Unposted, c.f. 21752: Expand insert-composed-char.

3 files changed, 387 insertions, 250 deletions

diff --git a/Functions/Zle/.distfiles b/Functions/Zle/.distfiles
index c6852732f..e6d1cf3d6 100644
--- a/Functions/Zle/.distfiles
+++ b/Functions/Zle/.distfiles
@@ -2,7 +2,8 @@ DISTFILES_SRC='
 .distfiles
 backward-kill-word-match       backward-word-match
 capitalize-word-match          copy-earlier-word
-cycle-completion-positions     delete-whole-word-match
+cycle-completion-positions     
+define-composed-chars          delete-whole-word-match
 down-case-word-match           down-line-or-beginning-search
 edit-command-line              forward-word-match
 history-pattern-search         history-search-end
diff --git a/Functions/Zle/define-composed-chars b/Functions/Zle/define-composed-chars
new file mode 100644
index 000000000..1a67add27
--- /dev/null
+++ b/Functions/Zle/define-composed-chars
@@ -0,0 +1,371 @@
+# This is not a widget function, it is only a helper for insert-composed-char
+# to cut down on resident memory use.
+
+# The associative array zsh_accent_chars is indexed by the
+# accent.  The values are sets of character / Unicode pairs for
+# the character with the given accent.  The Unicode value is
+# a hex index with no base discriminator; essentially a UCS-4 index
+# with the leading zeroes suppressed.
+typeset -gA zsh_accented_chars
+
+# Save quite a lot of space by using short names internally.
+local -A z
+local a
+
+# grave
+a=\!
+z[$a]="\
+A C0 E C8 I CC O D2 U D9 a E0 e E8 i EC o F2 u F9 N 1F8 n 1F9 \
+"
+# acute
+a=\'
+z[$a]="\
+A C1 E C9 I CD O D3 U DA Y DD a E1 e E9 i EC o F3 u FA y FD C 106 c 107 \
+L 139 l 13A N 143 n 144 R 154 r 155 S 15A s 15B Z 179 z 17A \
+"
+# circumflex
+a=\>
+z[$a]="\
+A C2 E CA I CE O D4 U DB a E2 e EA i EE o F4 u FB C 108 c 109 G 11C g 11d \
+H 124 h 125 J 134 j 135 S 15C s 15D W 174 w 175 Y 176 y 177 \
+"
+# tilde
+a=\?
+z[$a]="\
+A C3 E CB N D1 O D5 a E3 n F1 o F5 I 128 i 129 U 168 u 169 \
+"
+# macron (d-, D- give eth)
+a=-
+z[$a]="\
+A 100 a 101 d F0 D D0 E 112 e 113 I 12a i 12b O 14C o 14D U 16A u 16B \
+"
+# breve
+a=\(
+z[$a]="\
+A 102 a 103 E 114 e 115 G 11E g 11F I 12C i 12D O 14E o 14F U 16C u 16D "
+# dot above, small i with no dot, or l with middle dot
+a=.
+z[$a]="\
+\
+C 10A c 10b E 116 e 117 G 120 g 121 I 130 i 131 L 13F l 140 Z 17B z 17C \
+"
+# diaeresis / Umlaut
+a=:
+z[$a]="\
+A C4 I CF O D6 U DC a E4 e EB i EF o F6 u FC y FF Y 178 \
+"
+# cedilla
+a=,
+z[$a]="\
+C C7 c E7 G 122 g 123 K 136 k 137 L 13B l 13C N 145 n 146 R 156 r 157 \
+S 15E s 15F T 162 t 163 \
+"
+# underline (_) would go here
+# stroke through
+a=/
+z[$a]="\
+O D8 o F8 D 110 d 111 H 126 h 127 L 141 l 142 T 166 t 167 b 180 \
+"
+# double acute
+a=\"
+z[$a]="\
+O 150 o 151 U 170 u 171\
+"
+# ogonek
+a=\;
+z[$a]="\
+A 104 a 105 E 118 e 119 I 12E i 12F U 172 u 173 \
+"
+# caron
+a=\<
+z[$a]="\
+C 10C c 10D D 10E d 10F E 11A e 11B L 13D l 13E N 147 n 148 R 158 r 159 \
+S 160 s 161 T 164 t 165 Z 17D z 17E \
+"
+# ring above
+a=0
+z[$a]="\
+A C5 a E5 U 16E u 16F \
+"
+# hook above
+a=2
+z[$a]="\
+A 1EA2 a 1EA3 E 1EBA e 1EBA \
+"
+# horn, also right quotation marks
+a=9
+z[$a]="\
+O 1A0 o 1A1 U 1Af u 1b0 ' 2019 . 201A \" 201D : 201E \
+"
+# left quotation marks
+a=6
+z[$a]="\
+' 2018 \" 201C \
+"
+# reversed quotation marks for convenience
+a=\'
+z[$a]+=" \
+9 201B \
+"
+a=\"
+z[$a]+=" \
+9 201F \
+"
+
+# ligature with E
+a=e
+z[$a]="\
+A C6 O 152 \
+"
+# ligature with e
+a=e
+z[$a]="\
+a E6 o 153 \
+"
+# ligature with J
+a=J
+z[$a]="\
+I 132 \
+"
+# ligature with j
+a=j
+z[$a]="\
+i 133 \
+"
+# eszett
+a=s
+z[$a]="\
+s DF \
+"
+# upper case thorn
+a=H
+z[$a]="\
+T DE \
+"
+# lower case thorn
+a=h
+z[$a]="\
+t FE \
+"
+
+# Arabic characters
+a=\+
+z[$a]+=" \
+, 60C ; 61B ? 61F a 627 b 628 t 62A g 62C x 62E d 62F r 631 z 632 s 633 \
+c 635 e 639 i 63A + 640 f 641 q 642 k 643 l 644 m 645 n 646 h 647 w 648 \
+j 649 y 64A : 64B \" 64C = 64D / 64E ' 64F 1 650 3 651 0 652 p 67E v 6A4 \
+"
+a=\'
+z[$a]+=" H 621"
+z[a]+=" \
+0 6F0 1 6F1 2 6F2 3 6F3 4 6F4 5 6F5 6 6F6 7 6F7 8 6F8 9 6F9 \
+"
+z[d]+=" d 636"
+z[f]+=" g 6AF"
+z[H]+=" a 623 w 624 y 626 z 638"
+z[h]+=" a 625"
+z[j]+=" t 637"
+z[k]+=" t 62B h 62D d 630"
+z[M]+=" a 622"
+z[m]+=" t 629"
+z[n]+=" s 634"
+z[S]+=" a 670"
+
+# Cyrillic characters
+a=\=
+z[$a]+=" \
+A 410 B 411 V 412 G 413 D 414 E 415 Z 417 I 418 J 419 K 41A L 41B \
+M 41C N 41D O 41E P 41F R 420 S 421 T 422 U 423 F 424 H 425 C 426 \
+Y 42B
+a 430 b 431 v 432 g 433 d 434 e 435 z 437 i 438 j 439 k 43A l 43B \
+m 43C n 43D o 43E p 43F r 440 s 441 t 442 u 443 f 444 h 445 c 446 \
+y 44B
+"
+z[%]+=" \
+D 402 G 403 J 408 V 40E Z 416 C 427 S 428 z 436 c 447 s 448
+d 452 g 453 j 458 v 45E \
+"
+z[A]+=" J 42F"
+z[a]+=" j 44F"
+z[c]+=" S 429 s 449"
+z[E]+=" I 404 J 42D"
+z[e]+=" j 44D i 454"
+z[I]+=" I 406 Y 407"
+z[i]+=" i 456 y 457"
+z[J]+=" L 409 N 40A K 40C"
+z[j]+=" l 459 n 45A k 45C"
+z[O]+=" I 401"
+z[o]+=" i 451"
+z[S]+=" D 405"
+z[s]+=" T 40B d 455 t 45B"
+z[U]+=" J 42E"
+z[u]+=" j 44E"
+z[Z]+=" D 40F"
+z[z]+=" d 45F"
+a=\"
+z[$a]+=" = 42A % 42C"
+a=\'
+z[$a]+=" = 44A % 44C"
+z[3]+=" \
+Y 462 y 463 O 46A o 46B F 472 f 473 V 474 v 475 C 480 c 481 \
+G 490 g 491 \
+"
+
+# Greek characters
+a=%
+z[$a]+=" \
+A 386 E 388 Y 389 I 38A O 38C U 38E W 38F \
+a 3Ac e 3Ad y 3Ae i 3AF \
+o 3CC u 3CD w 3CE ' 3F4 \
+"
+a=\*
+z[$a]+=" \
+A 391 B 392 G 393 D 394 E 395 Z 396 Y 397 H 398 I 399 K 39A L 39B \
+M 39C N 39D C 39E O 39F P 3A0 R 3A1 S 3A3 T 3A4 U 3A5 F 3A6 X 3A7 \
+Q 3A8 W 3A9 J 3AA V 3Ab \
+a 3B1 b 3B2 g 3B3 d 3B4 e 3B5 z 3B6 y 3b7 h 3B8 i 3B9 k 3Ba l 3BB \
+m 3BC n 3BD c 3BE o 3BF p 3C0 r 3C1 s 3C3 t 3C4 u 3C5 f 3C6 x 3C7 \
+q 3C8 w 3C9 j 3CA v 3CB \
+"
+a=3
+z[$a]+=" \
+i 390 u 3B0 T 3DA t 3DB M 3DC m 3DD K 3DE k 3DF P 3E0 p 3E1 j 3F5 \
+"
+z[s]+=" * 3C2"
+z[G]+=" ' 3D8 , 3D9"
+
+# Hebrew characters
+a=+
+z[$a]+=" \
+A 5D0 B 5D1 G 5D2 D 5D3 H 5D4 W 5D5 Z 5D6 X 5D7 J 5D9 K 5DB L 5Dc M 5dE \
+N 5E0 S 5E1 E 5E2 P 5E4 Q 5E7 R 5E8 T 5EA \
+"
+a=j
+z[$a]+=" T 5D8 Z 5E5"
+a=%
+z[$a]+=" K 5DA M 5DD N 5DF P 5E3 "
+a=J
+z[$a]+=" Z 5e6"
+a=h
+z[$a]+=" S 5e9"
+
+# Remaining characters are handled as separate pairs.
+# We need to remember that the assoc array is keyed by the second character.
+# Left square bracket
+a=\(
+z[$a]+=" < 5B"
+# Reverse solidus (backslash to you and me).
+a=/
+z[$a]+=" / 5C"
+# Right square bracket, circumflex
+a=\>
+z[$a]+=" ) 5D ' 5E"
+# Grave a
+a=\!
+z[$a]+=" ' 60"
+# diglyphys for (usually) standard characters {, |, }, ~
+a=\!
+z[$a]+=" ( 7B"
+z[$a]+=" ! 7C"
+a=\)
+z[$a]+=" ! 7D"
+a=\?
+z[$a]+=" ' 7E"
+# non-breaking space
+z[S]+=" N A0"
+# inverted exclamation mark
+z[I]+=" ! A1"
+# cent
+z[t]+=" C A2"
+# pound sterling
+z[d]+=" P A3"
+# currency
+z[u]+=" C A4"
+# yen
+z[e]+=" Y A5"
+# broken bar
+z[B]+=" B A6"
+# section
+z[E]+=" S A7"
+# lonely diaeresis
+z[:]+=" ' A8"
+# copyright
+z[o]+=" C A9"
+# spanish feminine ordinal
+z[a]+=" - AA"
+# left guillemet
+a=\<
+z[$a]+=" < AB"
+z[O]+=" N AC"
+# soft hyphen
+z[-]+=" - AD"
+# registered
+z[g]+=" R AE"
+# lonely macron
+z[m]+=" ' AF"
+# degree
+z[G]+=" D B0"
+# +/-
+z[-]+=" + B1"
+# superscripts
+z[S]+=" 2 B2 3 B3"
+# lonely acute
+a=\'
+z[$a]+=" ' B4"
+# micro
+z[y]+=" M B5"
+# pilcrow (paragraph)
+z[I]+=" P B6"
+# Middle dot
+z[M]+=" . B7"
+# Lonely cedilla
+z[,]+=" ' B8"
+# Superscript one
+z[S]+=" 1 B9"
+# spanish masculine ordinal
+z[o]+=" - BA"
+# right guillemet
+a=\>
+z[$a]+=" > BB"
+# fractions
+z[4]+=" 1 BC 3 BE"
+z[2]+=" 1 BD"
+# inverted question mark
+z[I]+=" ? BF"
+# multiplication
+z[X]+=" * D7"
+# division
+z[:]+=" - F7"
+# kra
+z[k]+=" k 138"
+# apostrophe n
+z[n]+=" ' 149"
+# Lappish ng
+z[G]+=" N 14A"
+z[g]+=" n 14B"
+# OI
+z[I]+=" O 1A2"
+z[i]+=" o 1A3"
+# yr
+z[r]+=" y 1A6"
+# ezh
+z[D]+=" E 1B7"
+# euro (I invented this but it's logical)
+z[u]+=" E 20AC"
+# hyphen
+z[1]+=" - 2010"
+# en dash
+z[N]+=" - 2013"
+# em dash
+z[M]+=" - 2014"
+# horizontal bar, vertical and horizontal ellipsis
+z[3]+=" - 2015 : 22EE . 22EF"
+# double vertical line, double low line
+z[2]+=" ! 2016 = 2017"
+# dagger and double dagger
+z[-]+=" / 2020"
+z[=]+=" / 2021"
+# per mille
+z[0]+=" % 2030"
+
+zsh_accented_chars=("${(kv)z[@]}")
diff --git a/Functions/Zle/insert-composed-char b/Functions/Zle/insert-composed-char
index 60a42e089..7b97c96d4 100644
--- a/Functions/Zle/insert-composed-char
+++ b/Functions/Zle/insert-composed-char
@@ -86,6 +86,14 @@
 #  -o           Spanish masculine ordinal indicator
 #  >>           Right guillemet
 #  ?I           Inverted question mark
+#  -1           Hyphen
+#  -N           en dash
+#  -M           em dash
+#  -3           horizontal bar
+#  :3           vertical ellipsis
+#  .3           horizontal midline ellipsis
+#  !2           double vertical line
+#  =2           double low line
 #  '6           Left single quote
 #  '9           Right single quote
 #  .9           "Right" low quote
@@ -110,6 +118,7 @@
 #  34           Three quarters
 #  *X           Multiplication
 #  -:           Division
+#  %0           Per mille
 #
 # Accents with no base character
 # '>            Circumflex (caret)
@@ -131,255 +140,11 @@ else
 fi
 
 if (( ${+zsh_accented_chars} == 0 )); then
-  # The associative array zsh_accent_chars is indexed by the
-  # accent.  The values are sets of character / Unicode pairs for
-  # the character with the given accent.  The Unicode value is
-  # a hex index with no base discriminator; essentially a UCS-4 index
-  # with the leading zeroes suppressed.
-  typeset -gA zsh_accented_chars
-
-  # grave
-  accent=\!
-  zsh_accented_chars[$accent]="\
-A C0 E C8 I CC O D2 U D9 a E0 e E8 i EC o F2 u F9 N 1F8 n 1F9 \
-"
-  # acute
-  accent=\'
-  zsh_accented_chars[$accent]="\
-A C1 E C9 I CD O D3 U DA Y DD a E1 e E9 i EC o F3 u FA y FD C 106 c 107 \
-L 139 l 13A N 143 n 144 R 154 r 155 S 15A s 15B Z 179 z 17A \
-"
-  # circumflex
-  accent=\>
-  zsh_accented_chars[$accent]="\
-A C2 E CA I CE O D4 U DB a E2 e EA i EE o F4 u FB C 108 c 109 G 11C g 11d \
-H 124 h 125 J 134 j 135 S 15C s 15D W 174 w 175 Y 176 y 177 \
-"
-  # tilde
-  accent=\?
-  zsh_accented_chars[$accent]="\
-A C3 E CB N D1 O D5 a E3 n F1 o F5 I 128 i 129 U 168 u 169 \
-"
-  # macron (d-, D- give eth)
-  accent=-
-  zsh_accented_chars[$accent]="\
-A 100 a 101 d F0 D D0 E 112 e 113 I 12a i 12b O 14C o 14D U 16A u 16B \
-"
-  # breve
-  accent=\(
-  zsh_accented_chars[$accent]="\
-A 102 a 103 E 114 e 115 G 11E g 11F I 12C i 12D O 14E o 14F U 16C u 16D \
-"
-  # dot above, small i with no dot, or l with middle dot
-  accent=.
-  zsh_accented_chars[$accent]="\
-C 10A c 10b E 116 e 117 G 120 g 121 I 130 i 131 L 13F l 140 Z 17B z 17C \
-"
-  # diaeresis / Umlaut
-  accent=:
-  zsh_accented_chars[$accent]="\
-A C4 I CF O D6 U DC a E4 e EB i EF o F6 u FC y FF Y 178 \
-"
-  # cedilla
-  accent=,
-  zsh_accented_chars[$accent]="\
-C C7 c E7 G 122 g 123 K 136 k 137 L 13B l 13C N 145 n 146 R 156 r 157 \
-S 15E s 15F T 162 t 163 \
-"
-  # underline (_) would go here
-  # stroke through
-  accent=/
-  zsh_accented_chars[$accent]="\
-O D8 o F8 D 110 d 111 H 126 h 127 L 141 l 142 T 166 t 167 b 180 \
-"
-  # double acute
-  accent=\"
-  zsh_accented_chars[$accent]="\
-O 150 o 151 U 170 u 171\
-"
-  # ogonek
-  accent=\;
-  zsh_accented_chars[$accent]="\
-A 104 a 105 E 118 e 119 I 12E i 12F U 172 u 173 \
-"
-  # caron
-  accent=\<
-  zsh_accented_chars[$accent]="\
-C 10C c 10D D 10E d 10F E 11A e 11B L 13D l 13E N 147 n 148 R 158 r 159 \
-S 160 s 161 T 164 t 165 Z 17D z 17E \
-"
-  # ring above
-  accent=0
-  zsh_accented_chars[$accent]="\
-A C5 a E5 U 16E u 16F \
-"
-  # hook above
-  accent=2
-  zsh_accented_chars[$accent]="\
-A 1EA2 a 1EA3 E 1EBA e 1EBA \
-"
-  # horn, also right quotation marks
-  accent=9
-  zsh_accented_chars[$accent]="\
-O 1A0 o 1A1 U 1Af u 1b0 ' 2019 . 201A \" 201D : 201E \
-"
-  # left quotation marks
-  accent=6
-  zsh_accented_chars[$accent]="\
-' 2018 \" 201C \
-"
-  # reversed quotation marks for convenience
-  accent=\'
-  zsh_accented_chars[$accent]+=" \
-9 201B \
-"
-  accent=\"
-  zsh_accented_chars[$accent]+=" \
-9 201F \
-"
-
-  # ligature with E
-  accent=e
-  zsh_accented_chars[$accent]="\
-A C6 O 152 \
-"
-  # ligature with e
-  accent=e
-  zsh_accented_chars[$accent]="\
-a E6 o 153 \
-"
-  # ligature with J
-  accent=J
-  zsh_accented_chars[$accent]="\
-I 132 \
-"
-  # ligature with j
-  accent=j
-  zsh_accented_chars[$accent]="\
-i 133 \
-"
-  # eszett
-  accent=s
-  zsh_accented_chars[$accent]="\
-s DF \
-"
-  # upper case thorn
-  accent=H
-  zsh_accented_chars[$accent]="\
-T DE \
-"
-  # lower case thorn
-  accent=h
-  zsh_accented_chars[$accent]="\
-t FE \
-"
-
-  # Remaining characters are handled as separate pairs.
-  # We need to remember that the assoc array is keyed by the second character.
-  # Left square bracket
-  accent=\(
-  zsh_accented_chars[$accent]+=" < 5B"
-  # Reverse solidus (backslash to you and me).
-  accent=/
-  zsh_accented_chars[$accent]+=" / 5C"
-  # Right square bracket, circumflex
-  accent=\>
-  zsh_accented_chars[$accent]+=" ) 5D ' 5E"
-  # Grave accent
-  accent=\!
-  zsh_accented_chars[$accent]+=" ' 60"
-  # diglyphys for (usually) standard characters {, |, }, ~
-  accent=\!
-  zsh_accented_chars[$accent]+=" ( 7B"
-  zsh_accented_chars[$accent]+=" ! 7C"
-  accent=\)
-  zsh_accented_chars[$accent]+=" ! 7D"
-  accent=\?
-  zsh_accented_chars[$accent]+=" ' 7E"
-  # non-breaking space
-  zsh_accented_chars[S]+=" N A0"
-  # inverted exclamation mark
-  zsh_accented_chars[I]+=" ! A1"
-  # cent
-  zsh_accented_chars[t]+=" C A2"
-  # pound sterling
-  zsh_accented_chars[d]+=" P A3"
-  # currency
-  zsh_accented_chars[u]+=" C A4"
-  # yen
-  zsh_accented_chars[e]+=" Y A5"
-  # broken bar
-  zsh_accented_chars[B]+=" B A6"
-  # section
-  zsh_accented_chars[E]+=" S A7"
-  # lonely diaeresis
-  zsh_accented_chars[:]+=" ' A8"
-  # copyright
-  zsh_accented_chars[o]+=" C A9"
-  # spanish feminine ordinal
-  zsh_accented_chars[a]+=" - AA"
-  # left guillemet
-  accent=\<
-  zsh_accented_chars[$accent]+=" < AB"
-  zsh_accented_chars[O]+=" N AC"
-  # soft hyphen
-  zsh_accented_chars[-]+=" - AD"
-  # registered
-  zsh_accented_chars[g]+=" R AE"
-  # lonely macron
-  zsh_accented_chars[m]+=" ' AF"
-  # degree
-  zsh_accented_chars[G]+=" D B0"
-  # +/-
-  zsh_accented_chars[-]+=" + B1"
-  # superscripts
-  zsh_accented_chars[S]+=" 2 B2 3 B3"
-  # lonely acute
-  accent=\'
-  zsh_accented_chars[$accent]+=" ' B4"
-  # micro
-  zsh_accented_chars[y]+=" M B5"
-  # pilcrow (paragraph)
-  zsh_accented_chars[I]+=" P B6"
-  # Middle dot
-  zsh_accented_chars[M]+=" . B7"
-  # Lonely cedilla
-  zsh_accented_chars[,]+=" ' B8"
-  # Superscript one
-  zsh_accented_chars[S]+=" 1 B9"
-  # spanish masculine ordinal
-  zsh_accented_chars[o]+=" - BA"
-  # right guillemet
-  accent=\>
-  zsh_accented_chars[$accent]+=" > BB"
-  # fractions
-  zsh_accented_chars[4]+=" 1 BC 3 BE"
-  zsh_accented_chars[2]+=" 1 BD"
-  # inverted question mark
-  zsh_accented_chars[I]+=" ? BF"
-  # multiplication
-  zsh_accented_chars[X]+=" * D7"
-  # division
-  zsh_accented_chars[:]+=" - F7"
-  # kra
-  zsh_accented_chars[k]+=" k 138"
-  # apostrophe n
-  zsh_accented_chars[n]+=" ' 149"
-  # Lappish ng
-  zsh_accented_chars[G]+=" N 14A"
-  zsh_accented_chars[g]+=" n 14B"
-  # OI
-  zsh_accented_chars[I]+=" O 1A2"
-  zsh_accented_chars[i]+=" o 1A3"
-  # yr
-  zsh_accented_chars[r]+=" y 1A6"
-  # ezh
-  zsh_accented_chars[D]+=" E 1B7"
-  # euro (I invented this but it's logical)
-  zsh_accented_chars[u]+=" E 20AC"
-  # dagger and double dagger
-  zsh_accented_chars[-]+=" / 2020"
-  zsh_accented_chars[=]+=" / 2021"
+  # Save quite a lot of memory by running and then erasing
+  # the function that defines the characters.
+  autoload -U define-composed-chars
+  define-composed-chars
+  unfunction define-composed-chars
 fi
 
 read -k basechar || return 1