Added field extraction and model creation

views/editor.py

@@ -1,10 +1,12 @@
+import os
+
 from django.views.generic import View
 from django.http import HttpResponse
 from django.utils import simplejson
 import poppler
-import os
 
-class PDFFormFillerEditorView(View):
+
+class PDFFormFillerEditor(View):
     def get(self, request, pdf=""):
 
         doc = poppler.document_new_from_file('file:///home/mark/Bedraga.pdf', password=None)
@@ -20,6 +22,25 @@ class PDFFormFillerEditorView(View):
         if os.path.isfile(pdf_path):
             return poppler.document_new_from_file('file://' + pdf_path, password=None)
 
+
+    def generate_page_pngs(self, poppler_doc, pages=[]):
+        import cairo
+        import StringIO
+
+        page = poppler_doc.get_page(pages[0] if pages else 0)
+        width, height = page.get_size()
+        surface = cairo.ImageSurface(cairo.FORMAT_RGB24, int(width*2), int(height*2))
+        context = cairo.Context(surface)
+
+        for page_num in pages or xrange(poppler_doc.num_pages()):
+            page = poppler_doc.get_page(page_num)
+            page.render(context)
+            fd = StringIO.StringIO()
+            surface.write(fd)
+            fd.seek(0)
+            yield fd
+
+
     def get_fields(self, poppler_doc, page):
         fields = poppler_doc.get_page(page).get_form_field_mapping()