hardware.all() to api.model() would treat it as one big array rather than and array of sqlalchmey results,
I found @api.errorhandler but It works only with http error .
I want to catch error from business function and return httpexception depending of the error catched (by mappinp BusinessError - Httpexception)
do you have any idea how can I do it ?
Thank you
@api.errorhandler associates an exception of your choice with a message and http message to send back via a func. so@api.errorhandler(BusinessError)
def handle_business_error(error):
return {"message": "A logical error has occurred"}, 418
@api.errorhandler(AnotherBusiness):
def hand_another_business_error(error):
return {"message": "a different business logic error"}, 451provide api.errorhandler with the error you want need it to catch
db_scenario.to_domain() will raise, but just setup handlers for each one of those. You can see in my example, I'm just calling raise without an try/except code and it's caught, a 418 is returned, along with the message i passed it
db errors, and they all derive from a single err class MyDbError(Exception):
pass
class BadQuery(MyDbError):
pass
class BadFormatting(MyDbError):
passapi.errorhandler(MyDbError) and catch everything that has inherited MyDbError
what's it actually returning? What's the 500 and the stacktrace in the log?
I sent some screen , I reproduced your pastebin to catch an exception https://prnt.sc/123j36f and handle it with my function https://prnt.sc/123j4lg but when I test the request I have HttpError 500 instead of 418
what are the app logs saying in conjunction with the 500? Are the app logs quiet and it's just returning 500, or is the application throwing a stacktrace?
https://prnt.sc/1258amm
Here is the stacktrace
hello @j5awry
I successed to make the simple example on your pastebin working.
I tried to implement it on a very simple get routes but It didnt work , https://pastebin.ubuntu.com/p/8NWdRCgz38/
I made to many tests and it always return http error 500.
I'll continue my research
It looks flask raise http Error 500 before the exception get caught by the handler with debug mode on
I disabled debug mod and It works
I needed the debug mode on so I found another solution
I set PROPAGATE_EXCEPTIONS = False
https://prnt.sc/12956at
But It looks weird because we can see here https://github.com/python-restx/flask-restx/blob/66471c5223e0237762fc2722d680229bb68e0d16/flask_restx/api.py#L675
It says the opposite, so I am a little confused
@j5awry
flasgger is doing could be an interesting entry point. but it's going to take code change in flask-restx
We tracked it down to the fact that the error handler was registered on the flask api, rather than the api instance. Fixed that and it worked as expected, but my questions is:
Should it have ignored error handlers on the app and only looked at api error handlers?
@Gidgidonihah : Reading the code, the intent is that errorhandlers get checked for the restx.API and it's children (restx.API.namespace). It wouldn't make sense, to me, for the framework to go up the chain as well.
So self + children vs. self + parents + children. self + children makes sense to me for checking handlers. But I may be misinterpreting the question?
1 reply
I have a question about marshal models.
How can I make a self reference in a model ?
https://prnt.sc/12kvzl2
I want my model 'error' can contain himself recursively.
But i'm struggling with ...
Is it possible without using marshmallow ?
@Abdur-rahmaanJ If you don't mind me asking, is this an on-prem or cloud hosted? And is it possible to add authentication at other layers? It depends a lot on your setup, but something like using JWT with NGinx, or x509 based auth at Apache Webserver are other possibilities. But there's a lot of nuance there. I do think we need to figure out a better way of handling the pre-generated doc endpoint, and I think what
flasggeris doing could be an interesting entry point. but it's going to take code change inflask-restx
Cloud-hosted, server-based auth is a great option also but see the above
1 reply
