switched to mistune (over misaka) for markdown parsing, added table of contents

6 files changed, 52 insertions, 20 deletions

diff --git a/app.py b/app.py
index 08e45ff..8e1e703 100644
--- a/app.py
+++ b/app.py
@@ -57,7 +57,7 @@ def index():
             return flask.render_template(
                 "index.jinja", 
                 **get_template_items("eden's site :3", db),
-                markdown = parser.parse_text(f.read()),
+                markdown = parser.parse_text(f.read())[0],
                 featured_thoughts = db.get_featured_thoughts(),
                 tweets = db.get_cached_tweets(7) + [("view all tweets...", db.get_my_twitter())],
                 commits = db.get_cached_commits(since = datetime.datetime.now() - datetime.timedelta(days = 7))
@@ -93,7 +93,8 @@ def get_thought():
     thought_id = flask.request.args.get("id", type=int)
     with database.Database() as db:
         try:
-            category_name, title, dt, parsed = parser.get_thought_from_id(db, thought_id)
+            category_name, title, dt, parsed, headers = parser.get_thought_from_id(db, thought_id)
+            print(headers)
         except TypeError:
             flask.abort(404)
             return
@@ -102,6 +103,7 @@ def get_thought():
             **get_template_items(title, db),
             thought = True,
             dt = "published: " + str(dt),
+            headers = headers,
             category = category_name,
             othercategories = db.get_categories_not(category_name),
             related = db.get_similar_thoughts(category_name, thought_id)
diff --git a/database.py b/database.py
index 3dd2d9a..42833d5 100644
--- a/database.py
+++ b/database.py
@@ -247,6 +247,7 @@ def request_recent_commits(since = datetime.datetime.now() - datetime.timedelta(
     return sorted(out, key = lambda a: a["datetime"], reverse = True) 
 
 if __name__ == "__main__":
+    import parser
     with Database() as db:
         # print(db.get_similar_thoughts("about me", 5))
-        print(db.get_iso_cd_options())
\ No newline at end of file
+        print(parser.parse_file("cpus.md"))
\ No newline at end of file
diff --git a/parser.py b/parser.py
index d79d400..4044fb8 100755
--- a/parser.py
+++ b/parser.py
@@ -6,17 +6,19 @@ from pygments.formatters import HtmlFormatter, ClassNotFound
 from pygments.lexers import get_lexer_by_name
 import urllib.parse
 import webbrowser
+import lxml.etree
+import lxml.html
 import database
 import argparse
 import getpass
 import houdini
-import misaka
+import mistune
 import app
 import sys
 import re
 import os
 
-class HighlighterRenderer(misaka.SaferHtmlRenderer):
+class EdawebRenderer(mistune.HTMLRenderer):
     def blockcode(self, text, lang):
         try:
             lexer = get_lexer_by_name(lang, stripall=True)
@@ -29,42 +31,55 @@ class HighlighterRenderer(misaka.SaferHtmlRenderer):
         # default
         return '\n<pre><code>{}</code></pre>\n'.format(houdini.escape_html(text.strip()))
 
-    def blockquote(self, content):
+    def block_quote(self, content):
         content = content[3:-5] # idk why this is required...
         out = '\n<blockquote>'
         for line in houdini.escape_html(content.strip()).split("\n"):
             out += '\n<span class="quote">{}</span><br>'.format(line)
         return out + '\n</blockquote>'
 
-    def image(self, link, title, alt):
+    def image(self, link, text, title):
         return "<a href='%s' target='_blank'><img alt='%s' src='%s'></a>" % (
-            urlparse(link)._replace(query='').geturl(), alt, link
+            urlparse(link)._replace(query='').geturl(), text, link
         )
 
-    def header(self, content, level):
-        # if level > 1:
-        hash_ = urllib.parse.quote_plus(content)
+    def heading(self, text, level):
+        hash_ = urllib.parse.quote_plus(text)
         return "<h%d id='%s'>%s <a class='header_linker' href='#%s'>[#]</a></h%d>" % (
-            level, hash_, content, hash_, level
+            level, hash_, text, hash_, level
         )
-        # else:
-        #     return "<h1>%s</h1>" % content
 
 def get_thought_from_id(db, id_):
     category_name, title, dt, markdown = db.get_thought(id_)
-    return category_name, title, dt, parse_text(markdown)
+    html, headers = parse_text(markdown)
+    return category_name, title, dt, html, headers
 
 def parse_file(path):
     with open(path, "r") as f:
         unformatted = f.read()
 
-    return parse_text(unformatted)    
+    return parse_text(unformatted)[0]    
 
 def parse_text(unformatted):
-    renderer = HighlighterRenderer()
-    md = misaka.Markdown(renderer, extensions=('fenced-code', 'quote'))
+    md = mistune.create_markdown(
+        renderer = EdawebRenderer(), 
+        plugins = ["strikethrough", "table", "url", "task_lists"]
+    )
+    html = md(unformatted)
+    root = lxml.html.fromstring(html)
+
+    headers = []
+    for node in root.xpath('//h1|//h2|//h3|//h4|//h5//h6'):
+        headers.append((
+            # lxml.etree.tostring(node),
+            # "<p>%s</p>" % urllib.parse.unquote_plus(node.attrib["id"]),     # possibly insecure?
+            urllib.parse.unquote_plus(node.attrib["id"]),
+            int(node.tag[-1]),                                              #   -horrible hack
+            "#%s" % node.attrib["id"])
+        )
+    # print(headers)
 
-    return md(unformatted)
+    return html, headers
 
 def preview_markdown(path, title, category):
     def startBrowser():
diff --git a/requirements.txt b/requirements.txt
index c896a79..90f647f 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -16,4 +16,4 @@ PasteScript==3.2.0
 waitress
 houdini.py
 Pygments
-misaka
+mistune
diff --git a/static/style.css b/static/style.css
index 8fcfa6f..89d8814 100644
--- a/static/style.css
+++ b/static/style.css
@@ -108,6 +108,14 @@ aside {
     float: right;
 }
 
+#header_linkers {
+    font-size: xx-small;
+}
+
+#header_linkers li {
+    list-style: none;
+}
+
 #tags {
     font-size: xx-small;
 }
diff --git a/templates/template.jinja b/templates/template.jinja
index 77bf38c..f28163e 100644
--- a/templates/template.jinja
+++ b/templates/template.jinja
@@ -42,6 +42,12 @@
             {% if thought %}
                 <aside>
                     <h4>{{dt}}</h4>
+                    <h5>contents:</h5>
+                    <ul id="header_linkers">
+                        {% for text, depth, link in headers %}
+                            <li>{{ '#'*depth }} <a href="{{ link }}">{{ text }}</a></li>
+                        {% endfor %}
+                    </ul>
                     <h5>this category:</h5>
                     <ul>
                         <li><b><a href="{{'/thoughts#'+category.replace(' ', '_')}}">{{category}}</a></b></li>