Support py3

Makefile

@@ -2,7 +2,7 @@ all:
 	@echo 'please run "make publish" to publish'
 
 publish:
-	python2 setup.py sdist bdist_wheel
+	python3 setup.py sdist bdist_wheel
 	twine upload dist/*
 
 clean:

bin/pdf2md

@@ -1,19 +1,24 @@
-#!/usr/bin/env python2
+#!/usr/bin/env python3
 
 import sys
 import os
+
+PACKAGE_PARENT = '..'
+SCRIPT_DIR = os.path.dirname(os.path.realpath(os.path.join(os.getcwd(), os.path.expanduser(__file__))))
+sys.path.append(os.path.normpath(os.path.join(SCRIPT_DIR, PACKAGE_PARENT)))
+
 import pdf2md
 
 def main(argv):
-    if len(argv) == 2:
-        filename = argv[1]
-        title = os.path.splitext(os.path.basename(filename))[0]
-        print 'Parsing', filename
-    else:
-        print 'usage:'
-        print '    python main.py <pdf>'
+    if len(argv) < 2 or argv[1] in ['-h', '--help']:
+        print('usage:')
+        print('    {} <pdf>'.format(argv[0]))
         return
 
+    filename = argv[1]
+    title = os.path.splitext(os.path.basename(filename))[0]
+    print('Parsing', filename)
+
     parser = pdf2md.Parser(filename)
     parser.extract()
     piles = parser.parse()
@@ -26,7 +31,7 @@ def main(argv):
     writer.set_title(title)
     writer.write(piles)
 
-    print 'Your markdown is at', writer.get_location()
+    print('Your markdown is at', writer.get_location())
 
 if __name__ == '__main__':
     main(sys.argv)

pdf2md/__init__.py

@@ -1,4 +1,4 @@
-from parser import Parser
-from writer import Writer
-from syntax import BaseSyntax
-from syntax import UrbanSyntax
+from .parser import Parser
+from .writer import Writer
+from .syntax import BaseSyntax
+from .syntax import UrbanSyntax

pdf2md/parser.py

@@ -6,7 +6,7 @@
 from pdfminer.pdfdevice import PDFDevice
 from pdfminer.layout import LAParams
 from pdfminer.converter import PDFPageAggregator
-from pile import Pile
+from .pile import Pile
 
 
 class Parser(object):

pdf2md/pile.py

@@ -53,7 +53,7 @@ def parse_layout(self, layout):
             elif type(obj) == LTChar:
                 pass
             elif type(obj) == LTLine:
-                pass                    
+                pass
             else:
                 assert False, "Unrecognized type: %s" % type(obj)
 
@@ -63,7 +63,9 @@ def split_piles(self):
         paragraphs = self._find_paragraphs(tables)
         images = self._find_images()
 
-        piles = sorted(tables + paragraphs + images, reverse=True, key=lambda x: x._get_anything().y0)
+        piles = tables + paragraphs + images
+        piles = list(filter(lambda x: x._empty(), piles))
+        piles = sorted(piles, reverse=True, key=lambda x: x._get_anything().y0)
 
         return piles
 
@@ -99,7 +101,7 @@ def gen_html(self):
                 'height': text.y1 - text.y0,
                 'x': text.x0,
                 'y': text.y0,
-                'text': text.get_text().encode('utf8'),
+                'text': text.get_text(),
                 'fill': 'green',
             }
             html += rect.format(**info)
@@ -202,7 +204,7 @@ def _find_paragraphs(self, tables):
                     paragraphs[idx].texts.append(text)
                     break
 
-        paragraphs = filter(None, paragraphs)
+        paragraphs = list(filter(lambda x: x._empty(), paragraphs))
 
         return paragraphs
 
@@ -225,7 +227,7 @@ def _get_anything(self):
 
 
     def _is_overlap(self, top, bottom, obj):
-        assert top > bottom
+        assert top >= bottom
         return (bottom - self._SEARCH_DISTANCE) <= obj.y0 <= (top + self._SEARCH_DISTANCE) or \
                (bottom - self._SEARCH_DISTANCE) <= obj.y1 <= (top + self._SEARCH_DISTANCE)
 
@@ -350,9 +352,9 @@ def _is_ignore_cell(self, left, top, right, bottom):
 
 
     def _find_exist_coor(self, minimum, maximum, start_idx, line_coor, direction):
-        span = 0
+        span = -1
         line_exist = False
-        while not line_exist:
+        while not line_exist and start_idx+span+1 < len(line_coor):
             span += 1
             coor = line_coor[start_idx + span]
             line_exist = self._line_exists(coor, minimum, maximum, direction)
@@ -411,7 +413,7 @@ def _create_tag(self, tag_name, start, level):
 
     def _create_td_tag(self, cell):
         indent = '\t' * 2
-        texts = [text.get_text().encode('utf8').strip() for text in cell['texts']]
+        texts = [text.get_text().strip() for text in cell['texts']]
         texts = ' '.join(texts)
         colspan = ' colspan={}'.format(cell['colspan']) if 'colspan' in cell else ''
         rowspan = ' rowspan={}'.format(cell['rowspan']) if 'rowspan' in cell else ''
@@ -426,4 +428,6 @@ def _calc_coordinates(self, axes, attr, reverse):
         coor_list.sort(reverse=reverse)
         return coor_list
 
+    def _empty(self):
+        return bool(self.images) or bool(self.texts)
 

pdf2md/syntax.py

@@ -18,7 +18,7 @@ def newline(self, text):
 
 
     def purify(self, text):
-        return text.get_text().encode('utf8').strip()
+        return text.get_text().strip()
 
 
 class UrbanSyntax(BaseSyntax):
@@ -27,7 +27,7 @@ def __init__(self):
 
 
     def pattern(self, text):
-        content = text.get_text().encode('utf8').strip()
+        content = text.get_text().strip()
 
         if not content:
             return 'none'
@@ -62,7 +62,7 @@ def pattern(self, text):
         return 'plain-text'
 
     def newline(self, text):
-        content = text.get_text().encode('utf8').strip()
+        content = text.get_text().strip()
 
         if text.x0 < 90.1: # special case for neihu page 2
             return True
@@ -78,7 +78,7 @@ def newline(self, text):
 
 
     def purify(self, text):
-        content = text.get_text().encode('utf8').strip()
+        content = text.get_text().strip()
 
         mo = re.match('(一|二|三|四|五|六|七|八|九|十)、(.*)', content)
         if mo: