Clarify the implications of reloading routes and/or using symlinks in the applications folder. about web2py

[niphlod]

From [email protected] on June 10, 2012 21:28:46

routes.py is not a module that gets imported -- it is a file that is read and executed -- so track changes should not have any effect on it.

Reloading routes while web2py is running is supported. The only potential problem is with existing pages open in browsers at the time a change is made -- links, Ajax requests, and static file requests from those pages may not work as expected because of the change in the rewrite logic. However, this problem would not be resolved by stopping the webserver before reloading routes, and I don't think it's specific to the web2py rewrite system. It's also not easy to resolve because you can't know for how long existing web pages will be kept open and therefore might continue to make Ajax requests or generate link clicks.

[niphlod]

From [email protected] on June 10, 2012 21:34:13

Ok, where in The Book do I find that info that routes is not imported, but read and executed? I don't recall coming across that, so, while it is true, it would not be a safe conclusion to the reader unless it is called out somewhere.

As to the issue of even stopping the server not making it safe to change routes... Hmmm, then I wonder what it means for changing them to be safe. Maybe something about routes could be hashed into a cookie or session so that when the server comes back up, any open pages still generating traffic could be detected?

[niphlod]

From [email protected] on August 07, 2012 20:07:02

Labels: -Type-Defect Type-Enhancement