Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 02:59
    sqlalchemy-bot closed #6449
  • 02:59

    sqlalchemy-bot on master

    Ensure propagate_to_loaders hon… Merge "Ensure propagate_to_load… (compare)

  • 02:17
    zzzeek milestoned #6449
  • 02:17
    zzzeek labeled #6449
  • 02:17
    zzzeek labeled #6449
  • 02:17
    zzzeek labeled #6449
  • 02:17
    zzzeek unlabeled #6449
  • 02:17
    zzzeek labeled #6449
  • 02:17
    zzzeek unlabeled #6449
  • 02:17
    zzzeek edited #6449
  • May 08 18:49
    CaselIT labeled #6447
  • May 08 01:42
    zzzeek closed #6444
  • May 08 01:27
    zzzeek locked #6450
  • May 08 01:27
    zzzeek closed #6450
  • May 08 01:10
    zzzeek labeled #6449
  • May 08 01:10
    zzzeek unlabeled #6449
  • May 08 01:10
    zzzeek labeled #6449
  • May 07 22:19
    Dsinghbailey opened #6450
  • May 07 21:26
    awgymer labeled #6449
  • May 07 21:26
    awgymer opened #6449
Bryan Forbes
@bryanforbes
we can do that in the future, though
Federico Caselli
@CaselIT

@CaselIT I think I’m gonna hold off on the copied properties going into their own protocol

ok then

i can still answer questions though!

sorry, was repling to a pr

I was trying to came up with a solution for https://gerrit.sqlalchemy.org/c/sqlalchemy/sqlalchemy/+/2629
basically I would like to avoid leaving all the only_once callbacks there
mike bayer
@zzzeek
yeah i had not been able to figure out a way to do that
Federico Caselli
@CaselIT
but I'm not sure how the best way would be for it. My idea was to make it return something that identifies the fact that's a once, since is is cheap and should not cause too much overhead
that part is really hard to follow btw. I'm still not sure what half of the stuff does
mike bayer
@zzzeek
yah me neithr :)
Federico Caselli
@CaselIT
do you think the return thing would work?
mike bayer
@zzzeek
im not really sure i hitnk overall some system outside of the event.call() has to know to remove it
Federico Caselli
@CaselIT
yes. the next step was to recreate the event key from the callback
so I can call remove on it
not figured out that part yet
also is only_once public or can I move it around?
mike bayer
@zzzeek
yeah i would ahve to dig into that code again and im a little too headachy today
the...keyword arugment to event.listen() ? i think it's public but nobody uses it
Federico Caselli
@CaselIT
no the wrapper thing in utils
mike bayer
@zzzeek
oh, that's whatever
Federico Caselli
@CaselIT
ok.
I think I'll start by passing the stuff to create the event key when creating the wrapper, just to see what breaks with that change. when in review you can point to better alternative
mike bayer
@zzzeek
ok
Federico Caselli
@CaselIT
thanks
@bryanforbes are you working on sqlalchemy/sqlalchemy2-stubs#51 ? If not I think it can be merged and anything that I mentioned could be done in a later pr
Bryan Forbes
@bryanforbes
yeah, I’ll take care of it
Federico Caselli
@CaselIT
ok

yeah, I’ll take care of it

sorry, rereading I wonder if I understood wrong. are you working on it or you can take care of it later. sorry :)

Bryan Forbes
@bryanforbes
I’m working on it now
Federico Caselli
@CaselIT
ok
Bryan Forbes
@bryanforbes
@CaselIT pushed
Federico Caselli
@CaselIT

do you think the return thing would work?

it does not, since retvat=True is a thing, that while not really useful with once=True works at least once

mike bayer
@zzzeek
I would just be adding the listener to some collection somewhere
Federico Caselli
@CaselIT
oh retval=True event behave differently
meaning thay are not called by the event but explicetly
Federico Caselli
@CaselIT
Bryan Forbes
@bryanforbes
@zzzeek the update to strict mode is large, but no new functionality is added
mike bayer
@zzzeek
@bryanforbes whichj is this
I just pushed it
mike bayer
@zzzeek
wow mr fast
Bryan Forbes
@bryanforbes
lol
I’ve been working on it since late last week
mike bayer
@zzzeek
so, "strict mode", meaning the plugin itself passes mypy --strict ?
Bryan Forbes
@bryanforbes
yup
mike bayer
@zzzeek
failed orderinglist1
err 2
Bryan Forbes
@bryanforbes
lies
mike bayer
@zzzeek
well both
test_mypy[orderinglist1.py] FAILED [ 69%]
test/ext/mypy/test_mypy_plugin_py3k.py::MypyPluginTest::test_mypy[orderinglist2.py] FAILED [ 72%]
Bryan Forbes
@bryanforbes
ok, I’ll look into it tomorrow
mike bayer
@zzzeek
ok