From 3f3bbea7fa58cce8f96e26523e3b8172c1b5b109 Mon Sep 17 00:00:00 2001
From: Robert Schroll <rschroll@gmail.com>
Date: Thu, 9 Jan 2014 15:10:19 -0500
Subject: [PATCH 1/2] Use new pdfminer API if old one isn't available

---
 prsannots/pagetext.py | 36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

diff --git a/prsannots/pagetext.py b/prsannots/pagetext.py
index 130145c..66c9fc4 100644
--- a/prsannots/pagetext.py
+++ b/prsannots/pagetext.py
@@ -3,11 +3,39 @@
 # This file is part of prsannots and is distributed under the terms of
 # the LGPL license.  See the file COPYING for full details.
 
-from pdfminer.pdfparser import PDFParser, PDFDocument
+from pdfminer.pdfparser import PDFParser
 from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
 from pdfminer.layout import LAParams, LTAnon, LTTextBox
 from pdfminer.converter import PDFPageAggregator
 
+# pdfminer suddenly decided to change its API...
+try:
+    from pdfminer.pdfparser import PDFDocument
+    
+    def new_doc(parser):
+        doc = PDFDocument()
+        parser.set_document(doc)
+        doc.set_parser(parser)
+        return doc
+    
+    def get_pages(doc):
+        return doc.get_pages()
+
+except ImportError:
+    from pdfminer.pdfdocument import PDFDocument
+    from pdfminer.pdfpage import PDFPage
+
+    def new_doc(parser):
+        return PDFDocument(parser)
+    
+    def get_pages(doc):
+        return PDFPage.create_pages(doc)
+
+try:
+    from pdfminer.layout import LTAnon
+except ImportError:
+    from pdfminer.layout import LTAnno as LTAnon
+
 
 LIGATURES = {u"\ufb00": "ff",
              u"\ufb01": "fi",
@@ -27,9 +55,7 @@ def get_layouts(fd):
     """From an open PDF file, get the page layouts (of type pdfminer.layout.LTPage)."""
     
     parser = PDFParser(fd)
-    doc = PDFDocument()
-    parser.set_document(doc)
-    doc.set_parser(parser)
+    doc = new_doc(parser)
     doc.initialize()
     
     laparams = LAParams()
@@ -38,7 +64,7 @@ def get_layouts(fd):
     interpreter = PDFPageInterpreter(rsrcmgr, device)
     
     layouts = []
-    for page in doc.get_pages():
+    for page in get_pages(doc):
         interpreter.process_page(page)
         layouts.append(device.get_result())
     return layouts

From 070f404a15eca87e7b8f497dd81a8778aef4e364 Mon Sep 17 00:00:00 2001
From: Robert Schroll <rschroll@gmail.com>
Date: Thu, 9 Jan 2014 16:48:05 -0500
Subject: [PATCH 2/2] Remove all bits of old API from outside of try blocks

Thanks to Kris Thielemans for troubleshooting.  Closes #13.
---
 prsannots/pagetext.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/prsannots/pagetext.py b/prsannots/pagetext.py
index 66c9fc4..0f02753 100644
--- a/prsannots/pagetext.py
+++ b/prsannots/pagetext.py
@@ -5,7 +5,7 @@
 
 from pdfminer.pdfparser import PDFParser
 from pdfminer.pdfinterp import PDFResourceManager, PDFPageInterpreter
-from pdfminer.layout import LAParams, LTAnon, LTTextBox
+from pdfminer.layout import LAParams, LTTextBox
 from pdfminer.converter import PDFPageAggregator
 
 # pdfminer suddenly decided to change its API...