Expand internal links for split html documents

Internal links need to be converted to page.html#section_id
links so they work accross split html documents

1 files changed, 35 insertions, 1 deletions

diff --git a/panfry/document.py b/panfry/document.py
index aa29561..ce3a0ca 100755
--- a/panfry/document.py
+++ b/panfry/document.py
@@ -29,6 +29,39 @@ class Document:
     def set_templater(self, templater):
         self.templater = templater
 
+    def expand_int_links(self, source):
+        '''
+        Converts all internal links to full url links
+        so that links reference accross the split up html pages
+
+        Arguments:
+        source = original markdown source
+
+        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)
+        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):
+        print('get_link: %s' % m.groupdict())
+        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)
+
     @property
     def title(self):
         if self.header.firstline.startswith('%'):
@@ -165,7 +198,8 @@ class Document:
             doc.add_argument('toc')
             doc.add_argument('template=%s' % template_file.name)
             doc.add_argument('css=%s' % self.css_file)
-            doc.markdown = '%s\n%s' % (self.header.source, page.source)
+            doc.markdown = '%s\n%s' % (self.header.source, 
+                                       self.expand_int_links(page.source))
             content = doc.html
             write_file(os.path.join(pub_path, page.htmlfile),
                        unicode(content, 'utf-8'))