We have autotemplating! :D

articles/__init__.py

@@ -1,31 +1,69 @@
+"""Route requests for articles according to shoofle's rules.
+
+This is a simple module which is basically entirely here to provide access, in a sensible location,
+to the `bloop` object. It's a blueprint which describes how to route requests for articles on my website
+(possibly located at http://shoofle.net, possibly not located there). Most of the interesting stuff is in
+`render_article`; Go team!"""
+
 import os
 from flask import Blueprint, render_template, abort
-from jinja2.exceptions import TemplatesNotFound
 
 folder = "articles"
+article_base_template = os.path.join(folder, "article.template.html")
 bloop = Blueprint("articles", __name__, template_folder="")
 
 @bloop.route("/")
 def main_page():
+	"""Renders the list of articles/projects."""
 	return render_template("project_list.html")
 
-@bloop.route("/a/<page_name>/")
-def apply_base_template(page_name):
+@bloop.route("/raw/<path:page_name>/")
+def render_file(page_name):
+	"""Does nothing. Not really sure why I have this."""
 	file_name = os.path.join(folder, page_name.replace("-", "_"))
 	
 @bloop.route("/<page_name>/")
 def render_article(page_name):
-	# Arguably, the various options for how to render (templates, articles, flat html) could be stuck into various subdirectories.
-	# Ultimately I don't want to do this because it's supposed to be lightweight - I should be able to chuck this __init__.py file into
-	# any folder and start generating stuff correctly. But whatever!
+	"""Renders a requested article! This should always be @routed last, because it catches a 
+	wide variety of requests. As a result, other things need to be @routed first, because they 
+	might never get called if this catches them first."""
+	# Arguably, the various options for how to render (templates, articles, flat html) could be stuck into various 
+	# subdirectories. Ultimately I don't want to do this because I want this to be lightweight - this __init__.py file 
+	# can be chucked into any folder and start showing pages correctly. But whatever!
 
-	# Let's find some test filenames!
-	file_name = os.path.join(folder, page_name.replace("-", "_"))
-	
-	# Is there a template by that name? This list is in the priority order for rendering.
-	extensions = [".template.html", ".article.html", ".html", ""]
-	
-	try:
-		return render_template([ file_name + suffix for suffix in extensions ])
-	except TemplatesNotFound as e:
-		abort(404)
+	# In the examples, let's think about a request for "example.com/some-article".
+
+	# First, we convert the important part of the requested page into a filename
+	# "example.com/Some-Article/" => folder="Some-Article" => file_name = "articles/some_article"
+	file_name = os.path.join(folder, page_name.replace("-", "_").lower())
+
+	# Here's the priority list for file rendering!
+	if os.path.isfile(file_name + ".template.html"):
+		# If the file "articles/some_article.template.html" exists, then there's a specific {template} written 
+		# for this path. Specific page {templates} take priority.
+		return render_template(file_name + ".template.html")
+	if os.path.isfile(file_name + ".article.html"):
+		# If "articles/some_article.article.html" exists but there's no template, then we should render that
+		# {article fragment}, but using the {article base template}. In the future, this should possibly also
+		# extract the title from the {article fragment} and feed it into the {article base template} as well.
+		return render_template(article_base_template, target=file_name + ".article.html")
+	if os.path.isfile(file_name + ".html"):
+		# If we haven't found any results yet, check to see if "articles/some_article.html" exists. If it does,
+		# just display it plain. This also provides a clean way to access the raw form of files like
+		# "articles/some_article.template.html" - just make a request for "example.com/some-article.template/"
+		# and it will be caught by this rule and rendered.
+		return render_template(file_name + ".html")
+	if os.path.isfile(file_name):
+		# If it didn't match any other rules, then just render the file that has precisely the requested name.
+		return render_template(file_name);
+
+	# I *believe* there's one instance that can't be accessed by this kind of routing in any way:
+	# If the files "articles/some_article" and "articles/some_article.html" both exist, then no request will
+	# convince this blueprint to return the former. However, as sacrifices go, I don't think it's too bad, and 
+	# that should be the only case when this happens.
+
+	# If we didn't find any files, throw up a 404.
+
+	abort(404)
+
+"""Th-th-th-that's all, folks!"""

articles/article.template.html

@@ -11,6 +11,7 @@
 	</head>
 	<body>
 		{% block body -%}
+		{%- include target -%}
 		{%- endblock %}
 	</body>
 </html>

tornado_server.py

@@ -1,12 +0,0 @@
-import tornado.ioloop
-import tornado.web
-
-
-application = tornado.web.Application([
-	(r"/(.*)", tornado.web.StaticFileHandler, {"path":"./"})
-], debug=True)
-
-if __name__ == "__main__":
-	application.listen(80)
-	tornado.ioloop.IOLoop.instance().start()
-