From 9f9072d8e0eaa147b0498cb5c916e079ade9e3a4 Mon Sep 17 00:00:00 2001
From: Tom Ryder <tom@sanctum.geek.nz>
Date: Tue, 21 Mar 2023 18:57:58 +1300
Subject: Match [:space:] and all Unicode spaces

---
 autoload/strip_trailing_whitespace.vim | 49 ++++++++++++++++++++++++++++++++--
 1 file changed, 47 insertions(+), 2 deletions(-)

diff --git a/autoload/strip_trailing_whitespace.vim b/autoload/strip_trailing_whitespace.vim
index f9e9a1e..17dec7a 100644
--- a/autoload/strip_trailing_whitespace.vim
+++ b/autoload/strip_trailing_whitespace.vim
@@ -1,3 +1,48 @@
+" Set the pattern for trailing horizontal whitespace to match and remove.  The
+" `[:space:]` character class suffices for almost everything in practice, but
+" at the time of writing it still only includes ASCII characters.  I'm writing
+" this because I had a document with lines with a trailing NO-BREAK SPACE
+" (U+00A0) which `[:space:]` doesn't catch, so we'll round out the collection
+" by adding all the Unicode space characters, since that's easy to do.
+"
+" <https://jkorpela.fi/chars/spaces.html>
+" archived: <http://web.archive.org/web/20230318105634/https://jkorpela.fi/chars/spaces.html>
+"
+" * U+0020: SPACE
+" * U+00A0: NO-BREAK SPACE
+" * U+1680: OGHAM SPACE MARK
+" * U+180E: MONGOLIAN VOWEL SEPARATOR
+" * U+2000: EN QUAD
+" * U+2001: EM QUAD
+" * U+2002: EN SPACE
+" * U+2003: EM SPACE
+" * U+2004: THREE-PER-EM SPACE
+" * U+2005: FOUR-PER-EM SPACE
+" * U+2006: SIX-PER-EM SPACE
+" * U+2007: FIGURE SPACE
+" * U+2008: PUNCTUATION SPACE
+" * U+2009: THIN SPACE
+" * U+200A: HAIR SPACE
+" * U+200B: ZERO WIDTH SPACE
+" * U+202F: NARROW NO-BREAK SPACE
+" * U+205F: MEDIUM MATHEMATICAL SPACE
+" * U+3000: IDEOGRAPHIC SPACE
+" * U+FEFF: ZERO WIDTH NO-BREAK SPACE
+"
+let s:pattern
+      \ = '['
+      \ . '[:space:]'
+      \ . '\u0020'
+      \ . '\u00A0'
+      \ . '\u1680'
+      \ . '\u180E'
+      \ . '\u2000-\u200B'
+      \ . '\u202F'
+      \ . '\u205F'
+      \ . '\u3000'
+      \ . '\uFEFF'
+      \ . ']\+$'
+
 " Wrapper function to strip both horizontal and vertical trailing whitespace,
 " return the cursor to its previous position, and report changes
 function! strip_trailing_whitespace#(start, end) abort
@@ -46,8 +91,8 @@ function! s:StripHorizontal(start, end) abort
 
     " If the line has trailing whitespace, strip it off and bump the count
     let line = getline(num)
-    if line =~# '\s\+$'
-      call setline(num, substitute(line, '\s*$', '', ''))
+    if line =~# s:pattern
+      call setline(num, substitute(line, s:pattern, '', ''))
       let stripped = stripped + 1
     endif
 
-- 
cgit v1.2.3