Hello, thank you for your quick reply ! It is great that you invest so much time.
I cleaned up my configuration last night and started "fresh" with web2py Version 1.61.4. My problem with broken URLs behind my proxy is still there, but I think I went the wrong way by trying to fix it by using mod_proxy_html (http://apache.webthing.com/mod_proxy_html/). It should (and mostly does) work by "routes.py". My Config: Apache on Linux as a Proxy. It routes requests "https:/<server>:443/web2py/<appl>/.." to "http:/localhost:8000/ <appl>/..": -------- <VirtualHost _default_:443> .... <Location /web2py/> ProxyPass http://localhost:8000/ ProxyPassReverse / </Location> .... </VirtualHost> -------- On port 8000 I have the Standalone-CherryPy-Server with the "routes.py": -------- #!/usr/bin/python # -*- coding: utf-8 -*- routes_in = () routes_out = ( ('/(?P<file>[\w\./_-]+)', '/web2py/\g<file>'), ) -------- This makes every URL in the web-pages start with "/web2py/..", so if I click a link in an application, it correctly finds it's way to CherryPy via Apache. And here's the (remaining) problem: Some Pages are somehow "pre-generated" and therefore do not know about the changed "routes.py". Best I could do is change the "routes.py" immediately after unpacking the installation zip, before the "first-time-start" where the applications are generated. Page "Hello World" still contains a Link "/admin/" where "/web2py/ admin/" should be, In the Page "Installed Aplications": Link "Examples" points to "/ examples/default/index", but "Design" points to "/web2py/admin/default/design/examples" (this is correct behavior) The Green Menu on Top of the Page is generally not correct: for example "Logout" points to "/admin/default/logout" instead of "/web2py/admin/ default/logout" So the Question: How do I regenerate all the affected Files (js, css, html, whatever) after changes in "routes.py" ? It does not need to be automatic, because routes.py is seldom changed. If there is a simple way, then it would be nice to add it as a comment in "routes.py": After you change this file, issue the command "....." to update the relevant pages. Thank you for this great product, Bernd On 22 Apr., 18:09, mdipierro <mdipie...@cs.depaul.edu> wrote: > I am not sure I understand so I apologize if the answer is incomplete. > > I do not know what mdp_proxy_html does vs mod_proxy. > > mod_proxy is required if you want to have another database server run > as a proxy, but its job is not to rewrite URL, although you can do > that. > > You can also have web2py rewrite URL using routes.py (an example in > routes.example.py). > > routes supports both url mapping (routes_in) and reverse url mapping > (routes_out) and if they match there is no problem. routes.py does not > parse the generated for urls. So you should not do this: > > jQuery(document).ready(function(){ setTimeout("keepalive('/yourapp/ > default/action')",10000); }); > > You should do this instead > > jQuery(document).ready(function(){ setTimeout("keepalive('{{=URL > ('yourapp','default','action')}})'",10000); }); > > Anyway I have a feeling you know about this more than I do. If you can > propose any impeovement to the current mechanism, please let us know. > > Massimo > > On Apr 22, 5:41 am, Bernd das Brot <b.b...@gmx.at> wrote: > > > > > Hi web2py fans ! > > > I want to use web2py (current production version 1.61.1) behind an > > apache proxy, alongside other applications. > > Therefore I need a unique URL-Prefix to guide mod_proxy to the > > (standalone) web2py-Server > > (Application "cookbook":https://server:port/prefix/cookbook/...) > > > 1) I followed the instructions in the FAQs using "mod_proxy_html" and > > it kind-of-works: > > it rewrites URLs like "<a href="/cookbook/..>" to <a href="/prefix/ > > cookbook/">" BUT > > in Javascript it does not understand things like > > > jQuery(document).ready(function(){ setTimeout("keepalive('/admin/ > > default/keepalive')",10000); }); > > > and therefore does not rewrite it. Result: no editor-Keepalive, but > > (after 10 seconds) an red text > > "Communication Error". > > > How can I fix this in mod_proxy_html (No, I do not want to hardcode > > it into the framework) ? > > > 2) Can somebody with deeper knowledge of web2py explain, why I need > > mod_proxy_html anyway ? > > > IMHO the use of mod_proxy_html is a workaround for something a > > decent framework should do. > > "route.py" does "half" of it but "breaks your application" (!): > > The "URL()" function does not use the variables from "route.py", > > so the absolute URLs > > (in page code and in javascript) are not proxy-aware (but could be > > - I don't think that this is > > too hard). > > > Thanks for any replies, Bernd- Zitierten Text ausblenden - > > - Zitierten Text anzeigen - --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "web2py Web Framework" group. To post to this group, send email to web2py@googlegroups.com To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/web2py?hl=en -~----------~----~----~----~------~----~------~--~---
