release 0.2.0

Author: JarbasAl

CHANGELOG.md

@@ -0,0 +1,22 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.2.0]  - 2019-12-12
+
+### Changed
+
+- Transfered ownership to [OpenJarbas](https://github.com/OpenJarbas)
+- Made a changelog
+
+### Fixed
+
+- updated deprecated mycroft_lang_utils to lingua_franca
+
+[unreleased]: https://github.com/OpenJarbas/simple_NER/tree/dev
+[0.1.5]: https://github.com/OpenJarbas/simple_NER/tree/0.1.5

examples/neural.py

@@ -4,21 +4,15 @@
 ner.add_rule("name", "my name is {person}")
 
 for ent in ner.extract_entities("the name is jarbas"):
-    assert ent.as_json() == {'confidence': 0.5251495787186434,
-                             'data': {},
-                             'entity_type': 'person',
-                             'rules': [{'name': 'name',
-                                        'rules': ['my name is {person}']}],
-                             'source_text': 'the name is jarbas',
-                             'spans': [(12, 18)],
-                             'value': 'jarbas'}
+    assert ent.as_json()["value"] == 'jarbas'
 
 for ent in ner.extract_entities("name is kevin"):
-    assert ent.as_json() == {'confidence': 0.8363423970007801,
-                             'data': {},
-                             'entity_type': 'person',
-                             'rules': [{'name': 'name',
-                                        'rules': ['my name is {person}']}],
-                             'source_text': 'name is kevin',
-                             'spans': [(8, 13)],
-                             'value': 'kevin'}
+    # {'confidence': 0.8363423970007801,
+    #                              'data': {},
+    #                              'entity_type': 'person',
+    #                              'rules': [{'name': 'name',
+    #                                         'rules': ['my name is {person}']}],
+    #                              'source_text': 'name is kevin',
+    #                              'spans': [(8, 13)],
+    #                              'value': 'kevin'}
+    assert ent.as_json()["value"] == 'kevin'

requirements.txt

@@ -5,4 +5,4 @@ pyspotlight
 fann2==1.0.7
 padatious==0.4.5
 nltk
-mycroft_lang_utils
+lingua_franca==0.1.0

setup.py

@@ -14,17 +14,17 @@ def package_files(directory):
 
 setup(
     name='simple_NER',
-    version='0.1.14',
+    version='0.2.0',
     packages=['simple_NER', 'simple_NER.rules', 'simple_NER.annotators',
               'simple_NER.annotators.remote', 'simple_NER.annotators.utils',
               'simple_NER.annotators.utils.keywords'],
-    url='https://github.com/JarbasAl/simple_NER',
+    url='https://github.com/OpenJarbas/simple_NER',
     package_data={'': extra_files},
     include_package_data=True,
     license='MIT',
     author='jarbasAI',
     install_requires=["padaos", "fann2==1.0.7", "padatious", "nltk",
-                      "quantulum3", "requests", "mycroft_lang_utils",
+                      "quantulum3", "requests", "lingua_franca==0.1.0",
                       "pyspotlight"],
     author_email='jarbasai@mailfence.com',
     description='rule based NER'

simple_NER/annotators/date.py

@@ -4,10 +4,9 @@
 import re
 from datetime import timedelta, datetime
 from dateutil.relativedelta import relativedelta
-from mycroft_lang_utils.lang.parse_en import _convert_words_to_numbers, \
+from lingua_franca.lang.parse_en import _convert_words_to_numbers_en, \
     is_numeric, extractnumber_en
-from mycroft_lang_utils.format import nice_duration, nice_date
-from mycroft_lang_utils.time import now_local
+from lingua_franca.format import nice_duration, nice_date
 
 
 def _annotate_datetime_en(string, dateNow=None, default_time=None):
@@ -37,7 +36,7 @@ def _annotate_datetime_en(string, dateNow=None, default_time=None):
                          text consumed in the parsing, or None if no
                          date or time related text was found.
     """
-    dateNow = dateNow or now_local()
+    dateNow = dateNow or datetime.now()
 
     def clean_string(s):
         # clean unneeded punctuation and capitalization among other things.
@@ -854,7 +853,7 @@ def _annotate_duration_en(text):
     }
 
     pattern = r"(?P<value>\d+(?:\.?\d+)?)\s+{unit}s?"
-    norm_text = _convert_words_to_numbers(text)
+    norm_text = _convert_words_to_numbers_en(text)
     t = norm_text
     duration_text = text
 
@@ -915,7 +914,7 @@ def _annotate_duration_en(text):
 class DateTimeNER(NERWrapper):
     def __init__(self, anchor_date=None):
         super().__init__()
-        self.anchor_date = anchor_date or now_local()
+        self.anchor_date = anchor_date or datetime.now()
         self.add_detector(self.annotate_datetime)
         self.add_detector(self.annotate_duration)
 
@@ -972,24 +971,6 @@ def annotate(self, text):
             }
             yield Entity(value, "date", source_text=text, data=data)
 
-    def _old_annotate(self, text):
-        # deprecated
-        import datefinder
-        matches = datefinder.find_dates(text, index=True)
-        for date, span in matches:
-            value = text[span[0]:span[1]].strip()
-            data = {
-                "timestamp": date.timestamp(),
-                "isoformat": date.isoformat(),
-                "weekday": date.isoweekday(),
-                "month": date.month,
-                "day": date.day,
-                "hour": date.hour,
-                "minute": date.minute,
-                "year": date.year
-            }
-            yield Entity(value, "date", source_text=text, data=data)
-
 
 if __name__ == "__main__":
     from pprint import pprint