textutils: fix a bug in normalize_[rest_]paragraph which may cause

      infinite loop if an indent string containing some spaces is given

3 files changed, 28 insertions, 8 deletions

diff --git a/ChangeLog b/ChangeLog
index ede934f..dbf8742 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -4,7 +4,9 @@ ChangeLog for logilab.common
 2008-01-08  --  0.26.1
     * optparser: support --version at main command level
     * testlib: added man page for pytest
-
+    * textutils: fix a bug in normalize_[rest_]paragraph which may cause
+      infinite loop if an indent string containing some spaces is given
+	
 2008-01-07  --  0.26.0
     * db: binarywrap support
     * modutils: new LazyObject class
diff --git a/test/unittest_textutils.py b/test/unittest_textutils.py
index 2043e84..535052b 100644
--- a/test/unittest_textutils.py
+++ b/test/unittest_textutils.py
@@ -70,6 +70,22 @@ aller discuter avec les autres si c'est utile ou necessaire.""")
         self.assertEquals(ulines(tu.normalize_rest_paragraph("""**nico**: toto""")),
                           """**nico**: toto""")
 
+    def test_normalize_rest_paragraph2(self):
+        self.assertEquals(ulines(tu.normalize_rest_paragraph(""".. _tdm: http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Table-des-matieres/.20_adaa41fb-c125-4919-aece-049601e81c8e_0_0.pdf
+.. _extrait: http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Extrait-du-livre/.20_d6eed0be-0d36-4384-be59-2dd09e081012_0_0.pdf""", indent='> ')),
+                          """> .. _tdm:
+> http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Table-des-matieres/.20_adaa41fb-c125-4919-aece-049601e81c8e_0_0.pdf
+> .. _extrait:
+> http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Extrait-du-livre/.20_d6eed0be-0d36-4384-be59-2dd09e081012_0_0.pdf""")
+        
+    def test_normalize_paragraph2(self):
+        self.assertEquals(ulines(tu.normalize_paragraph(""".. _tdm: http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Table-des-matieres/.20_adaa41fb-c125-4919-aece-049601e81c8e_0_0.pdf
+.. _extrait: http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Extrait-du-livre/.20_d6eed0be-0d36-4384-be59-2dd09e081012_0_0.pdf""", indent='> ')),
+                          """> .. _tdm:
+> http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Table-des-matieres/.20_adaa41fb-c125-4919-aece-049601e81c8e_0_0.pdf
+> .. _extrait:
+> http://www.editions-eni.fr/Livres/Python-Les-fondamentaux-du-langage---La-programmation-pour-les-scientifiques-Extrait-du-livre/.20_d6eed0be-0d36-4384-be59-2dd09e081012_0_0.pdf""")
+
 class NormalizeParagraphTC(TestCase):
 
     def test_known_values(self):
diff --git a/textutils.py b/textutils.py
index cf73f8d..88167a8 100644
--- a/textutils.py
+++ b/textutils.py
@@ -13,7 +13,7 @@
 """Some text manipulation utility functions.
 
 :author:    Logilab
-:copyright: 2003-2007 LOGILAB S.A. (Paris, FRANCE)
+:copyright: 2003-2008 LOGILAB S.A. (Paris, FRANCE)
 :contact:   http://www.logilab.fr/ -- mailto:python-projects@logilab.org
 
 :group text formatting: normalize_text, normalize_paragraph, pretty_match,\
@@ -156,10 +156,11 @@ def normalize_paragraph(text, line_len=80, indent=''):
       indentation string
     """
     text = _NORM_SPACES_RGX.sub(' ', text)
+    line_len = line_len - len(indent)
     lines = []
     while text:
-        aline, text = splittext(indent + text.strip(), line_len)
-        lines.append(aline)
+        aline, text = splittext(text.strip(), line_len)
+        lines.append(indent + aline)
     return linesep.join(lines)
     
 def normalize_rest_paragraph(text, line_len=80, indent=''):
@@ -185,20 +186,21 @@ def normalize_rest_paragraph(text, line_len=80, indent=''):
     """
     toreport = ''
     lines = []
+    line_len = line_len - len(indent)
     for line in text.splitlines():
-        line = indent + toreport + _NORM_SPACES_RGX.sub(' ', line.strip())
+        line = toreport + _NORM_SPACES_RGX.sub(' ', line.strip())
         toreport = ''
         while len(line) > line_len:
             # too long line, need split
             line, toreport = splittext(line, line_len)
-            lines.append(line)
+            lines.append(indent + line)
             if toreport:
-                line = indent + toreport + ' '
+                line = toreport + ' '
                 toreport = ''
             else:
                 line = ''
         if line:
-            lines.append(line.strip())
+            lines.append(indent + line.strip())
     return linesep.join(lines)
 
 def splittext(text, line_len):