git-svn-id: https://pdfminerr.googlecode.com/svn/trunk/pdfminer@15 1aa58f4a-7d42-0410-adbc-911cccaed67c

pdf2txt.py

@@ -23,16 +23,18 @@ class TextConverter(PDFDevice):
     self.outfp.write('\n')
     return
   
-  def begin_page(self, name, (x0,y0,x1,y1)):
-    self.outfp.write('<page name="%s" x0="%d" y0="%d" x1="%d" y1="%d">\n' %
-                     (name,x0,y0,x1,y1))
+  def begin_page(self, page):
+    (x0,y0,x1,y1) = page.mediabox
+    self.outfp.write('<page id="%d" mediabox="%d,%d,%d,%d" rotate="%d">' %
+                     (page.pageid, x0,y0,x1,y1, page.rotate))
     return
   def end_page(self, _):
     self.outfp.write('</page>\n')
     return
 
-  def begin_figure(self, name, (x0,y0,x1,y1)):
-    self.outfp.write('<figure name="%s" x0="%d" y0="%d" x1="%d" y1="%d">\n' %
+  def begin_figure(self, name, bbox):
+    (x0,y0,x1,y1) = bbox
+    self.outfp.write('<figure name="%s" bbox="%d,%d,%d,%d">\n' %
                      (name, x0,y0,x1,y1))
     return
   def end_figure(self, _):
@@ -80,7 +82,7 @@ class TextConverter(PDFDevice):
 # pdf2txt
 def pdf2txt(outfp, rsrc, fname, pages, codec, debug=0):
   device = TextConverter(outfp, rsrc, codec)
-  outfp.write('<document>')
+  outfp.write('<document>\n')
   doc = PDFDocument(debug=debug)
   fp = file(fname)
   parser = PDFParser(doc, fp, debug=debug)
@@ -89,7 +91,7 @@ def pdf2txt(outfp, rsrc, fname, pages, codec, debug=0):
     if pages and (i not in pages): continue
     interpreter.process_page(page)
   fp.close()
-  outfp.write('</document>')
+  outfp.write('</document>\n')
   device.close()
   return
 

pdfinterp.py

@@ -443,9 +443,9 @@ class PDFDevice:
     self.ctm = ctm
     return
 
-  def begin_page(self, name, bbox):
+  def begin_page(self, page):
     return
-  def end_page(self, name):
+  def end_page(self, page):
     return
   def begin_figure(self, name, bbox):
     return
@@ -835,9 +835,9 @@ class PDFPageInterpreter:
   def process_page(self, page):
     if 1 <= self.debug:
       print >>stderr, 'Processing page: %r' % page
-    self.device.begin_page(page.pageid, page.mediabox)
+    self.device.begin_page(page)
     self.render_contents(page.resources, page.contents)
-    self.device.end_page(page.pageid)
+    self.device.end_page(page)
     return
 
   def render_contents(self, resources, contents, ctm=MATRIX_IDENTITY):

pdfparser.py

@@ -205,8 +205,16 @@ class PDFPage:
     self.doc = doc
     self.pageid = pageidx
     self.attrs = dict_value(attrs)
+    self.lastmod = self.attrs.get('LastModified')
     self.resources = resolve1(self.attrs['Resources'])
     self.mediabox = resolve1(self.attrs['MediaBox'])
+    if 'CropBox' in self.attrs:
+      self.cropbox = resolve1(self.attrs['CropBox'])
+    else:
+      self.cropbox = self.mediabox
+    self.rotate = self.attrs.get('Rotate', 0)
+    self.annots = self.attrs.get('Annots')
+    self.beads = self.attrs.get('B')
     contents = resolve1(self.attrs['Contents'])
     if not isinstance(contents, list):
       contents = [ contents ]
@@ -293,6 +301,7 @@ class PDFXRefStream:
 
 ##  PDFDocument
 ##
+INHERITABLE_ATTRS = set(['Resources', 'MediaBox', 'CropBox', 'Rotate'])
 class PDFDocument:
   
   def __init__(self, debug=0):
@@ -368,7 +377,7 @@ class PDFDocument:
     def search(obj, parent):
       tree = dict_value(obj).copy()
       for (k,v) in parent.iteritems():
-        if k not in tree:
+        if k in INHERITABLE_ATTRS and k not in tree:
           tree[k] = v
       if tree['Type'] == LITERAL_PAGES:
         if 1 <= debug: