Changeset 539 for pkpgcounter

Timestamp:

12/09/07 09:58:08 (17 years ago)

Author:

jerome

Message:

Added support for Structured Fax documents.

Location:

pkpgcounter/trunk

Files:

• bin/pkpgcounter (modified) (1 diff)
• man/pkpgcounter.1 (modified) (2 diffs)
• NEWS (modified) (1 diff)
• pkpgpdls/analyzer.py (modified) (2 diffs)
• pkpgpdls/cfax.py (copied) (copied from pkpgcounter/trunk/pkpgpdls/dvi.py) (2 diffs)
• pkpgpdls/version.py (modified) (1 diff)
• README (modified) (1 diff)

pkpgcounter/trunk/bin/pkpgcounter

@@ -56 +56 @@
     * Brother HBP
     * Hewlett-Packard LIDIL (hpijs)
+    * Structured Fax
 
-The seven latter ones, as well as some TIFF documents, are currently
+The eight latter ones, as well as some TIFF documents, are currently
 only supported in page counting mode.
 

pkpgcounter/trunk/man/pkpgcounter.1

@@ -1 +1 @@
 .\" DO NOT MODIFY THIS FILE!  It was generated by help2man 1.36.
-.TH PKPGCOUNTER "1" "November 2007" "C@LL - Conseil Internet & Logiciels Libres" "User Commands"
+.TH PKPGCOUNTER "1" "December 2007" "C@LL - Conseil Internet & Logiciels Libres" "User Commands"
 .SH NAME
 pkpgcounter \- count number of pages required to print various types of documents
 .SH DESCRIPTION
-pkpgcounter v3.40 (c) 2003, 2004, 2005, 2006, 2007 Jerome Alet
+pkpgcounter v3.50alpha (c) 2003, 2004, 2005, 2006, 2007 Jerome Alet
 .PP
 pkpgcounter is a generic Page Description Language parser.
@@ -33 +33 @@
 * Brother HBP
 * Hewlett\-Packard LIDIL (hpijs)
+* Structured Fax
 .PP
-The seven latter ones, as well as some TIFF documents, are currently
+The eight latter ones, as well as some TIFF documents, are currently
 only supported in page counting mode.
 .PP

pkpgcounter/trunk/NEWS

@@ -21 +21 @@
 pkpgcounter News :
 
+  * 3.50 :
+  
+    - Added support for Structured Fax documents in page counting mode.
+    
+    - Fixed a problem with the parsing of SPL1 (GDI) documents.
+    
   * 3.40 :
   

pkpgcounter/trunk/pkpgpdls/analyzer.py

@@ -28 +28 @@
 import tempfile
 
-import version, pdlparser, postscript, pdf, pcl345, pclxl, hbp, pil, mscrap, \
-       lidil, escp2, dvi, tiff, ooo, zjstream, qpdl, spl1, escpages03, plain
+import version, pdlparser, postscript, pdf, pcl345, pclxl, hbp, \
+       pil, mscrap, cfax, lidil, escp2, dvi, tiff, ooo, zjstream, \
+       qpdl, spl1, escpages03, plain
 import inkcoverage
 
@@ -155 +156 @@
                        dvi, \
                        tiff, \
+                       cfax, \
                        zjstream, \
                        ooo, \

pkpgcounter/trunk/pkpgpdls/cfax.py

@@ -21 +21 @@
 #
 
-"""This modules implements a page counter for DVI documents."""
+"""This modules implements a page counter for Structured Fax documents."""
 
-import sys
-import os
-import mmap
-from struct import unpack
+import struct
 
 import pdlparser
 
 class Parser(pdlparser.PDLParser) :
-    """A parser for DVI documents."""
-    totiffcommands = [ 'dvips -q -o - "%(infname)s" | gs -sDEVICE=tiff24nc -dPARANOIDSAFER -dNOPAUSE -dBATCH -dQUIET -r"%(dpi)i" -sOutputFile="%(outfname)s" -' ]
-    required = [ "dvips", "gs" ]
+    """A parser for Structured Fax documents."""
     def isValid(self) :        
-        """Returns True if data is DVI, else False."""
-        try :
-            if (ord(self.firstblock[0]) == 0xf7) \
-                and (ord(self.lastblock[-1]) == 0xdf) :
-                self.logdebug("DEBUG: Input file is in the DVI format.")
-                return True
-            else :    
-                return False
-        except IndexError :          
+        """Returns True if data is Structured Fax, else False."""
+        if self.firstblock.startswith("Sfff") :
+            self.logdebug("DEBUG: Input file is in the Structured Fax format.")
+            return True
+        else :    
             return False
             
     def getJobSize(self) :
-        """Counts pages in a DVI document.
+        """Counts pages in a Structured Fax document.
         
            Algorithm by Jerome Alet.
@@ -53 +44 @@
            The documentation used for this was :
          
-           http://www.math.umd.edu/~asnowden/comp-cont/dvi.html
+           http://delphi.pjh2.de/articles/graphic/sff_format.php
         """
-        infileno = self.infile.fileno()
-        minfile = mmap.mmap(infileno, os.fstat(infileno)[6], prot=mmap.PROT_READ, flags=mmap.MAP_SHARED)
+        unpack = struct.unpack
         pagecount = 0
-        pos = -1
-        eofchar = chr(0xdf)
-        postchar = chr(0xf8)
+        docheader = self.infile.read(20)
         try :
-            try :
-                while minfile[pos] == eofchar :
-                    pos -= 1
-                idbyte = minfile[pos]    
-                if idbyte != minfile[1] :
-                    raise IndexError, "Invalid DVI file."
-                pos = unpack(">I", minfile[pos - 4:pos])[0]
-                if minfile[pos] != postchar :
-                    raise IndexError, "Invalid DVI file."
-                pagecount = unpack(">H", minfile[pos + 27: pos + 29])[0]
-            except IndexError : # EOF ?
-                pass
-        finally :        
-            minfile.close() # reached EOF
-        return pagecount
+            (sffid,
+             version,
+             reserved,
+             userinfo,
+             docpagecount,
+             offsetfirstpage,
+             offsetlastpage,
+             offsetdocumentend) = unpack("<4sBBHHHII", docheader)
+            self.infile.seek(offsetfirstpage - len(docheader), 1)     
+            while True :
+                headerid = self.infile.read(1)
+                if not headerid :
+                    break
+                headerid = ord(headerid)    
+                if 1 <= headerid <= 216 : # Normal record header
+                    self.infile.seek(headerid, 1)
+                elif headerid == 255 :    # Illegal line / Additional user info
+                    additionalbyte = self.infile.read(1)
+                    if not additionalbyte :
+                        break
+                    additionalbyte = ord(additionalbyte)    
+                    if 1 <= additionalbyte <= 255 :
+                        # Skip additional user information (reserved)
+                        self.infile.seek(additionalbyte, 1)
+                elif not headerid :        
+                    # Record with more than 216 MH-coded bytes
+                    recordlen = self.infile.read(2)
+                    if not recordlen :
+                        break
+                    recordlen = unpack("<H", recordlen)[0]    
+                    self.infile.seek(recordlen, 1)
+                elif headerid == 254 : # Page header
+                    pageheader = self.infile.read(17)
+                    if not pageheader : 
+                        break
+                    headerlen = ord(pageheader[0])
+                    if not headerlen :
+                        break # End Of Document
+                    (vres,
+                     hres,
+                     coding,
+                     reserved,
+                     linelen,
+                     pagelen,
+                     offsetpreviouspage,
+                     offsetnextpage) = unpack("<4BHHII", pageheader[1:])
+                    pagecount += 1    
+                    if (offsetnextpage == 1) or (vres == 255) :
+                        break # End Of Document
+                    self.infile.seek(offsetnextpage, 1)    
+        except struct.error :     
+             raise pdlparser.PDLParserError, "Invalid Structured Fax datas"
+        return max(docpagecount, pagecount)     

pkpgcounter/trunk/pkpgpdls/version.py

@@ -22 +22 @@
 
 
-__version__ = "3.41alpha"
+__version__ = "3.50alpha"
 
 __doc__ = """pkpgcounter : a generic Page Description Languages parser."""

pkpgcounter/trunk/README

@@ -59 +59 @@
         - Hewlett-Packard Lightweight Imaging Device Interface Language
         
-The seven latter ones, as well as some TIFF documents, are currently 
+        - Structured Fax
+        
+The eight latter ones, as well as some TIFF documents, are currently 
 only supported in page counting mode.