sure, the only indicator for "while typing" being the `update_freq`; but well, we can't really predict what you're gonna do in the future, can we :)
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/ec4a5ea4f92f337d97b3ab2abae88e4ebf95d…
It can still happen while you're typing though, if you take the `update_freq` time to reach a next key, it will start parsing/updating even though you're still typing. Anyway, I'm not suggesting we actually change it since it obviously works fine in practice, it just seems like a weird way of doing it.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/ec4a5ea4f92f337d97b3ab2abae88e4ebf95d…
The point of not doing it when you type is that if you're actively typing enough and parsing would happen to be slow enough for you to notice (unlikely, but well), it seemed better to keep typing responsive instead of symbol updating -- which you're probably not looking at if you're typing so actively anyway.
So, yeah, `g_source_remove()` is possibly slow, but we never encountered a situation where it was *noticeably* slow, so I didn't bare. Possibly if we're worried, we could create a `GSource` explicitly and attach/detach it at will, not sure.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/ec4a5ea4f92f337d97b3ab2abae88e4ebf95d…
It's still never going to fire more than every `update_freq` ms though, and as you say, if you type fast, you don't get any highlighting updates. Also it feels like `g_source_remove` would do quite a bit of work (list patching, dynamic re-allocation, etc), more than is necessary.
FWIW, I only asked about this since I was doing something similar in a plugin and it had me wondering about it again.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/ec4a5ea4f92f337d97b3ab2abae88e4ebf95d…
The goal is to re-schedule the source `editor_prefs.autocompletion_update_freq`ms later, instead of it kicking effectively after exactly `editor_prefs.autocompletion_update_freq` -- so if you type fast, it's not called.
Yes, a better way would be to be able to amend the source's due time, but I don't know how to do that.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/commit/ec4a5ea4f92f337d97b3ab2abae88e4ebf95d…
A good example of why is the changes to `ui_menu_add_document_items_sorted()`: the *format* should contain markup, but the placeholders should not (they will generally be mostly unknown text).
This however has one side effect on the API: [`ui_frame_new_with_alignment()`](https://www.geany.org/manual/reference/ui__utils_8h.html#a740aa567ff6b9e3829565193e8317706)'s first argument is passed as-is to `ui_label_new_bold()`, which itself is implemented using `ui_label_set_markup()`. The documentation of `ui_frame_new_with_alignment()` never suggested it supported markup (and none of the users I could fine did expect that), but it's a change in how the API works.
IMO it's a bug fix (markup control character no longer break frame labels), but someone could see it as a feature removal (although it'd be somewhat dishonest as it never was advertized).
You can view, comment on, or merge this pull request online at:
https://github.com/geany/geany/pull/1174
-- Commit Summary --
* Escape format arguments passed to ui_label_set_markup()
-- File Changes --
M src/ui_utils.c (10)
-- Patch Links --
https://github.com/geany/geany/pull/1174.patchhttps://github.com/geany/geany/pull/1174.diff
---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/pull/1174
https://githubcom/geany/geany/blob/master/configureac#L35
this check is bogus and should be removed
it breaks usage for example when CXX is set to ```arm-linux-musleabi -isystem /sysroot/foo```
```
$ which g++ -isystem foo
/bin/which: unrecognized option: i
BusyBox v1202 (2015-04-23 14:57:49 CEST) multi-call binary
Usage: which [COMMAND]
Locate a COMMAND
```
---
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/issues/829
All the `#if` stuff could be isolated in a single module (vte.c), and not even spread out as much as the current runtime `dlopen()` support (we currently check at compile-time _and_ runtime for much stuff).
BTW, I'm currently tinkering with geany-vte.m4 to try and handle all the VTE versions including with and without gtk3, and whether `--enable-vte` was passed yes or auto, it's a combinatorial explosion, but it would make the code nicer/easier.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/issues/336#issuecomment-241004727