working version of the parser

1 files changed, 7 insertions, 5 deletions

diff --git a/parsepdftext.py b/parsepdftext.py
index 778d30b..51e237c 100644
--- a/parsepdftext.py
+++ b/parsepdftext.py
@@ -3,8 +3,10 @@ from xml.etree import ElementTree as ET
 
 document = ET.parse(sys.argv[1])
 ns = 'http://www.w3.org/1999/xhtml'
-for page, i in enumerate(document.findall('.//{{{0}}}page'.format(ns))):
-    for word in page.getchildren():
-        octalescapedtext = ''.join(["\{0:o}".format(c) if c>127 else chr(c) for c in map(ord,word.text.encode('utf8'))])
-        #escape quote character
-        print octalescapedtext
+corners = ['xMin', 'xMax', 'yMin', 'yMax']
+# coordinates are in dpi, and computed from the top left corner
+words = []
+coords = []
+for i, page in enumerate(document.findall('.//{{{0}}}page'.format(ns))):
+    words.append([word.text for word in page.getchildren()])
+    coords.append([tuple([float(word.attrib[c]) for c in corners]) for word in page.getchildren()])