From: Max Mikhanosha <max.mikhanosha@protonmail.com>
To: Eli Zaretskii <eliz@gnu.org>
Cc: emacs-devel@gnu.org
Subject: Re: Bugfix for utf-8 XTerm/MinTTY and (set-input-meta-mode t)
Date: Wed, 02 Jun 2021 09:22:08 +0000 [thread overview]
Message-ID: <l1dgWg2Mb2xfOS6txIRLWYsuUpWmBb0-q7KxuCT2VJq7QdYj4XBRvXMkOP8N1DXiD7kPmdSb_Om4C6aFM3TDQs0sGvKhvuyQQB1eakUNmKY=@protonmail.com> (raw)
In-Reply-To: <837djdzalk.fsf@gnu.org>
On Tuesday, June 1st, 2021 at 2:46 PM, Eli Zaretskii <eliz@gnu.org> wrote:
> > Date: Tue, 01 Jun 2021 18:35:43 +0000
>
> > From: Max Mikhanosha max.mikhanosha@protonmail.com
> >
> > Cc: emacs-devel@gnu.org
> >
> > Ok I'll rework the patch along these lines, and submit again later. Do you have a suggestion or input-meta-mode new enum value? My only idea is 'ENCODED, to mean same as T but that 8th bit will be checked after decoding the input with whatever keyboard coding system is.
>
> 'encoded' is okay (just in lower-case).
Updated patch pasted below. I've run it through the motions testing all combinations of utf8 xterm, non-utf-8 xterm, with input-meta-mode 'encoded' vs t and so on. In non-utf-8 xterm 'encoded' behaves exactly as t would.
diff --git a/src/keyboard.c b/src/keyboard.c
index aa3448439b..017e2176ed 100644
--- a/src/keyboard.c
+++ b/src/keyboard.c
@@ -2249,7 +2249,11 @@ read_decoded_event_from_main_queue (struct timespec *end_time,
int i;
if (meta_key != 2)
for (i = 0; i < n; i++)
- events[i] = make_number (XINT (events[i]) & ~0x80);
+ {
+ int c = XINT (events[i]);
+ int modifier = (meta_key == 3 && c < 0x100 && (c & 0x80)) ? meta_modifier : 0;
+ events[i] = make_number ((c & ~0x80) | modifier);
+ }
}
else
{
@@ -2258,7 +2262,7 @@ read_decoded_event_from_main_queue (struct timespec *end_time,
int i;
for (i = 0; i < n; i++)
src[i] = XINT (events[i]);
- if (meta_key != 2)
+ if (meta_key < 2) /* input-meta-mode T or NIL */
for (i = 0; i < n; i++)
src[i] &= ~0x80;
coding->destination = dest;
@@ -2276,7 +2280,15 @@ read_decoded_event_from_main_queue (struct timespec *end_time,
eassert (coding->carryover_bytes == 0);
n = 0;
while (n < coding->produced_char)
- events[n++] = make_number (STRING_CHAR_ADVANCE (p));
+ {
+ int c = STRING_CHAR_ADVANCE (p);
+ if (meta_key == 3)
+ {
+ int modifier = (c < 0x100 && (c & 0x80)) ? meta_modifier : 0;
+ c = (c & ~0x80) | modifier;
+ }
+ events[n++] = make_number (c);
+ }
}
}
}
@@ -7126,7 +7138,7 @@ tty_read_avail_input (struct terminal *terminal,
buf.modifiers = 0;
if (tty->meta_key == 1 && (cbuf[i] & 0x80))
buf.modifiers = meta_modifier;
- if (tty->meta_key != 2)
+ if (tty->meta_key < 2)
cbuf[i] &= ~0x80;
buf.code = cbuf[i];
@@ -10644,6 +10656,10 @@ DEFUN ("set-input-meta-mode", Fset_input_meta_mode, Sset_input_meta_mode, 1, 2,
If META is t, Emacs will accept 8-bit input, and interpret the 8th
bit as the Meta modifier.
+If META is 'encoded', Emacs will decode the input with accordance
+to current coding system, and then examine decoded input 8th bit
+and interpret it as the Meta modifier
+
If META is nil, Emacs will ignore the top bit, on the assumption it is
parity.
@@ -10671,8 +10687,10 @@ See also `current-input-mode'. */)
new_meta = 0;
else if (EQ (meta, Qt))
new_meta = 1;
- else
+ else if (!EQ (meta, Qencoded))
new_meta = 2;
+ else
+ new_meta = 3;
if (tty->meta_key != new_meta)
{
@@ -10754,6 +10772,7 @@ The value is a list of the form (INTERRUPT FLOW META QUIT), where
FLOW is non-nil if Emacs uses ^S/^Q flow control for output to the
terminal; this does not apply if Emacs uses interrupt-driven input.
META is t if accepting 8-bit input with 8th bit as Meta flag.
+ META 'encoded' means the same but 8th bit is checked after coding system
META nil means ignoring the top bit, on the assumption it is parity.
META is neither t nor nil if accepting 8-bit input and using
all 8 bits as the character code.
@@ -10771,7 +10790,8 @@ The elements of this list correspond to the arguments of
flow = FRAME_TTY (sf)->flow_control ? Qt : Qnil;
meta = (FRAME_TTY (sf)->meta_key == 2
? make_number (0)
- : (CURTTY ()->meta_key == 1 ? Qt : Qnil));
+ : (CURTTY ()->meta_key == 1 ? Qt :
+ (CURTTY ()->meta_key == 3 ? Qencoded : Qnil)));
}
else
{
@@ -11210,6 +11230,9 @@ syms_of_keyboard (void)
DEFSYM (Qecho_keystrokes, "echo-keystrokes");
+ /* input-meta-mode constant */
+ DEFSYM (Qencoded, "encoded");
+
Fset (Qinput_method_exit_on_first_char, Qnil);
Fset (Qinput_method_use_echo_area, Qnil);
next prev parent reply other threads:[~2021-06-02 9:22 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-01 16:19 Bugfix for utf-8 XTerm/MinTTY and (set-input-meta-mode t) Max Mikhanosha
2021-06-01 16:51 ` Eli Zaretskii
2021-06-01 17:28 ` Max Mikhanosha
2021-06-01 17:38 ` Eli Zaretskii
2021-06-01 18:01 ` Max Mikhanosha
2021-06-01 18:18 ` Eli Zaretskii
2021-06-01 18:35 ` Max Mikhanosha
2021-06-01 18:46 ` Eli Zaretskii
2021-06-02 9:22 ` Max Mikhanosha [this message]
2021-06-02 12:16 ` Andreas Schwab
2021-06-03 5:42 ` Max Mikhanosha
2021-06-05 14:20 ` Eli Zaretskii
2021-06-01 17:29 ` Eli Zaretskii
2021-06-01 17:45 ` Max Mikhanosha
2021-06-01 17:52 ` Eli Zaretskii
2021-06-01 18:10 ` Max Mikhanosha
2021-06-01 17:04 ` Andreas Schwab
2021-06-01 17:36 ` Max Mikhanosha
2021-06-01 20:06 ` Stefan Monnier
2021-06-02 10:21 ` Max Mikhanosha
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='l1dgWg2Mb2xfOS6txIRLWYsuUpWmBb0-q7KxuCT2VJq7QdYj4XBRvXMkOP8N1DXiD7kPmdSb_Om4C6aFM3TDQs0sGvKhvuyQQB1eakUNmKY=@protonmail.com' \
--to=max.mikhanosha@protonmail.com \
--cc=eliz@gnu.org \
--cc=emacs-devel@gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/emacs.git
https://git.savannah.gnu.org/cgit/emacs/org-mode.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.