Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 07:38
    pepe commented #613
  • 04:51
    johnwcowan opened #613
  • 04:36
    andrewchambers commented #475
  • Jan 24 23:37
    bakpakin closed #373
  • Jan 24 23:37
    bakpakin closed #359
  • Jan 24 23:25

    bakpakin on master

    Use a slightly clearer example … Merge pull request #611 from su… (compare)

  • Jan 24 23:25
    bakpakin closed #611
  • Jan 24 22:49

    bakpakin on master

    Update for sending streams to n… (compare)

  • Jan 24 20:45
    subsetpark opened #611
  • Jan 24 03:44
    andrewchambers commented #478
  • Jan 24 02:03
    sogaiu commented #478
  • Jan 24 00:28
    bakpakin commented #478
  • Jan 24 00:27
    bakpakin commented #478
  • Jan 24 00:26
    bakpakin closed #478
  • Jan 24 00:26
    bakpakin commented #478
  • Jan 23 23:50

    bakpakin on master

    Begin work on allowing small bi… Address #478 - Disable core in … (compare)

  • Jan 23 20:50

    bakpakin on v1.14.2

    (compare)

  • Jan 23 20:50

    bakpakin on master

    Prepare for patch release. (compare)

  • Jan 23 20:31
    pepe commented #610
  • Jan 23 20:28
    bakpakin commented #610
Michael Camilleri
@pyrmont
Oh, wait. I misread. Let me check.
Michael Camilleri
@pyrmont
Same error.
I have an alternative strategy I think I'll employ but am curious if there is a way if anyone knows.
yumaikas-
@yumaikas:matrix.org
[m]
Basically, if you're requiring "./build/my.so", even at runtime, I'd expect it to look for it in the build folder
relative to the current folder
Michael Camilleri
@pyrmont
Well, it's finding it.
At least that's what the stacktrace looks like to me. The problem is that the necessary machinery to actually process it isn't there.
But is there when using the REPL. So obviously there's something you can do when producing an executable artefact that makes this possible. Maybe it's like a debug flag or something.
yumaikas-
@yumaikas:matrix.org
[m]
You're not using --no-core or some such, are you?
Michael Camilleri
@pyrmont
Just declare-executable in my project.janet.
yumaikas-
@yumaikas:matrix.org
[m]
Hrm....
yumaikas:matrix.org @yumaikas:matrix.org is kinda tapped out then.
Michael Camilleri
@pyrmont
No worries! Thanks for the help regardless!
pp
@damnpepe:matrix.org
[m]
Executable cannot load native module on runtime
yumaikas-
@yumaikas:matrix.org
[m]
Why can the REPL do so then?
(It is a janet executable, after all)
pp
@damnpepe:matrix.org
[m]
No it is not.
This funcitonality may help I guess pyrmont Adds a :preload loader which allows one to manually put things into module/cache.
yumaikas-
@yumaikas:matrix.org
[m]
The REPL isn't a janet executable?
I thought it was?
pp
@damnpepe:matrix.org
[m]
not the same as when you declare-executable I meant
yumaikas-
@yumaikas:matrix.org
[m]
I mean, not the same, but, theoretically speaking, there's not a reason that a janet executable shouldn't be able to use the same underlying mechanisms to load .so or .dll files. I could see JPM's default build discouraging the practice, however
(Especially since it tends to make for fragile executables)
pp
@damnpepe:matrix.org
[m]
Indeed. REPL is more similar to when you run code with janet init.janet than to running declare-executable generated program
I have been bitten by it not that long time ago, but understanding it, made me better Janet again :-)
skuza
@skuza:matrix.skuz.xyz
[m]
can you create an AST from Janet PEGs?
yumaikas-
@yumaikas:matrix.org
[m]
If you were to use cmt and some functions, I think you'd be able to pull it off
skuza
@skuza:matrix.skuz.xyz
[m]
I'm trying to parse a document format quite similar to markdown
that I've created
yumaikas-
@yumaikas:matrix.org
[m]
The main thing to keep in mind is that PEGs return an array of the captures in them, so you have to balance things so that makes sense
I think spork/fmt is a good reference point for that style of PEG
skuza
@skuza:matrix.skuz.xyz
[m]
thanks I'll check that out
pp
@damnpepe:matrix.org
[m]
skuza: i just tuned one easy format parsing in jesty https://git.sr.ht/~pepe/jesty
Oh pushed just now 😛
John Gabriele
@uvtc
Was a Janet doc added somewhere recently specifically regarding docstrings?
Maybe to janet-lang.org docs? I thought I saw something, but now can't find it.
Michael Camilleri
@pyrmont
It's a PR.
John Gabriele
@uvtc
Ah! That was it! Thanks, @pyrmont . I just created janet-lang/janet-lang.org#101 . Didn't want that change to slip through the cracks, but unable to suggest a change myself right now.
yumaikas-
@yumaikas:matrix.org
[m]
Are there any ways to inspect a fibers signal mask?
yumaikas-
@yumaikas:matrix.org
[m]
Does thread/recieve still block?
1 reply
yumaikas-
@yumaikas:matrix.org
[m]
looooooool
Main thing I was thinking is it'd be nice to have thread supervisors, kinda like we have fiber supervisors
Dunno exactly how that'd work
But it'd be cool
pp
@damnpepe:matrix.org
[m]
I hope we got there, supervisors are such great concept
yumaikas-
@yumaikas:matrix.org
[m]
It'd be a little harder with threads. But, they do have an event loop with a fiber, so ultimately it's about making sure that fiber's supervisor uses the inter-thread comms to go back and forth
pp
@damnpepe:matrix.org
[m]
I guess it is a lot harder (but always mind my C-fu rank (very low)) and yet I guess bp has some trumps in his sleeve