Conversation

Notices

1. Embed this notice
   pancake :radare2: (pancake@infosec.exchange)'s status on Saturday, 10-May-2025 17:43:40 JST pancake :radare2:

   Now #Swift permits functions with spaces in the name. TIL that Kotlin, Zig, Scala and other languages already supported this “feature”.., https://github.com/swiftlang/swift-evolution/blob/main/proposals/0451-escaped-identifiers.md

   • Embed this notice
     Jeremiah Korreck (korreckj328@mastodon.social)'s status on Saturday, 10-May-2025 17:43:39 JST Jeremiah Korreck

     in reply to

     @pancake Maybe I'm just old but that just looks so wrong.

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Saturday, 10-May-2025 17:43:39 JST Alfred M. Szmidt

     in reply to

     • Jeremiah Korreck

     @korreckj328 @pancake 50+ year old languages support this ... frankly, this is a design flaw in how "modern" languages treat "symbols" / "names" or whatever you wish to call it.

     #Lisp did it right.

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Saturday, 10-May-2025 18:25:34 JST Alfred M. Szmidt

     in reply to

     • Jeremiah Korreck

     @pancake There are "sane restrictions" on symbols, if you do not quote them it should generally be the normal A-Za-z0-9 rule; but quoted ones should accept anything that can be accepted by the language.

     I see nothing wrong with having spaces, or emojis in file names. It is like lesser languages do not allowed for hyphens in symbols ... It is a silly limitation, that has no reason to exist.

     @korreckj328

   • Embed this notice
     pancake :radare2: (pancake@infosec.exchange)'s status on Saturday, 10-May-2025 18:25:35 JST pancake :radare2:

     in reply to

     • Alfred M. Szmidt
     • Jeremiah Korreck

     @amszmidt @korreckj328 imho defining some sane restrictions on symbols is somehow cleaner. It’s like having emojis or spaces in filenames. Surely we can even make files with slashes but it feels wrong

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Saturday, 10-May-2025 21:05:28 JST Alfred M. Szmidt

     in reply to

     • Jeremiah Korreck
     • smlckz

     @smlckz I mean, some languages already allow for that. E.g., in Lisp you could "trivially" have a readtable thing that would do something like auto translating snake_case to kebab-case.

     @pancake @korreckj328

   • Embed this notice
     smlckz (smlckz@fe.disroot.org)'s status on Saturday, 10-May-2025 21:05:35 JST smlckz

     in reply to

     • Alfred M. Szmidt
     • Jeremiah Korreck

     @amszmidt @pancake @korreckj328 I wonder if we’ll get to see the day when syntax qua syntax could be treated in the same manner as coding style (spaces vs. tabs, camelCase vs. snake_case vs. kebab-case vs. …), allowing for multiple “views” (in the database sense) of the same code for different choices of syntax, when each developer can locally use their preferred syntax and coding style and seamlessly collaborate with each other.. Is this too fantastical? We can already see a pattern of convergence towards certain features in the “most popular” programming languages: destructuring assignment, pattern matching, Option and Result types and so on.. it should be possible, right?

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Sunday, 11-May-2025 00:18:44 JST Alfred M. Szmidt

     in reply to

     • Jeremiah Korreck
     • smlckz

     @smlckz In Lisp this is very common, files can have separate read-tables, or bases, or whatever. Being able to change case of symbols, or the base or numbers is essentially the same in my book. Sometimes you _really_ want to print everything in octal or hex, or have things read in octal or hex. Most languages only allow extra syntax (and sometimes confusing, 0744 vs 744 ...). In Lisp you can change the read/print base of integers very easily.

     @pancake @korreckj328

   • Embed this notice
     smlckz (smlckz@fe.disroot.org)'s status on Sunday, 11-May-2025 00:18:45 JST smlckz

     in reply to

     • Alfred M. Szmidt
     • Jeremiah Korreck

     @amszmidt @pancake @korreckj328 Yeah, you can also “trivially” do something similar in Forth with SOURCE and >IN, here’s quite a fitting example: https://forth-standard.org/standard/core/PARSE-NAME

     You should understand that these are quite low-level mechanisms. What I’m imagining would need some sufficient “framework” or “infrastructure”, which would make it as easy as installing and tweaking editor themes or plugins. I wonder if there are people who are interested in making this possible; if so, who they are and how much progress was made.

   • Embed this notice
     Alfred M. Szmidt (amszmidt@mastodon.social)'s status on Sunday, 11-May-2025 00:19:36 JST Alfred M. Szmidt

     in reply to

     • Jeremiah Korreck
     • smlckz

     @smlckz You could do the same with case of symbols, by having special purpose read tables, and all you do is "Readtable: CamelCase", and ta da .. fooCase gets implicitly converted to foo-case or whatever is "canonical".

     @pancake @korreckj328