[Erp5-report] r40925 nicolas - /erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py

nobody at svn.erp5.org nobody at svn.erp5.org
Tue Nov 30 14:47:51 CET 2010


Author: nicolas
Date: Tue Nov 30 14:47:51 2010
New Revision: 40925

URL: http://svn.erp5.org?rev=40925&view=rev
Log:
Use subprocesstransform for better transformation handling.
changes in argument:
 '-enc' use UTF-8 by defaults
 '-layout' Maintain (as best as possible) the original physical layout of the text.
 '-nopgbrk' Don't insert page breaks


Modified:
    erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py

Modified: erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py
URL: http://svn.erp5.org/erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py?rev=40925&r1=40924&r2=40925&view=diff
==============================================================================
--- erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py [utf8] (original)
+++ erp5/trunk/products/PortalTransforms/transforms/pdf_to_text.py [utf8] Tue Nov 30 14:47:51 2010
@@ -1,3 +1,4 @@
+# -*- coding: utf-8 -*-
 """
 Uses the xpdf (www.foolabs.com/xpdf)
 """
@@ -6,10 +7,11 @@ from Products.PortalTransforms.interface
 from Products.PortalTransforms.libtransforms.utils import bin_search, sansext
 from Products.PortalTransforms.libtransforms.commandtransform import commandtransform
 from Products.PortalTransforms.libtransforms.commandtransform import popentransform
+from Products.PortalTransforms.libtransforms.commandtransform import subprocesstransform
 import os
 from zope.interface import implements
 
-class pdf_to_text(popentransform):
+class pdf_to_text(subprocesstransform):
     implements(itransform)
 
     __name__ = "pdf_to_text"
@@ -20,8 +22,8 @@ class pdf_to_text(popentransform):
     __version__ = '2004-07-02.01'
 
     binaryName = "pdftotext"
-    binaryArgs = "%(infile)s -enc UTF-8 -"
-    useStdin = False
+    binaryArgs = "-layout -nopgbrk %(infile)s -"
+    useStdin = True
 
 class old_pdf_to_text(commandtransform):
     implements(itransform)



More information about the Erp5-report mailing list