backport changes for upstream: #145, #95, #111, #117, #129, #132.

pdfminer/glyphlist.py

@@ -7,7 +7,7 @@ Unicode characters instead of using decimal/hex character code.
 
 The following data was taken by
 
-  $ wget http://www.adobe.com/devnet/opentype/archives/glyphlist.txt
+  $ wget https://partners.adobe.com/public/developer/en/opentype/glyphlist.txt
   $ python tools/conv_glyphlist.py glyphlist.txt > glyphlist.py
 
 """

pdfminer/latin_enc.py

@@ -162,6 +162,7 @@ ENCODING = [
   ('mu', None, 181, 181, 181),
   ('multiply', None, None, 215, 215),
   ('n', 110, 110, 110, 110),
+  ('nbspace', None, 202, 160, None),
   ('nine', 57, 57, 57, 57),
   ('ntilde', None, 150, 241, 241),
   ('numbersign', 35, 35, 35, 35),

pdfminer/pdftypes.py

@@ -229,7 +229,7 @@ class PDFStream(PDFObject):
         if not isinstance(filters, list):
             filters = [filters]
         if not isinstance(params, list):
-            params = [params]
+            params = [params] * len(filters)
         return zip(filters, params)
 
     def decode(self):

pdfminer/utils.py

@@ -46,8 +46,9 @@ def compatible_encode_method(bytesorstring, encoding='utf-8', erraction='ignore'
 def apply_png_predictor(pred, colors, columns, bitspercomponent, data):
     if bitspercomponent != 8:
         # unsupported
-        raise ValueError(bitspercomponent)
+        raise ValueError("Unsupported `bitspercomponent': %d" %
-    nbytes = colors*columns*bitspercomponent//8
+                         bitspercomponent)
+    nbytes = colors * columns * bitspercomponent // 8
     i = 0
     buf = b''
     line0 = b'\x00' * columns
@@ -86,7 +87,7 @@ def apply_png_predictor(pred, colors, columns, bitspercomponent, data):
                 line2 += six.int2byte(c)
         else:
             # unsupported
-            raise ValueError(ft)
+            raise ValueError("Unsupported predictor value: %d" % ft)
         buf += line2
         line0 = line2
     return buf

setup.py

@@ -4,15 +4,19 @@ from setuptools import setup
 from pdfminer import __version__
 import sys
 
+requires = ['six', 'pycrypto']
+if sys.version_info >= (3, 0):
+    requires.append('chardet')
+
 setup(
     name='pdfminer.six',
     version=__version__,
-    packages=['pdfminer',],
+    packages=['pdfminer'],
     package_data={'pdfminer': ['cmap/*.pickle.gz']},
-    install_requires=['six', 'chardet'] if sys.version_info >= (3, 0) else ['six'],
+    install_requires=requires,
     description='PDF parser and analyzer',
     long_description='''fork of PDFMiner using six for Python 2+3 compatibility
-    
+
 PDFMiner is a tool for extracting information from PDF documents.
 Unlike other PDF-related tools, it focuses entirely on getting
 and analyzing text data. PDFMiner allows to obtain