Hi,
I'm trying to call web2py.gluon.main.wsgibase depending on whether a
url path matches.
The code is like this:
[...]
os.chdir(web2py_dir)
import web2py.gluon.main
os.chdir(current_dir)
[...]
if [...]:
httpd = wsgiref.simple_server.make_server('', int(options.port),
self)
httpd.serve_forever()
else:
socket_name = options.socket
flup.server.fcgi.WSGIServer(self,
bindAddress=socket_name).run()
where "self" has __call__(self, environ, start_response) and will call
web2py.gluon.main.wsgibase or another wsgi app.
When I try to visit the page (simple_server or flup with lighttpd), I
get this error:
Traceback (most recent call last):
File "[redacted]/web2py/gluon/restricted.py", line 192, in
restricted
exec ccode in environment
File "[redacted]/web2py/applications/[redacted]/models/db.py", line
35, in <module>
auth = Auth(db) # authentication/
authorization
File "[redacted]/web2py/gluon/tools.py", line 805, in __init__
request = current.request
AttributeError: 'thread._local' object has no attribute 'request'
The admin application runs fine however. It says: Version 1.97.1
Running on WSGIServer/0.1 Python/2.7.1+
Will web2py work in call chains like this? I'm not sure how this is
different than running plain web2py with web2py built-in servers,
doesn't it end up calling wsgibase anyway?
Thanks for your help.
