From 429cc98556bd399268a375a4dfd58e387be9f6e0 Mon Sep 17 00:00:00 2001 From: Maurice van der Pot Date: Fri, 26 Feb 2016 19:32:38 +0100 Subject: [PATCH 1/3] Improve RawHtmlProcessor to have linear iso quadratic performance --- markdown/postprocessors.py | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) diff --git a/markdown/postprocessors.py b/markdown/postprocessors.py index 2d4dcb589..8b311b268 100644 --- a/markdown/postprocessors.py +++ b/markdown/postprocessors.py @@ -10,6 +10,7 @@ from __future__ import absolute_import from __future__ import unicode_literals +from collections import OrderedDict from . import util from . import odict import re @@ -50,6 +51,7 @@ class RawHtmlPostprocessor(Postprocessor): def run(self, text): """ Iterate over html stash and restore "safe" html. """ + replacements = OrderedDict() for i in range(self.markdown.htmlStash.html_counter): html, safe = self.markdown.htmlStash.rawHtmlBlocks[i] if self.markdown.safeMode and not safe: @@ -61,14 +63,15 @@ def run(self, text): html = self.markdown.html_replacement_text if (self.isblocklevel(html) and (safe or not self.markdown.safeMode)): - text = text.replace( - "

%s

" % - (self.markdown.htmlStash.get_placeholder(i)), + replacements["

%s

" % + (self.markdown.htmlStash.get_placeholder(i))] = \ html + "\n" - ) - text = text.replace( - self.markdown.htmlStash.get_placeholder(i), html - ) + replacements[self.markdown.htmlStash.get_placeholder(i)] = html + + if replacements: + pattern = re.compile("|".join(re.escape(k) for k in replacements)) + text = pattern.sub(lambda m: replacements[m.group(0)], text) + return text def escape(self, html): From e56f0cd9b0c6f934bac62751234c0f4df2050ac6 Mon Sep 17 00:00:00 2001 From: Maurice van der Pot Date: Sat, 27 Feb 2016 00:34:42 +0100 Subject: [PATCH 2/3] Added assertStartsWith to tests to give better failure messages The failure printed when self.assertTrue was used with str.startswith did not show the string that was being searched. --- tests/test_extensions.py | 64 +++++++++++++++++++++++----------------- 1 file changed, 37 insertions(+), 27 deletions(-) diff --git a/tests/test_extensions.py b/tests/test_extensions.py index 72ce21256..e9b611222 100644 --- a/tests/test_extensions.py +++ b/tests/test_extensions.py @@ -11,6 +11,15 @@ import unittest import markdown +class TestCaseWithAssertStartsWith(unittest.TestCase): + + def assertStartsWith(self, expectedPrefix, text, msg=None): + if not text.startswith(expectedPrefix): + if len(expectedPrefix) + 5 < len(text): + text = text[:len(expectedPrefix) + 5] + '...' + standardMsg = '%s not found at the start of %s' % (repr(expectedPrefix), + repr(text)) + self.fail(self._formatMessage(msg, standardMsg)) class TestExtensionClass(unittest.TestCase): """ Test markdown.extensions.Extension. """ @@ -85,8 +94,7 @@ def testNestedAbbr(self): 'and ABBR

' ) - -class TestCodeHilite(unittest.TestCase): +class TestCodeHilite(TestCaseWithAssertStartsWith): """ Test codehilite extension. """ def setUp(self): @@ -101,7 +109,7 @@ def testBasicCodeHilite(self): md = markdown.Markdown(extensions=['markdown.extensions.codehilite']) if self.has_pygments: # Pygments can use random lexer here as we did not specify the language - self.assertTrue(md.convert(text).startswith('
'))
+            self.assertStartsWith('
', md.convert(text))
         else:
             self.assertEqual(
                 md.convert(text),
@@ -117,10 +125,9 @@ def testLinenumsTrue(self):
             # Different versions of pygments output slightly different markup.
             # So we use 'startwith' and test just enough to confirm that
             # pygments received and processed linenums.
-            self.assertTrue(
-                md.convert(text).startswith(
-                    '
'
-                )
+            self.assertStartsWith(
+                '
',
+                md.convert(text)
             )
         else:
             self.assertEqual(
@@ -134,7 +141,7 @@ def testLinenumsFalse(self):
         md = markdown.Markdown(
             extensions=[markdown.extensions.codehilite.CodeHiliteExtension(linenums=False)])
         if self.has_pygments:
-            self.assertTrue(md.convert(text).startswith('
'))
+            self.assertStartsWith('
', md.convert(text))
         else:
             self.assertEqual(
                 md.convert(text),
@@ -164,10 +171,9 @@ def testLinenumsNoneWithShebang(self):
             # Differant versions of pygments output slightly different markup.
             # So we use 'startwith' and test just enough to confirm that
             # pygments received and processed linenums.
-            self.assertTrue(
-                md.convert(text).startswith(
-                    '
'
-                )
+            self.assertStartsWith(
+                '
',
+                md.convert(text)
             )
         else:
             self.assertEqual(
@@ -182,7 +188,7 @@ def testLinenumsNoneWithColon(self):
             extensions=[markdown.extensions.codehilite.CodeHiliteExtension(linenums=None)]
         )
         if self.has_pygments:
-            self.assertTrue(md.convert(text).startswith('
Header 1\n'
             '
Header 2
'
         )
-        self.assertTrue(md.toc.startswith('
'))
+        self.assertStartsWith('
', md.toc)
 
     def testReset(self):
         """ Test TOC Reset. """
         self.assertEqual(self.md.toc, '')
         self.md.convert('# Header 1\n\n## Header 2')
-        self.assertTrue(self.md.toc.startswith('
'))
+        self.assertStartsWith('
', self.md.toc)
         self.md.reset()
         self.assertEqual(self.md.toc, '')
 
@@ -771,7 +778,10 @@ def testTitle(self):
             extensions=[markdown.extensions.toc.TocExtension(title='Table of Contents')]
         )
         md.convert('# Header 1\n\n## Header 2')
-        self.assertTrue(md.toc.startswith('
Table of Contents
    '))
+        self.assertStartsWith(
+            '
    Table of Contents
      ',
+            md.toc
+        )
 
     def testWithAttrList(self):
         """ Test TOC with attr_list Extension. """

From 596601380fc2911e25394b7675aa34f1cab761a9 Mon Sep 17 00:00:00 2001
From: Maurice van der Pot 
Date: Sat, 27 Feb 2016 00:47:55 +0100
Subject: [PATCH 3/3] Added a few empty lines in the test to satisfy flake8

---
 tests/test_extensions.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/tests/test_extensions.py b/tests/test_extensions.py
index e9b611222..30b21bb50 100644
--- a/tests/test_extensions.py
+++ b/tests/test_extensions.py
@@ -11,6 +11,7 @@
 import unittest
 import markdown
 
+
 class TestCaseWithAssertStartsWith(unittest.TestCase):
 
     def assertStartsWith(self, expectedPrefix, text, msg=None):
@@ -21,6 +22,7 @@ def assertStartsWith(self, expectedPrefix, text, msg=None):
                                                                repr(text))
             self.fail(self._formatMessage(msg, standardMsg))
 
+
 class TestExtensionClass(unittest.TestCase):
     """ Test markdown.extensions.Extension. """
 
@@ -94,6 +96,7 @@ def testNestedAbbr(self):
             'and ABBR
      '
         )
 
+
 class TestCodeHilite(TestCaseWithAssertStartsWith):
     """ Test codehilite extension. """