From 01c03043e6e93dd12570a960ca54108d543578db Mon Sep 17 00:00:00 2001
From: Eli Zaretskii <eliz@gnu.org>
Date: Sat, 15 Mar 2025 11:48:36 +0200
Subject: [PATCH 1/2] ; Improve documentation of 'gui-get-selection' use on X

* doc/lispref/frames.texi (Window System Selections):
* lisp/select.el (gui-get-selection): Suggest to use an explicit
DATA-TYPE when calling 'gui-get-selection'.
---
 doc/lispref/frames.texi | 10 ++++++++--
 lisp/select.el          | 13 +++++++++----
 2 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/doc/lispref/frames.texi b/doc/lispref/frames.texi
index 39bfed48e4e..bc7a7e6c7c7 100644
--- a/doc/lispref/frames.texi
+++ b/doc/lispref/frames.texi
@@ -4079,8 +4079,14 @@ programs.  It takes two optional arguments, @var{type} and
 
 The @var{data-type} argument specifies the form of data conversion to
 use, to convert the raw data obtained from another program into Lisp
-data.  @xref{X Selections}, for an enumeration of data types valid under
-X, and @pxref{Other Selections} for those elsewhere.
+data.  It defaults to @code{STRING}.  @xref{X Selections}, for an
+enumeration of data types valid on X, and @pxref{Other Selections} for
+those elsewhere.  On X Window system, we recommend to always specify a
+particular @var{data-type}, especially if the selection is expected to
+be non-ASCII text (in which case Lisp programs should prefer
+@code{UTF8_STRING} as the value of @var{data-type}).  This is because
+the default @var{data-type} value, @code{STRING}, can only support
+Latin-1 text, which in many cases is nowadays inadequate.
 @end defun
 
 @defopt selection-coding-system
diff --git a/lisp/select.el b/lisp/select.el
index d8e97f33001..59eedc79263 100644
--- a/lisp/select.el
+++ b/lisp/select.el
@@ -372,10 +372,15 @@ all upper-case names.  The most often used ones, in addition to
 `PRIMARY', are `SECONDARY' and `CLIPBOARD'.
 
 DATA-TYPE is usually `STRING', but can also be one of the symbols
-in `selection-converter-alist', which see.  Window systems other
-than X usually support only a small subset of these symbols, in
-addition to `STRING'; MS-Windows supports `TARGETS', which reports
-the formats available in the clipboard if TYPE is `CLIPBOARD'."
+in `selection-converter-alist', which see.  On X, we recommend
+to always use a specific DATA-TYPE expected from the selection
+owner.  In particular, if the data is expected to be non-ASCII
+text, in many cases using \\='UTF8_STRING is the most reasonable
+value for DATA-TYPE.
+
+Window systems other than X usually support only a small subset of
+these symbols, in addition to `STRING'; MS-Windows supports `TARGETS',
+which reports the formats available in the clipboard if TYPE is `CLIPBOARD'."
   (let ((data (gui-backend-get-selection (or type 'PRIMARY)
                                          (or data-type 'STRING))))
     (when (and (stringp data)

From 1a8fbf69e3378996e633cf299b6a2d0716722d86 Mon Sep 17 00:00:00 2001
From: Eli Zaretskii <eliz@gnu.org>
Date: Sat, 15 Mar 2025 13:42:12 +0200
Subject: [PATCH 2/2] Fix 'whitespace-mode' in CJK locales

* lisp/international/characters.el (ambiguous-width-chars): Remove
U+00A4 and U+00B7 from the list of ambiguous-width characters.
(cjk-ambiguous-chars-are-wide): Doc fix.  (Bug#76852)
---
 lisp/international/characters.el | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/lisp/international/characters.el b/lisp/international/characters.el
index 318957c15b9..7be8bf32a93 100644
--- a/lisp/international/characters.el
+++ b/lisp/international/characters.el
@@ -1397,12 +1397,14 @@ with L, LRE, or LRO Unicode bidi character type.")
 
 ;; A: East Asian "Ambiguous" characters.
 (let ((l '((#x00A1 . #x00A1)
-	   (#x00A4 . #x00A4)
+	   ; (#x00A4 . #x00A4) whitespace-mode uses this
 	   (#x00A7 . #x00A8)
 	   (#x00AA . #x00AA)
 	   (#x00AD . #x00AE)
 	   (#x00B0 . #x00B4)
-	   (#x00B6 . #x00BA)
+	   ; (#x00B6 . #x00BA)  whitespace-mode uses U+00B7
+           (#x00B6 . #x00B6)
+           (#x00B8 . #x00BA)
 	   (#x00BC . #x00BF)
 	   (#x00C6 . #x00C6)
 	   (#x00D0 . #x00D0)
@@ -1620,6 +1622,10 @@ fonts being used.  In some CJK locales the fonts are set so that
 these characters are displayed as full-width.  This setting is most
 important for text-mode frames, because there Emacs cannot access the
 metrics of the fonts used by the console or the terminal emulator.
+You should configure the terminal emulator to behave consistently
+with the value of this option, by making sure it dispays ambiguous-width
+characters as half-width or full-width, depending on the value of this
+option.
 
 Do not set this directly via `setq'; instead, use `setopt' or the
 Customize commands.  Alternatively, call `update-cjk-ambiguous-char-widths'