Catch All Script in App Engine Python (APP.YAML) does not work in Static Files

Question

I have tried everything but it seems that you cannot get a catch all url...

- url: /.*
  script: not_found.py  

...to work on urls that are based on static directory paths. eg. I can type in www.foobar.com/asdas/asd/asd/asd/ad/sa/das/d and I can get a nice custom 404 page. But if I alter a static path url like www.foobar.com/mydir/mydir/mypage.html, I just get the horrible generic 404....

Error: Not Found

The requested URL /mydir/mydir/mypage.html was not found on this server.

... I would like to alter whatever catches the url in directory paths and writes the 404. This appears the only way to get a consistent custom 404 page in GAE Python.

Can anyone help? I have written my website from scratch and have a very limited knowledge of Python. Achieving a consistent custom 404 is the only thing I cannot seem to overcome.

EDIT/ADD : OK I've added the kind suggestion of @Lipis , and gone through getting started which which thankfully has given me a much better understanding of classes (I sadly can't vote it up yet). But! I am using a .py script found on the net and I think the NotFound class is interfering with the class that gives my index page, because now my index page is the 404 page specified by the Jinja! I have very little understanding of MainHandler so I may have to give up for now.

import os
from google.appengine.ext import webapp
from google.appengine.ext.webapp import util
from google.appengine.ext.webapp import template
from google.appengine.ext.webapp.util import run_wsgi_app

import jinja2


class MainHandler(webapp.RequestHandler):
  def get (self, q):
    if q is None:
      q = 'index.html'

    path = os.path.join (os.path.dirname (__file__), q)
    self.response.headers ['Content-Type'] = 'text/html'
    self.response.out.write (template.render (path, {}))


class NotFound(webapp.RequestHandler):
      def post(self):
         # you need to create the not_found.html file
         # check Using Templates from Getting Started for more

         jinja_environment = jinja2.Environment(
         loader=jinja2.FileSystemLoader(os.path.dirname(__file__)))

         template = jinja_environment.get_template('404.html')
         self.response.out.write(template.render(template_values))


def main ():
  application = webapp.WSGIApplication ([('/(.*html)?', MainHandler),('/.*', NotFound)], 
                                      debug=True)

  util.run_wsgi_app (application)


if __name__ == '__main__':
  main ()

Answer 1

For better understanding I'll make some modifications on the Getting Started example which I assume that you have done it and you made some experiments with it.

It's not a good idea to have the static file for all the not found pages in the app.yaml since most likely you would like to show something more dynamic and usually the - url: /.* should be handled within your app.

In this example we are going to add a new RequestHandler for all your not found pages

import jinja2
import os
# more imports

jinja_environment = jinja2.Environment(loader=jinja2.FileSystemLoader(os.path.dirname(__file__)))

class MainPage(webapp2.RequestHandler):
    def get(self):
         template = jinja_environment.get_template('index.html')
         self.response.out.write(template.render(template_values))

class NotFound(webapp.RequestHandler):
    def get(self):
         # you need to create the not_found.html file
         # check Using Templates from Getting Started for more
         template = jinja_environment.get_template('not_found.html')
         self.response.out.write(template.render(template_values))

application = webapp.WSGIApplication(
                                     [('/', MainPage),
                                      ('/.*', NotFound)], # <-- This line is important
                                      debug=True)

But in order to make the jinja2 templates work, follow carefully the modifications that you need to do in Using Templates section from the Getting Started.

The order in the URL mapping is very important so this catch all regular expression (/.*) should be always the last one, because otherwise all the other rules will be skipped.