每个环境的web2py配置不适用于计划程序/脚本

Question

I have been using the recommended method for setting up DB connections per environment in web2py:

settings = dict()
if request.env.http_host == 'http://127.0.0.1:8000'
     settings['development']=True
else:
     settings['development']=False
if settings['development']:
     db = DAL('sqlite://....')
else:
     db = DAL('mysql://....')

When using the scheduler in the non-development environment the db connections fail as it is using the port 8000 config.

Similarly when attempting to run a script I get the same errors. This is my last attempt:

python web2py.py -S app -M -N -R -L options.py -P 443 applications/testscript.py

"Can't connect to MySQL server on '127.0.0.1' (10061)"�

Is there a recommended way to inject a config per environment that will apply outside of web access?

Answer 1

Temporarily insert the following on a page:

{{=response.toolbar()}}

Then on both development and production, click the "request" button and inspect the values in request.env to identify things that differ between the two environments. Some likely candidates are "server_signature", "server_software", and "web2py_path".

Answer 2

I had the same concern and I first though of request.env.server_name . But:

• this variable does not exist when starting the scheduler
• it could be replaced by request.env.cmd_options.server_name but first, this one does not exist for the server and second, after its startup, the scheduler does not find it any more...

Then I found this Q/A and tried to solve it following Anthony's advice:

• "server_signature" does not exist when starting up the scheduler
• "server_software" does not exist when starting up the scheduler

Finally, "web2py_path" does the job even if for me, testing against a full installation path is not really a 'clean' solution...

if request.env.web2py_path == '/your/path/on/server/to/web2py'
    settings['development']=False
else:
    settings['development']=True

Now it should work, but I fell into this one : Can not run the web2py scheduler using postgresql database