Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • Jan 17 15:37
    ishehadeh commented #912
  • Jan 16 21:41
    andrewchambers commented #912
  • Jan 16 21:39
    andrewchambers commented #912
  • Jan 16 20:09
    ishehadeh commented #912
  • Jan 16 03:28
    andrewchambers commented #912
  • Jan 16 03:27
    andrewchambers commented #912
  • Jan 16 03:26
    andrewchambers commented #912
  • Jan 16 03:25
    andrewchambers commented #912
  • Jan 16 03:24
    andrewchambers commented #912
  • Jan 16 03:24
    andrewchambers commented #912
  • Jan 15 23:38
    eko234 commented #912
  • Jan 15 22:32
    bakpakin commented #912
  • Jan 15 20:12
    ishehadeh opened #912
  • Jan 14 23:14
    bakpakin closed #889
  • Jan 14 21:15
    sogaiu commented #911
  • Jan 14 18:14
    pyrmont edited #911
  • Jan 14 18:13
    pyrmont opened #911
  • Jan 14 18:06
    pyrmont commented #889
  • Jan 14 15:16
    elimisteve commented #898
  • Jan 14 14:16
    sogaiu commented #898
willfennel
@willfennel:matrix.org
[m]
I didn't know about match yet, btw. I like it, but it does make the macro its signature less descriptive in this case. 🤔
willfennel
@willfennel:matrix.org
[m]
Ah, I can change the signature and the match expression, of course:
(defmacro ifte
  [if-cond then-kw then-exp else-kw else-exp]
  (match
    @[if-cond then-kw then-exp else-kw else-exp]
    @[if-cond :then   then-exp :else   else-exp]
    ~(if ,if-cond ,then-exp ,else-exp)
    (error "Expected (ifte <if> :then <then> :else <else>"))
willfennel
@willfennel:matrix.org
[m]
Whoa, I didn't know that would result in compile time errors if the wrong keywords are supplied. That's awesome!
Will stop spamming now. Thanks again, Zach Smith (subsetpark)
andrewchambers
@andrewchambers
@willfennel:matrix.org macros run at compile time
so an error thrown in a macro must be a compile time error
willfennel
@willfennel:matrix.org
[m]
Ah yes, makes sense 👍️
pp
@damnpepe:matrix.org
[m]

bakpakin: the latest Janet is not showing error message only the stacktrace I guess after janet-lang/janet@e8c7380 Repro:

> janet -e '(empty? 1)'
  in empty? [boot.janet] on line 121, column 50
  in _thunk [eval-string] (tailcall) on line 1, column 1
  in eval1 [boot.janet] on line 2333, column 16
  in run-context [boot.janet] on line 2393, column 13
  in eval-string [boot.janet] on line 2432, column 3
  in e-switch [boot.janet] (tailcall) on line 3618, column 12
  in cli-main [boot.janet] on line 3644, column 13

Is it by design? Or is it error on my side?

bakpakin
@bakpakin:matrix.org
[m]
Hmm that is a bug. There were some changes there to handle a missing status line in some cases but I missed a really obvious case.
I will release a 1.19.2 patch today with a fix
pp
@damnpepe:matrix.org
[m]
thank you very much
It is fixed here, thank you
pp
@damnpepe:matrix.org
[m]
speaking of macros, I am playing with these two:
(defmacro vars [& bindings]
  ~(upscope
     ,;(seq [[n v] :in (partition 2 bindings)] (tuple 'var n v))))

(defmacro defs [& bindings]
  ~(upscope
     ,;(seq [[n v] :in (partition 2 bindings)] (tuple 'def n v))))
tionis
@tionis:matrix.org
[m]

How would one increase the value of a field in a 3-dimensional array by one (like the ++ macro does) the janet way?

To give an example I want to increase the 1 to a 2:

@(
  @(
    @(0 0 0) 
    @(0 1 0) 
    @(0 0 0)) 
  @(
    @(0 0 0) 
    @(0 0 0) 
    @(0 0 0))
)
1 reply
llmII
@llmii:matrix.org
[m]
(++ (get 1 (get 1 (get 0 3darray)))) possibly
I may have fudged the indexing a bit with the gets but that's sorta it
tionis
@tionis:matrix.org
[m]
this gives me the same error as the get-in macro
```
compile error: expected 2 element tuple for l-value to set
llmII
@llmii:matrix.org
[m]
what's the result of (get 1 (get 1 (get 0 3darray)))?
tionis
@tionis:matrix.org
[m]
nil
llmII
@llmii:matrix.org
[m]
that's why
tionis
@tionis:matrix.org
[m]
(get-in grid [0 0 0]) gives me 0
1 reply
llmII
@llmii:matrix.org
[m]
need to use a get that gets the thing you need gotten
oh hmm, what does (get-in wait, you're using@(instead of@[`?
tionis
@tionis:matrix.org
[m]
yes, they should be the same if I read the docs correctly
llmII
@llmii:matrix.org
[m]
(def arr [0 1 2 3 4])
(put arr 0 (+ 1 (get arr 0)))
# @[1 1 2 3 4]
so...
try put-in
tionis
@tionis:matrix.org
[m]
Oh I overlooked that
Maybe i should look into making my own ++in macro
llmII
@llmii:matrix.org
[m]
I guess it's put and alike for associative datastructures and set for vars
hmmm, no, set works on keyed data structures per the specials doc
but yes, I sorta would have assumed (++ (variant-amount-of-gets-here)) would have worked
tionis
@tionis:matrix.org
[m]
Ok, this does the trick for me:
(defmacro ++in [variable arr] ~(,put-in ,variable ,arr (+ (,get-in ,variable ,arr) 1)))
Zach Smith (subsetpark)
@subsetpark:matrix.org
[m]
I think update-in is probably the best way to do what you want using standard pieces
1 reply
Relevant to the recurring conversation about more powerful version resolution in jpm, this is apparently all the rage: https://nex3.medium.com/pubgrub-2fb6470504f
crocket
@crocket
On gentoo linux, I manage jpm packages as gentoo packages. On other linux distributions, jpm needs to take care of jpm package updates.
jpm doesn't seem to have a command for updating installed jpm packages.
bakpakin
@bakpakin:matrix.org
[m]
Reinstalling a package should update. No easy way to update all at once though besides iterating through the manifest
pp
@damnpepe:matrix.org
[m]
bakpakin: what is the easiest way to use u64 with string/from-bytes please?
or even better I have 64 bit number and I need to create string with bytes from it :-)
pp
@damnpepe:matrix.org
[m]
This is the best I have managed to do it :-)
(string/from-bytes 2r01111111
                           ;(seq [i :down-to [56 0 8]]
                              (scan-number (string (band (brshift (int/u64 dl) i) 0x000000FF)))))
the 64bits number is the seq
@saikyun: PS: finaly fixing trevor, as I got over the 16 bits size tooo :-)
pp
@damnpepe:matrix.org
[m]
maybe string and scan-number is good enough, at least for me now
saikyun
@saikyun

@saikyun: PS: finaly fixing trevor, as I got over the 16 bits size tooo :-)

haha, nice! :D

I took a break from working on my blog, hopefully I'll get to it again sometime soon :)
2 replies
I got a new job and will work 50% on creating a game using Freja :D starting in janetuary
yumaikas-
@yumaikas:matrix.org
[m]
Congratulations 🎉
1 reply