Changeset 491 for pkpgcounter/trunk/pkpgpdls/pdf.py

Timestamp:

11/20/07 00:04:09 (17 years ago)

Author:

jerome

Message:

Major code cleaning. Now clearer, although probably a bit slower since
a file can be opened several times.
Now universal line opening mode is only used when needed (PS, PDF and plain
text), and binary opening mode is used for the other formats.
This mean we will be able to remove mmap calls wherever possible, finally.

Files:

• pkpgcounter/trunk/pkpgpdls/pdf.py (modified) (8 diffs)

pkpgcounter/trunk/pkpgpdls/pdf.py

@@ -23 +23 @@
 """This modules implements a page counter for PDF documents."""
 
-import sys
 import re
 
@@ -34 +33 @@
         self.major = major
         self.minor = minor
+        self.majori = int(major)
+        self.minori = int(minor)
         self.description = description
         self.comments = []
@@ -43 +44 @@
     """A parser for PDF documents."""
     totiffcommands = [ 'gs -sDEVICE=tiff24nc -dPARANOIDSAFER -dNOPAUSE -dBATCH -dQUIET -r%(dpi)i -sOutputFile="%(fname)s" -' ]
+    openmode = "rU"
     def isValid(self) :    
         """Returns True if data is PDF, else False."""
@@ -62 +64 @@
         objre = re.compile(r"\s?(\d+)\s+(\d+)\s+obj[<\s/]?")
         for line in self.infile :
-            line = line.strip()
+            line = line.strip()    
             if line.startswith("% ") :    
                 if inobject :
@@ -72 +74 @@
                 result = objre.search(line)
                 if result is not None :
-                    (major, minor) = [int(num) for num in line[result.start():result.end()].split()[:2]]
+                    (major, minor) = line[result.start():result.end()].split()[:2]
                     obj = PDFObject(major, minor, lastcomment)
                     obj.content.append(line[result.end():])
@@ -86 +88 @@
                         # if we want the first one, just use > instead of >=
                         oldobject = objects.setdefault(major, obj)
-                        if minor >= oldobject.minor :
+                        if int(minor) >= oldobject.minori :
                             objects[major] = obj
+                            # self.logdebug("Object(%i, %i) overwritten with Object(%i, %i)" % (oldobject.majori, oldobject.minori, obj.majori, obj.minori))
+                        # self.logdebug("Object(%i, %i)" % (obj.majori, obj.minori))
                         inobject = 0        
                 else :    
@@ -94 +98 @@
                         
         # Now we check each PDF object we've just created.
-        # colorregexp = re.compile(r"(/ColorSpace) ?(/DeviceRGB|/DeviceCMYK)[/ \t\r\n]", re.I)
         newpageregexp = re.compile(r"(/Type)\s?(/Page)[/>\s]", re.I)
         pagecount = 0
@@ -103 +106 @@
                 pagecount += count
         return pagecount    
-        
-if __name__ == "__main__" :    
-    pdlparser.test(Parser)