2026-06-14 08:13:16 <xentrac     > lofty: what editor are you using?
2026-06-14 09:09:08 <lofty       > xentrac: VS Code
2026-06-14 11:14:20 <lofty       > https://gist.github.com/Ravenslofty/92c4908a7d6e8eff6a62a71e04a74913 <-- so, I ended up adding yet another stack to forth; a position stack that the position functions work on. this turns out to make the code substantially nicer than passing a position around everywhere ^^;;
2026-06-14 13:15:14 <Stalevar    > $ ./test.fs
2026-06-14 13:15:14 <Stalevar    > in file included from *OS command line*:-1
2026-06-14 13:15:15 <Stalevar    > $7F5D31A11A08 throw
2026-06-14 13:15:26 <Stalevar    > Why they didn't think about shebang?
2026-06-14 13:28:17 <lofty       > Who's "they"?
2026-06-14 13:29:38 <lofty       > Wikipedia's article on the Bourne shell says that "#" as a comment character comes from UNIX System III in 1981, so I don't think Moore was thinking of it :p
2026-06-14 15:08:15 <Stalevar    > lofty, gforth authors. They could make an exception for she-bang even if forth normal comment is backslash
2026-06-14 15:08:41 <kibty       > `#! /usr/bin/env gforth`
2026-06-14 15:08:50 <Stalevar    > Other languages tend to add special case # as comment if it's first line
2026-06-14 15:16:25 <kibty       > you have to separate the shebang from the binary path and then it works
2026-06-14 15:18:42 <Stalevar    > $ ./test.fs
2026-06-14 15:18:42 <Stalevar    > Hello world!
2026-06-14 15:18:53 <Stalevar    > #! /usr/bin/env gforth
2026-06-14 15:18:53 <Stalevar    > \ Use gforth 0.7.3 to run
2026-06-14 18:49:44 <sqvery      > I am sure you can add this to Forth.
2026-06-14 19:18:54 <xentrac     > lofty: surely it's not harder to change the font of some text in VS Code than it is in Emacs?
2026-06-14 22:46:24 <oak         > Hi. I was looking at COUNT (https://forth-standard.org/standard/core/COUNT) and it mentions the term "character string specification". I couldn't find this term defined in the spec though. Did I miss it, perhaps by another name?
2026-06-14 23:03:17 <oak         > Ok, I found "character string" in Chapter 2. I'm still confused though. Does the standard have terms to distinguish between 1) a counted string in memory (i.e. a char for the length followed by the char data), and 2) a counted string on the stack (i.e. a *cell* for the length, with a pointer to the char data under it)?
2026-06-14 23:08:53 <lofty       > Not that I'm aware of; these are both "counted strings"
2026-06-14 23:12:20 <xentrac     > no, a counted string is #1
2026-06-14 23:12:57 <xentrac     > #2, two values on the stack (a pointer to the char data and a length) is never called a "counted string" in the standard
2026-06-14 23:13:17 <lofty       > I think I might have misinterpreted what oak said
2026-06-14 23:14:02 <xentrac     > this is explained in https://forth-standard.org/standard/notation
2026-06-14 23:14:27 <xentrac     > counted string: A data structure consisting of one character containing a length followed by zero or more contiguous data characters. Normally, counted strings contain text.
2026-06-14 23:14:54 <lofty       > The standard seems to just refer to #2 as "c-addr u" format
2026-06-14 23:15:04 <xentrac     > right
2026-06-14 23:15:21 <xentrac     > it would be helpful to have a term for it I think
2026-06-14 23:16:25 <xentrac     > because of course you could have a situation where you have a c-addr on the stack with an unsigned number on top of it where the unsigned number is not the length of a string beginning at the c-addr
2026-06-14 23:19:22 <xentrac     > I think it would be even better to just not have counted strings (according to the standard definition) at all.  they're a pain in the ass, and they're a big source of confusion for new Forthers
2026-06-14 23:19:52 <lofty       > I think the standard is definitely moving in that direction
2026-06-14 23:20:38 <xentrac     > it's moving in the direction of not having to use them (for example with the addition of find-name) but not in the direction of removing the existing standardized words that use them