Sat 15 Oct 2016 03:22:40 PM UTC, comment #11:
Stupid mistake with double carets. Fixed now in git, c92982d.
Thanks for reporting the original bug and trying to fix it.
|
Sat 15 Oct 2016 03:15:06 PM UTC, comment #10:
Fixed in git, fdee0df.
And the replacements with arrows in 7a274d6, including proper alignment. ... Oops! So I thought! But the latter commit breaks the Ctrl+Arrow keys -- I just looked whether the help text was okay, and never tested the keys themselves. :|
What went wrong there? I don't get it, yet.
|
Thu 13 Oct 2016 02:20:39 PM UTC, comment #9:
Ignoring the alignment, I am fine with representation.
|
Wed 12 Oct 2016 03:25:19 PM UTC, comment #8:
Oops. Forgot to attach the patch, 0002, preceded by the modified first one, 0001.
(file #38711, file #38712)
|
Wed 12 Oct 2016 10:29:08 AM UTC, comment #7:
Well, I was thinking of using ^← ^→ ^↑ ^↓ to represent the Ctrl+Arrow keys. See the attached patch. However, it breaks the alignment of the help text -- the help stuff counts bytes instead of characters. But ignoring that for a moment, what do you think of this?
|
Wed 12 Oct 2016 08:41:47 AM UTC, comment #6:
It works fine but the old issue still persists: "^Left" and "^Right" are added to help text.[1] I was considering to skip these two specific string in src/help.c after you alluded me towards this solution. I have attached a patch[2] doing so on the top of your patch, if required. I have forgot to sign, I will send one if required.
1. http://lists.gnu.org/archive/html/nano-devel/2016-09/msg00082.html
2. https://savannah.gnu.org/bugs/download.php?file_id=38708
|
Tue 11 Oct 2016 05:57:51 PM UTC, comment #5:
I'm suggesting the attached patch. What do you think?
(file #38704)
|
Tue 20 Sep 2016 11:19:34 AM UTC, comment #4:
Partially fixed in git, 8ba57cf -- when completely unbinding a function, also the corresponding modified Arrow key is unbound. This is not entirely what we would like, but at least it prevents unexpected movements.
|
Mon 12 Sep 2016 05:17:06 PM UTC, comment #3:
Thanks for the help!
Somehow, I had figured the first step. For second step, I will think of few possible alternatives and post/mail again.
|
Sun 11 Sep 2016 04:08:14 PM UTC, comment #2:
Bug exists since nano-2.3.5 -- since the introduction of the Ctrl+Left/Right combos.
The fix will have to consist of multiple steps: first, let the sc_seq_or() calls not return zero when a lookup fails, but the current keycode. That will prevent Ctrl+Left from doing the /wrong/ thing when M-Space is unbound. As a second step, find a way to hard-bind Ctrl+Left/Right to their functions.
|
Sun 11 Sep 2016 12:56:43 PM UTC, comment #1:
I will try and fix this bug.
|
Sun 11 Sep 2016 12:40:10 PM UTC, original submission:
To repoduce, put "unbind M-Space all" in your ~/.nanorc and do src/nano src/browser.c. Press Ctrl-Left and Ctrl-Right one after the another, both makes cursor jump to the next word.
Expected behaviour is: Ctrl-Left should lead to jump to the previous word (as it does normally and ) as rcfile doesn't mention change its in behaviour.
|