The whole value 99999 doesn't fit into the entry which looks ugly and, to me, somehow a rounded value like 1000 (which is still sufficient in terms of "unlimited" number of characters in a tab) looks more pleasing.
I'm adding this nit-picky change for consideration for 2.0 since it should be low-risk, and, also, because once the default value "99999" gets written to user's config files, it will stay there forever even if we change it in future releases.
Now: <img width="812" alt="Screenshot 2023-10-18 at 10 59 21" src="https://github.com/geany/geany/assets/713965/fb12a455-10e0-4e74-bb82-f283ac3...;
After: <img width="812" alt="Screenshot 2023-10-18 at 10 59 02" src="https://github.com/geany/geany/assets/713965/653b9751-6bdb-4eb0-9bfb-d00c0b2...;
You can view, comment on, or merge this pull request online at:
https://github.com/geany/geany/pull/3612
-- Commit Summary --
* Change default tab_label_length from 99999 to 1000
-- File Changes --
M src/ui_utils.c (2)
-- Patch Links --
https://github.com/geany/geany/pull/3612.patch https://github.com/geany/geany/pull/3612.diff
Although I might have chosen 9999, 1000 is fine, and after all its the default, not the maximum the user can set. So LGBI.
An aside: @eht16 congrats the Windows CI is working again. Interesting to see the relative speeds, meson/ninja vs Autotools/gmake vs Autotools in a container.
BTW, what's the unit of the setting? pixels, chars, apples?
BTW, what's the unit of the setting? pixels, chars, apples?
Good question, probably glyphs if its done by GTK, or bytes if its done by Geany :-)
@kugel- @techee looks like it's code points if cursory reading is to be trusted (looks like it's based on `g_utf8_strlen()`).
Anyway, LGBI and agree that it's not pretty in the prefs dialog as is.
@b4n approved this pull request.
BTW, what's the unit of the setting? pixels, chars, apples?
The tooltip of the entry says "Characters to be visible on the tab label of the file". While it's a bit inaccurate wrt international text as @b4n noted, it's probably fine for "normal" projects and files.
Merged #3612 into master.
An aside: @eht16 congrats the Windows CI is working again. Interesting to see the relative speeds, meson/ninja vs Autotools/gmake vs Autotools in a container.
Comparing Meson vs Autotools was already possible before. The Windows CI builds should be excluded from such comparisons because the cross compiling toolchain is somewhat slower than compiling natively on Windows with mingw64 and this is much slower than compiling natively on Linux. Factors like the build system or the Docker container overhead are less significant than the used compiler with regard to mingw64.
github-comments@lists.geany.org