From e4396d657dd69364ad13c96baed36fa051fd366e Mon Sep 17 00:00:00 2001
From: Yuri Takhteyev <yuri@freewisdom.org>
Date: Fri, 17 Oct 2008 09:27:11 -0700
Subject: Adding Jack Miller's TOC extension and a test file for it.

(I can't get it to work, though.)
---
 markdown_extensions/toc.py | 127 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 127 insertions(+)
 create mode 100644 markdown_extensions/toc.py

(limited to 'markdown_extensions/toc.py')

diff --git a/markdown_extensions/toc.py b/markdown_extensions/toc.py
new file mode 100644
index 0000000..ccc927c
--- /dev/null
+++ b/markdown_extensions/toc.py
@@ -0,0 +1,127 @@
+"""
+Table of Contents Extension for Python-Markdown
+* * *
+
+(c) 2008 [Jack Miller](http://codezen.org)
+
+Dependencies:
+* [Markdown 2.0+](http://www.freewisdom.org/projects/python-markdown/)
+
+"""
+import markdown
+from markdown import etree
+import re
+
+class TocPostprocessor (markdown.Postprocessor):
+    # Iterator wrapper to get parent and child all at once
+    def iterparent(self, root):
+        for parent in root.getiterator():
+            for child in parent:
+                yield parent, child
+
+    def run(self, doc) :
+        div = etree.Element("div")
+        div.attrib["class"] = "toc"
+        last_li = None
+
+        # Add title to the div
+        if self.config["title"][0]:
+            header = etree.SubElement(div, "span")
+            header.attrib["class"] = "toctitle"
+            header.text = self.config["title"][0]
+
+        level = 0
+        list_stack=[div]
+        header_rgx = re.compile("[Hh][123456]")
+
+        for (p, c) in self.iterparent(doc):
+            if not c.text:
+                continue
+
+            # To keep the output from screwing up the
+            # validation by putting a <div> inside of a <p>
+            # we actually replace the <p> in its entirety.
+
+            if c.text.find(self.config["marker"][0]) > -1:
+                for i in range(len(p)):
+                    if p[i] == c:
+                        p[i] = div
+                        break
+                    
+            if header_rgx.match(c.tag):
+                tag_level = int(c.tag[-1])
+                
+                # Regardless of how many levels we jumped
+                # only one list should be created, since
+                # empty lists containing lists are illegal.
+    
+                if tag_level < level:
+                    list_stack.pop()
+                    level = tag_level
+
+                if tag_level > level:
+                    newlist = etree.Element("ul")
+                    if last_li:
+                        last_li.append(newlist)
+                    else:
+                        list_stack[-1].append(newlist)
+                    list_stack.append(newlist)
+                    level = tag_level
+
+                # Do not override pre-existing ids 
+                if c.attrib.has_key("id"):
+                    id = c.attrib["id"]
+                else:
+                    id = self.config["slugify"][0](c.text)
+
+                # List item link, to be inserted into the toc div
+                last_li = etree.Element("li")
+                link = etree.SubElement(last_li, "a")
+                link.text = c.text
+                link.attrib["href"] = '#' + id
+
+                if int(self.config["anchorlink"][0]):
+                    anchor = etree.SubElement(c, "a")
+                    anchor.text = c.text
+                    anchor.attrib["id"] = id
+                    anchor.attrib["href"] = "#" + id
+                    anchor.attrib["class"] = "toclink"
+                    c.text = ""
+                else:
+                    c.attrib["id"] = id
+
+                list_stack[-1].append(last_li)
+
+class TocExtension (markdown.Extension):
+    def __init__(self, configs):
+        self.config = { "marker" : ["[TOC]", 
+                            "Text to find and replace with Table of Contents -"
+                            "Defaults to \"[TOC]\""],
+                        "slugify" : [self.slugify,
+                            "Function to generate anchors based on header text-"
+                            "Defaults to a built in slugify function."],
+                        "title" : [None,
+                            "Title to insert into TOC <div> - "
+                            "Defaults to None"],
+                        "anchorlink" : [0,
+                            "1 if header should be a self link"
+                            "Defaults to 0"]}
+
+        for key, value in configs:
+            self.setConfig(key, value)
+
+    # This is exactly the same as Django's slugify
+    def slugify(self, value):
+        """ Slugify a string, to make it URL friendly. """
+        import unicodedata
+        value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore')
+        value = unicode(re.sub('[^\w\s-]', '', value).strip().lower())
+        return re.sub('[-\s]+','-',value)
+
+    def extendMarkdown(self, md, md_globals) :
+        tocext = TocPostprocessor(md)
+        tocext.config = self.config
+        md.postprocessors.add("toc", tocext, "_begin")
+	
+def makeExtension(configs={}) :
+    return TocExtension(configs=configs)
-- 
cgit v1.2.3