patch 7.4.1604v7.4.1604

Problem:    Although emoji characters are ambiguous width, best is to treat
            them as full width.
Solution:   Update the Unicode character tables. Add the 'emoji' options.
            (Yasuhiro Matsumoto)

3 files changed, 51 insertions, 13 deletions

diff --git a/runtime/doc/options.txt b/runtime/doc/options.txt
index 024e621e5..35cef9de6 100644
--- a/runtime/doc/options.txt
+++ b/runtime/doc/options.txt
@@ -1,4 +1,4 @@
-*options.txt*	For Vim version 7.4.  Last change: 2016 Mar 08
+*options.txt*	For Vim version 7.4.  Last change: 2016 Mar 19
 
 
 		  VIM REFERENCE MANUAL	  by Bram Moolenaar
@@ -69,7 +69,7 @@ achieve special effects.  These options come in three forms:
 :se[t] {option}:{value}
 			Set string or number option to {value}.
 			For numeric options the value can be given in decimal,
- 			hex (preceded with 0x) or octal (preceded with '0').
+			hex (preceded with 0x) or octal (preceded with '0').
 			The old value can be inserted by typing 'wildchar' (by
 			default this is a <Tab> or CTRL-E if 'compatible' is
 			set).  See |cmdline-completion|.
@@ -1002,7 +1002,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	- The backup file will be created in the first directory in the list
 	  where this is possible.  The directory must exist, Vim will not
 	  create it for you.
-	- Empty means that no backup file will be created ( 'patchmode' is
+	- Empty means that no backup file will be created ('patchmode' is
 	  impossible!).  Writing may fail because of this.
 	- A directory "." means to put the backup file in the same directory
 	  as the edited file.
@@ -1792,7 +1792,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	when CTRL-P or CTRL-N are used.  It is also used for whole-line
 	completion |i_CTRL-X_CTRL-L|.  It indicates the type of completion
 	and the places to scan.  It is a comma separated list of flags:
-	.	scan the current buffer ( 'wrapscan' is ignored)
+	.	scan the current buffer ('wrapscan' is ignored)
 	w	scan buffers from other windows
 	b	scan other loaded buffers that are in the buffer list
 	u	scan the unloaded buffers that are in the buffer list
@@ -2644,6 +2644,15 @@ A jump table for the options with a short description can be found at |Q_op|.
 	also 'gdefault' option.
 	Switching this option on is discouraged!
 
+						*'emoji'* *'emo'*
+'emoji' 'emo'	boolean (default: on)
+			global
+			{not in Vi}
+			{only available when compiled with the |+multi_byte|
+			feature}
+	When on all Unicode emoji characters are considered to be full width.
+
+
 					*'encoding'* *'enc'* *E543*
 'encoding' 'enc'	string (default: "latin1" or value from $LANG)
 			global
@@ -3732,7 +3741,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 		The same applies to the modeless selection.
 								*'go-P'*
 	  'P'	Like autoselect but using the "+ register instead of the "*
-	  	register.
+		register.
 								*'go-A'*
 	  'A'	Autoselect for the modeless selection.  Like 'a', but only
 		applies to the modeless selection.
@@ -4841,7 +4850,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	reset this option. |-u| |--noplugin|
 
 						*'luadll'*
-'luadll' 		string	(default depends on the build)
+'luadll'		string	(default depends on the build)
 			global
 			{not in Vi}
 			{only available when compiled with the |+lua/dyn|
@@ -5298,7 +5307,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	respectively; see |CTRL-A| for more info on these commands.
 	alpha	If included, single alphabetical characters will be
 		incremented or decremented.  This is useful for a list with a
-		letter index a), b), etc.	  	*octal-nrformats*
+		letter index a), b), etc.		*octal-nrformats*
 	octal	If included, numbers that start with a zero will be considered
 		to be octal.  Example: Using CTRL-A on "007" results in "010".
 	hex	If included, numbers starting with "0x" or "0X" will be
@@ -5328,7 +5337,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	relative to the cursor.  Together with 'number' there are these
 	four combinations (cursor in line 3):
 
-              	'nonu'          'nu'            'nonu'          'nu'
+		'nonu'          'nu'            'nonu'          'nu'
 		'nornu'         'nornu'         'rnu'           'rnu'
 
 	    |apple          |  1 apple      |  2 apple      |  2 apple
@@ -5567,7 +5576,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	this doesn't work when $INCL contains a comma or white space.
 
 						*'perldll'*
-'perldll' 		string	(default depends on the build)
+'perldll'		string	(default depends on the build)
 			global
 			{not in Vi}
 			{only available when compiled with the |+perl/dyn|
@@ -5704,7 +5713,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	|ins-completion-menu|.
 
 						*'pythondll'*
-'pythondll' 		string	(default depends on the build)
+'pythondll'		string	(default depends on the build)
 			global
 			{not in Vi}
 			{only available when compiled with the |+python/dyn|
@@ -5945,7 +5954,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	The 'rightleft' option must be set for 'rightleftcmd' to take effect.
 
 						*'rubydll'*
-'rubydll' 		string	(default: depends on the build)
+'rubydll'		string	(default: depends on the build)
 			global
 			{not in Vi}
 			{only available when compiled with the |+ruby/dyn|
@@ -7439,7 +7448,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 	mapping which should not change the tagstack.
 
 						*'tcldll'*
-'tcldll' 		string	(default depends on the build)
+'tcldll'		string	(default depends on the build)
 			global
 			{not in Vi}
 			{only available when compiled with the |+tcl/dyn|
@@ -7494,7 +7503,7 @@ A jump table for the options with a short description can be found at |Q_op|.
 			{not in Vi}
 	Encoding used for the terminal.  This specifies what character
 	encoding the keyboard produces and the display will understand.  For
-	the GUI it only applies to the keyboard ( 'encoding' is used for the
+	the GUI it only applies to the keyboard ('encoding' is used for the
 	display).  Except for the Mac when 'macatsui' is off, then
 	'termencoding' should be "macroman".
 								*E617*
diff --git a/runtime/optwin.vim b/runtime/optwin.vim
index 46a5ee034..c44d73a34 100644
--- a/runtime/optwin.vim
+++ b/runtime/optwin.vim
@@ -1254,6 +1254,8 @@ if has("multi_byte")
   endif
   call append("$", "ambiwidth\twidth of ambiguous width characters")
   call <SID>OptionG("ambw", &ambw)
+  call append("$", "emoji\temoji characters are full width")
+  call <SID>BinOptionG("emo", &emo)
 endif
 
 
diff --git a/runtime/tools/unicode.vim b/runtime/tools/unicode.vim
index d733fe867..dfe9cef41 100644
--- a/runtime/tools/unicode.vim
+++ b/runtime/tools/unicode.vim
@@ -251,6 +251,27 @@ func! BuildWidthTable(pattern, tableName)
   wincmd p
 endfunc
 
+" Build the amoji width table in a new buffer.
+func! BuildEmojiTable(pattern, tableName)
+  let ranges = []
+  for line in map(filter(filter(getline(1, '$'), 'v:val=~"^[1-9]"'), 'v:val=~a:pattern'), 'matchstr(v:val,"^\\S\\+")')
+    let token = split(line, '\.\.')
+    if len(token) == 1
+      call add(token, token[0])
+    endif
+    call add(ranges, printf("\t{0x%04x, 0x%04x},", "0x".token[0], "0x".token[1]))
+  endfor
+
+  " New buffer to put the result in.
+  new
+  exe "file " . a:tableName
+  call setline(1, "    static struct interval " . a:tableName . "[] =")
+  call setline(2, "    {")
+  call append('$', ranges)
+  call setline('$', getline('$')[:-2])  " remove last comma
+  call setline(line('$') + 1, "    };")
+  wincmd p
+endfunc
 
 " Try to avoid hitting E36
 set equalalways
@@ -290,3 +311,9 @@ call BuildWidthTable('[WF]', 'doublewidth')
 
 " Build the ambiguous width table.
 call BuildWidthTable('A', 'ambiguous')
+
+" Edit the emoji text file.  Requires the netrw plugin.
+edit http://www.unicode.org/Public/emoji/3.0/emoji-data.txt
+
+" Build the emoji table. Ver. 1.0 - 6.0
+call BuildEmojiTable('; Emoji\s\+# [1-6]\.[0-9]', 'emoji')