Fix Marshmallow Schema's Meta exclude field error

[blagasz]

Mixing tuple and list when collecting fields in the Marshmallow extension may result in error.

[lafrech]

Good.

Do you think you could test all combinations ([dump_only; exclude] x [tuple; list])?

Also, there are CI failures.

You should use a set, not a tuple, in the test, as the order doesn't matter (or perhaps better add ordered=True to the schema Meta).

(Note: no need to close/recreate a PR, you can amend/rebase and then push -f.)

[blagasz]

Ok, I check that. (Yes, I know, but I'm not a big git magician and accidentally added a second commit and it would have been very complicated for me to remove. From now on I will do the rebase and push -f combo.)

…

On Fri, 12 Apr 2019 at 09:22, Jérôme Lafréchoux ***@***.***> wrote: Good. Do you think you could test all combinations ([dump_only; exclude] x [tuple; list])? Also, there are CI failures. You should use a set, not a tuple, in the test, as the order doesn't matter. (Note: no need to close/recreate a PR, you can amend/rebase and then push -f.) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#431 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFQNeLxeLtuxN-a3tZEM8Md3IYREFRFLks5vgDRGgaJpZM4cq9sU> .

[blagasz]

pyupgrade and black fails but I don't really know from the log what their problem is.

[lafrech]

Yes, the log is not verbose enough.

If you're using Python 3.6+ you can install pre-commit to do that locally before pushing.
https://github.com/marshmallow-code/apispec/blob/dev/CONTRIBUTING.rst

[lafrech]

Sorry, I was not notified on the force-push.

I'd be tempted to factorize the tests using some obfuscating pytest parametrization trickery, but that's just personal taste/obsession. I can do that later on if I have spare time to waste.

I think this is good to merge.

Thanks.

[blagasz]

Ok, as you wish :) Thanks for your support in this!

…

On Thu, 18 Apr 2019 at 16:18, Jérôme Lafréchoux ***@***.***> wrote: Sorry, I was not notified on the force-push. I'd be tempted to factorize the tests using some obfuscating pytest parametrization trickery, but that's just personal taste/obsession. I can do that later on if I have spare time to waste. I think this is good to merge. Thanks. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#431 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABKA26BK6JHFNCOY5HHWA5TPRB7KFANCNFSM4HFL3MKA> .

[lafrech]

I did the rework and released in 1.2.1.