Adding after and next keywords with basic functionality.

dateparser/data/date_translation_data/en.py

@@ -190,6 +190,16 @@
             "in (\\d+) years",
             "in (\\d+) yr"
         ],
+        "next \\1 year": [
+            "next (\\d+) year",
+            "next (\\d+) years",
+            "next (\\d+) yr"
+        ],
+        "after \\1 year": [
+            "after (\\d+) year",
+            "after (\\d+) years",
+            "after (\\d+) yr"
+        ],
         "\\1 year ago": [
             "(\\d+) year ago",
             "(\\d+) years ago",
@@ -200,6 +210,16 @@
             "in (\\d+) months",
             "in (\\d+) mo"
         ],
+        "next \\1 month": [
+            "next (\\d+) month",
+            "next (\\d+) months",
+            "next (\\d+) mo"
+        ],
+        "after \\1 month": [
+            "after (\\d+) month",
+            "after (\\d+) months",
+            "after (\\d+) mo"
+        ],
         "\\1 month ago": [
             "(\\d+) month ago",
             "(\\d+) months ago",
@@ -210,6 +230,16 @@
             "in (\\d+) weeks",
             "in (\\d+) wk"
         ],
+        "next \\1 week": [
+            "next (\\d+) week",
+            "next (\\d+) weeks",
+            "next (\\d+) wk"
+        ],
+        "after \\1 week": [
+            "after (\\d+) week",
+            "after (\\d+) weeks",
+            "after (\\d+) wk"
+        ],
         "\\1 week ago": [
             "(\\d+) week ago",
             "(\\d+) weeks ago",
@@ -219,6 +249,14 @@
             "in (\\d+) day",
             "in (\\d+) days"
         ],
+        "next \\1 day": [
+            "next (\\d+) day",
+            "next (\\d+) days"
+        ],
+        "after \\1 day": [
+            "after (\\d+) day",
+            "after (\\d+) days"
+        ],
         "\\1 day ago": [
             "(\\d+) day ago",
             "(\\d+) days ago"
@@ -228,6 +266,16 @@
             "in (\\d+) hours",
             "in (\\d+) hr"
         ],
+        "next \\1 hour": [
+            "next (\\d+) hour",
+            "next (\\d+) hours",
+            "next (\\d+) hr"
+        ],
+        "after \\1 hour": [
+            "after (\\d+) hour",
+            "after (\\d+) hours",
+            "after (\\d+) hr"
+        ],
         "\\1 hour ago": [
             "(\\d+) hour ago",
             "(\\d+) hours ago",
@@ -238,6 +286,16 @@
             "in (\\d+) minutes",
             "in (\\d+) min"
         ],
+        "next \\1 minute": [
+            "next (\\d+) minute",
+            "next (\\d+) minutes",
+            "next (\\d+) min"
+        ],
+        "after \\1 minute": [
+            "after (\\d+) minute",
+            "after (\\d+) minutes",
+            "after (\\d+) min"
+        ],
         "\\1 minute ago": [
             "(\\d+) minute ago",
             "(\\d+) minutes ago",
@@ -248,6 +306,16 @@
             "in (\\d+) seconds",
             "in (\\d+) sec"
         ],
+        "next \\1 second": [
+            "next (\\d+) second",
+            "next (\\d+) seconds",
+            "next (\\d+) sec"
+        ],
+        "after \\1 second": [
+            "after (\\d+) second",
+            "after (\\d+) seconds",
+            "after (\\d+) sec"
+        ],
         "\\1 second ago": [
             "(\\d+) second ago",
             "(\\d+) seconds ago",
@@ -318,36 +386,72 @@
                 "in \\1 year": [
                     "in (\\d+) yrs"
                 ],
+                "next \\1 year": [
+                    "next (\\d+) yrs"
+                ],
+                "after \\1 year": [
+                    "after (\\d+) yrs"
+                ],
                 "\\1 year ago": [
                     "(\\d+) yrs ago"
                 ],
                 "in \\1 month": [
                     "in (\\d+) mos"
                 ],
+                "next \\1 month": [
+                    "next (\\d+) mos"
+                ],
+                "after \\1 month": [
+                    "after (\\d+) mos"
+                ],
                 "\\1 month ago": [
                     "(\\d+) mos ago"
                 ],
                 "in \\1 week": [
                     "in (\\d+) wks"
                 ],
+                "next \\1 week": [
+                    "next (\\d+) wks"
+                ],
+                "after \\1 week": [
+                    "after (\\d+) wks"
+                ],
                 "\\1 week ago": [
                     "(\\d+) wks ago"
                 ],
                 "in \\1 hour": [
                     "in (\\d+) hrs"
                 ],
+                "next \\1 hour": [
+                    "next (\\d+) hrs"
+                ],
+                "after \\1 hour": [
+                    "after (\\d+) hrs"
+                ],
                 "\\1 hour ago": [
                     "(\\d+) hrs ago"
                 ],
                 "in \\1 minute": [
                     "in (\\d+) mins"
                 ],
+                "next \\1 minute": [
+                    "next (\\d+) mins"
+                ],
+                "after \\1 minute": [
+                    "after (\\d+) mins"
+                ],
                 "\\1 minute ago": [
                     "(\\d+) mins ago"
                 ],
                 "in \\1 second": [
                     "in (\\d+) secs"
                 ],
+                "next \\1 second": [
+                    "next (\\d+) secs"
+                ],
+                "after \\1 second": [
+                    "after (\\d+) secs"
+                ],
                 "\\1 second ago": [
                     "(\\d+) secs ago"
                 ]
@@ -761,6 +865,9 @@
         "in",
         "from now"
     ],
+    "next": [
+        "next"
+    ],
     "simplifications": [
         {
             "an": "1"
@@ -820,4 +927,4 @@
             "twelve": "12"
         }
     ]
-}
+}

dateparser/freshness_date_parser.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import regex as re
+import calendar as cal
 from datetime import datetime
 from datetime import time
 from tzlocal import get_localzone
@@ -15,6 +16,7 @@
 
 _UNITS = r'year|month|week|day|hour|minute|second'
 PATTERN = re.compile(r'(\d+)\s*(%s)\b' % _UNITS, re.I | re.S | re.U)
+_WEEKDAYS = r'monday|tuesday|wednesday|thursday|friday|saturday|sunday'
 
 
 class FreshnessDateDataParser(object):
@@ -24,7 +26,7 @@ def __init__(self):
 
     def _are_all_words_units(self, date_string):
         skip = [_UNITS,
-                r'ago|in|\d+',
+                r'ago|in|next|after|\d+',
                 r':|[ap]m']
 
         date_string = re.sub(r'\s+', ' ', date_string.strip())
@@ -36,7 +38,7 @@ def _are_all_words_units(self, date_string):
     def _parse_time(self, date_string, settings):
         """Attempts to parse time part of date strings like '1 day ago, 2 PM' """
         date_string = PATTERN.sub('', date_string)
-        date_string = re.sub(r'\b(?:ago|in)\b', '', date_string)
+        date_string = re.sub(r'\b(?:ago|in|next|after)\b', '', date_string)
         try:
             return time_parser(date_string)
         except:
@@ -111,23 +113,38 @@ def apply_time(dateobj, timeobj):
         return date, period
 
     def _parse_date(self, date_string, prefer_dates_from):
-        if not self._are_all_words_units(date_string):
+
+        _weekday = self.get_weekday_data(date_string)
+
+        if not self._are_all_words_units(date_string) and not _weekday:
             return None, None
 
         kwargs = self.get_kwargs(date_string)
-        if not kwargs:
+
+        if not kwargs and not _weekday:
             return None, None
 
         period = 'day'
-        if 'days' not in kwargs:
-            for k in ['weeks', 'months', 'years']:
-                if k in kwargs:
-                    period = k[:-1]
-                    break
+        if _weekday:
+            day = getattr(cal, _weekday.upper())
+            day_ahead = day - self.now.weekday()
+            if day_ahead <= 0:
+                day_ahead += 7
 
-        td = relativedelta(**kwargs)
+            td = relativedelta(days=day_ahead)
+
+        else:
+            if 'days' not in kwargs:
+                for k in ['weeks', 'months', 'years']:
+                    if k in kwargs:
+                        period = k[:-1]
+                        break
+
+            td = relativedelta(**kwargs)
         if (
             re.search(r'\bin\b', date_string) or
+            re.search(r'\bnext\b', date_string) or
+            re.search(r'\bafter\b', date_string) or
             ('future' in prefer_dates_from and
              not re.search(r'\bago\b', date_string))
         ):
@@ -151,5 +168,14 @@ def get_date_data(self, date_string, settings=None):
         date, period = self.parse(date_string, settings)
         return dict(date_obj=date, period=period)
 
+    def get_weekday_data(self, date_string):
+        words = re.split(r"\s+", date_string)
+        for word in words:
+            if re.search(r'\b'+word, _WEEKDAYS) and 'next' in date_string:
+                return word
+
+        else:
+            return None
+
 
 freshness_date_parser = FreshnessDateDataParser()

dateparser/languages/dictionary.py

@@ -11,11 +11,11 @@
 PARSER_HARDCODED_TOKENS = [":", ".", " ", "-", "/"]
 PARSER_KNOWN_TOKENS = ["am", "pm", "UTC", "GMT", "Z"]
 ALWAYS_KEEP_TOKENS = ["+"] + PARSER_HARDCODED_TOKENS
-KNOWN_WORD_TOKENS = ['monday', 'tuesday', 'wednesday', 'thursday', 'friday',
-                     'saturday', 'sunday', 'january', 'february', 'march',
-                     'april', 'may', 'june', 'july', 'august', 'september',
-                     'october', 'november', 'december', 'year', 'month', 'week',
-                     'day', 'hour', 'minute', 'second', 'ago', 'in', 'am', 'pm']
+KNOWN_WORD_TOKENS = ['monday', 'tuesday', 'wednesday', 'thursday', 'friday', 'saturday', 
+                     'sunday', 'january', 'february', 'march', 'april', 'may', 'june', 
+                     'july', 'august', 'september', 'october', 'november', 'december', 
+                     'year', 'month', 'week', 'day', 'hour', 'minute', 'second', 'ago', 
+                     'in', 'next', 'after', 'am', 'pm']
 
 PARENTHESES_PATTERN = re.compile(r'[\(\)]')
 NUMERAL_PATTERN = re.compile(r'(\d+)')

dateparser/languages/locale.py

@@ -221,7 +221,7 @@ def translate_search(self, search_string, settings=None):
                 translated.append(translated_chunk)
                 original.append(original_chunk)
         for i in range(len(translated)):
-            if "in" in translated[i]:
+            if any(key in translated[i] for key in ("in", "next", "after")):
                 translated[i] = self._clear_future_words(translated[i])
             translated[i] = self._join_chunk(list(filter(bool, translated[i])), settings=settings)
             original[i] = self._join_chunk(list(filter(bool, original[i])), settings=settings)
@@ -426,7 +426,9 @@ def _generate_simplifications(self, normalize=False):
     def _clear_future_words(self, words):
         freshness_words = {'day', 'week', 'month', 'year', 'hour', 'minute', 'second'}
         if set(words).isdisjoint(freshness_words):
-            words.remove("in")
+            for future_word in ["in", "next", "after"]:
+                if future_word in words:
+                    words.remove(future_word)
         return words
 
     def _join(self, tokens, separator=" ", settings=None):

dateparser/languages/validation.py

@@ -8,7 +8,7 @@ class LanguageValidator(object):
     logger = None
 
     VALID_KEYS = [
-        'name', 'skip', 'pertain', 'simplifications', 'no_word_spacing', 'ago', 'in',
+        'name', 'skip', 'pertain', 'simplifications', 'no_word_spacing', 'ago', 'in', 'next', 'after',
         'monday', 'tuesday', 'wednesday', 'thursday', 'friday', 'saturday', 'sunday',
         'january', 'february', 'march', 'april', 'may', 'june', 'july', 'august',
         'september', 'october', 'november', 'december', 'year', 'month', 'week', 'day',

dateparser/search/search.py

@@ -13,7 +13,7 @@
     # Python 2.7
     from collections import Set
 
-RELATIVE_REG = re.compile("(ago|in|from now|tomorrow|today|yesterday)")
+RELATIVE_REG = re.compile("(ago|in|next|after|from now|tomorrow|today|yesterday)")
 
 
 def date_is_relative(translation):