On 30 May 2013, at 9:40 AM, Marian <[email protected]> wrote: > Sorry if it's a repost: > > I don't use routes_onerror or at least not when testing the parameter-based > routing. > > I tried to explicitly write the imagecontest app in the router but the I get > a 500.0 server error: > routers = dict( > BASE = dict( > path_prefix='applications', > default_application = 'imagecontest', > ), > imagecontest = dict(), > )
There's no need for an imagecontest dict if it's empty. It's possible that an empty router dict is causing the 500; I'm not sure it's been tested. The thing about routes_onerror is that the redirects are conditioned on it being present (the default is empty). So I'm thinking that the 303 messages below must be coming from a routes.py with routes_onerror defined (so maybe not a parametric router). Note that you need to restart web2py after changing routes.py. The reload-routes function in admin isn't useful if there are multiple server processes. > > > On Thursday, May 30, 2013 6:13:42 PM UTC+2, Jonathan Lundell wrote: > On 30 May 2013, at 8:46 AM, Marian <[email protected]> wrote: >> In rewrite.py I see two 303 responses in try_rewrite_on_error and >> try_redirect_on_error, I think one is responsible for the automatic redirect. > > I didn't notice that you had routes_onerror defined. What's its value? > > >> >> >> >> On Thursday, May 30, 2013 5:31:54 PM UTC+2, Jonathan Lundell wrote: >> On 30 May 2013, at 8:24 AM, Marian <[email protected]> wrote: >>> When I use the prefix w2p and try to open /applications I get: >>> >>> GET /applications /303 SEE OTHER >>> GET welcome 404 Not Found >>> >>> The endless loop are 303 SEE OTHERs >>> >>> Response: You are being redirected <a href="/applications/welcome">here</a> >> >> Any idea who's generating the redirects? The router does not. >> >>> >>> >>> >>> On Thursday, May 30, 2013 3:54:51 PM UTC+2, Jonathan Lundell wrote: >>> On 30 May 2013, at 6:36 AM, Marian <[email protected]> wrote: >>>> When I change the prefix to e.g. 'w2p' it directs me to /w2p/welcome an >>>> there is no endless loop. >>> >>> What are the details of the invalid-response message? >>> >>>> >>>> On Thursday, May 30, 2013 3:12:21 PM UTC+2, Massimo Di Pierro wrote: >>>> Do you get the endless loop if the prefix is different from "applications"? >>>> >>>> On Thursday, 30 May 2013 02:12:01 UTC-5, Marian wrote: >>>> Nobody had issues like this before? :( >>>> >>>> On Thursday, May 23, 2013 10:06:06 AM UTC+2, Marian wrote: >>>> I wanted to switch from pattern-based routing on a Windows IIS7 Server >>>> with isapi_wsgi to a parameter-based routing to benefit from nicer urls >>>> but it is not working. >>>> It either says invalid request or the server stucks in an endless loop. >>>> >>>> My functioning pattern based routing. As you can see I need to remove the >>>> prefix 'applications'. >>>> default_application = 'imagecontest' >>>> default_controller = 'default' >>>> default_function = 'index' >>>> >>>> routes_in=(('/applications/(?P<any>.*)','/\g<any>'),) >>>> routes_out=(('/(?P<any>.*)','/applications/\g<any>'),) >>>> >>>> >>>> What I tried... >>>> Invalid request (even without the path_prefix): >>>> routers = dict( >>>> BASE=dict( >>>> default_application = 'imagecontest', >>>> path_prefix = 'applications', >>>> ) >>>> ) >>>> >>>> >>>> >>>> Endless loop: >>>> routers = dict( >>>> BASE=dict( >>>> path_prefix = 'applications', >>>> ) >>>> ) >>>> >>>> Am I doing something wrong? >>>> >>>> >>> >>> >>> >>> -- >>> >> >> >> >> >> -- >> > > > > > -- > -- --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
