Summary: | Allow kioclient exec to open kwrite/kate documents at specified cursor | ||
---|---|---|---|
Product: | [Applications] kate | Reporter: | Gleb Popov <6yearold> |
Component: | general | Assignee: | KWrite Developers <kwrite-bugs-null> |
Status: | RESOLVED FIXED | ||
Severity: | wishlist | CC: | elvis.angelaccio, nate, P.Suetterlin |
Priority: | NOR | ||
Version: | Git | ||
Target Milestone: | --- | ||
Platform: | unspecified | ||
OS: | All | ||
See Also: | https://bugs.kde.org/show_bug.cgi?id=408632 | ||
Latest Commit: | https://commits.kde.org/kde-cli-tools/8072a6acf221b8b609a781290ee2870ca2eea3dd | Version Fixed In: | Applications 19.04.0 with Frameworks 5.58 or later |
Description
Gleb Popov
2018-09-23 19:43:54 UTC
I'd say best way to deal with it would be to use url semantics (e.g. file.cpp?l=23). Either way, this needs to be decided by kate maintainers. (In reply to Aleix Pol from comment #1) > I'd say best way to deal with it would be to use url semantics (e.g. > file.cpp?l=23). Either way, this needs to be decided by kate maintainers. This is less comfortable to type and as I said, the main idea is to be able to copy-paste error line produced by the compiler, like: /absolute/path/to/foo.cpp:123:1: Couldn't find #include <foo.h> How about this: We add KRun::setCursorInformation(int,int) method and in KRun::runApplicationImpl() or whatever actually runs the program do if(serviceSupportsPassingCursorInfo()) args << "--line " << line << "--column" << column; where bool serviceSupportsPassingCursorInfo() { return service.name() == "KWrite" || service.name() == "Kate; } Looks pretty clean and non-invasive to me. Moving to Kate for now. (In reply to Gleb Popov from comment #2) > (In reply to Aleix Pol from comment #1) > > I'd say best way to deal with it would be to use url semantics (e.g. > > file.cpp?l=23). Either way, this needs to be decided by kate maintainers. > > This is less comfortable to type and as I said, the main idea is to be able > to copy-paste error line produced by the compiler, like: > > /absolute/path/to/foo.cpp:123:1: > Couldn't find #include <foo.h> What if kioclient converts from "foo.cpp:123:1" to url parameters? This way you can just copy-paste the thing and the patch in KIO would be less invasive (no need to add/change the KRun API). Actually, if Kate adds support for row/column as URL parameters, we probably wouldn't need to patch KRun at all. (In reply to Elvis Angelaccio from comment #6) > Actually, if Kate adds support for row/column as URL parameters, we probably > wouldn't need to patch KRun at all. Sounds OK, I'll try implementing it. Git commit 1fa3a0e85e7146a7a3535980aadda6ebbab11ee6 by Gleb Popov. Committed on 15/01/2019 at 11:16. Pushed by arrowdodger into branch 'master'. Add ability to get cursor data from URL query arguments, like file://foo.txt?line=123&column=456 Summary: CCBUG: 398998 Test Plan: Opening a file with such URL works in both KWrite and Kate. Reviewers: #kate Subscribers: kwrite-devel Tags: #kate Differential Revision: https://phabricator.kde.org/D18099 M +15 -0 kate/kateapp.cpp M +16 -0 kwrite/main.cpp https://commits.kde.org/kate/1fa3a0e85e7146a7a3535980aadda6ebbab11ee6 Git commit 8072a6acf221b8b609a781290ee2870ca2eea3dd by Gleb Popov. Committed on 11/04/2019 at 07:41. Pushed by arrowdodger into branch 'master'. Add support for passing cursor information via URL parameters when running kioclient exec. Summary: Second part of fix for a BUG 398998 ( https://bugs.kde.org/show_bug.cgi?id=398998 ). Kate changes are in D18099. Test Plan: Printed resulted URL with `qDebug()`. Reviewers: #plasma, #ktexteditor, broulik, cullmann, #frameworks, pino, cfeck, dfaure, elvisangelaccio Reviewed By: dfaure Subscribers: apol, cullmann, plasma-devel Tags: #plasma Differential Revision: https://phabricator.kde.org/D18296 M +16 -5 kioclient/kioclient.cpp M +1 -2 kioclient/kioclient.h A +108 -0 kioclient/urlinfo.h [License: LGPL] https://commits.kde.org/kde-cli-tools/8072a6acf221b8b609a781290ee2870ca2eea3dd Please either revert or fix this patch, it is BREAKING OTHER APPLICATIONS, see https://bugs.kde.org/show_bug.cgi?id=408632 Since it was me who made this change, I'll look at it ASAP. Sorry for missing initial bug report. Fixed with https://phabricator.kde.org/D22525 |