Changeset 519

Timestamp:

11/27/07 21:55:29 (17 years ago)

Author:

jerome

Message:

Added a skeleton for Microsoft Word (c) (tm) (r) (etc...) documents.

Location:

pkpgcounter/trunk/pkpgpdls

Files:

• analyzer.py (modified) (2 diffs)
• mstrash.py (copied) (copied from pkpgcounter/trunk/pkpgpdls/plain.py) (3 diffs)

pkpgcounter/trunk/pkpgpdls/analyzer.py

@@ -28 +28 @@
 import tempfile
 
-import version, pdlparser, postscript, pdf, pcl345, pclxl, hbp, pil, \
+import version, pdlparser, postscript, pdf, pcl345, pclxl, hbp, pil, mstrash, \
        lidil, escp2, dvi, tiff, ooo, zjstream, qpdl, spl1, escpages03, plain
 import inkcoverage
@@ -159 +159 @@
                        escpages03, \
                        pil, \
+                       mstrash, \
                        plain) :     # IMPORTANT : don't move this one up !
             try :               

pkpgcounter/trunk/pkpgpdls/mstrash.py

@@ -21 +21 @@
 #
 
-"""This modules implements a page counter for plain text documents."""
+"""This module implements a page counter for Microsoft Word (r) (tm) (c) (etc...) documents"""
+
+import os
+import urllib2
 
 import pdlparser
@@ -27 +30 @@
 
 class Parser(pdlparser.PDLParser) :
-    """A parser for plain text documents."""
-    totiffcommands = [ 'enscript --quiet --portrait --no-header --columns 1 --output - "%(infname)s" | gs -sDEVICE=tiff24nc -dPARANOIDSAFER -dNOPAUSE -dBATCH -dQUIET -r"%(dpi)i" -sOutputFile="%(outfname)s" -',
-                       'a2ps --borders 0 --quiet --portrait --no-header --columns 1 --output - "%(infname)s" | gs -sDEVICE=tiff24nc -dPARANOIDSAFER -dNOPAUSE -dBATCH -dQUIET -r"%(dpi)i" -sOutputFile="%(outfname)s" -',
-                     ]  
-    openmode = "rU"                 
+    """A parser for that MS crap thing."""
+    totiffcommands = [ ]
     def isValid(self) :    
-        """Returns True if data is plain text, else False.
+        """Returns True if data is MS crap, else False.
         
-           It's hard to detect a plain text file, so we just try to
-           extract lines from the first block (sufficiently large).
-           If it's impossible to find one we consider it's not plain text.
+           Identifying datas taken from the file command's magic database.
+           IMPORTANT : some magic values are not reused here because they
+           IMPORTANT : seem to be specific to some particular i18n release.
         """   
-        lines = self.firstblock.split("\r\n")
-        if len(lines) == 1 :
-            lines = lines[0].split("\r")
-            if len(lines) == 1 :
-                lines = lines[0].split("\n")
-        if len(lines) > 1 :
-            self.logdebug("DEBUG: Input file seems to be in the plain text format.")
+        if self.firstblock.startswith("PO^Q`") \
+           or self.firstblock.startswith("\xd0\xcf\x11\xe0\xa1\xb1\x1a\xe1") \
+           or self.firstblock.startswith("\xfe7\x00#") \
+           or self.firstblock.startswith("\xdb\xa5-\x00\x00\x00") \
+           or self.firstblock.startswith("\x31\xbe\x00\x00") \
+           or self.firstblock[2112:].startswith("MSWordDoc") :
+            self.logdebug("DEBUG: Input file seems to be in a Microsoft shitty file format.")
             return True
         else :    
@@ -51 +51 @@
             
     def getJobSize(self) :
-        """Counts pages in a plain text document."""
-        pagesize = 66   # TODO : Does this vary wrt the default page size ?
-                        # TODO : /etc/papersize and /etc/paper.config
-        pagecount = 0
-        linecount = 0
-        for line in self.infile :
-            if line.endswith("\n") :
-                linecount += 1    
-                if (linecount > pagesize) :
-                    pagecount += 1
-                    linecount = 0
-                else :    
-                    cnt = line.count("\f")
-                    if cnt :
-                        pagecount += cnt
-                        linecount = 0
-            else :        
-                raise pdlparser.PDLParserError, "Unsupported file format. Please send the file to %s" % version.__authoremail__
-        return pagecount + 1    # NB : empty files are catched in isValid()
+        """Counts pages in a Microsoft Word (r) (tm) (c) (etc...) document."""
+        return 0