Merge branch 'master' of https://github.com/pdfminer/pdfminer.six.git

README.md

@@ -16,7 +16,6 @@ PDF parser that can be used for other purposes than text analysis.
 
  * Webpage: https://github.com/pdfminer/
  * Download (PyPI): https://pypi.python.org/pypi/pdfminer.six/
- * Demo WebApp: http://pdf2html.tabesugi.net:8080/ (broken?)
 
 
 Features
@@ -36,14 +35,12 @@ Features
 How to Install
 --------------
 
- * Install Python 2.7 or newer. (Python 3.4 is supported in pdfminer.six)
- * Download the source code.
- * Unpack it.
- * Run `setup.py`:
+ * Install Python 2.7 or newer. (Python 3.x is supported in pdfminer.six)
+ * Install
 
-    $ python setup.py install
+    $ pip install pdfminer.six
 
- * Do the following test:
+ * Run the following test:
 
     $ pdf2txt.py samples/simple1.pdf
 
@@ -76,35 +73,11 @@ but it's also possible to extract some meaningful contents (e.g. images).
 (For details, refer to the html document.)
 
 
-API Changes
------------
-
-As of November 2013, there were a few changes made to the PDFMiner API
-prior to October 2013. This is the result of code restructuring.  Here
-is a list of the changes:
-
- * PDFDocument class is moved to pdfdocument.py.
- * PDFDocument class now takes a PDFParser object as an argument.
-   PDFDocument.set_parser() and PDFParser.set_document() is removed.
- * PDFPage class is moved to pdfpage.py
- * process_pdf function is implemented as a class method PDFPage.get_pages.
-
-
 TODO
 ----
 
  * PEP-8 and PEP-257 conformance.
  * Better documentation.
- * Crypt stream filter support.
-
-
-Related Projects
-----------------
-
- * <a href="http://pybrary.net/pyPdf/">pyPdf</a>
- * <a href="http://www.foolabs.com/xpdf/">xpdf</a>
- * <a href="http://pdfbox.apache.org/">pdfbox</a>
- * <a href="http://mupdf.com/">mupdf</a>
 
 
 Terms and Conditions

pdfminer/__init__.py

@@ -1,6 +1,6 @@
 
 # -*- coding: utf-8 -*-
-__version__ = '20170119'
+__version__ = '20170418'
 
 if __name__ == '__main__':
     print (__version__)

pdfminer/ccitt.py

@@ -13,7 +13,17 @@
 import sys
 import array
 
-import six #Python 2+3 compatibility
+import six  #Python 2+3 compatibility
+
+if six.PY3:
+    def get_bytes(data):
+        for byte in data:
+            yield byte
+else:
+    def get_bytes(data):
+        for char in data:
+            yield ord(char)
+
 
 ##  BitParser
 ##
@@ -40,10 +50,9 @@ class BitParser(object):
         return
 
     def feedbytes(self, data):
-        for c in data:
-            b = ord(c)
+        for byte in get_bytes(data):
             for m in (128, 64, 32, 16, 8, 4, 2, 1):
-                self._parse_bit(b & m)
+                self._parse_bit(byte & m)
         return
 
     def _parse_bit(self, x):
@@ -328,11 +337,10 @@ class CCITTG4Parser(BitParser):
         return
 
     def feedbytes(self, data):
-        for c in data:
-            b = ord(c)
+        for byte in get_bytes(data):
             try:
                 for m in (128, 64, 32, 16, 8, 4, 2, 1):
-                    self._parse_bit(b & m)
+                    self._parse_bit(byte & m)
             except self.ByteSkip:
                 self._accept = self._parse_mode
                 self._state = self.MODE

pdfminer/pdftypes.py

@@ -175,7 +175,7 @@ def stream_value(x):
     if not isinstance(x, PDFStream):
         if settings.STRICT:
             raise PDFTypeError('PDFStream required: %r' % x)
-        return PDFStream({}, '')
+        return PDFStream({}, b'')
     return x
 
 

setup.py

@@ -27,7 +27,7 @@ PDF parser that can be used for other purposes instead of text analysis.''',
     license='MIT/X',
     author='Yusuke Shinyama + Philippe Guglielmetti',
     author_email='pdfminer@goulu.net',
-    url='http://github.com/goulu/pdfminer',
+    url='http://github.com/pdfminer/pdfminer',
     scripts=[
     'tools/pdf2txt.py',
     'tools/dumppdf.py',

tests/test_tools_pdf2txt.py

@@ -19,7 +19,6 @@ def run(datapath,filename,options=None):
 
 class TestDumpPDF():
 
-
     def test_1(self):
         run('../samples/','jo')
         run('../samples/','simple1')
@@ -41,5 +40,12 @@ class TestDumpPDF():
     def test_6(self):
         run('../samples/nonfree/','naacl06-shinyama')
 
+    # this test works on Windows but on Linux & Travis-CI it says
+    # PDFSyntaxError: No /Root object! - Is this really a PDF?
+    # TODO: Find why
+    """
+    def test_7(self):
+        run('../samples/contrib/','stamp-no')
+    """
 if __name__ == '__main__':
     nose.runmodule()