Merge branch 'master' of github.com:waylan/Python-Markdown

15 files changed, 89 insertions, 60 deletions

diff --git a/.travis.yml b/.travis.yml
index 7471163..1931a20 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -4,6 +4,7 @@ env:
   - TOXENV=py27
   - TOXENV=py32
   - TOXENV=py33
+  - TOXENV=py34
 before_install:
   - sudo apt-get update -qq
   - sudo apt-get install libtidy-0.99-0
diff --git a/docs/extensions/smarty.txt b/docs/extensions/smarty.txt
index b96b82e..56e9bd5 100644
--- a/docs/extensions/smarty.txt
+++ b/docs/extensions/smarty.txt
@@ -15,11 +15,12 @@ their HTML entity equivalents.
 
 ASCII symbol | Replacements    | HTML Entities
 ------------ | --------------- | -------------------
-'        | ‘ ’ | `‘` `’`
-"       | “ ” | `“` `”`
-\...         | …        | `…`
-\--          | –         | `–`
--\--         | —         | `—`
+`'`          | ‘ ’ | `‘` `’`
+`"`          | “ ” | `“` `”`
+`<< >>`      | &laquo; &raquo; | `&laquo;` `&raquo;`
+`...`        | &hellip;        | `&hellip;`
+`--`         | &ndash;         | `&ndash;`
+`---`        | &mdash;         | `&mdash;`
 
 !!! note
     This extension reimplements the Python [SmartyPants] 
@@ -42,13 +43,14 @@ as the name of the extension.
 See the [Library Reference](../reference.html#extensions) for information about
 configuring extensions.
 
-The following options are provided to configure the output (all three are set to `True` by default):
+The following options are provided to configure the output:
 
-Option           | Description
-------           | -----------
-`smart_dashes`   | whether to convert dashes
-`smart_quotes`   | whether to convert quotes
-`smart_ellipses` | whether to convert ellipses
+Option                | Default value | Description
+------                | ------------- | -----------
+`smart_dashes`        | enabled       | whether to convert dashes
+`smart_quotes`        | enabled       | whether to convert straight quotes
+`smart_angled_quotes` | disabled      | whether to convert angled quotes
+`smart_ellipses`      | enabled       | whether to convert ellipses
 
 Further reading
 ---------------
diff --git a/markdown/extensions/abbr.py b/markdown/extensions/abbr.py
index 3f8a443..bed3bb4 100644
--- a/markdown/extensions/abbr.py
+++ b/markdown/extensions/abbr.py
@@ -13,7 +13,7 @@ Simple Usage:
     ... *[ABBR]: Abbreviation
     ... *[REF]: Abbreviation Reference
     ... """
-    >>> print markdown.markdown(text, ['abbr'])
+    >>> print(markdown.markdown(text, ['abbr']))
     <p>Some text with an <abbr title="Abbreviation">ABBR</abbr> and a <abbr title="Abbreviation Reference">REF</abbr>. Ignore REFERENCE and ref.</p>
 
 Copyright 2007-2008
diff --git a/markdown/extensions/fenced_code.py b/markdown/extensions/fenced_code.py
index d6e043c..fe1559c 100644
--- a/markdown/extensions/fenced_code.py
+++ b/markdown/extensions/fenced_code.py
@@ -13,14 +13,14 @@ This extension adds Fenced Code Blocks to Python-Markdown.
     ... ~~~
     ... '''
     >>> html = markdown.markdown(text, extensions=['fenced_code'])
-    >>> print html
+    >>> print(html)
     <p>A paragraph before a fenced code block:</p>
     <pre><code>Fenced code block
     </code></pre>
 
 Works with safe_mode also (we check this because we are using the HtmlStash):
 
-    >>> print markdown.markdown(text, extensions=['fenced_code'], safe_mode='replace')
+    >>> print(markdown.markdown(text, extensions=['fenced_code'], safe_mode='replace'))
     <p>A paragraph before a fenced code block:</p>
     <pre><code>Fenced code block
     </code></pre>
@@ -32,7 +32,7 @@ Include tilde's in a code block and wrap with blank lines:
     ...
     ... ~~~~
     ... ~~~~~~~~'''
-    >>> print markdown.markdown(text, extensions=['fenced_code'])
+    >>> print(markdown.markdown(text, extensions=['fenced_code']))
     <pre><code>
     ~~~~
     </code></pre>
@@ -43,7 +43,7 @@ Language tags:
     ... ~~~~{.python}
     ... # Some python code
     ... ~~~~'''
-    >>> print markdown.markdown(text, extensions=['fenced_code'])
+    >>> print(markdown.markdown(text, extensions=['fenced_code']))
     <pre><code class="python"># Some python code
     </code></pre>
 
@@ -54,7 +54,7 @@ Optionally backticks instead of tildes as per how github's code block markdown i
     ... # Arbitrary code
     ... ~~~~~ # these tildes will not close the block
     ... `````'''
-    >>> print markdown.markdown(text, extensions=['fenced_code'])
+    >>> print(markdown.markdown(text, extensions=['fenced_code']))
     <pre><code># Arbitrary code
     ~~~~~ # these tildes will not close the block
     </code></pre>
@@ -67,11 +67,11 @@ If the codehighlite extension and Pygments are installed, lines can be highlight
     ... line 2
     ... line 3
     ... ```'''
-    >>> print markdown.markdown(text, extensions=['codehilite', 'fenced_code'])
-    <pre><code><span class="hilight">line 1</span>
-    line 2
-    <span class="hilight">line 3</span>
-    </code></pre>
+    >>> print(markdown.markdown(text, extensions=['codehilite', 'fenced_code']))
+    <div class="codehilite"><pre><span class="hll"><span class="n">line</span> <span class="mi">1</span>
+    </span><span class="n">line</span> <span class="mi">2</span>
+    <span class="hll"><span class="n">line</span> <span class="mi">3</span>
+    </span></pre></div>
 
 Copyright 2007-2008 [Waylan Limberg](http://achinghead.com/).
 
diff --git a/markdown/extensions/headerid.py b/markdown/extensions/headerid.py
index 8221fe1..07351cc 100644
--- a/markdown/extensions/headerid.py
+++ b/markdown/extensions/headerid.py
@@ -9,7 +9,7 @@ Basic usage:
     >>> import markdown
     >>> text = "# Some Header #"
     >>> md = markdown.markdown(text, ['headerid'])
-    >>> print md
+    >>> print(md)
     <h1 id="some-header">Some Header</h1>
 
 All header IDs are unique:
@@ -19,7 +19,7 @@ All header IDs are unique:
     ... #Header
     ... #Header'''
     >>> md = markdown.markdown(text, ['headerid'])
-    >>> print md
+    >>> print(md)
     <h1 id="header">Header</h1>
     <h1 id="header_1">Header</h1>
     <h1 id="header_2">Header</h1>
@@ -30,7 +30,7 @@ To fit within a html template's hierarchy, set the header base level:
     ... #Some Header
     ... ## Next Level'''
     >>> md = markdown.markdown(text, ['headerid(level=3)'])
-    >>> print md
+    >>> print(md)
     <h3 id="some-header">Some Header</h3>
     <h4 id="next-level">Next Level</h4>
 
@@ -38,7 +38,7 @@ Works with inline markup.
 
     >>> text = '#Some *Header* with [markup](http://example.com).'
     >>> md = markdown.markdown(text, ['headerid'])
-    >>> print md
+    >>> print(md)
     <h1 id="some-header-with-markup">Some <em>Header</em> with <a href="http://example.com">markup</a>.</h1>
 
 Turn off auto generated IDs:
@@ -47,7 +47,7 @@ Turn off auto generated IDs:
     ... # Some Header
     ... # Another Header'''
     >>> md = markdown.markdown(text, ['headerid(forceid=False)'])
-    >>> print md
+    >>> print(md)
     <h1>Some Header</h1>
     <h1>Another Header</h1>
 
@@ -58,7 +58,7 @@ Use with MetaData extension:
     ...
     ... # A Header'''
     >>> md = markdown.markdown(text, ['headerid', 'meta'])
-    >>> print md
+    >>> print(md)
     <h2>A Header</h2>
 
 Copyright 2007-2011 [Waylan Limberg](http://achinghead.com/).
diff --git a/markdown/extensions/meta.py b/markdown/extensions/meta.py
index c4a4b21..9063188 100644
--- a/markdown/extensions/meta.py
+++ b/markdown/extensions/meta.py
@@ -15,16 +15,16 @@ Basic Usage:
     ... The body. This is paragraph one.
     ... '''
     >>> md = markdown.Markdown(['meta'])
-    >>> print md.convert(text)
+    >>> print(md.convert(text))
     <p>The body. This is paragraph one.</p>
-    >>> print md.Meta
-    {u'blank_data': [u''], u'author': [u'Waylan Limberg', u'John Doe'], u'title': [u'A Test Doc.']}
+    >>> print(md.Meta) # doctest: +SKIP
+    {'blank_data': [''], 'author': ['Waylan Limberg', 'John Doe'], 'title': ['A Test Doc.']}
 
 Make sure text without Meta Data still works (markdown < 1.6b returns a <p>).
 
     >>> text = '    Some Code - not extra lines of meta data.'
     >>> md = markdown.Markdown(['meta'])
-    >>> print md.convert(text)
+    >>> print(md.convert(text))
     <pre><code>Some Code - not extra lines of meta data.
     </code></pre>
     >>> md.Meta
diff --git a/markdown/extensions/nl2br.py b/markdown/extensions/nl2br.py
index da4b339..028561c 100644
--- a/markdown/extensions/nl2br.py
+++ b/markdown/extensions/nl2br.py
@@ -8,7 +8,7 @@ GitHub-flavored Markdown does.
 Usage:
 
     >>> import markdown
-    >>> print markdown.markdown('line 1\\nline 2', extensions=['nl2br'])
+    >>> print(markdown.markdown('line 1\\nline 2', extensions=['nl2br']))
     <p>line 1<br />
     line 2</p>
 
diff --git a/markdown/extensions/smart_strong.py b/markdown/extensions/smart_strong.py
index 4818cf9..c7ac9f1 100644
--- a/markdown/extensions/smart_strong.py
+++ b/markdown/extensions/smart_strong.py
@@ -7,14 +7,14 @@ This extention adds smarter handling of double underscores within words.
 Simple Usage:
 
     >>> import markdown
-    >>> print markdown.markdown('Text with double__underscore__words.',
-    ...                   extensions=['smart_strong'])
+    >>> print(markdown.markdown('Text with double__underscore__words.',
+    ...                         extensions=['smart_strong']))
     <p>Text with double__underscore__words.</p>
-    >>> print markdown.markdown('__Strong__ still works.',
-    ...                   extensions=['smart_strong'])
+    >>> print(markdown.markdown('__Strong__ still works.',
+    ...                         extensions=['smart_strong']))
     <p><strong>Strong</strong> still works.</p>
-    >>> print markdown.markdown('__this__works__too__.',
-    ...                   extensions=['smart_strong'])
+    >>> print(markdown.markdown('__this__works__too__.',
+    ...                         extensions=['smart_strong']))
     <p><strong>this__works__too</strong>.</p>
 
 Copyright 2011
diff --git a/markdown/extensions/smarty.py b/markdown/extensions/smarty.py
index 2f946f8..8131591 100644
--- a/markdown/extensions/smarty.py
+++ b/markdown/extensions/smarty.py
@@ -68,6 +68,8 @@
 from __future__ import unicode_literals
 from . import Extension
 from ..inlinepatterns import HtmlPattern
+from ..odict import OrderedDict
+from ..treeprocessors import InlineProcessor
 from ..util import parseBoolValue
 
 # Constants for quote education.
@@ -135,6 +137,7 @@ class SmartyExtension(Extension):
     def __init__(self, configs):
         self.config = {
             'smart_quotes': [True, 'Educate quotes'],
+            'smart_angled_quotes': [False, 'Educate angled quotes'],
             'smart_dashes': [True, 'Educate dashes'],
             'smart_ellipses': [True, 'Educate ellipses']
         }
@@ -145,20 +148,28 @@ class SmartyExtension(Extension):
         for ind, pattern in enumerate(patterns):
             pattern += (md,)
             pattern = SubstituteTextPattern(*pattern)
-            after = ('>smarty-%s-%d' % (serie, ind - 1) if ind else '>entity')
+            after = ('>smarty-%s-%d' % (serie, ind - 1) if ind else '_begin')
             name = 'smarty-%s-%d' % (serie, ind)
-            md.inlinePatterns.add(name, pattern, after)
+            self.inlinePatterns.add(name, pattern, after)
 
     def educateDashes(self, md):
         emDashesPattern = SubstituteTextPattern(r'(?<!-)---(?!-)', ('&mdash;',), md)
         enDashesPattern = SubstituteTextPattern(r'(?<!-)--(?!-)', ('&ndash;',), md)
-        md.inlinePatterns.add('smarty-em-dashes', emDashesPattern, '>entity')
-        md.inlinePatterns.add('smarty-en-dashes', enDashesPattern,
+        self.inlinePatterns.add('smarty-em-dashes', emDashesPattern, '_begin')
+        self.inlinePatterns.add('smarty-en-dashes', enDashesPattern,
             '>smarty-em-dashes')
 
     def educateEllipses(self, md):
         ellipsesPattern = SubstituteTextPattern(r'(?<!\.)\.{3}(?!\.)', ('&hellip;',), md)
-        md.inlinePatterns.add('smarty-ellipses', ellipsesPattern, '>entity')
+        self.inlinePatterns.add('smarty-ellipses', ellipsesPattern, '_begin')
+
+    def educateAngledQuotes(self, md):
+        leftAngledQuotePattern = SubstituteTextPattern(r'\<\<', ('&laquo;',), md)
+        rightAngledQuotePattern = SubstituteTextPattern(r'\>\>', ('&raquo;',), md)
+        self.inlinePatterns.add('smarty-left-angle-quotes',
+                                leftAngledQuotePattern, '_begin')
+        self.inlinePatterns.add('smarty-right-angle-quotes',
+                                rightAngledQuotePattern, '>smarty-left-angle-quotes')
 
     def educateQuotes(self, md):
         patterns = (
@@ -179,12 +190,18 @@ class SmartyExtension(Extension):
 
     def extendMarkdown(self, md, md_globals):
         configs = self.getConfigs()
+        self.inlinePatterns = OrderedDict()
+        if configs['smart_ellipses']:
+            self.educateEllipses(md)
         if configs['smart_quotes']:
             self.educateQuotes(md)
+        if configs['smart_angled_quotes']:
+            self.educateAngledQuotes(md)
         if configs['smart_dashes']:
             self.educateDashes(md)
-        if configs['smart_ellipses']:
-            self.educateEllipses(md)
+        inlineProcessor = InlineProcessor(md)
+        inlineProcessor.inlinePatterns = self.inlinePatterns
+        md.treeprocessors.add('smarty', inlineProcessor, '_end')
         md.ESCAPED_CHARS.extend(['"', "'"])
 
 def makeExtension(configs=None):
diff --git a/markdown/extensions/wikilinks.py b/markdown/extensions/wikilinks.py
index ba1947c..6d34173 100644
--- a/markdown/extensions/wikilinks.py
+++ b/markdown/extensions/wikilinks.py
@@ -9,21 +9,21 @@ Basic usage:
     >>> import markdown
     >>> text = "Some text with a [[WikiLink]]."
     >>> html = markdown.markdown(text, ['wikilinks'])
-    >>> print html
+    >>> print(html)
     <p>Some text with a <a class="wikilink" href="/WikiLink/">WikiLink</a>.</p>
 
 Whitespace behavior:
 
-    >>> print markdown.markdown('[[ foo bar_baz ]]', ['wikilinks'])
+    >>> print(markdown.markdown('[[ foo bar_baz ]]', ['wikilinks']))
     <p><a class="wikilink" href="/foo_bar_baz/">foo bar_baz</a></p>
-    >>> print markdown.markdown('foo [[ ]] bar', ['wikilinks'])
+    >>> print(markdown.markdown('foo [[ ]] bar', ['wikilinks']))
     <p>foo  bar</p>
 
 To define custom settings the simple way:
 
-    >>> print markdown.markdown(text, 
+    >>> print(markdown.markdown(text,
     ...     ['wikilinks(base_url=/wiki/,end_url=.html,html_class=foo)']
-    ... )
+    ... ))
     <p>Some text with a <a class="foo" href="/wiki/WikiLink.html">WikiLink</a>.</p>
     
 Custom settings the complex way:
@@ -35,7 +35,7 @@ Custom settings the complex way:
     ...                                 ('end_url', '.html'),
     ...                                 ('html_class', '') ]},
     ...     safe_mode = True)
-    >>> print md.convert(text)
+    >>> print(md.convert(text))
     <p>Some text with a <a href="http://example.com/WikiLink.html">WikiLink</a>.</p>
 
 Use MetaData with mdx_meta.py (Note the blank html_class in MetaData):
@@ -46,12 +46,12 @@ Use MetaData with mdx_meta.py (Note the blank html_class in MetaData):
     ...
     ... Some text with a [[WikiLink]]."""
     >>> md = markdown.Markdown(extensions=['meta', 'wikilinks'])
-    >>> print md.convert(text)
+    >>> print(md.convert(text))
     <p>Some text with a <a href="http://example.com/WikiLink.html">WikiLink</a>.</p>
 
 MetaData should not carry over to next document:
 
-    >>> print md.convert("No [[MetaData]] here.")
+    >>> print(md.convert("No [[MetaData]] here."))
     <p>No <a class="wikilink" href="/MetaData/">MetaData</a> here.</p>
 
 Define a custom URL builder:
@@ -60,7 +60,7 @@ Define a custom URL builder:
     ...     return '/bar/'
     >>> md = markdown.Markdown(extensions=['wikilinks'], 
     ...         extension_configs={'wikilinks' : [('build_url', my_url_builder)]})
-    >>> print md.convert('[[foo]]')
+    >>> print(md.convert('[[foo]]'))
     <p><a class="wikilink" href="/bar/">foo</a></p>
 
 From the command line:
diff --git a/markdown/treeprocessors.py b/markdown/treeprocessors.py
index ef0a2aa..32b73b1 100644
--- a/markdown/treeprocessors.py
+++ b/markdown/treeprocessors.py
@@ -53,6 +53,7 @@ class InlineProcessor(Treeprocessor):
                                       + len(self.__placeholder_suffix)
         self.__placeholder_re = util.INLINE_PLACEHOLDER_RE
         self.markdown = md
+        self.inlinePatterns = md.inlinePatterns
 
     def __makePlaceholder(self, type):
         """ Generate a placeholder """
@@ -99,9 +100,9 @@ class InlineProcessor(Treeprocessor):
         """
         if not isinstance(data, util.AtomicString):
             startIndex = 0
-            while patternIndex < len(self.markdown.inlinePatterns):
+            while patternIndex < len(self.inlinePatterns):
                 data, matched, startIndex = self.__applyPattern(
-                    self.markdown.inlinePatterns.value_for_index(patternIndex),
+                    self.inlinePatterns.value_for_index(patternIndex),
                     data, patternIndex, startIndex)
                 if not matched:
                     patternIndex += 1
diff --git a/tests/extensions/smarty.html b/tests/extensions/smarty.html
index 6b5e698..c0459d9 100644
--- a/tests/extensions/smarty.html
+++ b/tests/extensions/smarty.html
@@ -13,6 +13,9 @@ one two &lsquo;60s<br />
 &lsquo;quoted&rsquo; text and <strong>bold &lsquo;quoted&rsquo; text</strong><br />
 em-dashes (&mdash;) and ellipes (&hellip;)<br />
 &ldquo;<a href="http://example.com">Link</a>&rdquo; &mdash; she said.</p>
+<p>&ldquo;Ellipsis within quotes&hellip;&rdquo;</p>
+<p>Кавычки-&laquo;ёлочки&raquo;<br />
+Anführungszeichen-&raquo;Chevrons&laquo;</p>
 <hr />
 <p>Escaped -- ndash<br />
 'Escaped' "quotes"<br />
diff --git a/tests/extensions/smarty.txt b/tests/extensions/smarty.txt
index fbf4d03..1cee803 100644
--- a/tests/extensions/smarty.txt
+++ b/tests/extensions/smarty.txt
@@ -15,6 +15,11 @@ It's fun. What's fun?
 em-dashes (---) and ellipes (...)  
 "[Link](http://example.com)" --- she said.
 
+"Ellipsis within quotes..."
+
+Кавычки-<<ёлочки>>  
+Anführungszeichen->>Chevrons<<
+
 --- -- ---
 
 Escaped \-- ndash  
diff --git a/tests/extensions/test.cfg b/tests/extensions/test.cfg
index 494d79b..c05a64e 100644
--- a/tests/extensions/test.cfg
+++ b/tests/extensions/test.cfg
@@ -1,5 +1,5 @@
 [attr_list]
-extensions=attr_list,def_list
+extensions=attr_list,def_list,smarty
 
 [codehilite]
 extensions=codehilite
@@ -40,4 +40,4 @@ extensions=nl2br,attr_list
 extensions=admonition
 
 [smarty]
-extensions=smarty
+extensions=smarty(smart_angled_quotes=1)
diff --git a/tox.ini b/tox.ini
index 791a462..b1833da 100644
--- a/tox.ini
+++ b/tox.ini
@@ -1,5 +1,5 @@
 [tox]
-envlist = py26, py27, py31, py32, py33
+envlist = py26, py27, py31, py32, py33, py34
 
 [testenv]
 downloadcache = {toxworkdir}/cache