Fix generation of cross page header referencing

2 files changed, 25 insertions, 19 deletions

diff --git a/panfry/document.py b/panfry/document.py
index 969d5f4..4379f6e 100755
--- a/panfry/document.py
+++ b/panfry/document.py
@@ -40,26 +40,24 @@ class Document:
         returns: markdown source with expanded links
         '''
         re_link1 = re.compile(r'(\[(?P<s>[^\]^]+)\])\((?P<l>#.*)\)', re.S)
-        re_link2 = re.compile(r'(\[(?P<s>[^\]^]+)\]): (?P<l>#[a-z0-9-]+)',
-                              re.S)
-        re_link3 = re.compile(r'(\[(?P<s>[^\]^]+)\][^(])', re.S)
+        re_link2 = re.compile(r'(\[(?P<s>[^\]^]+)\][^(])', re.S)
         source = re_link1.sub(self._get_cross_link, source)
         source = re_link2.sub(self._get_cross_link, source)
-        source = re_link3.sub(self._get_cross_link, source)
 
         return source
 
     def _get_cross_link(self, m):
         ref_txt = m.group('s')
-        if 'l' in m.groupdict():
-            header = m.group('l')[1:]
-        else:
-            header = ref_txt.lower().replace(' ', '-')
 
         for page in self.pages:
             for tocitem in page.toc:
-                if tocitem.section_id == header:
-                    return '[%s](%s#%s)' % (ref_txt, page.htmlfile, header)
+                if tocitem.heading == ref_txt.replace('\n', ' '):
+                    return '[%s](%s#%s)' % (ref_txt,
+                                            page.htmlfile,
+                                            tocitem.header_id,
+                                            )
+
+        return '[%s]' % ref_txt
 
     @property
     def title(self):
@@ -136,7 +134,7 @@ class Document:
                 toc = [page.toc[0]]
 
             for tocitem in toc:
-                link = '%s#%s' % (page.htmlfile, tocitem.section_id)
+                link = '%s#%s' % (page.htmlfile, tocitem.header_id)
                 links.append(dict(link=unicode(link, "utf8"),
                                   text=unicode(tocitem.heading, "utf8"),
                                   level=tocitem.level,
@@ -227,7 +225,7 @@ class Document:
             doc.add_argument('template=%s' % template_file.name)
             doc.add_argument('css=%s' % self.css_file)
             doc.markdown = '%s\n%s' % (self.header.source, 
-                                       self.expand_int_links(page.source))
+                                       self.expand_int_links(page.markdown))
             content = doc.html
             write_file(os.path.join(pub_path, page.htmlfile),
                        unicode(content, 'utf-8'))
diff --git a/panfry/page.py b/panfry/page.py
index 76620bc..fdf7c86 100755
--- a/panfry/page.py
+++ b/panfry/page.py
@@ -1,6 +1,7 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 import re
+import pandoc
 
 
 class TextFile:
@@ -32,9 +33,10 @@ class TOCItem:
         self.heading = heading
         self.level = level
 
-    @property
-    def section_id(self):
-        return self.heading.lower().replace(' ', '-')
+    def header_id(self):
+        remove = re.compile('[+~!@#$%^&*\(\){}\[\];:"\',.<>?/\`]')
+        header_id = self.heading.lower().replace('\n', ' ').replace(' ', '-')
+        return remove.sub('', header_id)
 
 
 class Page(TextFile):
@@ -42,6 +44,15 @@ class Page(TextFile):
     Represents single page source file.
     '''
 
+    def __init__(self, filename, source, format='markdown'):
+        TextFile.__init__(self, filename, source)
+        self.doc = pandoc.Document()
+        self.doc._input(self.source, format=format)
+
+    @property
+    def markdown(self):
+        return self.doc.markdown
+
     @property
     def title(self):
         '''
@@ -74,10 +85,7 @@ class Page(TextFile):
     @property
     def toc(self):
         '''
-        Returns a list of tuples representing the headings/sub-heading
-        structure of the page.
-        The first element of each tuple is the heading text.
-        The second element of each tuple is the heading level.
+        Returns a list of TOCItems of the page.
         '''
         toc = []
         heading = ''