[PATCH] [gdb/cli] Don't emit emojis in MI

[PATCH] [gdb/cli] Don't emit emojis in MI

[Tom de Vries]

PR mi/34501 reports the following:
...
$ gdb -q \
    -ex 'set charset UTF-8' \
    -ex 'interpreter-exec mi2 "-break-insert -f foo' \
    -ex quit
&"�\235\214�\217 No symbol table is loaded.  Use the \"file\" command.\n"
  ...
$
...

The output is a bit odd, but that gets better if we use
'set print sevenbit-strings on':
...
&"\342\235\214\357\270\217 No symbol table is loaded.  Use the \"file\" command.\n"
...

The output we see there is the error emoji:
...
$ gdb
(gdb) b foo
❌️ No symbol table is loaded.  Use the "file" command.
...

More specifically, two utf-8 encoded unicode characters:
- Cross Mark [1]: 0xE2 0x9D 0x8C
- Variation Selector-16 (VS16) [2]: 0xEF 0xB8 0x8F

Now the question: is GDB doing something wrong?

I think we probably should encode unicode characters in MI error strings as
octal escapes, independent of the sevenbit-strings setting.  This patch does
not address this part.

Then there's the question whether we should emit emojis in MI error strings in
the first place [1].  In principle they're unicode characters encoded in UTF-8,
and we can expect other such unicode characters in translated error strings.

But, given that MI has can_emit_style_escape () == false, and already filters
out ANSI escape sequences, I think it's reasonable to also disable emojis.

As for implementation, I introduced a function emoji_allowed alongside
can_emit_style_escape, which defaults to the value of can_emit_style_escape.

Tested on x86_64-linux.

Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=34501

[1] https://www.compart.com/en/unicode/U+274C
[2] https://www.compart.com/en/unicode/U+FE0F
[3] https://sourceware.org/bugzilla/show_bug.cgi?id=33920#c1
---
 gdb/cli/cli-style.c                          |  4 +--
 gdb/testsuite/gdb.base/style-mi-no-emoji.exp | 30 ++++++++++++++++++++
 gdb/ui-file.h                                |  8 ++++++
 3 files changed, 40 insertions(+), 2 deletions(-)
 create mode 100644 gdb/testsuite/gdb.base/style-mi-no-emoji.exp

diff --git a/gdb/cli/cli-style.c b/gdb/cli/cli-style.c
index 03a887232c7..cd67a6f382c 100644
--- a/gdb/cli/cli-style.c
+++ b/gdb/cli/cli-style.c
@@ -467,7 +467,7 @@ show_warning_prefix (struct ui_file *file, int from_tty,
 void
 print_warning_prefix (ui_file *file)
 {
-  if (emojis_ok ())
+  if (file->emoji_allowed () && emojis_ok ())
     gdb_puts (warning_prefix.c_str (), file);
 }
 
@@ -489,7 +489,7 @@ show_error_prefix (struct ui_file *file, int from_tty,
 void
 print_error_prefix (ui_file *file)
 {
-  if (emojis_ok ())
+  if (file->emoji_allowed () && emojis_ok ())
     gdb_puts (error_prefix.c_str (), file);
 }
 
diff --git a/gdb/testsuite/gdb.base/style-mi-no-emoji.exp b/gdb/testsuite/gdb.base/style-mi-no-emoji.exp
new file mode 100644
index 00000000000..c46c35b04a6
--- /dev/null
+++ b/gdb/testsuite/gdb.base/style-mi-no-emoji.exp
@@ -0,0 +1,30 @@
+# Copyright (C) 2026 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+# Check that emojis are not printed in MI error messages.
+
+with_ansi_styling_terminal {
+    clean_restart
+
+    gdb_test "set style emoji on"
+
+    # Check that there is no emoji printed before the error.  Regression test
+    # for PR34501.
+    set re_line \
+	[string_to_regexp \
+	     {&"No symbol table is loaded.  Use the \"file\" command.\n"}]
+    gdb_test {interpreter-exec mi2 "-break-insert -f foo"} \
+	"\r\n${re_line}(?=\r\n).*"
+}
diff --git a/gdb/ui-file.h b/gdb/ui-file.h
index 76185bbedf0..62671c2af81 100644
--- a/gdb/ui-file.h
+++ b/gdb/ui-file.h
@@ -93,6 +93,14 @@ class ui_file
   virtual bool can_emit_style_escape ()
   { return false; }
 
+  /* True if emojis are allowed on STREAM.  */
+  virtual bool emoji_allowed ()
+  {
+    /* By default, assume that emojis are not allowed on streams that don't
+       support ANSI escapes.  */
+    return can_emit_style_escape ();
+  }
+
   virtual void flush ()
   {}
 

base-commit: 4e8ba93fdb24f9bcf31f706e62e67c57ae91168f
-- 
2.51.0

Re: [PATCH] [gdb/cli] Don't emit emojis in MI

[Tom Tromey]

>>>>> "Tom" == Tom de Vries <tdevries@suse.de> writes:

Tom> I think we probably should encode unicode characters in MI error strings as
Tom> octal escapes, independent of the sevenbit-strings setting.  This patch does
Tom> not address this part.

I think there are some other bugs concerning this FWIW.

I tend to think MI should not set sevenbit-strings and should just use
the host encoding.  That way, clients could simply use UTF-8 and not
have to deal with the oddities here.

sevenbit-strings also seems pretty questionable to me.  Like maybe it
made sense in the ancient times before the charset code was written, but
nowadays I think it's just a mistake.

Tom> As for implementation, I introduced a function emoji_allowed alongside
Tom> can_emit_style_escape, which defaults to the value of can_emit_style_escape.

Seems reasonable though:

Tom> +  /* True if emojis are allowed on STREAM.  */
Tom> +  virtual bool emoji_allowed ()
Tom> +  {
Tom> +    /* By default, assume that emojis are not allowed on streams that don't
Tom> +       support ANSI escapes.  */
Tom> +    return can_emit_style_escape ();
Tom> +  }

...there's no present need for 'virtual' here, and if we do ever need
it, it is easy to add.  So I would suggest removing that.

Approved-By: Tom Tromey <tom@tromey.com>

Tom