01CA\u01CD\u01CF\u01D1\u01D3\u01D5\u01D7\u01D9\u01DB" \ + r"\u01DE\u01E0\u01E2\u01E4\u01E6\u01E8\u01EA\u01EC\u01EE\u01F1\u01F4\u01F6-\u01F8\u01FA" \ + r"\u01FC\u01FE\u0200\u0202\u0204\u0206\u0208\u020A\u020C\u020E\u0210\u0212\u0214\u0216" \ + r"\u0218\u021A\u021C\u021E\u0220\u0222\u0224\u0226\u0228\u022A\u022C\u022E\u0230\u0232" \ + r"\u023A\u023B\u023D\u023E\u0241\u0243-\u0246\u0248\u024A\u024C\u024E" +_latin_l_extendedB = r"\u0180\u0183\u0185\u0188\u018C\u018D\u0192\u0195\u0199-\u019B\u019E\u01A1\u01A3\u01A5" \ + r"\u01A8\u01AA\u01AB\u01AD\u01B0\u01B4\u01B6\u01B9\u01BA\u01BD-\u01BF\u01C6\u01C9\u01CC" \ + r"\u01CE\u01D0\u01D2\u01D4\u01D6\u01D8\u01DA\u01DC\u01DD\u01DF\u01E1\u01E3\u01E5\u01E7" \ + r"\u01E9\u01EB\u01ED\u01EF\u01F0\u01F3\u01F5\u01F9\u01FB\u01FD\u01FF\u0201\u0203\u0205" \ + r"\u0207\u0209\u020B\u020D\u020F\u0211\u0213\u0215\u0217\u0219\u021B\u021D\u021F\u0221" \ + r"\u0223\u0225\u0227\u0229\u022B\u022D\u022F\u0231\u0233-\u0239\u023C\u023F\u0240\u0242" \ + r"\u0247\u0249\u024B\u024D\u024F" +_latin_extendedB = r"\u0180-\u01BF\u01C4-\u024F" + +# special characters - Uighur, Uralic Phonetic +_latin_u_extendedC = r"\u2C60\u2C62-\u2C64\u2C67\u2C69\u2C6B\u2C6D-\u2C70\u2C72\u2C75\u2C7E\u2C7F" +_latin_l_extendedC = r"\u2C61\u2C65\u2C66\u2C68\u2C6A\u2C6C\u2C71\u2C73\u2C74\u2C76-\u2C7B" +_latin_extendedC = r"\u2C60-\u2C7B\u2C7E\u2C7F" + +# special characters - phonetic, Mayan, Medieval +_latin_u_extendedD = r"\uA722\uA724\uA726\uA728\uA72A\uA72C\uA72E\uA732\uA734\uA736\uA738\uA73A\uA73C" \ + r"\uA73E\uA740\uA742\uA744\uA746\uA748\uA74A\uA74C\uA74E\uA750\uA752\uA754\uA756\uA758" \ + r"\uA75A\uA75C\uA75E\uA760\uA762\uA764\uA766\uA768\uA76A\uA76C\uA76E\uA779\uA77B\uA77D" \ + r"\uA77E\uA780\uA782\uA784\uA786\uA78B\uA78D\uA790\uA792\uA796\uA798\uA79A\uA79C\uA79E" \ + r"\uA7A0\uA7A2\uA7A4\uA7A6\uA7A8\uA7AA-\uA7AE\uA7B0-\uA7B4\uA7B6\uA7B8" +_latin_l_extendedD = r"\uA723\uA725\uA727\uA729\uA72B\uA72D\uA72F-\uA731\uA733\uA735\uA737\uA739\uA73B\uA73D" \ + r"\uA73F\uA741\uA743\uA745\uA747\uA749\uA74B\uA74D\uA74F\uA751\uA753\uA755\uA757\uA759" \ + r"\uA75B\uA75D\uA75F\uA761\uA763\uA765\uA767\uA769\uA76B\uA76D\uA76F\uA771-\uA778\uA77A" \ + r"\uA77C\uA77F\uA781\uA783\uA785\uA787\uA78C\uA78E\uA791\uA793-\uA795\uA797\uA799\uA79B" \ + r"\uA79D\uA79F\uA7A1\uA7A3\uA7A5\uA7A7\uA7A9\uA7AF\uA7B5\uA7B7\uA7B9\uA7FA" +_latin_extendedD = r"\uA722-\uA76F\uA771-\uA787\uA78B-\uA78E\uA790-\uA7B9\uA7FA" + +# special characters - phonetic Teuthonista and Sakha +_latin_l_extendedE = r"\uAB30-\uAB5A\uAB60-\uAB64" +_latin_extendedE = _latin_l_extendedE + +# phonetic letters - Greek, Latin, Cyrillic +_latin_l_phonetic = r"\u0250-\u02AF\u1D00-\u1D25\u1D6B-\u1D77\u1D79-\u1D9A" +_latin_phonetic = _latin_l_phonetic + +# letters with multiple diacritics - Vietnamese +_latin_u_diacritics = r"\u1E00\u1E02\u1E04\u1E06\u1E08\u1E0A\u1E0C\u1E0E\u1E10\u1E12\u1E14\u1E16\u1E18\u1E1A" \ + r"\u1E1C\u1E1E\u1E20\u1E22\u1E24\u1E26\u1E28\u1E2A\u1E2C\u1E2E\u1E30\u1E32\u1E34\u1E36" \ + r"\u1E38\u1E3A\u1E3C\u1E3E\u1E40\u1E42\u1E44\u1E46\u1E48\u1E4A\u1E4C\u1E4E\u1E50\u1E52" \ + r"\u1E54\u1E56\u1E58\u1E5A\u1E5C\u1E5E\u1E60\u1E62\u1E64\u1E66\u1E68\u1E6A\u1E6C\u1E6E" \ + r"\u1E70\u1E72\u1E74\u1E76\u1E78\u1E7A\u1E7C\u1E7E\u1E80\u1E82\u1E84\u1E86\u1E88\u1E8A" \ + r"\u1E8C\u1E8E\u1E90\u1E92\u1E94\u1E9E\u1EA0\u1EA2\u1EA4\u1EA6\u1EA8\u1EAA\u1EAC\u1EAE" \ + r"\u1EB0\u1EB2\u1EB4\u1EB6\u1EB8\u1EBA\u1EBC\u1EBE\u1EC0\u1EC2\u1EC4\u1EC6\u1EC8" \ + r"\u1ECA\u1ECC\u1ECE\u1ED0\u1ED2\u1ED4\u1ED6\u1ED8\u1EDA\u1EDC\u1EDE\u1EE0\u1EE2\u1EE4" \ + r"\u1EE6\u1EE8\u1EEA\u1EEC\u1EEE\u1EF0\u1EF2\u1EF4\u1EF6\u1EF8\u1EFA\u1EFC\u1EFE" +_latin_l_diacritics = r"\u1E01\u1E03\u1E05\u1E07\u1E09\u1E0B\u1E0D\u1E0F\u1E11\u1E13\u1E15\u1E17\u1E19\u1E1B" \ + r"\u1E1D\u1E1F\u1E21\u1E23\u1E25\u1E27\u1E29\u1E2B\u1E2D\u1E2F\u1E31\u1E33\u1E35\u1E37" \ + r"\u1E39\u1E3B\u1E3D\u1E3F\u1E41\u1E43\u1E45\u1E47\u1E49\u1E4B\u1E4D\u1E4F\u1E51\u1E53" \ + r"\u1E55\u1E57\u1E59\u1E5B\u1E5D\u1E5F\u1E61\u1E63\u1E65\u1E67\u1E69\u1E6B\u1E6D\u1E6F" \ + r"\u1E71\u1E73\u1E75\u1E77\u1E79\u1E7B\u1E7D\u1E7F\u1E81\u1E83\u1E85\u1E87\u1E89\u1E8B" \ + r"\u1E8D\u1E8F\u1E91\u1E93\u1E95-\u1E9D\u1E9F\u1EA1\u1EA3\u1EA5\u1EA7\u1EA9\u1EAB\u1EAD" \ + r"\u1EAF\u1EB1\u1EB3\u1EB5\u1EB7\u1EB9\u1EBB\u1EBD\u1EBF\u1EC1\u1EC3\u1EC5\u1EC7\u1EC9" \ + r"\u1ECB\u1ECD\u1ECF\u1ED1\u1ED3\u1ED5\u1ED7\u1ED9\u1EDB\u1EDD\u1EDF\u1EE1\u1EE3\u1EE5" \ + r"\u1EE7\u1EE9\u1EEB\u1EED\u1EEF\u1EF1\u1EF3\u1EF5\u1EF7\u1EF9\u1EFB\u1EFD\u1EFF" +_latin_diacritics = r"\u1E00-\u1EFF" + +# all lower latin classes +LATIN_LOWER_BASIC = _latin_l_standard + _latin_l_standard_fullwidth + _latin_l_supplement + _latin_l_extendedA +LATIN_LOWER = LATIN_LOWER_BASIC + _latin_l_extendedB + _latin_l_extendedC + _latin_l_extendedD + _latin_l_extendedE \ + + _latin_l_phonetic + _latin_l_diacritics + +# all upper latin classes +LATIN_UPPER_BASIC = _latin_u_standard + _latin_u_standard_fullwidth + _latin_u_supplement + _latin_u_extendedA +LATIN_UPPER = LATIN_UPPER_BASIC + _latin_u_extendedB + _latin_u_extendedC + _latin_u_extendedD + _latin_u_diacritics + +# all latin classes +LATIN_BASIC = _latin_standard + _latin_standard_fullwidth + _latin_supplement + _latin_extendedA +LATIN = LATIN_BASIC + _latin_extendedB + _latin_extendedC + _latin_extendedD + _latin_extendedE \ + + _latin_phonetic + _latin_diacritics + +_persian = r"\u0620-\u064A\u066E-\u06D5\u06E5-\u06FF\u0750-\u077F\u08A0-\u08BD" \ + r"\uFB50-\uFBB1\uFBD3-\uFD3D\uFD50-\uFDC7\uFDF0-\uFDFB\uFE70-\uFEFC\U0001EE00-\U0001EEBB" _russian_lower = r"ёа-я" _russian_upper = r"ЁА-Я" -# \p{L}&&\p{Sinhala} +_russian = r"ёа-яЁА-Я" + _sinhala = r"\u0D85-\u0D96\u0D9A-\u0DB1\u0DB3-\u0DBB\u0DBD\u0DC0-\u0DC6" + _tatar_lower = r"әөүҗңһ" _tatar_upper = r"ӘӨҮҖҢҺ" +_tatar = r"әөүҗңһӘӨҮҖҢҺ" + _greek_lower = r"α-ωάέίόώήύ" _greek_upper = r"Α-ΩΆΈΊΌΏΉΎ" +_greek = r"α-ωάέίόώήύΑ-ΩΆΈΊΌΏΉΎ" + _ukrainian_lower = r"а-щюяіїєґ" _ukrainian_upper = r"А-ЩЮЯІЇЄҐ" +_ukrainian = r"а-щюяіїєґА-ЩЮЯІЇЄҐ" + +_upper = LATIN_UPPER + _russian_upper + _tatar_upper + _greek_upper + _ukrainian_upper +_lower = LATIN_LOWER + _russian_lower + _tatar_lower + _greek_lower + _ukrainian_lower -_upper = _latin_upper + _russian_upper + _tatar_upper + _greek_upper + _ukrainian_upper -_lower = _latin_lower + _russian_lower + _tatar_lower + _greek_lower + _ukrainian_lower _uncased = _bengali + _hebrew + _persian + _sinhala -ALPHA = group_chars(_upper + _lower + _uncased) +ALPHA = group_chars(LATIN + _russian + _tatar + _greek + _ukrainian + _uncased) ALPHA_LOWER = group_chars(_lower + _uncased) ALPHA_UPPER = group_chars(_upper + _uncased) @@ -63,7 +165,33 @@ _hyphens = "- – — -- --- —— ~" # Various symbols like dingbats, but also emoji # Details: https://www.compart.com/en/unicode/category/So -_other_symbols = r"\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D" +_other_symbols = r"\u00A6\u00A9\u00AE\u00B0\u0482\u058D\u058E\u060E\u060F\u06DE\u06E9\u06FD\u06FE\u07F6\u09FA\u0B70" \ + r"\u0BF3-\u0BF8\u0BFA\u0C7F\u0D4F\u0D79\u0F01-\u0F03\u0F13\u0F15-\u0F17\u0F1A-\u0F1F\u0F34" \ + r"\u0F36\u0F38\u0FBE-\u0FC5\u0FC7-\u0FCC\u0FCE\u0FCF\u0FD5-\u0FD8\u109E\u109F\u1390-\u1399" \ + r"\u1940\u19DE-\u19FF\u1B61-\u1B6A\u1B74-\u1B7C\u2100\u2101\u2103-\u2106\u2108\u2109\u2114\u2116" \ + r"\u2117\u211E-\u2123\u2125\u2127\u2129\u212E\u213A\u213B\u214A\u214C\u214D\u214F\u218A\u218B" \ + r"\u2195-\u2199\u219C-\u219F\u21A1\u21A2\u21A4\u21A5\u21A7-\u21AD\u21AF-\u21CD\u21D0\u21D1\u21D3" \ + r"\u21D5-\u21F3\u2300-\u2307\u230C-\u231F\u2322-\u2328\u232B-\u237B\u237D-\u239A\u23B4-\u23DB" \ + r"\u23E2-\u2426\u2440-\u244A\u249C-\u24E9\u2500-\u25B6\u25B8-\u25C0\u25C2-\u25F7\u2600-\u266E" \ + r"\u2670-\u2767\u2794-\u27BF\u2800-\u28FF\u2B00-\u2B2F\u2B45\u2B46\u2B4D-\u2B73\u2B76-\u2B95" \ + r"\u2B98-\u2BC8\u2BCA-\u2BFE\u2CE5-\u2CEA\u2E80-\u2E99\u2E9B-\u2EF3\u2F00-\u2FD5\u2FF0-\u2FFB" \ + r"\u3004\u3012\u3013\u3020\u3036\u3037\u303E\u303F\u3190\u3191\u3196-\u319F\u31C0-\u31E3" \ + r"\u3200-\u321E\u322A-\u3247\u3250\u3260-\u327F\u328A-\u32B0\u32C0-\u32FE\u3300-\u33FF\u4DC0-\u4DFF" \ + r"\uA490-\uA4C6\uA828-\uA82B\uA836\uA837\uA839\uAA77-\uAA79\uFDFD\uFFE4\uFFE8\uFFED\uFFEE\uFFFC" \ + r"\uFFFD\U00010137-\U0001013F\U00010179-\U00010189\U0001018C-\U0001018E\U00010190-\U0001019B" \ + r"\U000101A0\U000101D0-\U000101FC\U00010877\U00010878\U00010AC8\U0001173F\U00016B3C-\U00016B3F" \ + r"\U00016B45\U0001BC9C\U0001D000-\U0001D0F5\U0001D100-\U0001D126\U0001D129-\U0001D164" \ + r"\U0001D16A-\U0001D16C\U0001D183\U0001D184\U0001D18C-\U0001D1A9\U0001D1AE-\U0001D1E8" \ + r"\U0001D200-\U0001D241\U0001D245\U0001D300-\U0001D356\U0001D800-\U0001D9FF\U0001DA37-\U0001DA3A" \ + r"\U0001DA6D-\U0001DA74\U0001DA76-\U0001DA83\U0001DA85\U0001DA86\U0001ECAC\U0001F000-\U0001F02B" \ + r"\U0001F030-\U0001F093\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF" \ + r"\U0001F0D1-\U0001F0F5\U0001F110-\U0001F16B\U0001F170-\U0001F1AC\U0001F1E6-\U0001F202" \ + r"\U0001F210-\U0001F23B\U0001F240-\U0001F248\U0001F250\U0001F251\U0001F260-\U0001F265" \ + r"\U0001F300-\U0001F3FA\U0001F400-\U0001F6D4\U0001F6E0-\U0001F6EC\U0001F6F0-\U0001F6F9" \ + r"\U0001F700-\U0001F773\U0001F780-\U0001F7D8\U0001F800-\U0001F80B\U0001F810-\U0001F847" \ + r"\U0001F850-\U0001F859\U0001F860-\U0001F887\U0001F890-\U0001F8AD\U0001F900-\U0001F90B" \ + r"\U0001F910-\U0001F93E\U0001F940-\U0001F970\U0001F973-\U0001F976\U0001F97A\U0001F97C-\U0001F9A2" \ + r"\U0001F9B0-\U0001F9B9\U0001F9C0-\U0001F9C2\U0001F9D0-\U0001F9FF\U0001FA60-\U0001FA6D" UNITS = merge_chars(_units) CURRENCY = merge_chars(_currency) diff --git a/spacy/lang/fr/_tokenizer_exceptions_list.py b/spacy/lang/fr/_tokenizer_exceptions_list.py index 504bbbef1..c99ed06b4 100644 --- a/spacy/lang/fr/_tokenizer_exceptions_list.py +++ b/spacy/lang/fr/_tokenizer_exceptions_list.py @@ -7076,14 +7076,6 @@ FR_BASE_EXCEPTIONS = [ "au-lof", "au-tour", "aube-vigne", -"audio-numérique", -"audio-numériques", -"audio-prothésiste", -"audio-prothésistes", -"audio-visuel", -"audio-visuelle", -"audio-visuelles", -"audio-visuels", "aujourd'hui", "aulnaie-frênaie", "aulnaies-frênaies", @@ -14400,7 +14392,6 @@ FR_BASE_EXCEPTIONS = [ "attaques surprises", "attaques-surprises", "attrape-con", -"audio-oral", "auriculo-cardiaque", "auriculo-temporal", "austro-bavarois", diff --git a/spacy/lang/fr/tokenizer_exceptions.py b/spacy/lang/fr/tokenizer_exceptions.py index 29813d913..9b4385dc7 100644 --- a/spacy/lang/fr/tokenizer_exceptions.py +++ b/spacy/lang/fr/tokenizer_exceptions.py @@ -3,12 +3,15 @@ from __future__ import unicode_literals import re -from ._tokenizer_exceptions_list import FR_BASE_EXCEPTIONS from .punctuation import ELISION, HYPHENS from ..tokenizer_exceptions import URL_PATTERN from ..char_classes import ALPHA_LOWER, ALPHA from ...symbols import ORTH, LEMMA, TAG +# not using the large _tokenizer_exceptions_list by default as it slows down the tokenizer +# from ._tokenizer_exceptions_list import FR_BASE_EXCEPTIONS +FR_BASE_EXCEPTIONS = ["aujourd'hui", "Aujourd'hui"] + def upper_first_letter(text): if len(text) == 0: @@ -128,6 +131,7 @@ _hyphen_prefix = [ "arcs?", "archi", "arrières?", + "audio", "avant", "avion", "auto", diff --git a/spacy/lang/hu/tokenizer_exceptions.py b/spacy/lang/hu/tokenizer_exceptions.py index 8637e378f..c18a2cec2 100644 --- a/spacy/lang/hu/tokenizer_exceptions.py +++ b/spacy/lang/hu/tokenizer_exceptions.py @@ -636,17 +636,17 @@ for orth in [ _exc[orth] = [{ORTH: orth}] -_ord_num_or_date = "([A-Z0-9]+[./-])*(\d+\.?)" -_num = "[+\-]?\d+([,.]\d+)*" -_ops = "[=<>+\-\*/^()÷%²]" -_suffixes = "-[{al}]+".format(al=ALPHA_LOWER) -_numeric_exp = "({n})(({o})({n}))*[%]?".format(n=_num, o=_ops) -_time_exp = "\d+(:\d+)*(\.\d+)?" +_ord_num_or_date = r"([A-Z0-9]+[./-])*(\d+\.?)" +_num = r"[+\-]?\d+([,.]\d+)*" +_ops = r"[=<>+\-\*/^()÷%²]" +_suffixes = r"-[{al}]+".format(al=ALPHA_LOWER) +_numeric_exp = r"({n})(({o})({n}))*[%]?".format(n=_num, o=_ops) +_time_exp = r"\d+(:\d+)*(\.\d+)?" -_nums = "(({ne})|({t})|({on})|({c}))({s})?".format( +_nums = r"(({ne})|({t})|({on})|({c}))({s})?".format( ne=_numeric_exp, t=_time_exp, on=_ord_num_or_date, c=CURRENCY, s=_suffixes ) TOKENIZER_EXCEPTIONS = _exc -TOKEN_MATCH = re.compile("^({u})|({n})$".format(u=URL_PATTERN, n=_nums)).match +TOKEN_MATCH = re.compile(r"^({u})|({n})$".format(u=URL_PATTERN, n=_nums)).match diff --git a/spacy/lang/it/__init__.py b/spacy/lang/it/__init__.py index 28ce2f145..b858f7327 100644 --- a/spacy/lang/it/__init__.py +++ b/spacy/lang/it/__init__.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals from .stop_words import STOP_WORDS from .lemmatizer import LOOKUP from .tag_map import TAG_MAP +from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS from ..tokenizer_exceptions import BASE_EXCEPTIONS from ..norm_exceptions import BASE_NORMS @@ -20,7 +21,7 @@ class ItalianDefaults(Language.Defaults): lex_attr_getters[NORM] = add_lookups( Language.Defaults.lex_attr_getters[NORM], BASE_NORMS ) - tokenizer_exceptions = update_exc(BASE_EXCEPTIONS) + tokenizer_exceptions = update_exc(BASE_EXCEPTIONS, TOKENIZER_EXCEPTIONS) stop_words = STOP_WORDS lemma_lookup = LOOKUP tag_map = TAG_MAP diff --git a/spacy/lang/it/tokenizer_exceptions.py b/spacy/lang/it/tokenizer_exceptions.py new file mode 100644 index 000000000..cada9c65c --- /dev/null +++ b/spacy/lang/it/tokenizer_exceptions.py @@ -0,0 +1,9 @@ +# coding: utf8 +from __future__ import unicode_literals +from ...symbols import ORTH, LEMMA + +_exc = { + "po'": [{ORTH: "po'", LEMMA: 'poco'}] +} + +TOKENIZER_EXCEPTIONS = _exc diff --git a/spacy/lang/punctuation.py b/spacy/lang/punctuation.py index 17e20fa0c..ed2127d2c 100644 --- a/spacy/lang/punctuation.py +++ b/spacy/lang/punctuation.py @@ -39,10 +39,10 @@ _infixes = ( + LIST_ICONS + [ r"(?<=[0-9])[+\-\*^](?=[0-9-])", - r"(?<=[{al}])\.(?=[{au}])".format(al=ALPHA_LOWER, au=ALPHA_UPPER), + r"(?<=[{al}{q}])\.(?=[{au}{q}])".format(al=ALPHA_LOWER, au=ALPHA_UPPER, q=CONCAT_QUOTES), r"(?<=[{a}]),(?=[{a}])".format(a=ALPHA), r"(?<=[{a}])(?:{h})(?=[{a}])".format(a=ALPHA, h=HYPHENS), - r"(?<=[{a}])[:<>=/](?=[{a}])".format(a=ALPHA), + r"(?<=[{a}0-9])[:<>=/](?=[{a}])".format(a=ALPHA), ] ) diff --git a/spacy/matcher/matcher.pyx b/spacy/matcher/matcher.pyx index f0783df3f..5f17ec867 100644 --- a/spacy/matcher/matcher.pyx +++ b/spacy/matcher/matcher.pyx @@ -44,7 +44,7 @@ cdef find_matches(TokenPatternC** patterns, int n, Doc doc, extensions=None, cdef Pool mem = Pool() predicate_cache = mem.alloc(doc.length * len(predicates), sizeof(char)) if extensions is not None and len(extensions) >= 1: - nr_extra_attr = max(extensions.values()) + nr_extra_attr = max(extensions.values()) + 1 extra_attr_values = mem.alloc(doc.length * nr_extra_attr, sizeof(attr_t)) else: nr_extra_attr = 0 @@ -60,9 +60,8 @@ cdef find_matches(TokenPatternC** patterns, int n, Doc doc, extensions=None, for i in range(doc.length): for j in range(n): states.push_back(PatternStateC(patterns[j], i, 0)) - transition_states(states, matches, predicate_cache, + transition_states(states, matches, &predicate_cache[i], doc[i], extra_attr_values, predicates) - predicate_cache += nr_predicate extra_attr_values += nr_extra_attr # Handle matches that end in 0-width patterns finish_states(matches, states) @@ -74,6 +73,7 @@ cdef find_matches(TokenPatternC** patterns, int n, Doc doc, extensions=None, matches[i].start, matches[i].start+matches[i].length ) + # We need to deduplicate, because we could otherwise arrive at the same # match through two paths, e.g. .?.? matching 'a'. Are we matching the # first .?, or the second .? -- it doesn't matter, it's just one match. @@ -89,7 +89,8 @@ cdef attr_t get_ent_id(const TokenPatternC* pattern) nogil: # showed this wasn't the case when we had a reject-and-continue before a # match. I still don't really understand what's going on here, but this # workaround does resolve the issue. - while pattern.attrs.attr != ID and pattern.nr_attr > 0: + while pattern.attrs.attr != ID and \ + (pattern.nr_attr > 0 or pattern.nr_extra_attr > 0 or pattern.nr_py > 0): pattern += 1 return pattern.attrs.value @@ -101,13 +102,17 @@ cdef void transition_states(vector[PatternStateC]& states, vector[MatchC]& match cdef vector[PatternStateC] new_states cdef int nr_predicate = len(py_predicates) for i in range(states.size()): - if states[i].pattern.nr_py != 0: + if states[i].pattern.nr_py >= 1: update_predicate_cache(cached_py_predicates, states[i].pattern, token, py_predicates) + for i in range(states.size()): action = get_action(states[i], token.c, extra_attrs, - cached_py_predicates, nr_predicate) + cached_py_predicates) if action == REJECT: continue + # Keep only a subset of states (the active ones). Index q is the + # states which are still alive. If we reject a state, we overwrite + # it in the states list, because q doesn't advance. state = states[i] states[q] = state while action in (RETRY, RETRY_ADVANCE, RETRY_EXTEND): @@ -126,7 +131,7 @@ cdef void transition_states(vector[PatternStateC]& states, vector[MatchC]& match update_predicate_cache(cached_py_predicates, states[q].pattern, token, py_predicates) action = get_action(states[q], token.c, extra_attrs, - cached_py_predicates, nr_predicate) + cached_py_predicates) if action == REJECT: pass elif action == ADVANCE: @@ -154,8 +159,8 @@ cdef void transition_states(vector[PatternStateC]& states, vector[MatchC]& match states.push_back(new_states[i]) -cdef void update_predicate_cache(char* cache, - const TokenPatternC* pattern, Token token, predicates): +cdef int update_predicate_cache(char* cache, + const TokenPatternC* pattern, Token token, predicates) except -1: # If the state references any extra predicates, check whether they match. # These are cached, so that we don't call these potentially expensive # Python functions more than we need to. @@ -192,7 +197,7 @@ cdef void finish_states(vector[MatchC]& matches, vector[PatternStateC]& states) cdef action_t get_action(PatternStateC state, const TokenC* token, const attr_t* extra_attrs, - const char* predicate_matches, int nr_predicate) nogil: + const char* predicate_matches) nogil: '''We need to consider: a) Does the token match the specification? [Yes, No] @@ -252,7 +257,7 @@ cdef action_t get_action(PatternStateC state, Problem: If a quantifier is matching, we're adding a lot of open partials ''' cdef char is_match - is_match = get_is_match(state, token, extra_attrs, predicate_matches, nr_predicate) + is_match = get_is_match(state, token, extra_attrs, predicate_matches) quantifier = get_quantifier(state) is_final = get_is_final(state) if quantifier == ZERO: @@ -303,9 +308,9 @@ cdef action_t get_action(PatternStateC state, cdef char get_is_match(PatternStateC state, const TokenC* token, const attr_t* extra_attrs, - const char* predicate_matches, int nr_predicate) nogil: - for i in range(nr_predicate): - if predicate_matches[i] == -1: + const char* predicate_matches) nogil: + for i in range(state.pattern.nr_py): + if predicate_matches[state.pattern.py_predicates[i]] == -1: return 0 spec = state.pattern for attr in spec.attrs[:spec.nr_attr]: @@ -333,7 +338,7 @@ DEF PADDING = 5 cdef TokenPatternC* init_pattern(Pool mem, attr_t entity_id, object token_specs) except NULL: pattern = mem.alloc(len(token_specs) + 1, sizeof(TokenPatternC)) - cdef int i + cdef int i, index for i, (quantifier, spec, extensions, predicates) in enumerate(token_specs): pattern[i].quantifier = quantifier pattern[i].attrs = mem.alloc(len(spec), sizeof(AttrValueC)) @@ -356,11 +361,13 @@ cdef TokenPatternC* init_pattern(Pool mem, attr_t entity_id, object token_specs) pattern[i].attrs[0].attr = ID pattern[i].attrs[0].value = entity_id pattern[i].nr_attr = 0 + pattern[i].nr_extra_attr = 0 + pattern[i].nr_py = 0 return pattern cdef attr_t get_pattern_key(const TokenPatternC* pattern) nogil: - while pattern.nr_attr != 0: + while pattern.nr_attr != 0 or pattern.nr_extra_attr != 0 or pattern.nr_py != 0: pattern += 1 id_attr = pattern[0].attrs[0] if id_attr.attr != ID: @@ -384,7 +391,6 @@ def _preprocess_pattern(token_specs, string_store, extensions_table, extra_predi extra_predicates. """ tokens = [] - seen_predicates = {} for spec in token_specs: if not spec: # Signifier for 'any token' @@ -393,7 +399,7 @@ def _preprocess_pattern(token_specs, string_store, extensions_table, extra_predi ops = _get_operators(spec) attr_values = _get_attr_values(spec, string_store) extensions = _get_extensions(spec, string_store, extensions_table) - predicates = _get_extra_predicates(spec, extra_predicates, seen_predicates) + predicates = _get_extra_predicates(spec, extra_predicates) for op in ops: tokens.append((op, list(attr_values), list(extensions), list(predicates))) return tokens @@ -430,6 +436,7 @@ class _RegexPredicate(object): self.value = re.compile(value) self.predicate = predicate self.is_extension = is_extension + self.key = (attr, self.predicate, srsly.json_dumps(value, sort_keys=True)) assert self.predicate == 'REGEX' def __call__(self, Token token): @@ -447,6 +454,7 @@ class _SetMemberPredicate(object): self.value = set(get_string_id(v) for v in value) self.predicate = predicate self.is_extension = is_extension + self.key = (attr, self.predicate, srsly.json_dumps(value, sort_keys=True)) assert self.predicate in ('IN', 'NOT_IN') def __call__(self, Token token): @@ -459,6 +467,9 @@ class _SetMemberPredicate(object): else: return value not in self.value + def __repr__(self): + return repr(('SetMemberPredicate', self.i, self.attr, self.value, self.predicate)) + class _ComparisonPredicate(object): def __init__(self, i, attr, value, predicate, is_extension=False): @@ -467,6 +478,7 @@ class _ComparisonPredicate(object): self.value = value self.predicate = predicate self.is_extension = is_extension + self.key = (attr, self.predicate, srsly.json_dumps(value, sort_keys=True)) assert self.predicate in ('==', '!=', '>=', '<=', '>', '<') def __call__(self, Token token): @@ -488,7 +500,7 @@ class _ComparisonPredicate(object): return value < self.value -def _get_extra_predicates(spec, extra_predicates, seen_predicates): +def _get_extra_predicates(spec, extra_predicates): predicate_types = { 'REGEX': _RegexPredicate, 'IN': _SetMemberPredicate, @@ -499,6 +511,7 @@ def _get_extra_predicates(spec, extra_predicates, seen_predicates): '>': _ComparisonPredicate, '<': _ComparisonPredicate, } + seen_predicates = {pred.key: pred.i for pred in extra_predicates} output = [] for attr, value in spec.items(): if isinstance(attr, basestring): @@ -516,16 +529,15 @@ def _get_extra_predicates(spec, extra_predicates, seen_predicates): if isinstance(value, dict): for type_, cls in predicate_types.items(): if type_ in value: - key = (attr, type_, srsly.json_dumps(value[type_], sort_keys=True)) + predicate = cls(len(extra_predicates), attr, value[type_], type_) # Don't create a redundant predicates. # This helps with efficiency, as we're caching the results. - if key in seen_predicates: - output.append(seen_predicates[key]) + if predicate.key in seen_predicates: + output.append(seen_predicates[predicate.key]) else: - predicate = cls(len(extra_predicates), attr, value[type_], type_) extra_predicates.append(predicate) output.append(predicate.i) - seen_predicates[key] = predicate.i + seen_predicates[predicate.key] = predicate.i return output diff --git a/spacy/pipeline/pipes.pyx b/spacy/pipeline/pipes.pyx index 153e053e2..100b5abfd 100644 --- a/spacy/pipeline/pipes.pyx +++ b/spacy/pipeline/pipes.pyx @@ -145,9 +145,7 @@ class Pipe(object): """Serialize the pipe to a bytestring.""" serialize = OrderedDict() serialize["cfg"] = lambda: srsly.json_dumps(self.cfg) - if self.model in (True, False, None): - serialize["model"] = lambda: self.model - else: + if self.model not in (True, False, None): serialize["model"] = self.model.to_bytes serialize["vocab"] = self.vocab.to_bytes return util.to_bytes(serialize, exclude) @@ -538,9 +536,7 @@ class Tagger(Pipe): def to_bytes(self, **exclude): serialize = OrderedDict() - if self.model in (None, True, False): - serialize['model'] = lambda: self.model - else: + if self.model not in (None, True, False): serialize['model'] = self.model.to_bytes serialize['vocab'] = self.vocab.to_bytes serialize['cfg'] = lambda: srsly.json_dumps(self.cfg) diff --git a/spacy/syntax/_parser_model.pyx b/spacy/syntax/_parser_model.pyx index 657e30f41..5bfb0b410 100644 --- a/spacy/syntax/_parser_model.pyx +++ b/spacy/syntax/_parser_model.pyx @@ -261,13 +261,14 @@ class ParserStepModel(Model): # lowest score. # numpy's nan_to_num function doesn't take a value, and nan is replaced # by 0...-inf is replaced by minimum, so we go via that. Ugly to the max. - scores[self.ops.xp.isnan(scores)] = -self.ops.xp.inf - self.ops.xp.nan_to_num(scores, copy=False) + # Note that scores is always a numpy array! Should fix #3112 + scores[numpy.isnan(scores)] = -numpy.inf + numpy.nan_to_num(scores, copy=False) def backprop_parser_step(d_scores, sgd=None): # If we have a non-zero gradient for a previously unseen class, # replace the weight with 0. - new_classes = self.ops.xp.logical_and( + new_classes = self.vec2scores.ops.xp.logical_and( self.vec2scores.ops.xp.isnan(self.vec2scores.b), d_scores.any(axis=0) ) diff --git a/spacy/tests/lang/en/test_customized_tokenizer.py b/spacy/tests/lang/en/test_customized_tokenizer.py index ad371bf47..fdac32a90 100644 --- a/spacy/tests/lang/en/test_customized_tokenizer.py +++ b/spacy/tests/lang/en/test_customized_tokenizer.py @@ -13,10 +13,10 @@ def custom_en_tokenizer(en_vocab): prefix_re = compile_prefix_regex(English.Defaults.prefixes) suffix_re = compile_suffix_regex(English.Defaults.suffixes) custom_infixes = [ - "\.\.\.+", - "(?<=[0-9])-(?=[0-9])", - "[0-9]+(,[0-9]+)+", - "[\[\]!&:,()\*—–\/-]", + r"\.\.\.+", + r"(?<=[0-9])-(?=[0-9])", + r"[0-9]+(,[0-9]+)+", + r"[\[\]!&:,()\*—–\/-]", ] infix_re = compile_infix_regex(custom_infixes) return Tokenizer( diff --git a/spacy/tests/lang/fr/test_exceptions.py b/spacy/tests/lang/fr/test_exceptions.py index 3240af717..d0cafb7b6 100644 --- a/spacy/tests/lang/fr/test_exceptions.py +++ b/spacy/tests/lang/fr/test_exceptions.py @@ -22,9 +22,9 @@ import pytest u"11-septembre", u"11-Septembre", u"refox-trottâmes", - u"K-POP", - u"K-Pop", - u"K-pop", + # u"K-POP", + # u"K-Pop", + # u"K-pop", u"z'yeutes", u"black-outeront", u"états-unienne", diff --git a/spacy/tests/matcher/test_matcher_api.py b/spacy/tests/matcher/test_matcher_api.py index 7f7ebfc73..6ece07482 100644 --- a/spacy/tests/matcher/test_matcher_api.py +++ b/spacy/tests/matcher/test_matcher_api.py @@ -207,14 +207,13 @@ def test_matcher_set_value(en_vocab): assert len(matches) == 0 -@pytest.mark.xfail def test_matcher_set_value_operator(en_vocab): matcher = Matcher(en_vocab) pattern = [{"ORTH": {"IN": ["a", "the"]}, "OP": "?"}, {"ORTH": "house"}] matcher.add("DET_HOUSE", None, pattern) doc = Doc(en_vocab, words=["In", "a", "house"]) matches = matcher(doc) - assert len(matches) == 1 + assert len(matches) == 2 doc = Doc(en_vocab, words=["my", "house"]) matches = matcher(doc) assert len(matches) == 1 diff --git a/spacy/tests/regression/test_issue1-1000.py b/spacy/tests/regression/test_issue1-1000.py index 00024d74a..eec26ae20 100644 --- a/spacy/tests/regression/test_issue1-1000.py +++ b/spacy/tests/regression/test_issue1-1000.py @@ -358,6 +358,7 @@ def test_issue850_basic(): assert end == 4 +@pytest.mark.skip(reason="French exception list is not enabled in the default tokenizer anymore") @pytest.mark.parametrize( "text", ["au-delàs", "pair-programmâmes", "terra-formées", "σ-compacts"] ) diff --git a/spacy/tests/regression/test_issue1001-1500.py b/spacy/tests/regression/test_issue1001-1500.py index 0471502de..45bbc65aa 100644 --- a/spacy/tests/regression/test_issue1001-1500.py +++ b/spacy/tests/regression/test_issue1001-1500.py @@ -13,6 +13,22 @@ from spacy.lemmatizer import Lemmatizer from spacy.symbols import ORTH, LEMMA, POS, VERB, VerbForm_part +@pytest.mark.xfail( + reason="g is split of as a unit, as the suffix regular expression can not look back further (variable-width)" +) +def test_issue1235(): + """Test that g is not split of if preceded by a number and a letter""" + nlp = English() + testwords = u'e2g 2g 52g' + doc = nlp(testwords) + assert len(doc) == 5 + assert doc[0].text == "e2g" + assert doc[1].text == "2" + assert doc[2].text == "g" + assert doc[3].text == "52" + assert doc[4].text == "g" + + def test_issue1242(): nlp = English() doc = nlp("") diff --git a/spacy/tests/regression/test_issue1971.py b/spacy/tests/regression/test_issue1971.py index ecc7ebda1..e7273a5b0 100644 --- a/spacy/tests/regression/test_issue1971.py +++ b/spacy/tests/regression/test_issue1971.py @@ -6,7 +6,6 @@ from spacy.matcher import Matcher from spacy.tokens import Token, Doc -@pytest.mark.xfail def test_issue1971(en_vocab): # Possibly related to #2675 and #2671? matcher = Matcher(en_vocab) @@ -22,21 +21,20 @@ def test_issue1971(en_vocab): # We could also assert length 1 here, but this is more conclusive, because # the real problem here is that it returns a duplicate match for a match_id # that's not actually in the vocab! - assert all(match_id in en_vocab.strings for match_id, start, end in matcher(doc)) + matches = matcher(doc) + assert all([match_id in en_vocab.strings for match_id, start, end in matches]) -@pytest.mark.xfail def test_issue_1971_2(en_vocab): matcher = Matcher(en_vocab) - pattern1 = [{"LOWER": {"IN": ["eur"]}}, {"LIKE_NUM": True}] - pattern2 = list(reversed(pattern1)) + pattern1 = [{"ORTH": "EUR", "LOWER": {"IN": ["eur"]}}, {"LIKE_NUM": True}] + pattern2 = [{"LIKE_NUM": True}, {"ORTH": "EUR"}] #{"IN": ["EUR"]}}] doc = Doc(en_vocab, words=["EUR", "10", "is", "10", "EUR"]) - matcher.add("TEST", None, pattern1, pattern2) + matcher.add("TEST1", None, pattern1, pattern2) matches = matcher(doc) assert len(matches) == 2 -@pytest.mark.xfail def test_issue_1971_3(en_vocab): """Test that pattern matches correctly for multiple extension attributes.""" Token.set_extension("a", default=1) @@ -50,7 +48,6 @@ def test_issue_1971_3(en_vocab): assert matches == sorted([("A", 0, 1), ("A", 1, 2), ("B", 0, 1), ("B", 1, 2)]) -# @pytest.mark.xfail def test_issue_1971_4(en_vocab): """Test that pattern matches correctly with multiple extension attribute values on a single token. diff --git a/spacy/tests/regression/test_issue2001-2500.py b/spacy/tests/regression/test_issue2001-2500.py index 3922db6d3..81b7afa87 100644 --- a/spacy/tests/regression/test_issue2001-2500.py +++ b/spacy/tests/regression/test_issue2001-2500.py @@ -7,10 +7,22 @@ from spacy.displacy import render from spacy.gold import iob_to_biluo from spacy.lang.it import Italian import numpy +from spacy.lang.en import English from ..util import add_vecs_to_vocab, get_doc +@pytest.mark.xfail( + reason="The dot is now properly split off, but the prefix/suffix rules are not applied again afterwards." + "This means that the quote will still be attached to the remaining token." +) +def test_issue2070(): + """Test that checks that a dot followed by a quote is handled appropriately.""" + nlp = English() + doc = nlp('First sentence."A quoted sentence" he said ...') + assert len(doc) == 11 + + def test_issue2179(): """Test that spurious 'extra_labels' aren't created when initializing NER.""" nlp = Italian() diff --git a/spacy/tests/regression/test_issue2656.py b/spacy/tests/regression/test_issue2656.py new file mode 100644 index 000000000..ef51a10ce --- /dev/null +++ b/spacy/tests/regression/test_issue2656.py @@ -0,0 +1,24 @@ +# coding: utf8 +from __future__ import unicode_literals +from spacy.lang.en import English + + +def test_issue2656(): + """ Test that tokenizer correctly splits of punctuation after numbers with decimal points """ + text = "I went for 40.3, and got home by 10.0." + nlp = English() + doc = nlp(text) + + assert len(doc) == 11 + + assert doc[0].text == "I" + assert doc[1].text == "went" + assert doc[2].text == "for" + assert doc[3].text == "40.3" + assert doc[4].text == "," + assert doc[5].text == "and" + assert doc[6].text == "got" + assert doc[7].text == "home" + assert doc[8].text == "by" + assert doc[9].text == "10.0" + assert doc[10].text == "." diff --git a/spacy/tests/regression/test_issue2822.py b/spacy/tests/regression/test_issue2822.py new file mode 100644 index 000000000..f35f903df --- /dev/null +++ b/spacy/tests/regression/test_issue2822.py @@ -0,0 +1,21 @@ +# coding: utf8 +from __future__ import unicode_literals +from spacy.lang.it import Italian + + +def test_issue2822(): + """ Test that the abbreviation of poco is kept as one word """ + nlp = Italian() + text = "Vuoi un po' di zucchero?" + + doc = nlp(text) + + assert len(doc) == 6 + + assert doc[0].text == "Vuoi" + assert doc[1].text == "un" + assert doc[2].text == "po'" + assert doc[2].lemma_ == "poco" + assert doc[3].text == "di" + assert doc[4].text == "zucchero" + assert doc[5].text == "?" diff --git a/spacy/tests/regression/test_issue2926.py b/spacy/tests/regression/test_issue2926.py new file mode 100644 index 000000000..294b910de --- /dev/null +++ b/spacy/tests/regression/test_issue2926.py @@ -0,0 +1,21 @@ +# coding: utf8 +from __future__ import unicode_literals +from spacy.lang.fr import French + + +def test_issue2926(): + """ Test that the tokenizer correctly splits tokens separated by a slash (/) ending in a digit """ + nlp = French() + text = "Learn html5/css3/javascript/jquery" + doc = nlp(text) + + assert len(doc) == 8 + + assert doc[0].text == "Learn" + assert doc[1].text == "html5" + assert doc[2].text == "/" + assert doc[3].text == "css3" + assert doc[4].text == "/" + assert doc[5].text == "javascript" + assert doc[6].text == "/" + assert doc[7].text == "jquery" diff --git a/spacy/tests/regression/test_issue3002.py b/spacy/tests/regression/test_issue3002.py new file mode 100644 index 000000000..54e661d1f --- /dev/null +++ b/spacy/tests/regression/test_issue3002.py @@ -0,0 +1,11 @@ +# coding: utf8 +from __future__ import unicode_literals + +from spacy.lang.de import German + + +def test_issue3002(): + """Test that the tokenizer doesn't hang on a long list of dots""" + nlp = German() + doc = nlp('880.794.982.218.444.893.023.439.794.626.120.190.780.624.990.275.671 ist eine lange Zahl') + assert len(doc) == 5 diff --git a/spacy/tests/regression/test_issue3288.py b/spacy/tests/regression/test_issue3288.py index 5379a585a..f196fcc05 100644 --- a/spacy/tests/regression/test_issue3288.py +++ b/spacy/tests/regression/test_issue3288.py @@ -8,7 +8,6 @@ from spacy import displacy from ..util import get_doc -@pytest.mark.xfail def test_issue3288(en_vocab): """Test that retokenization works correctly via displaCy when punctuation is merged onto the preceeding token and tensor is resized.""" diff --git a/spacy/tests/regression/test_issue3289.py b/spacy/tests/regression/test_issue3289.py index 92b4ec853..ff423ddda 100644 --- a/spacy/tests/regression/test_issue3289.py +++ b/spacy/tests/regression/test_issue3289.py @@ -5,7 +5,6 @@ import pytest from spacy.lang.en import English -@pytest.mark.xfail def test_issue3289(): """Test that Language.to_bytes handles serializing a pipeline component with an uninitialized model.""" diff --git a/spacy/tests/tokenizer/test_naughty_strings.py b/spacy/tests/tokenizer/test_naughty_strings.py index b383c8fc8..36c69611e 100644 --- a/spacy/tests/tokenizer/test_naughty_strings.py +++ b/spacy/tests/tokenizer/test_naughty_strings.py @@ -9,104 +9,104 @@ import pytest NAUGHTY_STRINGS = [ # ASCII punctuation - ",./;'[]\-=", - '<>?:"{}|_+', - '!@#$%^&*()`~"', + r",./;'[]\-=", + r'<>?:"{}|_+', + r'!@#$%^&*()`~"', # Unicode additional control characters, byte order marks - "­؀؁؂؃؄؅؜۝܏᠎​‌‍‎‏‪", - "￾", + r"­؀؁؂؃؄؅؜۝܏᠎​‌‍‎‏‪", + r"￾", # Unicode Symbols - "Ω≈ç√∫˜µ≤≥÷", - "åß∂ƒ©˙∆˚¬…æ", + r"Ω≈ç√∫˜µ≤≥÷", + r"åß∂ƒ©˙∆˚¬…æ", "œ∑´®†¥¨ˆøπ“‘", - "¡™£¢∞§¶•ªº–≠", - "¸˛Ç◊ı˜Â¯˘¿", - "ÅÍÎÏ˝ÓÔÒÚÆ☃", - "Œ„´‰ˇÁ¨ˆØ∏”’", - "`⁄€‹›fifl‡°·‚—±", - "⅛⅜⅝⅞", - "ЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯабвгдежзийклмнопрстуфхцчшщъыьэюя", - "٠١٢٣٤٥٦٧٨٩", + r"¡™£¢∞§¶•ªº–≠", + r"¸˛Ç◊ı˜Â¯˘¿", + r"ÅÍÎÏ˝ÓÔÒÚÆ☃", + r"Œ„´‰ˇÁ¨ˆØ∏”’", + r"`⁄€‹›fifl‡°·‚—±", + r"⅛⅜⅝⅞", + r"ЁЂЃЄЅІЇЈЉЊЋЌЍЎЏАБВГДЕЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯабвгдежзийклмнопрстуфхцчшщъыьэюя", + r"٠١٢٣٤٥٦٧٨٩", # Unicode Subscript/Superscript/Accents - "⁰⁴⁵", - "₀₁₂", - "⁰⁴⁵₀₁₂", - "ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็ ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็ ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็", + r"⁰⁴⁵", + r"₀₁₂", + r"⁰⁴⁵₀₁₂", + r"ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็ ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็ ด้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็็้้้้้้้้็็็็็้้้้้็็็็", # Two-Byte Characters - "田中さんにあげて下さい", - "パーティーへ行かないか", - "和製漢語", - "部落格", - "사회과학원 어학연구소", - "찦차를 타고 온 펲시맨과 쑛다리 똠방각하", - "社會科學院語學研究所", - "울란바토르", - "𠜎𠜱𠝹𠱓𠱸𠲖𠳏", + r"田中さんにあげて下さい", + r"パーティーへ行かないか", + r"和製漢語", + r"部落格", + r"사회과학원 어학연구소", + r"찦차를 타고 온 펲시맨과 쑛다리 똠방각하", + r"社會科學院語學研究所", + r"울란바토르", + r"𠜎𠜱𠝹𠱓𠱸𠲖𠳏", # Japanese Emoticons - "ヽ༼ຈل͜ຈ༽ﾉ ヽ༼ຈل͜ຈ༽ﾉ", - "(｡◕ ∀ ◕｡)", - "｀ｨ(´∀｀∩", - "__ﾛ(,_,*)", - "・(￣∀￣)・:*:", - "ﾟ･✿ヾ╲(｡◕‿◕｡)╱✿･ﾟ", - ",。・:*:・゜’( ☻ ω ☻ )。・:*:・゜’", - "(╯°□°）╯︵ ┻━┻)" "(ﾉಥ益ಥ）ﾉ ┻━┻", - "┬─┬ノ( º _ ºノ)", - "( ͡° ͜ʖ ͡°)", + r"ヽ༼ຈل͜ຈ༽ﾉ ヽ༼ຈل͜ຈ༽ﾉ", + r"(｡◕ ∀ ◕｡)", + r"｀ｨ(´∀｀∩", + r"__ﾛ(,_,*)", + r"・(￣∀￣)・:*:", + r"ﾟ･✿ヾ╲(｡◕‿◕｡)╱✿･ﾟ", + r",。・:*:・゜’( ☻ ω ☻ )。・:*:・゜’", + r"(╯°□°）╯︵ ┻━┻)" "(ﾉಥ益ಥ）ﾉ ┻━┻", + r"┬─┬ノ( º _ ºノ)", + r"( ͡° ͜ʖ ͡°)", # Emoji - "😍", - "👩🏽", - "👾 🙇 💁 🙅 🙆 🙋 🙎 🙍", - "🐵 🙈 🙉 🙊", - "❤️ 💔 💌 💕 💞 💓 💗 💖 💘 💝 💟 💜 💛 💚 💙", - "✋🏿 💪🏿 👐🏿 🙌🏿 👏🏿 🙏🏿", - "🚾 🆒 🆓 🆕 🆖 🆗 🆙 🏧", - "0️⃣ 1️⃣ 2️⃣ 3️⃣ 4️⃣ 5️⃣ 6️⃣ 7️⃣ 8️⃣ 9️⃣ 🔟", + r"😍", + r"👩🏽", + r"👾 🙇 💁 🙅 🙆 🙋 🙎 🙍", + r"🐵 🙈 🙉 🙊", + r"❤️ 💔 💌 💕 💞 💓 💗 💖 💘 💝 💟 💜 💛 💚 💙", + r"✋🏿 💪🏿 👐🏿 🙌🏿 👏🏿 🙏🏿", + r"🚾 🆒 🆓 🆕 🆖 🆗 🆙 🏧", + r"0️⃣ 1️⃣ 2️⃣ 3️⃣ 4️⃣ 5️⃣ 6️⃣ 7️⃣ 8️⃣ 9️⃣ 🔟", # Regional Indicator Symbols - "🇺🇸🇷🇺🇸 🇦🇫🇦🇲🇸", - "🇺🇸🇷🇺🇸🇦🇫🇦🇲", - "🇺🇸🇷🇺🇸🇦", + r"🇺🇸🇷🇺🇸 🇦🇫🇦🇲🇸", + r"🇺🇸🇷🇺🇸🇦🇫🇦🇲", + r"🇺🇸🇷🇺🇸🇦", # Unicode Numbers - "１２３", - "١٢٣", + r"１２３", + r"١٢٣", # Right-To-Left Strings - "ثم نفس سقطت وبالتحديد،, جزيرتي باستخدام أن دنو. إذ هنا؟ الستار وتنصيب كان. أهّل ايطاليا، بريطانيا-فرنسا قد أخذ. سليمان، إتفاقية بين ما, يذكر الحدود أي بعد, معاملة بولندا، الإطلاق عل إيو.", - "إيو.", - "בְּרֵאשִׁית, בָּרָא אֱלֹהִים, אֵת הַשָּׁמַיִם, וְאֵת הָאָרֶץ", - "הָיְתָהtestالصفحات التّحول", - "﷽", - "ﷺ", - "مُنَاقَشَةُ سُبُلِ اِسْتِخْدَامِ اللُّغَةِ فِي النُّظُمِ الْقَائِمَةِ وَفِيم يَخُصَّ التَّطْبِيقَاتُ الْحاسُوبِيَّةُ،", + r"ثم نفس سقطت وبالتحديد،, جزيرتي باستخدام أن دنو. إذ هنا؟ الستار وتنصيب كان. أهّل ايطاليا، بريطانيا-فرنسا قد أخذ. سليمان، إتفاقية بين ما, يذكر الحدود أي بعد, معاملة بولندا، الإطلاق عل إيو.", + r"إيو.", + r"בְּרֵאשִׁית, בָּרָא אֱלֹהִים, אֵת הַשָּׁמַיִם, וְאֵת הָאָרֶץ", + r"הָיְתָהtestالصفحات التّحول", + r"﷽", + r"ﷺ", + r"مُنَاقَشَةُ سُبُلِ اِسْتِخْدَامِ اللُّغَةِ فِي النُّظُمِ الْقَائِمَةِ وَفِيم يَخُصَّ التَّطْبِيقَاتُ الْحاسُوبِيَّةُ،", # Trick Unicode - "‪‪test‪", - "‫test", - "
test
", - "test⁠test", - "⁦test⁧", + r"‪‪test‪", + r"‫test", + r"
test
", + r"test⁠test", + r"⁦test⁧", # Zalgo Text - "Ṱ̺̺̕o͞ ̷i̲̬͇̪͙n̝̗͕v̟̜̘̦͟o̶̙̰̠kè͚̮̺̪̹̱̤ ̖t̝͕̳̣̻̪͞h̼͓̲̦̳̘̲e͇̣̰̦̬͎ ̢̼̻̱̘h͚͎͙̜̣̲ͅi̦̲̣̰̤v̻͍e̺̭̳̪̰-m̢iͅn̖̺̞̲̯̰d̵̼̟͙̩̼̘̳ ̞̥̱̳̭r̛̗̘e͙p͠r̼̞̻̭̗e̺̠̣͟s̘͇̳͍̝͉e͉̥̯̞̲͚̬͜ǹ̬͎͎̟̖͇̤t͍̬̤͓̼̭͘ͅi̪̱n͠g̴͉ ͏͉ͅc̬̟h͡a̫̻̯͘o̫̟̖͍̙̝͉s̗̦̲.̨̹͈̣", - "̡͓̞ͅI̗̘̦͝n͇͇͙v̮̫ok̲̫̙͈i̖͙̭̹̠̞n̡̻̮̣̺g̲͈͙̭͙̬͎ ̰t͔̦h̞̲e̢̤ ͍̬̲͖f̴̘͕̣è͖ẹ̥̩l͖͔͚i͓͚̦͠n͖͍̗͓̳̮g͍ ̨o͚̪͡f̘̣̬ ̖̘͖̟͙̮c҉͔̫͖͓͇͖ͅh̵̤̣͚͔á̗̼͕ͅo̼̣̥s̱͈̺̖̦̻͢.̛̖̞̠̫̰", - "̗̺͖̹̯͓Ṯ̤͍̥͇͈h̲́e͏͓̼̗̙̼̣͔ ͇̜̱̠͓͍ͅN͕͠e̗̱z̘̝̜̺͙p̤̺̹͍̯͚e̠̻̠͜r̨̤͍̺̖͔̖̖d̠̟̭̬̝͟i̦͖̩͓͔̤a̠̗̬͉̙n͚͜ ̻̞̰͚ͅh̵͉i̳̞v̢͇ḙ͎͟-҉̭̩̼͔m̤̭̫i͕͇̝̦n̗͙ḍ̟ ̯̲͕͞ǫ̟̯̰̲͙̻̝f ̪̰̰̗̖̭̘͘c̦͍̲̞͍̩̙ḥ͚a̮͎̟̙͜ơ̩̹͎s̤.̝̝ ҉Z̡̖̜͖̰̣͉̜a͖̰͙̬͡l̲̫̳͍̩g̡̟̼̱͚̞̬ͅo̗͜.̟", - "̦H̬̤̗̤͝e͜ ̜̥̝̻͍̟́w̕h̖̯͓o̝͙̖͎̱̮ ҉̺̙̞̟͈W̷̼̭a̺̪͍į͈͕̭͙̯̜t̶̼̮s̘͙͖̕ ̠̫̠B̻͍͙͉̳ͅe̵h̵̬͇̫͙i̹͓̳̳̮͎̫̕n͟d̴̪̜̖ ̰͉̩͇͙̲͞ͅT͖̼͓̪͢h͏͓̮̻e̬̝̟ͅ ̤̹̝W͙̞̝͔͇͝ͅa͏͓͔̹̼̣l̴͔̰̤̟͔ḽ̫.͕", - "Z̮̞̠͙͔ͅḀ̗̞͈̻̗Ḷ͙͎̯̹̞͓G̻O̭̗̮", + r"Ṱ̺̺̕o͞ ̷i̲̬͇̪͙n̝̗͕v̟̜̘̦͟o̶̙̰̠kè͚̮̺̪̹̱̤ ̖t̝͕̳̣̻̪͞h̼͓̲̦̳̘̲e͇̣̰̦̬͎ ̢̼̻̱̘h͚͎͙̜̣̲ͅi̦̲̣̰̤v̻͍e̺̭̳̪̰-m̢iͅn̖̺̞̲̯̰d̵̼̟͙̩̼̘̳ ̞̥̱̳̭r̛̗̘e͙p͠r̼̞̻̭̗e̺̠̣͟s̘͇̳͍̝͉e͉̥̯̞̲͚̬͜ǹ̬͎͎̟̖͇̤t͍̬̤͓̼̭͘ͅi̪̱n͠g̴͉ ͏͉ͅc̬̟h͡a̫̻̯͘o̫̟̖͍̙̝͉s̗̦̲.̨̹͈̣", + r"̡͓̞ͅI̗̘̦͝n͇͇͙v̮̫ok̲̫̙͈i̖͙̭̹̠̞n̡̻̮̣̺g̲͈͙̭͙̬͎ ̰t͔̦h̞̲e̢̤ ͍̬̲͖f̴̘͕̣è͖ẹ̥̩l͖͔͚i͓͚̦͠n͖͍̗͓̳̮g͍ ̨o͚̪͡f̘̣̬ ̖̘͖̟͙̮c҉͔̫͖͓͇͖ͅh̵̤̣͚͔á̗̼͕ͅo̼̣̥s̱͈̺̖̦̻͢.̛̖̞̠̫̰", + r"̗̺͖̹̯͓Ṯ̤͍̥͇͈h̲́e͏͓̼̗̙̼̣͔ ͇̜̱̠͓͍ͅN͕͠e̗̱z̘̝̜̺͙p̤̺̹͍̯͚e̠̻̠͜r̨̤͍̺̖͔̖̖d̠̟̭̬̝͟i̦͖̩͓͔̤a̠̗̬͉̙n͚͜ ̻̞̰͚ͅh̵͉i̳̞v̢͇ḙ͎͟-҉̭̩̼͔m̤̭̫i͕͇̝̦n̗͙ḍ̟ ̯̲͕͞ǫ̟̯̰̲͙̻̝f ̪̰̰̗̖̭̘͘c̦͍̲̞͍̩̙ḥ͚a̮͎̟̙͜ơ̩̹͎s̤.̝̝ ҉Z̡̖̜͖̰̣͉̜a͖̰͙̬͡l̲̫̳͍̩g̡̟̼̱͚̞̬ͅo̗͜.̟", + r"̦H̬̤̗̤͝e͜ ̜̥̝̻͍̟́w̕h̖̯͓o̝͙̖͎̱̮ ҉̺̙̞̟͈W̷̼̭a̺̪͍į͈͕̭͙̯̜t̶̼̮s̘͙͖̕ ̠̫̠B̻͍͙͉̳ͅe̵h̵̬͇̫͙i̹͓̳̳̮͎̫̕n͟d̴̪̜̖ ̰͉̩͇͙̲͞ͅT͖̼͓̪͢h͏͓̮̻e̬̝̟ͅ ̤̹̝W͙̞̝͔͇͝ͅa͏͓͔̹̼̣l̴͔̰̤̟͔ḽ̫.͕", + r"Z̮̞̠͙͔ͅḀ̗̞͈̻̗Ḷ͙͎̯̹̞͓G̻O̭̗̮", # Unicode Upsidedown - "˙ɐnbᴉlɐ ɐuƃɐɯ ǝɹolop ʇǝ ǝɹoqɐl ʇn ʇunpᴉpᴉɔuᴉ ɹodɯǝʇ poɯsnᴉǝ op pǝs 'ʇᴉlǝ ƃuᴉɔsᴉdᴉpɐ ɹnʇǝʇɔǝsuoɔ 'ʇǝɯɐ ʇᴉs ɹolop ɯnsdᴉ ɯǝɹo˥", - "00˙Ɩ$-", + r"˙ɐnbᴉlɐ ɐuƃɐɯ ǝɹolop ʇǝ ǝɹoqɐl ʇn ʇunpᴉpᴉɔuᴉ ɹodɯǝʇ poɯsnᴉǝ op pǝs 'ʇᴉlǝ ƃuᴉɔsᴉdᴉpɐ ɹnʇǝʇɔǝsuoɔ 'ʇǝɯɐ ʇᴉs ɹolop ɯnsdᴉ ɯǝɹo˥", + r"00˙Ɩ$-", # Unicode font - "Ｔｈｅ ｑｕｉｃｋ ｂｒｏｗｎ ｆｏｘ ｊｕｍｐｓ ｏｖｅｒ ｔｈｅ ｌａｚｙ ｄｏｇ", - "𝐓𝐡𝐞 𝐪𝐮𝐢𝐜𝐤 𝐛𝐫𝐨𝐰𝐧 𝐟𝐨𝐱 𝐣𝐮𝐦𝐩𝐬 𝐨𝐯𝐞𝐫 𝐭𝐡𝐞 𝐥𝐚𝐳𝐲 𝐝𝐨𝐠", - "𝕿𝖍𝖊 𝖖𝖚𝖎𝖈𝖐 𝖇𝖗𝖔𝖜𝖓 𝖋𝖔𝖝 𝖏𝖚𝖒𝖕𝖘 𝖔𝖛𝖊𝖗 𝖙𝖍𝖊 𝖑𝖆𝖟𝖞 𝖉𝖔𝖌", - "𝑻𝒉𝒆 𝒒𝒖𝒊𝒄𝒌 𝒃𝒓𝒐𝒘𝒏 𝒇𝒐𝒙 𝒋𝒖𝒎𝒑𝒔 𝒐𝒗𝒆𝒓 𝒕𝒉𝒆 𝒍𝒂𝒛𝒚 𝒅𝒐𝒈", - "𝓣𝓱𝓮 𝓺𝓾𝓲𝓬𝓴 𝓫𝓻𝓸𝔀𝓷 𝓯𝓸𝔁 𝓳𝓾𝓶𝓹𝓼 𝓸𝓿𝓮𝓻 𝓽𝓱𝓮 𝓵𝓪𝔃𝔂 𝓭𝓸𝓰", - "𝕋𝕙𝕖 𝕢𝕦𝕚𝕔𝕜 𝕓𝕣𝕠𝕨𝕟 𝕗𝕠𝕩 𝕛𝕦𝕞𝕡𝕤 𝕠𝕧𝕖𝕣 𝕥𝕙𝕖 𝕝𝕒𝕫𝕪 𝕕𝕠𝕘", - "𝚃𝚑𝚎 𝚚𝚞𝚒𝚌𝚔 𝚋𝚛𝚘𝚠𝚗 𝚏𝚘𝚡 𝚓𝚞𝚖𝚙𝚜 𝚘𝚟𝚎𝚛 𝚝𝚑𝚎 𝚕𝚊𝚣𝚢 𝚍𝚘𝚐", - "⒯⒣⒠ ⒬⒰⒤⒞⒦ ⒝⒭⒪⒲⒩ ⒡⒪⒳ ⒥⒰⒨⒫⒮ ⒪⒱⒠⒭ ⒯⒣⒠ ⒧⒜⒵⒴ ⒟⒪⒢", + r"Ｔｈｅ ｑｕｉｃｋ ｂｒｏｗｎ ｆｏｘ ｊｕｍｐｓ ｏｖｅｒ ｔｈｅ ｌａｚｙ ｄｏｇ", + r"𝐓𝐡𝐞 𝐪𝐮𝐢𝐜𝐤 𝐛𝐫𝐨𝐰𝐧 𝐟𝐨𝐱 𝐣𝐮𝐦𝐩𝐬 𝐨𝐯𝐞𝐫 𝐭𝐡𝐞 𝐥𝐚𝐳𝐲 𝐝𝐨𝐠", + r"𝕿𝖍𝖊 𝖖𝖚𝖎𝖈𝖐 𝖇𝖗𝖔𝖜𝖓 𝖋𝖔𝖝 𝖏𝖚𝖒𝖕𝖘 𝖔𝖛𝖊𝖗 𝖙𝖍𝖊 𝖑𝖆𝖟𝖞 𝖉𝖔𝖌", + r"𝑻𝒉𝒆 𝒒𝒖𝒊𝒄𝒌 𝒃𝒓𝒐𝒘𝒏 𝒇𝒐𝒙 𝒋𝒖𝒎𝒑𝒔 𝒐𝒗𝒆𝒓 𝒕𝒉𝒆 𝒍𝒂𝒛𝒚 𝒅𝒐𝒈", + r"𝓣𝓱𝓮 𝓺𝓾𝓲𝓬𝓴 𝓫𝓻𝓸𝔀𝓷 𝓯𝓸𝔁 𝓳𝓾𝓶𝓹𝓼 𝓸𝓿𝓮𝓻 𝓽𝓱𝓮 𝓵𝓪𝔃𝔂 𝓭𝓸𝓰", + r"𝕋𝕙𝕖 𝕢𝕦𝕚𝕔𝕜 𝕓𝕣𝕠𝕨𝕟 𝕗𝕠𝕩 𝕛𝕦𝕞𝕡𝕤 𝕠𝕧𝕖𝕣 𝕥𝕙𝕖 𝕝𝕒𝕫𝕪 𝕕𝕠𝕘", + r"𝚃𝚑𝚎 𝚚𝚞𝚒𝚌𝚔 𝚋𝚛𝚘𝚠𝚗 𝚏𝚘𝚡 𝚓𝚞𝚖𝚙𝚜 𝚘𝚟𝚎𝚛 𝚝𝚑𝚎 𝚕𝚊𝚣𝚢 𝚍𝚘𝚐", + r"⒯⒣⒠ ⒬⒰⒤⒞⒦ ⒝⒭⒪⒲⒩ ⒡⒪⒳ ⒥⒰⒨⒫⒮ ⒪⒱⒠⒭ ⒯⒣⒠ ⒧⒜⒵⒴ ⒟⒪⒢", # File paths - "../../../../../../../../../../../etc/passwd%00", - "../../../../../../../../../../../etc/hosts", + r"../../../../../../../../../../../etc/passwd%00", + r"../../../../../../../../../../../etc/hosts", # iOS Vulnerabilities - "Powerلُلُصّبُلُلصّبُررً ॣ ॣh ॣ ॣ冗", - "🏳0🌈️", + r"Powerلُلُصّبُلُلصّبُررً ॣ ॣh ॣ ॣ冗", + r"🏳0🌈️", ] diff --git a/spacy/tokenizer.pyx b/spacy/tokenizer.pyx index 7fcff8d30..15a5f7274 100644 --- a/spacy/tokenizer.pyx +++ b/spacy/tokenizer.pyx @@ -407,7 +407,7 @@ cdef class Tokenizer: if data.get('infix_finditer'): self.infix_finditer = re.compile(data['infix_finditer']).finditer if data.get('token_match'): - self.token_match = re.compile(data['token_match']).search + self.token_match = re.compile(data['token_match']).match for string, substrings in data.get('rules', {}).items(): self.add_special_case(string, substrings) return self diff --git a/spacy/tokens/_retokenize.pyx b/spacy/tokens/_retokenize.pyx index 0355e3200..fd621922f 100644 --- a/spacy/tokens/_retokenize.pyx +++ b/spacy/tokens/_retokenize.pyx @@ -222,7 +222,7 @@ def _bulk_merge(Doc doc, merges): # whether the row is to be deleted, then use numpy.delete if doc.tensor is not None and doc.tensor.size != 0: doc.tensor = _resize_tensor(doc.tensor, - [(m[1][0].start, m[1][0].end) for m in merges]) + [(m[0].start, m[0].end) for m in merges]) # Memorize span roots and sets dependencies of the newly merged # tokens to the dependencies of their roots. span_roots = [] diff --git a/website/docs/usage/adding-languages.md b/website/docs/usage/adding-languages.md index ca7d26b74..6718bc7fd 100644 --- a/website/docs/usage/adding-languages.md +++ b/website/docs/usage/adding-languages.md @@ -102,11 +102,12 @@ language and training a language model. In order for the tokenizer to split suffixes, prefixes and infixes, spaCy needs to know the language's character set. If the language you're adding uses -non-latin characters, you might need to add the required character classes to +non-latin characters, you might need to define the required character classes in the global [`char_classes.py`](https://github.com/explosion/spaCy/tree/master/spacy/lang/char_classes.py). -spaCy uses the [`regex` library](https://pypi.python.org/pypi/regex/) to keep -this simple and readable. If the language requires very specific punctuation +For efficiency, spaCy uses hard-coded unicode ranges to define character classes, +the definitions of which can be found on [Wikipedia](https://en.wikipedia.org/wiki/Unicode_block). +If the language requires very specific punctuation rules, you should consider overwriting the default regular expressions with your own in the language's `Defaults`. diff --git a/website/docs/usage/v2-1.md b/website/docs/usage/v2-1.md index e000d73d7..bdf0cfa1f 100644 --- a/website/docs/usage/v2-1.md +++ b/website/docs/usage/v2-1.md @@ -215,6 +215,22 @@ if all of your models are up to date, you can run the means that the `Matcher` in v2.1.x may produce different results compared to the `Matcher` in v2.0.x. +- The deprecated [`Doc.merge`](/api/doc#merge) and + [`Span.merge`](/api/span#merge) methods still work, but you may notice that + they now run slower when merging many objects in a row. That's because the + merging engine was rewritten to be more reliable and to support more efficient + merging **in bulk**. To take advantage of this, you should rewrite your logic + to use the [`Doc.retokenize`](/api/doc#retokenize) context manager and perform + as many merges as possible together in the `with` block. + + ```diff + - doc[1:5].merge() + - doc[6:8].merge() + + with doc.retokenize() as retokenizer: + + retokenizer.merge(doc[1:5]) + + retokenizer.merge(doc[6:8]) + ``` + - For better compatibility with the Universal Dependencies data, the lemmatizer now preserves capitalization, e.g. for proper nouns. See [this issue](https://github.com/explosion/spaCy/issues/3256) for details. @@ -227,6 +243,11 @@ if all of your models are up to date, you can run the + sentence_splitter = nlp.create_pipe("sentencizer") ``` +- The keyword argument `n_threads` on the `.pipe` methods is now deprecated, as + the v2.x models cannot release the global interpreter lock. (Future versions + may introduce a `n_process` argument for parallel inference via + multiprocessing.) + - The `Doc.print_tree` method is now deprecated. If you need a custom nested JSON representation of a `Doc` object, you might want to write your own helper function. For a simple and consistent JSON representation of the `Doc` object diff --git a/website/netlify.toml b/website/netlify.toml index 7774e67cf..066bd94ef 100644 --- a/website/netlify.toml +++ b/website/netlify.toml @@ -1,4 +1,7 @@ redirects = [ + # Netlify + # TODO: uncomment once the site is switched over + # {from = "https://spacy.netlify.com/*", to="https://spacy.io/:splat" }, # Old demos {from = "/demos/*", to = "https://explosion.ai/demos/:splat"}, # Old blog diff --git a/website/src/components/seo.js b/website/src/components/seo.js index 4da6ab63a..4fe45299c 100644 --- a/website/src/components/seo.js +++ b/website/src/components/seo.js @@ -71,11 +71,11 @@ const SEO = ({ description, lang, title, section, sectionTitle, bodyClass }) => }, { name: 'twitter:creator', - content: `@${data.site.siteMetadata.social.twitter}`, + content: `@${siteMetadata.social.twitter}`, }, { name: 'twitter:site', - content: `@${data.site.siteMetadata.social.twitter}`, + content: `@${siteMetadata.social.twitter}`, }, { name: 'twitter:title', @@ -126,8 +126,6 @@ const query = graphql` title description slogan - siteUrl - email social { twitter }