https://pantsbuild.org/ logo
Powered by
#development
  • a

    aloof-angle-91616

    09/26/2019, 11:03 PM
    i obviously think this is all very possible, just wondering what that would be
  • a

    aloof-angle-91616

    09/26/2019, 11:03 PM
    oh
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:03 PM
    so we'd pass them into python as a field on the 
    HttpResponse
    type
  • a

    aloof-angle-91616

    09/26/2019, 11:03 PM
    yeah
  • a

    aloof-angle-91616

    09/26/2019, 11:03 PM
    of course it's serializable that's the point of http
  • a

    aloof-angle-91616

    09/26/2019, 11:03 PM
    amazing
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:03 PM
    and then other rules could do arbitrary things wiht the bytes by consuming that type
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:04 PM
    but that would happen in python because at that point the http request is done and nothing's blocking anymore
  • a

    aloof-angle-91616

    09/26/2019, 11:04 PM
    thank you for elevating my consciousness and talking it through with me, i would possibly have wasted time on that if i'd done this myself
  • a

    aloof-angle-91616

    09/26/2019, 11:04 PM
    yes
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:04 PM
    heh, np
  • a

    aloof-angle-91616

    09/26/2019, 11:04 PM
    and we could possibly specifically
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:04 PM
    I'll probalby get started on this soon-ish becuase it's something that we at toolchain are gonna need for our own use
  • a

    aloof-angle-91616

    09/26/2019, 11:04 PM
    make a rule 
    HttpResponse -> ParsedHttpResponse
    or something in python as well
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    as well as like some constructor for an 
    HttpRequest
    ? but like
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    these aren't blockers
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:05 PM
    I just wanted to make sure I was thinking about how to implement this the right way, and that more-or-less following what 
    UrlToFetch
    does is the right approach
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    i'm just excited about all of this and wondering how it'll work
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    i'm excited to see more of your great code
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    yes
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:05 PM
    ok cool
  • a

    aloof-angle-91616

    09/26/2019, 11:05 PM
    thanks again for being patient with me and talking it through in this thread!
  • h

    hundreds-breakfast-49010

    09/26/2019, 11:05 PM
    no worries 🙂
  • a

    aloof-angle-91616

    09/26/2019, 11:17 PM
    i love love love how easy it is to partition build steps into independent reusable rules!!! i’m so glad we’ve built enough infrastructure to start solving real problems with v2 rulesets!!
  • w

    witty-crayon-22786

    09/27/2019, 12:25 AM
    so... would be good to discuss in a ticket probably
  • w

    witty-crayon-22786

    09/27/2019, 12:26 AM
    that is a big hole to bust in things.
  • w

    witty-crayon-22786

    09/27/2019, 12:27 AM
    my knee jerk reaction is: should that maybe be in a separate process...?
  • h

    hundreds-breakfast-49010

    09/27/2019, 12:52 AM
  • h

    hundreds-breakfast-49010

    09/27/2019, 12:52 AM
    does it need to be in a separate process if we make the request asynchronously? which IIRC reqwest lets us do?
  • w

    witty-crayon-22786

    09/27/2019, 12:54 AM
    it's not about synchronous vs asynchronous so much as about being reproducible/sideeffect free
1...653654655...906Latest