Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 18:54
    datwaft opened #245
  • 17:48
    m00qek commented #53
  • Sep 14 18:50
    mixedCase opened #244
  • Sep 12 11:59

    Olical on develop

    Refresh the REPL if we're evalu… Refresh the REPL if we eval the… (compare)

  • Sep 12 11:47

    Olical on develop

    Sync sponsors, +@Fedreg <3 Recompile with latest Fennel, a… More tests + fix fennel evals, … (compare)

  • Sep 08 18:11
    EdwardIII closed #243
  • Sep 08 18:11
    EdwardIII commented #243
  • Sep 08 17:57
    EdwardIII commented #243
  • Sep 08 17:47
    EdwardIII edited #243
  • Sep 08 17:45
    EdwardIII opened #243
  • Sep 08 13:06
    grav updated the wiki
  • Sep 08 12:20
    Olical updated the wiki
  • Sep 08 09:54
    tomekw commented #5
  • Sep 08 09:51
    Olical commented #5
  • Sep 08 09:50
    Olical commented #5
  • Sep 08 09:41
    tomekw commented #5
  • Sep 07 21:52

    Olical on develop

    Bash the Aniseed REPL a lot, ju… Fix flaky test, remove newlines… (compare)

  • Sep 06 21:37

    Olical on develop

    Recompiling with latest Aniseed… (compare)

  • Sep 05 13:23

    Olical on develop

    Try to guess the module name fr… (compare)

  • Sep 05 12:14

    Olical on develop

    Slight aniseed completions fixe… Remove some dev comments from t… _LOCALS, _TESTS -> aniseed/{loc… (compare)

Oliver Caldwell
@Olical
And thank you so much for the kind words, I'm so happy it's working well for you and you're enjoying it :heart:
Michael Camilleri
@pyrmont
@Olical: I might be 'using it wrong' but I've been using Conjure as I work through the Advent of Code challenges. I tend to work in a single file where I will often evaluate the entire buffer.
Quite a few times, a large amount of output cause error messages when evaluating function definitions not to be visible. Since these functions are often defined, subsequent function calls simply use the old definition that's in memory. Is there a way to avoid this?
Oh, and I'm using Janet via netrepl.
Oliver Caldwell
@Olical
Hmm so you mean the function has some sort of error inside it, so the eval failed, but then large output pushed that error off screen, so you don't realise?
@pyrmont (I should've mentioned!)
Michael Camilleri
@pyrmont
Yep!
And then there's no later failure because the binding from the earlier evaluation still exists.
Oliver Caldwell
@Olical
Not sure how I'd get around this... might require some more Janet knowledge. Maybe I could wrap entire evals in some sort of "do" block, but that might have unintended consequences.
Hmm if you use <prefix>ef rather than eb it should have the desired behavior? If not, then it's just sort of how Janet behaves, a failure doesn't mean execution will stop, I wonder if there's a config thing we can set to get it to bail early, kinda like bash's set -xe
@pyrmont Hmm I just tried it, looks like Janet as a language just skips over errors, so unless there's a way to configure that there's not much I can do
sogaiu
@sogaiu

not sure if this is related, but saw this on the janet/community channel from bakpakin

also the behavior of continuing after a compilation error is old - you need to add :exit true to make import fail. This makes the repl experience better but is probably annoying for scripts

sogaiu
@sogaiu
Michael Camilleri
@pyrmont
@Olical I must be missing something obvious but how does the log reset described in Olical/conjure#144 work?
Oliver Caldwell
@Olical
It has two approaches, the lower case reset will just set the buffer content lines to an empty list (the same as ggdG really) and the upper case Reset will wipeout the buffer entirely and rebuild it next time it's require from scratch.
@pyrmont
I always forget to mention
@sogaiu That might be helpful at some point! So thanks! :D
Michael Camilleri
@pyrmont
@Olical: And the keyboard shortcut should be <localleader>lR?
Oliver Caldwell
@Olical
@pyrmont yep, which is the hard reset.
Michael Camilleri
@pyrmont
@Olical I'm not sure what I'm doing wrong but when I press <localleader>lR, my Vim moves into Replace Mode. Do you have any suggestions?
Oliver Caldwell
@Olical
Hmm are you definitely up to date? Because it sounds like you don't have the mapping to me
@pyrmont
Michael Camilleri
@pyrmont
@Olical I feel like I'm missing something obvious but when I update my plugins (using vim-plug), I'm told that the plug-in update is 'pending'. Do you have any ideas?
Michael Camilleri
@pyrmont
I've gone to my plugin directory, and the Conjure plugin is there and when I run git status, I see it's using the head detached at tag 4.10.0.
Michael Camilleri
@pyrmont
I don't know if it helps, but here's my init.vim file: https://gist.github.com/pyrmont/0c14be6eabe22eca55b8d1fc043afd49
Hmmm. Looks like soft resetting is working.
Oh wait. It is working. I think I was just too slow in pressing the key combination. Sorry for all
…the trouble :(
I do have a new problem. Since upgrading to 4.10.0 (I was previously on 3.4.0), I'm noticing that there appears to be escape sequences for colour. That wasn't there previously. Any ideas on this issue?
Oliver Caldwell
@Olical
@pyrmont Glad that part is working! And not a problem
You're seeing ansi escape sequences in Janet?
Janet results specifically?
The stripping of ansi codes is controlled by the following option g:conjure#log#strip_ansi_escape_sequences_line_limit, just in case that's involved here
If there's a bug there we can get an issue together and fixed ASAP
Oliver Caldwell
@Olical
Hmm Janet is looking okay to me, no ansi here.
sogaiu
@sogaiu
fwiw, similar story here (no ansi afaict) -- i'm on conjure 4.10.0 and aniseed 3.12.0
this is a linux box
Michael Camilleri
@pyrmont
@Olical: I'm only seeing the behaviour with certain output (I think that's over a certain length). Here's a gist with files that produce the behaviour: https://gist.github.com/pyrmont/a2607776bd5df012214e760b4461f40b
If I comment out the last line of test.janet, the behaviour doesn't occur. It's only when I split the string on newlines that I see it.
I have an example string (https://github.com/pyrmont/advent/blob/994041522365a6d13c429eda342eb9853aa92339/2020/day03.janet#L17-L29) which doesn't produce this behaviour so it's not simply an array of strings that causes this to happen.
Oliver Caldwell
@Olical
@pyrmont Oh! Yeah! So that's the g:conjure#log#strip_ansi_escape_sequences_line_limit option! I set a limit on this feature because I don't want to end up running regular expressions over 10000s of lines over and over grinding your nvim to a halt. The default is 100 lines, so maybe bump that to 1000 and I should raise the default?
So, legitimately, that's not a bug. It's a feature :sweat_smile:
Michael Camilleri
@pyrmont
Got it!
Updated my init.vim and it's working fine :)
Thanks!
Michael Camilleri
@pyrmont
Congrats on the release, @Olical :)
I'm really excited to try out virtual text!
Oliver Caldwell
@Olical
:D glad to hear it!
Michael Camilleri
@pyrmont
@Olical Gave it a bit of a crack trying to get something to work with the new stuff in Janet but no such luck. Am not in a big hurry so will leave it for now :)