Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Sep 23 23:00
    lucafs commented #21101
  • Sep 23 22:33
    lucafs commented #21101
  • Sep 23 22:33
    lucafs commented #21101
  • Sep 23 21:51
    timhoffm commented #21126
  • Sep 23 20:08
    QuLogic milestoned #21161
  • Sep 23 19:55
    jklymak commented #21162
  • Sep 23 19:34
    LinasNas labeled #21162
  • Sep 23 19:34
    LinasNas opened #21162
  • Sep 23 19:26
    QuLogic commented #21126
  • Sep 23 19:14
    aitikgupta edited #20740
  • Sep 23 19:13
    story645 commented #21126
  • Sep 23 16:45
    ericpre commented #20977
  • Sep 23 15:22
    jklymak commented #18011
  • Sep 23 15:17
    ttarabula commented #18011
  • Sep 23 15:16
    ttarabula commented #18011
  • Sep 23 14:40
    ttarabula commented #18011
  • Sep 23 14:39
    ttarabula commented #18011
  • Sep 23 12:39
    jklymak unlabeled #20914
  • Sep 23 12:38
    jklymak commented #21050
Thomas A Caswell
@tacaswell
(running 5 min late for meeting, sorry)
Jody Klymak
@jklymak
MeanGithub.png
Github is so judgemental
Thomas A Caswell
@tacaswell
indeed it is! I also get a "you seem to have too many notifications. Have you tried making better life choices and not over-committing?" popup
Jody Klymak
@jklymak
Ha ha, I’ve not quite seen that one 😀
Thomas A Caswell
@tacaswell
Screenshot_20210911_160955.png
anutosh491
@anutosh491
Hello all , this is Anutosh here . I'm a beginner/amateur in open source . I have contributed to easy issues on math based organizations/projects like sympy,sage,networkx before and I plan to contribute to matplotlib too . Hence this is a request to be added to #incubator !
hannah
@story645
Does anyone have any opinions on NFTs being made using Matplotlib (specifically with respect to acknowledging 'em on social media)
Jody Klymak
@jklymak
No doubt nfts are a bit crazy but if you see a nice one made w matplotlib I don’t see why we couldn’t share it.
hannah
@story645
There are a couple of folks who make 'em (one of whom has been featured on the Instagram & discourse -makes a lot of cool animated stuff)
Elliott Sales de Andrade
@QuLogic
I mean, NFTs are kind of dumb, but cool art is cool art
hannah
@story645
That's sort of my view, & I do tend to RT w/ a "this is not an endorsement to buy this thing" - and I guess also generally like there isn't a policy on paid media/self promotion (that may be an open issue) cause folks also advertise paid courses & like I am kind of openly shilling Nicholas Rougiers new book...
Till Stensitzki
@Tillsten
Are there still kerning issues in matplotlib 3.4.2?
grafik.png
the VA distance looks rather large
Aitik Gupta
@aitikgupta
@Tillsten which backend are you using to generate this..?
Noah
@nasgold
Hello hello... could I please be added to the "#incubator" channel? I'm looking to make some contributions to matplotlib as I want to give back a bit after grad school, but I have a few questions before submitting my first PR. Thanks in advance!
Thomas A Caswell
@tacaswell
Now that NASA has published the list and confirmed the information was never embargoed, happy to report that we got the NASA roses grant!
Ian Hunt-Isaak
@ianhi
Is the callback registry stable to use in 3rd party packages? I know a lot of cbook is/has been moving to _api
I want to make a widget-esque object that doesn't actually inherit from axeswidget because that clutters up autocomplete with things that I don't really think are necessary for this object
woops actually you can't inherit it from axeswidget anyway, every widget class just makes their own.
and to be consistent (with disconnecting cid etc) I'd like to also use it
Elliott Sales de Andrade
@QuLogic
image.png
thoughts instead of the bullet points?
some of the spacing might be a bit off
Antony Lee
@anntzer
@mwaskom perhaps(?) you can "fake" +/-0.5 margins by actually setting the margins to <very large value> and adding a do-nothing artist in the children list that has a sticky_edges property that is set to whatever margins you actually want (being a property it can dynamically recompute itself); it a bit ugly/API abuse but heh :)
4 replies
Jody Klymak
@jklymak
@QuLogic I like w/o the bullet points.
Some of those could be shorter as well while you are at it
- :ref:`Intermediate <tutorials-intermediate>` and 
      :ref:`Advanced <tutorials-advanced>` tutorials
Jody Klymak
@jklymak
Meeting today, 15:00 CDT (20:00 UTC). Please add to the agenda at https://hackmd.io/YA_MIqDlTRKTQuZENX8djA?both
(I may not be able to make it)
Antony Lee
@anntzer
I won't make it today, but added an item (consider whether #20990 requires post-merge changes, as it was merged with a single review).
Thomas A Caswell
@tacaswell
to mac people who use conda: are you building Matplotlib with the system clang / xcode or installing the compilers from conda-(forge) as well?
Ryan May
@dopplershift
I've been building using the system compilers.
Thomas A Caswell
@tacaswell
sigh, I am using the macstadium machine for some mac-debugging, was having the tests hard-fail, assumed it was a compiler incompatibility, but it was actually an unhandled attribute error in a Python side signal handler that was causing PyQt to yeet the application out of existence.
Ryan May
@dopplershift
wow, that's some kind of failure mode
Thomas A Caswell
@tacaswell
to a large degree I actually agree with PyQt's choice here. If there is a Python exception raised someplace it has no way to get back to you, it exits (by intentionally de-referencing NULL) to let you know :shrug:
also found a maybe-bug in the CPpthon docs https://docs.python.org/3/library/signal.html#signal.SIGCLD says it is an alias for SIGCHLD which says it is available on unix, however SIGCLD does not appear to be available on OSX (or at least conda's builds of Python on OSX)
Thomas A Caswell
@tacaswell
and I can not sort out where in the CPython source the alias gets defined...
Thomas A Caswell
@tacaswell
ok, well that ended in reading the private includes of signal.h...
Thomas A Caswell
@tacaswell
https://bugs.python.org/issue45251 is the net result of that rabbit hole...
martinRenou
@martinRenou
ipympl 0.8.0 is out! Plots are now saved as static images in Notebooks! https://twitter.com/martinRenou/status/1440617060640247815?s=20
1 reply
Thomas A Caswell
@tacaswell
I am going to have to miss the call tomorrow
Antony Lee
@anntzer
same here
Tim Hoffmann
@timhoffm
I'm also away tomorrow.
hannah
@story645
Does the afm in afmhot stand for atomic force microscopy?
1 reply
Jody Klymak
@jklymak
Re the calls. I can sometimes make them but my attendance is going to be spotty. I have need form numerous meetings in the evenings because of 9 h time zone difference with home, so can’t really commit to keep that time slot clear.
I don’t think there is any particular fix possible for this and still keeping Hawaii a viable time zone. If we did want to go earlier 9-11 edt will be possible some days.
Jody Klymak
@jklymak
Note I'm not 100% sure what the plan is for the font handling changes - I've been assuming there is a GSOC team working to get those through