4 ms·

This looks really cool, I'd like to give it a go. The idea of taking a screenshot of the terminal and then parsing that to determine the true colour support is

by tfsh 1y ago

This looks really cool, I'd like to give it a go. The idea of taking a screenshot of the terminal and then parsing that to determine the true colour support is definitely novel, though perhaps so, because for me I can't get it to work. Are there any debug flags I can enable? So far it was able to take the screenshot correctly (https://ibin.co/8kaRr8TIanv2.png https://ibin.co/8kaRr8TIanv2.png), however the parsing of that fails with the non-descript "Palette parsing failed." error. Edit: enabled tracing at got this: https://paste.ee/p/ZyNxG9FK https://paste.ee/p/ZyNxG9FK

shiomiru 1y ago
> The idea of taking a screenshot of the terminal and then parsing that to determine the true colour support is definitely novel, A better way to do this is to send `OSC 1 0 ; ? ST` (query foreground color), `OSC 1 1 ; ? ST` (background color), then `OSC 4 ; {n} ; ? ST` where {n} is the nth XTerm color. See: https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h4-Operating-System-Commands:OSC-Ps;Pt-ST.E65 https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h4-O...
- tombh 1y ago
  OMG really!? That link is blocked for me for some reason. If that OSC code is widely supported it's going to make things sooooo much easier.
  - hnlmorg 1y ago
    It’s supported by any xterm compatible terminal emulator. But like with most things in this domain, expect plenty of edge cases where it should work but doesn’t.
  - ku1ik 1y ago
    It’s very widely supported from my experience. This is how asciinema captures terminal palette.
tombh 1y ago
Thanks for trying it out. It looks like either your terminal or screenshotter isn't faithfully rendering the pure red marker column (it's needed for calibration in the parser). The red should be #ff0000, but the screenshot is using #ea3323. I've made a Github issue to keep track https://github.com/tattoy-org/tattoy/issues/98 https://github.com/tattoy-org/tattoy/issues/98 If you can add more details it'd really useful, I'm sure there'll be more people like you.
- mkesper 1y ago
  Maybe that's night mode (or whatever your DE calls it)?
  - tombh 1y ago
    Interesting, I'd never thought of that.

by tfsh 1y ago

shiomiru 1y ago

> The idea of taking a screenshot of the terminal and then parsing that to determine the true colour support is definitely novel, A better way to do this is to send `OSC 1 0 ; ? ST` (query foreground color), `OSC 1 1 ; ? ST` (background color), then `OSC 4 ; {n} ; ? ST` where {n} is the nth XTerm color. See: https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h4-Operating-System-Commands:OSC-Ps;Pt-ST.E65 https://invisible-island.net/xterm/ctlseqs/ctlseqs.html#h4-O...

tombh 1y ago
OMG really!? That link is blocked for me for some reason. If that OSC code is widely supported it's going to make things sooooo much easier.
- hnlmorg 1y ago
  It’s supported by any xterm compatible terminal emulator. But like with most things in this domain, expect plenty of edge cases where it should work but doesn’t.
- ku1ik 1y ago
  It’s very widely supported from my experience. This is how asciinema captures terminal palette.

tombh 1y ago

Thanks for trying it out. It looks like either your terminal or screenshotter isn't faithfully rendering the pure red marker column (it's needed for calibration in the parser). The red should be #ff0000, but the screenshot is using #ea3323. I've made a Github issue to keep track https://github.com/tattoy-org/tattoy/issues/98 https://github.com/tattoy-org/tattoy/issues/98 If you can add more details it'd really useful, I'm sure there'll be more people like you.

mkesper 1y ago
Maybe that's night mode (or whatever your DE calls it)?
- tombh 1y ago
  Interesting, I'd never thought of that.