Revert "Added published date to the extractor+article"

.gitignore

@@ -4,10 +4,6 @@
 .idea
 .pypirc
 
-# VIM
-*.swp
-*.swo
-
 # C extensions
 *.so
 
@@ -43,4 +39,4 @@ nosetests.xml
 .mr.developer.cfg
 .project
 .pydevproject
-venv
+venv

newspaper/article.py

@@ -11,7 +11,6 @@
 import copy
 import os
 import glob
-import datetime
 
 from . import nlp
 from . import images
@@ -169,9 +168,6 @@ def parse(self):
         title = self.extractor.get_title(self)
         self.set_title(title)
 
-        published_date = self.extractor.get_published_date(self)
-        self.set_published_date(published_date)
-
         authors = self.extractor.get_authors(self)
         self.set_authors(authors)
 
@@ -459,15 +455,6 @@ def set_keywords(self, keywords):
         if keywords:
             self.keywords = [encodeValue(k) for k in keywords[:self.config.MAX_KEYWORDS]]
 
-    def set_published_date(self, date):
-      """
-      Datetime object of the timestamp on the article
-      """
-      if date and not isinstance(date, datetime.datetime):
-          raise Exception("published date must be a datetime object")
-      if date:
-          self.published_date = date
-
     def set_authors(self, authors):
         """
         Authors are in ["firstName lastName", "firstName lastName"] format.

newspaper/extractors.py

@@ -14,7 +14,6 @@
 import re
 import copy
 import urlparse
-import datetime
 from collections import defaultdict
 
 from .packages.tldextract import tldextract
@@ -208,44 +207,6 @@ def get_title(self, article):
         title = MOTLEY_REPLACEMENT.replaceAll(title_text)
         return title
 
-    def get_published_date(self, article):
-        """
-        Find the article published date form either in the page or the url
-        """
-        timestamps = article.doc.xpath(
-            ("//meta[@property='article:published_time']/@content|"
-            "//meta[@name='date']/@content|"
-            "//meta[@name='sailthru.date']/@content")
-        )
-        timestamp = None
-        if timestamps:
-            timestamp = timestamps.pop(0).replace(
-                "T", " ").split(" ").pop(0)
-            if len(timestamp) != 10:
-                timestamp = None
-        if not timestamp:
-            parts = urlparse.urlparse(article.url)
-            path_parts = parts.path.strip("/").split("/")
-            numbered_parts = [part for part in path_parts 
-                if part.isdigit() and len(part) < 5]
-            if len(numbered_parts) == 3:
-                timestamp = "-".join(numbered_parts) 
-
-        if not timestamp:
-            return None
-
-        formats = (
-            "%Y-%m-%d",  # ISO 
-            "%m-%d-%Y",  # US
-            "%d-%m-%Y"   # UK
-        )
-        for format in formats:
-            try:
-                return datetime.datetime.strptime(
-                    timestamp, format)
-            except ValueError:
-                pass
-
     def split_title(self, title, splitter):
         """
         Split the title to best part possible.