[Python-modules-commits] r13412 - in packages/pdfminer/trunk/debian (4 files)

[jwilk at users.alioth.debian.org]

    Date: Sunday, June 13, 2010 @ 10:27:29
  Author: jwilk
Revision: 13412

Backport two upstream patches.

Added:
  packages/pdfminer/trunk/debian/patches/layout.diff
  packages/pdfminer/trunk/debian/patches/nested-tags.diff
Modified:
  packages/pdfminer/trunk/debian/changelog
  packages/pdfminer/trunk/debian/patches/series

Modified: packages/pdfminer/trunk/debian/changelog
===================================================================
--- packages/pdfminer/trunk/debian/changelog	2010-06-13 09:53:58 UTC (rev 13411)
+++ packages/pdfminer/trunk/debian/changelog	2010-06-13 10:27:29 UTC (rev 13412)
@@ -8,5 +8,8 @@
     [psparser-testsuite.diff]
   * Store encoding data in gzipped pickles rather than in Python modules.
     This way we can save lots of disk space. [encoding-data.diff]
+  * Backport upstream patches:
+    + to fix a bug in layout analysis [layout.diff];
+    + to allow extraction of nested tags [nested-tags.diff].
 
- -- Jakub Wilk <jwilk at debian.org>  Sat, 12 Jun 2010 16:54:14 +0200
+ -- Jakub Wilk <jwilk at debian.org>  Sun, 13 Jun 2010 11:58:54 +0200

Added: packages/pdfminer/trunk/debian/patches/layout.diff
===================================================================
--- packages/pdfminer/trunk/debian/patches/layout.diff	                        (rev 0)
+++ packages/pdfminer/trunk/debian/patches/layout.diff	2010-06-13 10:27:29 UTC (rev 13412)
@@ -0,0 +1,18 @@
+Description: Fix a bug in layout analysis.
+Origin: backport, http://code.google.com/p/pdfminerr/source/detail?r=226
+Last-Update: 2010-06-13
+
+--- a/pdfminer/layout.py
++++ b/pdfminer/layout.py
+@@ -527,8 +527,9 @@
+ 
+ ##  group_boxes
+ ##
+-def group_boxes(groupfunc, objs, distfunc, debug=0):
+-    assert objs
++def group_boxes(groupfunc, objs0, distfunc, debug=0):
++    assert objs0
++    objs = objs0[:]
+     while 2 <= len(objs):
+         mindist = INF
+         minpair = None

Added: packages/pdfminer/trunk/debian/patches/nested-tags.diff
===================================================================
--- packages/pdfminer/trunk/debian/patches/nested-tags.diff	                        (rev 0)
+++ packages/pdfminer/trunk/debian/patches/nested-tags.diff	2010-06-13 10:27:29 UTC (rev 13412)
@@ -0,0 +1,46 @@
+Description: Allow extraction of nested tags.
+Origin: backport, http://code.google.com/p/pdfminerr/source/detail?r=226
+Last-Update: 2010-06-13
+
+--- a/pdfminer/converter.py
++++ b/pdfminer/converter.py
+@@ -6,7 +6,7 @@
+ from pdfcolor import LITERAL_DEVICE_GRAY, LITERAL_DEVICE_RGB
+ from layout import LTContainer, LTPage, LTText, LTLine, LTRect, LTPolygon
+ from layout import LTFigure, LTImage, LTChar, LTTextLine, LTTextBox, LTTextGroup
+-from utils import apply_matrix_pt, mult_matrix
++from utils import apply_matrix_pt, mult_matrix, translate_matrix
+ from utils import enc, bbox2str, create_bmp
+ 
+ 
+@@ -354,7 +354,7 @@
+         self.outfp = outfp
+         self.codec = codec
+         self.pageno = 0
+-        self.tag = None
++        self.stack = []
+         return
+ 
+     def render_string(self, textstate, seq):
+@@ -388,16 +388,16 @@
+             s = ''.join( ' %s="%s"' % (enc(k), enc(str(v))) for (k,v)
+                          in sorted(props.iteritems()) )
+         self.outfp.write('<%s%s>' % (enc(tag.name), s))
+-        self.tag = tag
++        self.stack.append(tag)
+         return
+ 
+     def end_tag(self):
+-        assert self.tag
+-        self.outfp.write('</%s>' % enc(self.tag.name))
+-        self.tag = None
++        assert self.stack
++        tag = self.stack.pop(-1)
++        self.outfp.write('</%s>' % enc(tag.name))
+         return
+ 
+     def do_tag(self, tag, props=None):
+         self.begin_tag(tag, props)
+-        self.tag = None
++        self.stack.pop(-1)
+         return

Modified: packages/pdfminer/trunk/debian/patches/series
===================================================================
--- packages/pdfminer/trunk/debian/patches/series	2010-06-13 09:53:58 UTC (rev 13411)
+++ packages/pdfminer/trunk/debian/patches/series	2010-06-13 10:27:29 UTC (rev 13412)
@@ -1,3 +1,5 @@
+layout.diff
+nested-tags.diff
 pdf-testsuite.diff
 psparser-testsuite.diff
 encoding-data.diff