Sat 30 Jul 2016 09:06:40 AM UTC, comment #11:
Released in 2.6.2.
|
Sun 24 Jul 2016 11:49:21 AM UTC, comment #10:
Pushed to git, fc0ddab.
Thanks!
|
Sat 23 Jul 2016 08:04:35 PM UTC, comment #9:
Okay, I suppose I'm just paranoid about error checking sometimes. The attached version is fine. You go ahead and commit it; I'm not sure I still have commit rights, and I'm also rusty on some parts of git and would prefer not to screw things up unnecessarily.
Thanks in advance.
|
Sat 23 Jul 2016 07:30:45 PM UTC, comment #8:
The 'if (uni_digits < 6)' is not necessary, because when the unicode is complete, the program doesn't get there. Also, I don't like the cursor being displayed in the edit window while things are happening on the statusbar, so I've removed that part.
If you're okay with the attached version, you can commit it. (Or I can do that, if you prefer.)
(file #38004)
|
Sat 23 Jul 2016 04:58:50 PM UTC, comment #7:
Good to know. I've moved the printing into get_unicode_kbinput() in the new version (unicode-with-dots-improved.patch). I also tweaked the right-padding display code so that it doesn't need a second variable (it's still ugly, but printf()'s built-in support for right-padding seems to be very limited).
(file #38001)
|
Sat 23 Jul 2016 09:05:49 AM UTC, comment #6:
Whether it's parse_verbatim_kbinput() or get_unicode_kbinput() that does the printing doesn't make much difference, intertwiningwise. It's nicer when the routine that's called "get_unicode..." actually prints the code.
I don't want to do the same for <Esc> <Esc> three digits. Most people can keep track of three steps. But with Unicode input... In the beginning I often aborted, because I had forgotten what I had typed; I had to train myself to keep track of the sequence.
Also, in a UTF8 locale, the usefulness of Esc-Esc is very limited. When people need higher codes, they will have to use Unicode Input, and once they know that, they will likely use it for everything, because they will know the hexadecimal codes, not the decimal ones. Also, not giving feedback for Esc-Esc makes for extra contrast between the two modes.
|
Fri 22 Jul 2016 08:24:41 PM UTC, comment #5:
Looks good to me. I don't know how to do it in a single printf() either, and printf()'s options can get horribly complex. Indeed, maybe it would be nicer to do it in get_unicode_kbinput(), but I was trying to keep the input and output routines separate when I could.
Incidentally, should this type of feedback also be done for nano's byte input mode (Esc Esc [3-digit number sequence between 000-255] produces a byte with the associated value)? Currently, it doesn't, because the same functionality in Pico doesn't.
|
Fri 22 Jul 2016 06:41:01 PM UTC, comment #4:
Better. But I don't like the sideways expansion during input: a digit, once typed, doesn't stay in the exact same spot. And the lack of dots doesn't show the user how many more digits there are left to type. So I've tweaked it to the attached version.
It's a bit ugly, because I don't know how to do that in a single printf(), and the extra variable is just a silly name. Also, probably it would be nicer to let get_unicode_kbinput() itself print the statusbar and not pass the partial result back.
(file #37990)
|
Fri 22 Jul 2016 02:45:56 PM UTC, comment #3:
Okay. Attached an improved version against current git which should do this.
(file #37989)
|
Fri 22 Jul 2016 08:31:27 AM UTC, comment #2:
Hi David,
Thanks for the patch! It works well. But... I don't like the look of "000000" when I've just typed the first zero. When I've typed, say, 0120, it is impossible to see how many digits I have typed. So it would be much nicer if after the first digit, the display would look like "Unicode Input: 0.....", and after 0120: "Unicode Input: 0120..". This will be harder to make, but it makes it so much clearer what has been typed and how many more digits are expected.
|
Wed 20 Jul 2016 05:57:14 PM UTC, comment #1:
Attached an attempt at this, as of today's git. Note that the statusbar message is "Unicode Input: XXXXXX" in all cases, so that there isn't both a "Unicode Input" message and an "Entering Unicode: XXXXXX" message. This also attempts to guarantee that, when Unicode input is used in the edit window, the cursor remains visible in the edit window.
(file #37958)
|
Mon 06 Jun 2016 07:29:54 PM UTC, original submission:
When you type M-V 0, the statusbar will show: "Unicode Input". The other five hex digits you will have to type blind. It would be much nicer if the statusbar would show "Entering Unicode: 0" and then add the digits as you type them.
|