rkletr
@merkletr.ee
cryptography liker. ctf player. array languager? music enjoyer.
SPDX-License-Identifier: 0BSD
SPDX-License-Identifier: 0BSD
lesson learned: haskell in docker absolutely destroys iteration speed
![root@redacted:~# docker compose up -d --build
[+] Building 1037.0s (14/14) FINISHED](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ezhjhbzqt32bqprrn6qjlkri/bafkreicyygbwees4x4bi4xp3ooon5p6iqosleswzp7l5wv6vzx32nh3gby@jpeg)
November 12, 2025 at 8:36 PM
lesson learned: haskell in docker absolutely destroys iteration speed
the effect applied on the same image with a few different sets of parameters
November 8, 2025 at 10:37 PM
the effect applied on the same image with a few different sets of parameters
this doesnt feel great either but i guess it works
November 2, 2025 at 11:53 PM
this doesnt feel great either but i guess it works
successfully consuming jetstream from haskell
now comes the fun part
now comes the fun part
November 2, 2025 at 4:59 PM
successfully consuming jetstream from haskell
now comes the fun part
now comes the fun part
remade my "ctf team" "logo" in handwritten svg
October 28, 2025 at 2:29 AM
remade my "ctf team" "logo" in handwritten svg
October 22, 2025 at 10:10 PM
just a bunch of fun little noises
highly recommend
highly recommend
October 4, 2025 at 1:47 AM
just a bunch of fun little noises
highly recommend
highly recommend
sqlite is so insane sometimes i can't believe it's the best database
![Extract from https://www.sqlite.org/flextypegood.html
4.2. Rigid type enforcement helps prevent application bugs
It has become a point of doctrine among many programmers that the best way to prevent application bugs is strict type enforcement. But I find no evidence in support of this.
To be sure, strict type enforcement does help prevent some kinds of bugs in lower-level languages like C and C++ that present a model that is close to machine hardware. But this does not seem to be the case for higher-abstraction languages in which all data is passed around in a "Value" superclass of some kind which is subclassed for the various lower-level data types. When everything is a Value object, specific datatypes cease to be important.
This technical note is authored by the original author of SQLite. I have been writing TCL programs for 27 years. TCL has no type enforcement whatsoever. The "Value" class in TCL (called Tcl_Obj) can hold many different datatypes, but it presents the content to the program and to the application user as a string. And I've had a lot of bugs in those TCL programs over the years. But I do not recall a single instance where the bugs might have been caught by a rigid type system. I have also written a lot of C code over a span of 35 years, not the least of which is SQLite itself. I have found the type system in C to be very helpful at finding and preventing problems. For the Fossil Version Control System, which is written in C, I have even implemented supplemental static analysis programs that scan the Fossil source code prior to compilation, looking for problems that compilers miss. This works well for compiled programs.
[...alt text character limit too short...]](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ezhjhbzqt32bqprrn6qjlkri/bafkreiear2d5kzxrur7otvkrvsranfv6l5cgmxyxayjey32v27ddcan7nu@jpeg)
September 22, 2025 at 2:47 PM
sqlite is so insane sometimes i can't believe it's the best database
FAKE ID is like, peak, and i'm literally the only one who has it on bandcamp. that's fucking crazy to me. i can and have listened to this for hours
September 9, 2025 at 1:30 AM
FAKE ID is like, peak, and i'm literally the only one who has it on bandcamp. that's fucking crazy to me. i can and have listened to this for hours
i came up with interface{*A;B} and thought it was disgusting and there had to be a better way so i googled it... and nope that's like the correct way to do this???? that's insane
how does go make a function which is effectively an instantiation of `map` have more complicated typing than haskell 😭
how does go make a function which is effectively an instantiation of `map` have more complicated typing than haskell 😭
![except from https://go.dev/blog/generic-interfaces
// OrderedSet[E] does not satisfy Set[E] (method All has pointer receiver)
Unique[E, OrderedSet[E]](slices.Values(s))
// panic: invalid memory address or nil pointer dereference
Unique[E, *OrderedSet[E]](slices.Values(s))
The first problem is clear from the error message: Our type constraint says that the type argument for S needs to implement the Set[E] interface. And as the methods on OrderedSet use a pointer receiver, the type argument also has to be the pointer type.
When trying to do that, we run into the second problem. This stems from the fact that we declare a variable in the implementation:
var seen S
If S is *OrderedSet[E], the variable is initialized with nil, as before. Calling seen.Insert panics.
If we only have the pointer type, we cannot get a valid variable of the value type. And if we only have the value type, we cannot call pointer-methods on it. The consequence is that we need both the value and the pointer type. So we have to introduce an additional type parameter PS with a new constraint PtrToSet:
// PtrToSet is implemented by a pointer type implementing the Set[E] interface.
type PtrToSet[S, E any] interface {
*S
Set[E]
}
// Unique removes duplicate elements from the input sequence, yielding only
// the first instance of any element.
func Unique[E, S any, PS PtrToSet[S, E]](input iter.Seq[E]) iter.Seq[E] {
return func(yield func(E) bool) {
// We convert to PS, as only that is constrained to have the methods.
// The conversion is allowed, because the type set of PS only contains *S.
seen := PS(new(S))
for v := range input {
if seen.Has(v) {
continue
}
if !yield(v) {
return
}
seen.Insert(v)
}
}
}](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ezhjhbzqt32bqprrn6qjlkri/bafkreiatujkvncpdnwtmtdgcrlfjfhka3heuvxhyilglobwloixvvzmee4@jpeg)
![type pView[A, B any] interface {
*A
repo.View[B]
}
func ViewAll[A, B any, PA pView[A, B]](db repo.DB, as []A) []B {
var bs []B
for _, a := range as {
bs = append(bs, PA(&a).View(db))
}
return bs
}](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ezhjhbzqt32bqprrn6qjlkri/bafkreidcxyujz3edq4fbsnrneqxgeuokyoy4xmtp6nosmiptjh6smh5j4u@jpeg)
September 8, 2025 at 8:05 PM
i came up with interface{*A;B} and thought it was disgusting and there had to be a better way so i googled it... and nope that's like the correct way to do this???? that's insane
how does go make a function which is effectively an instantiation of `map` have more complicated typing than haskell 😭
how does go make a function which is effectively an instantiation of `map` have more complicated typing than haskell 😭
thanks go docs for recognizing this one but you didn't really need to tell me that statement kind of describes go as a whole
September 8, 2025 at 4:46 PM
thanks go docs for recognizing this one but you didn't really need to tell me that statement kind of describes go as a whole
damnit how did this one happen
September 5, 2025 at 3:26 PM
damnit how did this one happen
oh wait lol python doesnt allow indexed assign expressions so the above wouldn't work even in the insane universe
daul solves this
daul solves this
![screenshot of a terminal showing that this daul source code:
setmetatable([], ["__index":\s,k->s[k]=[]])
compiles to the following lua source code:
return setmetatable({},{["__index"]=(function(s,k)local _0={};(s)[k]=_0;return _0;end)});](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:ezhjhbzqt32bqprrn6qjlkri/bafkreigrx4p4gjgm3ipufn234henooilbre6m33prmegiqipmkfjrkgxcq@jpeg)
September 4, 2025 at 2:49 AM
oh wait lol python doesnt allow indexed assign expressions so the above wouldn't work even in the insane universe
daul solves this
daul solves this
huh this is actually a pretty interesting concept
September 1, 2025 at 4:50 PM
huh this is actually a pretty interesting concept
it's always seeing these that make me really think im back in quebec
August 25, 2025 at 1:49 AM
it's always seeing these that make me really think im back in quebec
sometimes i wonder why i bother pretending 300 characters of rambling above an album will make people more likely to listen to the music i post
this goes especially so for this album, for which i think going in blind was definitely part of what made me like it so much from the first time i heard it
this goes especially so for this album, for which i think going in blind was definitely part of what made me like it so much from the first time i heard it
August 18, 2025 at 5:13 AM
sometimes i wonder why i bother pretending 300 characters of rambling above an album will make people more likely to listen to the music i post
this goes especially so for this album, for which i think going in blind was definitely part of what made me like it so much from the first time i heard it
this goes especially so for this album, for which i think going in blind was definitely part of what made me like it so much from the first time i heard it
new kaytranada album is pretty damn good. seems to have much fewer features than the previous albums, and i appreciate the greater focus. not quite sure where it ranks within the discography yet but i like it.
August 14, 2025 at 9:19 PM
new kaytranada album is pretty damn good. seems to have much fewer features than the previous albums, and i appreciate the greater focus. not quite sure where it ranks within the discography yet but i like it.
okay so as predicted, nitepunk is fixed, and also as predicted, i have a bunch of new duplicate artists to deal with
gonna take forever to get through these but eh this seems like as good a reason as any to go contribute to a public good database
gonna take forever to get through these but eh this seems like as good a reason as any to go contribute to a public good database
August 8, 2025 at 7:14 PM
okay so as predicted, nitepunk is fixed, and also as predicted, i have a bunch of new duplicate artists to deal with
gonna take forever to get through these but eh this seems like as good a reason as any to go contribute to a public good database
gonna take forever to get through these but eh this seems like as good a reason as any to go contribute to a public good database
it'll probably eliminate this case, i think nitepunk is popular enough to have everything on MB? but i'm definitely gonna be splitting apart more artists than i join together by doing this
guess i'll have to actually become an MB contributor to add the rest of my stuff huh
guess i'll have to actually become an MB contributor to add the rest of my stuff huh
August 7, 2025 at 3:05 PM
it'll probably eliminate this case, i think nitepunk is popular enough to have everything on MB? but i'm definitely gonna be splitting apart more artists than i join together by doing this
guess i'll have to actually become an MB contributor to add the rest of my stuff huh
guess i'll have to actually become an MB contributor to add the rest of my stuff huh
the only thing keeping me sane right now is how good this album is
August 4, 2025 at 2:51 AM
the only thing keeping me sane right now is how good this album is