add meta.extract_text

oml/media/pdf.py

@@ -229,6 +229,22 @@ def extract_text(pdf):
             stdout = ''
     return stdout.strip()
 
+
+def pypdf_extract_text(path):
+    '''
+        slow and bad results
+    '''
+    pdf = PdfFileReader(path)
+    content = []
+    for i in range(0, pdf.getNumPages()):
+        try:
+            extracted_text = pdf.getPage(i).extractText()
+            content.append(extracted_text)
+        except:
+            pass
+    content = "\n".join(content).replace("\xa0", " ").strip()
+    return content
+
 def extract_isbn(text):
     isbns = find_isbns(text)
     if isbns: