grangier#142 - extract authors

Author: Xavier Grangier

goose/article.py

@@ -74,6 +74,9 @@ def __init__(self):
         # holds links found in the main article
         self.links = []
 
+        # hold author names
+        self.authors = []
+
         # stores the final URL that we're going to try
         # and fetch content against, this would be expanded if any
         self.final_url = u""

goose/crawler.py

@@ -105,6 +105,7 @@ def crawl(self, crawl_candidate):
         self.article.canonical_link = self.extractor.get_canonical_link()
         self.article.domain = self.extractor.get_domain()
         self.article.tags = self.extractor.extract_tags()
+        self.article.authors = self.extractor.extract_authors()
 
         # opengraph
         self.article.opengraph = self.extractor.extract_opengraph()

goose/extractors.py

@@ -259,6 +259,25 @@ def extract_links(self):
                 links.append(attr)
         return links
 
+    def extract_authors(self):
+        authors = []
+        author_nodes = self.parser.getElementsByTag(
+                            self.article.doc,
+                            attr='itemprop',
+                            value='author')
+
+        for author in author_nodes:
+            name_nodes = self.parser.getElementsByTag(
+                            author,
+                            attr='itemprop',
+                            value='name')
+
+            if len(name_nodes) > 0:
+                name = self.parser.getText(name_nodes[0])
+                authors.append(name)
+
+        return list(set(authors))
+
     def extract_tags(self):
         node = self.article.doc