#pyramid

but you will wnat to use SignedCookieSessionFactory

UnencryptoedCookiesessionwhatever is deprecetad.

thanks, I'll use it with a random secret.

I've found a crazy long discussion about sessions on the google group, https://groups.google.com/forum/#!topic/pylons-...

I've given up reading it after 20 mins, SignedCookieSessionFactory seems to be ok, that big red warning is a bit too much possibly. as read in the discussion, CSRF shouldn't be a problem if it's in a cookie

x58: https://gist.github.com/mmerickel/d83ebca5a6ad1...

[pyramid] stevepiercy pushed 2 new commits to master: http://git.io/vlPlJ

pyramid/master 8aa1c2a Steve Piercy: minor grammar, fix .rst markup, rewrap to 79 columns

pyramid/master bad95cd Steve Piercy: Merge pull request #2072 from stevepiercy/master...

[pyramid] stevepiercy pushed 2 new commits to 1.6-branch: http://git.io/vlP87

pyramid/1.6-branch 5fa4e31 Steve Piercy: minor grammar, fix .rst markup, rewrap to 79 columns...

pyramid/1.6-branch 9130d55 Steve Piercy: Merge pull request #2073 from stevepiercy/1.6-branch...

Good morning.

morning

[pyramid] stevepiercy pushed 2 new commits to 1.5-branch: http://git.io/vlPze

pyramid/1.5-branch d88b3fa Steve Piercy: minor grammar, fix .rst markup, rewrap to 79 columns...

pyramid/1.5-branch 06587db Steve Piercy: Merge pull request #2074 from stevepiercy/1.5-branch...

Ergo: nice

morning

Hi all!

I have problem trying to understand this comments – https://github.com/piotr-dobrogost/pyramid/blob...

Specifically, it seems to me that to enable overriding settings further (by user code) we _should_ call .commit() at the end of this method witch is opposite of what's stated in these comments.

"(...) the following registrations should be treated as analogues of methods called by the user after configurator construction" If so then shouldn't we call submit() to give a chance to override those? If we do not call commit() that further registrations would "compete" with default ones thus creating conflicts instead of overriding default registrations.

What am I missing here?

to override user stuff with user stuff you should commit() in-between

see lines 385-385?

*385-386

if there were no self.commit(), then calling config.add_view(... Exception) would result in conflict

or you had to call config.commit() right after constructing it

Zart: Ok, but what's different about the rest of calls that commit is not called "for them"?

those are provided by user itself

exceptionresponse_view is added by framework, while other stuff is taken from Configurator constructor arguments

you are not expected to register same stuff twice 8)

ie, you'll get conflict if you try to add something you already provided to configurator

pdobrogost, are you writing your own command handlers?

Zart: No, I don't think so :) I'm trying to find a way to fix https://github.com/Kotti/Kotti/issues/475

The comment says something about add-ons? How do they enter picture here?

iirc there is also a problem with translation dirs ordering

Yes, there is https://github.com/Pylons/pyramid/issues/1474

asset overriding works on import level afair

And that means that...?

Zart: Do you have idea why code at the end of comment https://github.com/Kotti/Kotti/issues/475#issue... does not trigger "pyramid.exceptions.ConfigurationError: "<path-to-override>/locale" is not a directory" error?