Formatting

akoumjian · Dec 27, 2018 · 5e24e8d · 5e24e8d
1 parent 6602b1f
commit 5e24e8d
Show file tree

Hide file tree

Showing 4 changed files with 108 additions and 97 deletions.
diff --git a/README.rst b/README.rst
@@ -21,9 +21,8 @@ datefinder - extract dates from text
 A python module for locating dates inside text. Use this package to extract all sorts 
 of date like strings from a document and turn them into datetime objects.
 
-This module finds the likely datetime strings and then uses the  
-`dateparser <https://github.com/scrapinghub/dateparser>`_ package to convert 
-to the datetime object.
+This module finds the likely datetime strings and then uses  
+`dateutil`to convert to the datetime object.
 
 
 Installation

diff --git a/datefinder/__init__.py b/datefinder/__init__.py
@@ -3,9 +3,15 @@
 import regex as re
 from dateutil import tz, parser
 
-from .constants import REPLACEMENTS, TIMEZONE_REPLACEMENTS, STRIP_CHARS, DATE_REGEX, RANGE_REGEX
+from .constants import (
+    REPLACEMENTS,
+    TIMEZONE_REPLACEMENTS,
+    STRIP_CHARS,
+    DATE_REGEX,
+    RANGE_REGEX,
+)
 
-logger = logging.getLogger('datefinder')
+logger = logging.getLogger("datefinder")
 
 
 class DateFinder(object):
@@ -18,7 +24,9 @@ def __init__(self, base_date=None):
 
     def find_dates(self, text, source=False, index=False, strict=False):
 
-        for date_string, indices, captures in self.extract_date_strings(text, strict=strict):
+        for date_string, indices, captures in self.extract_date_strings(
+            text, strict=strict
+        ):
 
             as_dt = self.parse_date_string(date_string, captures)
             if as_dt is None:
@@ -44,8 +52,8 @@ def _find_and_replace(self, date_string, captures):
         """
         # add timezones to replace
         cloned_replacements = copy.copy(REPLACEMENTS)  # don't mutate
-        for tz_string in captures.get('timezones', []):
-            cloned_replacements.update({tz_string: ' '})
+        for tz_string in captures.get("timezones", []):
+            cloned_replacements.update({tz_string: " "})
 
         date_string = date_string.lower()
         for key, replacement in cloned_replacements.items():
@@ -55,9 +63,14 @@ def _find_and_replace(self, date_string, captures):
             # 2. match ' to'
             # 3. match ' to '
             # but never match r'(\s|)to(\s|)' which would make 'october' > 'ocber'
-            date_string = re.sub(r'(^|\s)' + key + r'(\s|$)', replacement, date_string, flags=re.IGNORECASE)
+            date_string = re.sub(
+                r"(^|\s)" + key + r"(\s|$)",
+                replacement,
+                date_string,
+                flags=re.IGNORECASE,
+            )
 
-        return date_string, self._pop_tz_string(sorted(captures.get('timezones', [])))
+        return date_string, self._pop_tz_string(sorted(captures.get("timezones", [])))
 
     def _pop_tz_string(self, list_of_timezones):
         try:
@@ -66,7 +79,7 @@ def _pop_tz_string(self, list_of_timezones):
             # want replaced with better abbreviation
             return TIMEZONE_REPLACEMENTS.get(tz_string, tz_string)
         except IndexError:
-            return ''
+            return ""
 
     def _add_tzinfo(self, datetime_obj, tz_string):
         """
@@ -98,7 +111,7 @@ def parse_date_string(self, date_string, captures):
                 return None
 
             try:
-                logger.debug('Parsing {0} with dateutil'.format(date_string))
+                logger.debug("Parsing {0} with dateutil".format(date_string))
                 as_dt = parser.parse(date_string, default=self.base_date)
             except Exception as e:
                 logger.debug(e)
@@ -127,12 +140,12 @@ def extract_date_strings(self, text, strict=False):
 
             for dt2_str in dt2:
                 range_strings.extend(self.extract_date_strings(dt2_str, strict=strict))
-            
+
             found_range = True
 
         for range_string in range_strings:
             yield range_string
-        
+
         # Try to match regular datetimes if no ranges have been found
         if not found_range:
             for match in DATE_REGEX.finditer(text):
@@ -142,10 +155,10 @@ def extract_date_strings(self, text, strict=False):
                 ## Get individual group matches
                 captures = match.capturesdict()
                 # time = captures.get('time')
-                digits = captures.get('digits')
+                digits = captures.get("digits")
                 # digits_modifiers = captures.get('digits_modifiers')
                 # days = captures.get('days')
-                months = captures.get('months')
+                months = captures.get("months")
                 # timezones = captures.get('timezones')
                 # delimiters = captures.get('delimiters')
                 # time_periods = captures.get('time_periods')
@@ -155,28 +168,24 @@ def extract_date_strings(self, text, strict=False):
                     complete = False
                     if len(digits) == 3:  # 12-05-2015
                         complete = True
-                    elif (len(months) == 1) and (len(digits) == 2):  # 19 February 2013 year 09:10
+                    elif (len(months) == 1) and (
+                        len(digits) == 2
+                    ):  # 19 February 2013 year 09:10
                         complete = True
 
                     if not complete:
                         continue
 
                 ## sanitize date string
                 ## replace unhelpful whitespace characters with single whitespace
-                match_str = re.sub(r'[\n\t\s\xa0]+', ' ', match_str)
+                match_str = re.sub(r"[\n\t\s\xa0]+", " ", match_str)
                 match_str = match_str.strip(STRIP_CHARS)
 
                 ## Save sanitized source string
-                yield match_str, indices, captures            
+                yield match_str, indices, captures
 
 
-def find_dates(
-    text,
-    source=False,
-    index=False,
-    strict=False,
-    base_date=None
-    ):
+def find_dates(text, source=False, index=False, strict=False, base_date=None):
     """
     Extract datetime strings from text
 

diff --git a/datefinder/constants.py b/datefinder/constants.py
@@ -1,37 +1,48 @@
 import regex as re
 
-NUMBERS_PATTERN = r'first|second|third|fourth|fifth|sixth|seventh|eighth|nineth|tenth'
-POSITIONNAL_TOKENS= r'next|last'
-DIGITS_PATTERN = r'\d+'
-DIGITS_SUFFIXES= r'st|th|rd|nd'
-DAYS_PATTERN = 'monday|tuesday|wednesday|thursday|friday|saturday|sunday|mon|tue|tues|wed|thur|thurs|fri|sat|sun'
-MONTHS_PATTERN = r'january|february|march|april|may|june|july|august|september|october|november|december|jan\.?|feb\.?|mar\.?|apr\.?|may\.?|jun\.?|jul\.?|aug\.?|sep\.?|sept\.?|oct\.?|nov\.?|dec\.?'
-TIMEZONES_PATTERN = 'ACDT|ACST|ACT|ACWDT|ACWST|ADDT|ADMT|ADT|AEDT|AEST|AFT|AHDT|AHST|AKDT|AKST|AKTST|AKTT|ALMST|ALMT|AMST|AMT|ANAST|ANAT|ANT|APT|AQTST|AQTT|ARST|ART|ASHST|ASHT|AST|AWDT|AWST|AWT|AZOMT|AZOST|AZOT|AZST|AZT|BAKST|BAKT|BDST|BDT|BEAT|BEAUT|BIOT|BMT|BNT|BORT|BOST|BOT|BRST|BRT|BST|BTT|BURT|CANT|CAPT|CAST|CAT|CAWT|CCT|CDDT|CDT|CEDT|CEMT|CEST|CET|CGST|CGT|CHADT|CHAST|CHDT|CHOST|CHOT|CIST|CKHST|CKT|CLST|CLT|CMT|COST|COT|CPT|CST|CUT|CVST|CVT|CWT|CXT|ChST|DACT|DAVT|DDUT|DFT|DMT|DUSST|DUST|EASST|EAST|EAT|ECT|EDDT|EDT|EEDT|EEST|EET|EGST|EGT|EHDT|EMT|EPT|EST|ET|EWT|FET|FFMT|FJST|FJT|FKST|FKT|FMT|FNST|FNT|FORT|FRUST|FRUT|GALT|GAMT|GBGT|GEST|GET|GFT|GHST|GILT|GIT|GMT|GST|GYT|HAA|HAC|HADT|HAE|HAP|HAR|HAST|HAT|HAY|HDT|HKST|HKT|HLV|HMT|HNA|HNC|HNE|HNP|HNR|HNT|HNY|HOVST|HOVT|HST|ICT|IDDT|IDT|IHST|IMT|IOT|IRDT|IRKST|IRKT|IRST|ISST|IST|JAVT|JCST|JDT|JMT|JST|JWST|KART|KDT|KGST|KGT|KIZST|KIZT|KMT|KOST|KRAST|KRAT|KST|KUYST|KUYT|KWAT|LHDT|LHST|LINT|LKT|LMT|LMT|LMT|LMT|LRT|LST|MADMT|MADST|MADT|MAGST|MAGT|MALST|MALT|MART|MAWT|MDDT|MDST|MDT|MEST|MET|MHT|MIST|MIT|MMT|MOST|MOT|MPT|MSD|MSK|MSM|MST|MUST|MUT|MVT|MWT|MYT|NCST|NCT|NDDT|NDT|NEGT|NEST|NET|NFT|NMT|NOVST|NOVT|NPT|NRT|NST|NT|NUT|NWT|NZDT|NZMT|NZST|OMSST|OMST|ORAST|ORAT|PDDT|PDT|PEST|PET|PETST|PETT|PGT|PHOT|PHST|PHT|PKST|PKT|PLMT|PMDT|PMMT|PMST|PMT|PNT|PONT|PPMT|PPT|PST|PT|PWT|PYST|PYT|QMT|QYZST|QYZT|RET|RMT|ROTT|SAKST|SAKT|SAMT|SAST|SBT|SCT|SDMT|SDT|SET|SGT|SHEST|SHET|SJMT|SLT|SMT|SRET|SRT|SST|STAT|SVEST|SVET|SWAT|SYOT|TAHT|TASST|TAST|TBIST|TBIT|TBMT|TFT|THA|TJT|TKT|TLT|TMT|TOST|TOT|TRST|TRT|TSAT|TVT|ULAST|ULAT|URAST|URAT|UTC|UYHST|UYST|UYT|UZST|UZT|VET|VLAST|VLAT|VOLST|VOLT|VOST|VUST|VUT|WARST|WART|WAST|WAT|WDT|WEDT|WEMT|WEST|WET|WFT|WGST|WGT|WIB|WIT|WITA|WMT|WSDT|WSST|WST|WT|XJT|YAKST|YAKT|YAPT|YDDT|YDT|YEKST|YEKST|YEKT|YEKT|YERST|YERT|YPT|YST|YWT|zzz'
+NUMBERS_PATTERN = r"first|second|third|fourth|fifth|sixth|seventh|eighth|nineth|tenth"
+POSITIONNAL_TOKENS = r"next|last"
+DIGITS_PATTERN = r"\d+"
+DIGITS_SUFFIXES = r"st|th|rd|nd"
+DAYS_PATTERN = "monday|tuesday|wednesday|thursday|friday|saturday|sunday|mon|tue|tues|wed|thur|thurs|fri|sat|sun"
+MONTHS_PATTERN = r"january|february|march|april|may|june|july|august|september|october|november|december|jan\.?|feb\.?|mar\.?|apr\.?|may\.?|jun\.?|jul\.?|aug\.?|sep\.?|sept\.?|oct\.?|nov\.?|dec\.?"
+TIMEZONES_PATTERN = "ACDT|ACST|ACT|ACWDT|ACWST|ADDT|ADMT|ADT|AEDT|AEST|AFT|AHDT|AHST|AKDT|AKST|AKTST|AKTT|ALMST|ALMT|AMST|AMT|ANAST|ANAT|ANT|APT|AQTST|AQTT|ARST|ART|ASHST|ASHT|AST|AWDT|AWST|AWT|AZOMT|AZOST|AZOT|AZST|AZT|BAKST|BAKT|BDST|BDT|BEAT|BEAUT|BIOT|BMT|BNT|BORT|BOST|BOT|BRST|BRT|BST|BTT|BURT|CANT|CAPT|CAST|CAT|CAWT|CCT|CDDT|CDT|CEDT|CEMT|CEST|CET|CGST|CGT|CHADT|CHAST|CHDT|CHOST|CHOT|CIST|CKHST|CKT|CLST|CLT|CMT|COST|COT|CPT|CST|CUT|CVST|CVT|CWT|CXT|ChST|DACT|DAVT|DDUT|DFT|DMT|DUSST|DUST|EASST|EAST|EAT|ECT|EDDT|EDT|EEDT|EEST|EET|EGST|EGT|EHDT|EMT|EPT|EST|ET|EWT|FET|FFMT|FJST|FJT|FKST|FKT|FMT|FNST|FNT|FORT|FRUST|FRUT|GALT|GAMT|GBGT|GEST|GET|GFT|GHST|GILT|GIT|GMT|GST|GYT|HAA|HAC|HADT|HAE|HAP|HAR|HAST|HAT|HAY|HDT|HKST|HKT|HLV|HMT|HNA|HNC|HNE|HNP|HNR|HNT|HNY|HOVST|HOVT|HST|ICT|IDDT|IDT|IHST|IMT|IOT|IRDT|IRKST|IRKT|IRST|ISST|IST|JAVT|JCST|JDT|JMT|JST|JWST|KART|KDT|KGST|KGT|KIZST|KIZT|KMT|KOST|KRAST|KRAT|KST|KUYST|KUYT|KWAT|LHDT|LHST|LINT|LKT|LMT|LMT|LMT|LMT|LRT|LST|MADMT|MADST|MADT|MAGST|MAGT|MALST|MALT|MART|MAWT|MDDT|MDST|MDT|MEST|MET|MHT|MIST|MIT|MMT|MOST|MOT|MPT|MSD|MSK|MSM|MST|MUST|MUT|MVT|MWT|MYT|NCST|NCT|NDDT|NDT|NEGT|NEST|NET|NFT|NMT|NOVST|NOVT|NPT|NRT|NST|NT|NUT|NWT|NZDT|NZMT|NZST|OMSST|OMST|ORAST|ORAT|PDDT|PDT|PEST|PET|PETST|PETT|PGT|PHOT|PHST|PHT|PKST|PKT|PLMT|PMDT|PMMT|PMST|PMT|PNT|PONT|PPMT|PPT|PST|PT|PWT|PYST|PYT|QMT|QYZST|QYZT|RET|RMT|ROTT|SAKST|SAKT|SAMT|SAST|SBT|SCT|SDMT|SDT|SET|SGT|SHEST|SHET|SJMT|SLT|SMT|SRET|SRT|SST|STAT|SVEST|SVET|SWAT|SYOT|TAHT|TASST|TAST|TBIST|TBIT|TBMT|TFT|THA|TJT|TKT|TLT|TMT|TOST|TOT|TRST|TRT|TSAT|TVT|ULAST|ULAT|URAST|URAT|UTC|UYHST|UYST|UYT|UZST|UZT|VET|VLAST|VLAT|VOLST|VOLT|VOST|VUST|VUT|WARST|WART|WAST|WAT|WDT|WEDT|WEMT|WEST|WET|WFT|WGST|WGT|WIB|WIT|WITA|WMT|WSDT|WSST|WST|WT|XJT|YAKST|YAKT|YAPT|YDDT|YDT|YEKST|YEKST|YEKT|YEKT|YERST|YERT|YPT|YST|YWT|zzz"
 ## explicit north american timezones that get replaced
-NA_TIMEZONES_PATTERN = 'pacific|eastern|mountain|central'
-ALL_TIMEZONES_PATTERN = TIMEZONES_PATTERN + '|' + NA_TIMEZONES_PATTERN
-DELIMITERS_PATTERN = r'[/\:\-\,\s\_\+\@]+'
+NA_TIMEZONES_PATTERN = "pacific|eastern|mountain|central"
+ALL_TIMEZONES_PATTERN = TIMEZONES_PATTERN + "|" + NA_TIMEZONES_PATTERN
+DELIMITERS_PATTERN = r"[/\:\-\,\s\_\+\@]+"
 
 # Allows for straightforward datestamps e.g 2017, 201712, 20171223. Created with:
 #  YYYYMM_PATTERN = '|'.join(['19\d\d'+'{:0>2}'.format(mon)+'|20\d\d'+'{:0>2}'.format(mon) for mon in range(1, 13)])
 #  YYYYMMDD_PATTERN = '|'.join(['19\d\d'+'{:0>2}'.format(mon)+'[0123]\d|20\d\d'+'{:0>2}'.format(mon)+'[0123]\d' for mon in range(1, 13)])
-YYYY_PATTERN = r'19\d\d|20\d\d'
-YYYYMM_PATTERN = r'19\d\d01|20\d\d01|19\d\d02|20\d\d02|19\d\d03|20\d\d03|19\d\d04|20\d\d04|19\d\d05|20\d\d05|19\d\d06|20\d\d06|19\d\d07|20\d\d07|19\d\d08|20\d\d08|19\d\d09|20\d\d09|19\d\d10|20\d\d10|19\d\d11|20\d\d11|19\d\d12|20\d\d12'
-YYYYMMDD_PATTERN = r'19\d\d01[0123]\d|20\d\d01[0123]\d|19\d\d02[0123]\d|20\d\d02[0123]\d|19\d\d03[0123]\d|20\d\d03[0123]\d|19\d\d04[0123]\d|20\d\d04[0123]\d|19\d\d05[0123]\d|20\d\d05[0123]\d|19\d\d06[0123]\d|20\d\d06[0123]\d|19\d\d07[0123]\d|20\d\d07[0123]\d|19\d\d08[0123]\d|20\d\d08[0123]\d|19\d\d09[0123]\d|20\d\d09[0123]\d|19\d\d10[0123]\d|20\d\d10[0123]\d|19\d\d11[0123]\d|20\d\d11[0123]\d|19\d\d12[0123]\d|20\d\d12[0123]\d'
-YYYYMMDDHHMMSS_PATTERN = '|'.join([r'19\d\d' + '{:0>2}'.format(mon) + r'[0-3]\d[0-5]\d[0-5]\d[0-5]\d|20\d\d' + '{:0>2}'.format(mon) + r'[0-3]\d[0-5]\d[0-5]\d[0-5]\d' for mon in range(1, 13)])
-ISO8601_PATTERN = r'(?P<years>-?(\:[1-9][0-9]*)?[0-9]{4})\-(?P<months>1[0-2]|0[1-9])\-(?P<days>3[01]|0[1-9]|[12][0-9])T(?P<hours>2[0-3]|[01][0-9])\:(?P<minutes>[0-5][0-9]):(?P<seconds>[0-5][0-9])(?:[\.,]+(?P<microseconds>[0-9]+))?(?P<offset>(?:Z|[+-](?:2[0-3]|[01][0-9])\:[0-5][0-9]))?'
-UNDELIMITED_STAMPS_PATTERN = '|'.join([YYYYMMDDHHMMSS_PATTERN, YYYYMMDD_PATTERN, YYYYMM_PATTERN, ISO8601_PATTERN])
-DELIMITERS_PATTERN = r'[/\:\-\,\.\s\_\+\@]+'
-TIME_PERIOD_PATTERN = r'a\.m\.|am|p\.m\.|pm'
+YYYY_PATTERN = r"19\d\d|20\d\d"
+YYYYMM_PATTERN = r"19\d\d01|20\d\d01|19\d\d02|20\d\d02|19\d\d03|20\d\d03|19\d\d04|20\d\d04|19\d\d05|20\d\d05|19\d\d06|20\d\d06|19\d\d07|20\d\d07|19\d\d08|20\d\d08|19\d\d09|20\d\d09|19\d\d10|20\d\d10|19\d\d11|20\d\d11|19\d\d12|20\d\d12"
+YYYYMMDD_PATTERN = r"19\d\d01[0123]\d|20\d\d01[0123]\d|19\d\d02[0123]\d|20\d\d02[0123]\d|19\d\d03[0123]\d|20\d\d03[0123]\d|19\d\d04[0123]\d|20\d\d04[0123]\d|19\d\d05[0123]\d|20\d\d05[0123]\d|19\d\d06[0123]\d|20\d\d06[0123]\d|19\d\d07[0123]\d|20\d\d07[0123]\d|19\d\d08[0123]\d|20\d\d08[0123]\d|19\d\d09[0123]\d|20\d\d09[0123]\d|19\d\d10[0123]\d|20\d\d10[0123]\d|19\d\d11[0123]\d|20\d\d11[0123]\d|19\d\d12[0123]\d|20\d\d12[0123]\d"
+YYYYMMDDHHMMSS_PATTERN = "|".join(
+    [
+        r"19\d\d"
+        + "{:0>2}".format(mon)
+        + r"[0-3]\d[0-5]\d[0-5]\d[0-5]\d|20\d\d"
+        + "{:0>2}".format(mon)
+        + r"[0-3]\d[0-5]\d[0-5]\d[0-5]\d"
+        for mon in range(1, 13)
+    ]
+)
+ISO8601_PATTERN = r"(?P<years>-?(\:[1-9][0-9]*)?[0-9]{4})\-(?P<months>1[0-2]|0[1-9])\-(?P<days>3[01]|0[1-9]|[12][0-9])T(?P<hours>2[0-3]|[01][0-9])\:(?P<minutes>[0-5][0-9]):(?P<seconds>[0-5][0-9])(?:[\.,]+(?P<microseconds>[0-9]+))?(?P<offset>(?:Z|[+-](?:2[0-3]|[01][0-9])\:[0-5][0-9]))?"
+UNDELIMITED_STAMPS_PATTERN = "|".join(
+    [YYYYMMDDHHMMSS_PATTERN, YYYYMMDD_PATTERN, YYYYMM_PATTERN, ISO8601_PATTERN]
+)
+DELIMITERS_PATTERN = r"[/\:\-\,\.\s\_\+\@]+"
+TIME_PERIOD_PATTERN = r"a\.m\.|am|p\.m\.|pm"
 ## can be in date strings but not recognized by dateutils
-EXTRA_TOKENS_PATTERN = r'due|by|on|during|standard|daylight|savings|time|date|dated|of|to|through|between|until|at|day'
+EXTRA_TOKENS_PATTERN = r"due|by|on|during|standard|daylight|savings|time|date|dated|of|to|through|between|until|at|day"
 
 ## TODO: Get english numbers?
 ## http://www.rexegg.com/regex-trick-numbers-in-english.html
 
-RELATIVE_PATTERN = 'before|after|next|last|ago'
-TIME_SHORTHAND_PATTERN = 'noon|midnight|today|yesterday'
-UNIT_PATTERN = 'second|minute|hour|day|week|month|year'
+RELATIVE_PATTERN = "before|after|next|last|ago"
+TIME_SHORTHAND_PATTERN = "noon|midnight|today|yesterday"
+UNIT_PATTERN = "second|minute|hour|day|week|month|year"
 
 ## Time pattern is used independently, so specified here.
 TIME_PATTERN = r"""
@@ -60,8 +71,7 @@
     )
 )
 """.format(
-    time_periods=TIME_PERIOD_PATTERN,
-    timezones=ALL_TIMEZONES_PATTERN
+    time_periods=TIME_PERIOD_PATTERN, timezones=ALL_TIMEZONES_PATTERN
 )
 
 DATES_PATTERN = """
@@ -111,7 +121,7 @@
     months=MONTHS_PATTERN,
     delimiters=DELIMITERS_PATTERN,
     positionnal_tokens=POSITIONNAL_TOKENS,
-    extra_tokens=EXTRA_TOKENS_PATTERN
+    extra_tokens=EXTRA_TOKENS_PATTERN,
 )
 
 RANGE_PATTERN = r"""
@@ -120,13 +130,21 @@
     [\s]?(to|through)[\s]?
     (?P<dt2>{date_pattern})
 )
-""".format(date_pattern=DATES_PATTERN)
+""".format(
+    date_pattern=DATES_PATTERN
+)
 
-DATE_REGEX = re.compile(DATES_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE)
+DATE_REGEX = re.compile(
+    DATES_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE
+)
 
-TIME_REGEX = re.compile(TIME_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE)
+TIME_REGEX = re.compile(
+    TIME_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE
+)
 
-RANGE_REGEX = re.compile(RANGE_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE)
+RANGE_REGEX = re.compile(
+    RANGE_PATTERN, re.IGNORECASE | re.MULTILINE | re.UNICODE | re.DOTALL | re.VERBOSE
+)
 
 ## These tokens can be in original text but dateutil
 ## won't handle them without modification
@@ -151,4 +169,5 @@
 }
 
 ## Characters that can be removed from ends of matched strings
-STRIP_CHARS = ' \n\t:-.,_'
+STRIP_CHARS = " \n\t:-.,_"
+