Re: Desktop Entry Specification: Add new ExecArg key
Manuel Schneider <[email protected]> Tue, 16 Jul 2024 17:06:39 +0200
| Newsgroups | gmane.linux.xdg.devel |
|---|---|
| Message-ID | <[email protected]> |
--Apple-Mail=_0749C769-2536-4B4F-80B2-FAF796E2B2A6 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii I commented = <https://gitlab.freedesktop.org/terminal-wg/specifications/-/merge_request= s/3#note_2490069> on the PR. I copy it for the mailing list here:=20 To get an impression of the current state check the hard coded list of = exec args of popular terms = <https://github.com/albertlauncher/albert/blob/6783257df7e15dcfbae49d03023= 414b7c3692dbd/src/app/terminalprovider.cpp#L120-L150> I (try to) = maintain for albert launcher. -e is indeed common for a majority of = terminals. I guess that this -e majority just emerged out of the absence of a real = standard with the common focal point of the ancient xterm interface. = Nothing that was ever really thought out. I think we should not default to -e despite being quasi standard. I = think that we should not default at all. Some terminals don't support = command execution. We have to handle this. Imho the absence of the ExecArgs should indicate that the terminal does = not support command execution. This is compatible with existing desktop = entries. ExecArgs should be allowed to exist but be empty, which will = fit the foot and kgx interface. Generally defaulting to anything would define an arbitrary standard = terminal command execution interface. Nothing sophisticated. Nothing = terminal developers should be imposed on. Imho the self-reporting of the command execution interface in the = desktop entry is a feature on its own. A standard (your proposal) could = then build on top of it. Later fancy extensions on how to collate = windows under X or Wayland can be discussed. They are rather opinionated = and hinder the realization of the former, rather simple and objectively = useful ideas. Also I wonder if this PR makes sense at all in the terminal-wg. These = proposals are not related to terminal standards. It is rather a general = desktop environment thing. Especially since I skimmed through #24 = <https://gitlab.freedesktop.org/terminal-wg/specifications/-/issues/24> = and none of the admins left a comment in the past 5 years. Shouldn't it = go to https://gitlab.freedesktop.org/xdg/xdg-specs?= --Apple-Mail=_0749C769-2536-4B4F-80B2-FAF796E2B2A6 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=us-ascii <html><head><meta http-equiv=3D"content-type" content=3D"text/html; = charset=3Dus-ascii"></head><body style=3D"overflow-wrap: break-word; = -webkit-nbsp-mode: space; line-break: after-white-space;">I <a = href=3D"https://gitlab.freedesktop.org/terminal-wg/specifications/-/merge_= requests/3#note_2490069">commented</a> on the PR. I copy it for the = mailing list here: <div><br></div><blockquote style=3D"margin: 0 0 = 0 40px; border: none; padding: 0px;"><div>To get an impression of the = current state check the hard coded<a = href=3D"https://github.com/albertlauncher/albert/blob/6783257df7e15dcfbae4= 9d03023414b7c3692dbd/src/app/terminalprovider.cpp#L120-L150"> list of = exec args of popular terms</a> I (try to) maintain for albert = launcher. -e is indeed common for a majority of = terminals.</div><div><br></div><div><div>I guess that this -e majority = just emerged out of the absence of a real standard with the common focal = point of the ancient xterm interface. Nothing that was ever really = thought out.</div></div><div><div><br></div></div><div><div>I think we = should not default to -e despite being quasi standard. I think that we = should not default at all. Some terminals don't support command = execution. We have to handle = this.</div></div><div><div><br></div></div><div><div>Imho the absence of = the ExecArgs should indicate that the terminal does not support command = execution. This is compatible with existing desktop entries. ExecArgs = should be allowed to exist but be empty, which will fit the foot and kgx = interface.</div></div><div><br></div><div><div>Generally defaulting to = anything would define an arbitrary standard terminal command execution = interface. Nothing sophisticated. Nothing terminal developers should be = imposed on.</div></div><div><br></div></blockquote><div>Imho the = self-reporting of the command execution interface in the desktop entry = is a feature on its own. A standard (your proposal) could then build on = top of it. Later fancy extensions on how to collate windows under X or = Wayland can be discussed. They are rather opinionated and hinder the = realization of the former, rather simple and objectively useful = ideas.</div><div><br></div><div>Also I wonder if this PR makes sense at = all in the terminal-wg. These proposals are not related to terminal = standards. It is rather a general desktop environment thing. Especially = since I skimmed through <a = href=3D"https://gitlab.freedesktop.org/terminal-wg/specifications/-/issues= /24">#24</a> and none of the admins left a comment in the past 5 = years. Shouldn't it go = to https://gitlab.freedesktop.org/xdg/xdg-specs?</div></body></html>= --Apple-Mail=_0749C769-2536-4B4F-80B2-FAF796E2B2A6--