Skip to main content

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

8
  • 1
    +1 for a good answer; Alongside the already existing user interfaces, you can create your own aliases and small shellscripts to perform actions that you need often (and you can use short (yet unique) names to make them easy to type). Commented Jun 15, 2023 at 9:22
  • 1
    Additionally, it seems the wanted features can largely be provided by tab completion Commented Jun 15, 2023 at 16:47
  • @egmond these are all good question. For sure the solution should be smart enough to understand the scope of the execution. There is a lot of obstacles to be solved if we want a high class support for some app, but for some basic stuff it may be just a set of regexps mapped to the command to run. As to "where to run the command" question - the simplest would be to disable the interactive widgets if the terminal is not in prompt mode (yes, not very trivial task but solvable). As a last resort that can be a problem to be solved by ML. IMO, none of the issues listed above are hard blockers. Commented Jun 15, 2023 at 21:56
  • The best you can achieve is to invest a giant (as in, like, really giant) amount of work and get something that sometimes works, and by its very nature, can hardly be improved to work correctly a little bit more frequently. I, for one, would hate every second of having to work towards such a goal. Also, at many places the way to overcome the obstacles is to invent new stuff (e.g. helping escape sequences to be emitted by the shell): maybe easy if you do it just for yourself, but extremely hard if you want to get that accepted mainstream and ship for others out of the box... Commented Jun 16, 2023 at 5:02
  • Terminal emulation is sort of an "exact" science. Not that there aren't differences, disagreements, bugs etc., but if a TE does something, you can pretty much count on it doing this a 100% correctly. We've designed OSC 8 hyperlinking in a way that's 100% reliable. I've improved gnome-terminal's URL autodetection, spent one week on it, it's extremely complicated, works very well, but isn't 100% correct -- by the very nature of parsing a text flow for URLs isn't an exact science, there's no 100% correct solution, and just addressing all of the the already reported bugs at once took me one week. Commented Jun 16, 2023 at 5:07