2025/05/28

Newest at the top

2025-05-28 23:24:31 +0200 <EvanR> *sql-simple packages are solid
2025-05-28 23:24:12 +0200Otong(~Otong@user/Otong) (Ping timeout: 265 seconds)
2025-05-28 23:24:01 +0200merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 252 seconds)
2025-05-28 23:23:45 +0200 <[exa]> bwe: there's always a better library. stay with the expectable one. lack of surprise is good.
2025-05-28 23:22:53 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-05-28 23:22:43 +0200 <bwe> ... over and over I get the feeling that I tend to pick the wrong libraries in Haskell because I don't know about the existing right ones! Sadly, there is no current awesome-haskell list :(.
2025-05-28 23:22:31 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Read error: Connection reset by peer)
2025-05-28 23:21:03 +0200 <[exa]> bwe: the "core" part should be db-agnostic, yes. API there will need an update. I'm interested in having the postgres part working (not now but for future project) so if you have the API updated somewhere I'll just join in and port the postgres part if possible
2025-05-28 23:20:31 +0200 <dibblego> waargonaut is better than aeson anyway
2025-05-28 23:20:01 +0200 <EvanR> it's just unconventional
2025-05-28 23:19:19 +0200merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) merijn
2025-05-28 23:19:07 +0200 <EvanR> haskell can skip the second stage and be just as brittle
2025-05-28 23:18:57 +0200 <[exa]> json schemas yay
2025-05-28 23:18:15 +0200 <EvanR> other languages have the second stage too and it's just as complex
2025-05-28 23:18:09 +0200 <bwe> [exa]: atm I'd like to work on getting streaming into selda for SQLite; however isn't it agnostic to whatever DB selda supports?
2025-05-28 23:18:08 +0200 <EvanR> aeson is doing double duty, reading in "dumb" json values but also giving a way to parse-don't-validate it to useful types
2025-05-28 23:16:40 +0200 <[exa]> bwe: anyway yeah well the streaming pgsql mostly solves it, great. :D
2025-05-28 23:15:37 +0200 <[exa]> bwe: yeah that's mostly why I'm asking, it's not really maintained (but works :D), taking over is a possibility.
2025-05-28 23:15:19 +0200 <EvanR> or CBOR
2025-05-28 23:15:07 +0200 <[exa]> EvanR: or YAML!!!11
2025-05-28 23:14:56 +0200 <[exa]> sm: rofl.
2025-05-28 23:14:56 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr)
2025-05-28 23:14:50 +0200 <EvanR> shoulda used XML
2025-05-28 23:14:40 +0200 <EvanR> who thought that JSON would be so complicated
2025-05-28 23:14:35 +0200peterbecich(~Thunderbi@syn-047-229-123-186.res.spectrum.com) peterbecich
2025-05-28 23:14:33 +0200sabathan2(~sabathan@amarseille-159-1-12-107.w86-203.abo.wanadoo.fr) (Remote host closed the connection)
2025-05-28 23:14:27 +0200 <bwe> [exa]: ... and then, what about the fate of selda? Is the creator=maintainer still alive?
2025-05-28 23:13:45 +0200 <EvanR> that's ridiculous
2025-05-28 23:13:38 +0200 <bwe> [exa]: are you aware of https://github.com/haskell-streaming/streaming-postgresql-simple ?
2025-05-28 23:13:25 +0200 <haskellbridge> <sm> also: "Another short-coming was a cross-platform terminal IO with unicode support – it doesn’t really work out of the box."
2025-05-28 23:13:14 +0200 <haskellbridge> <sm> this great interview: https://serokell.io/blog/haskell-in-production-simplex
2025-05-28 23:13:01 +0200 <bwe> yeah, that's my gut feeling.. that I need to make my own flavour of constructing the result list. Which essentially boils down to finding the right place of inserting S.yield and getting the type signatures right.
2025-05-28 23:12:45 +0200 <[exa]> sm: lol where's that from.
2025-05-28 23:12:33 +0200 <[exa]> EvanR: anyway true, I read "through" too literally, there you could do by bits. Selda is more like "from"
2025-05-28 23:12:17 +0200 <EvanR> it's kind of like how do I incrementally read JSON if all I have is js JSON.parse
2025-05-28 23:12:10 +0200 <haskellbridge> <sm> "The worst part of Haskell is its handling of JSON encoding – the Aeson library that everybody uses is quite complex to understand and use effectively. I believe that the unintuitive handling of JSON is one of the primary reasons that slows down Haskell adoption."
2025-05-28 23:11:42 +0200 <bwe> [exa]: yeah, it's selda
2025-05-28 23:11:24 +0200 <[exa]> ah yes it's selda
2025-05-28 23:11:19 +0200 <[exa]> bwe: btw is this for the selda streams? if you start the work somewhere please ping me, I'd try to find time to write some code (at least for postgres)
2025-05-28 23:11:15 +0200 <EvanR> [exa], like if it's this selda library which only produces whole list of results at once and no way to get 1 row at a time
2025-05-28 23:10:31 +0200 <EvanR> but database APIs if you go close enough to the raw API, support streaming
2025-05-28 23:10:25 +0200 <[exa]> well you can but you have to provide a bit of glue
2025-05-28 23:10:08 +0200 <EvanR> if you are going through a library that doesn't support streaming, you can't
2025-05-28 23:09:41 +0200 <bwe> EvanR: sorry if I was not able to convey my point - I want to use streaming but the point for me is how to take whatever IO input from some yet non-streaming library like DB stuff.
2025-05-28 23:08:19 +0200 <EvanR> then it should play well with the rest of the library
2025-05-28 23:08:08 +0200merijn(~merijn@host-vr.cgnat-g.v4.dfn.nl) (Ping timeout: 252 seconds)
2025-05-28 23:08:06 +0200 <EvanR> if you want to feed in some data from a source, produce a stream, not a lazy I/O list
2025-05-28 23:07:55 +0200 <EvanR> if you're using this, then I would suggest not using lazy I/O
2025-05-28 23:07:03 +0200 <[exa]> yes this one is good too. A bit less universal approach but covers the usual usecase space :D
2025-05-28 23:06:14 +0200Unicorn_Princess(~Unicorn_P@user/Unicorn-Princess/x-3540542) (Remote host closed the connection)