# HG changeset patch # User Bram Moolenaar # Date 1264613366 -3600 # Node ID 1378bc45ebe5754c0d3ec64d37810eaad3a544f3 # Parent dae4cd29a0b7d899201c2e6a10f0441e7173bcfd updated for version 7.2.348 Problem: Unicode double-width characters are not up-to date. Solution: Produce the double-width table like the others. diff --git a/runtime/tools/unicode.vim b/runtime/tools/unicode.vim --- a/runtime/tools/unicode.vim +++ b/runtime/tools/unicode.vim @@ -187,16 +187,27 @@ func! BuildCombiningTable() wincmd p endfunc -" Build the ambiguous table in a new buffer. +" Build the double width or ambiguous width table in a new buffer. " Uses s:widthprops and s:dataprops. -func! BuildAmbiguousTable() +func! BuildWidthTable(pattern, tableName) let start = -1 let end = -1 let ranges = [] let dataidx = 0 for p in s:widthprops - if p[1][0] == 'A' - let n = ('0x' . p[0]) + 0 + if p[1][0] =~ a:pattern + if p[0] =~ '\.\.' + " It is a range. we don't check for composing char then. + let rng = split(p[0], '\.\.') + if len(rng) != 2 + echoerr "Cannot parse range: '" . p[0] . "' in width table" + endif + let n = ('0x' . rng[0]) + 0 + let n_last = ('0x' . rng[1]) + 0 + else + let n = ('0x' . p[0]) + 0 + let n_last = n + endif " Find this char in the data table. while 1 let dn = ('0x' . s:dataprops[dataidx][0]) + 0 @@ -205,27 +216,23 @@ func! BuildAmbiguousTable() endif let dataidx += 1 endwhile - if dn != n + if dn != n && n_last == n echoerr "Cannot find character " . n . " in data table" endif " Only use the char when it's not a composing char. + " But use all chars from a range. let dp = s:dataprops[dataidx] - if dp[2] != 'Mn' && dp[2] != 'Mc' && dp[2] != 'Me' + if n_last > n || (dp[2] != 'Mn' && dp[2] != 'Mc' && dp[2] != 'Me') if start >= 0 && end + 1 == n " continue with same range. - let end = n else if start >= 0 " produce previous range call add(ranges, printf("\t{0x%04x, 0x%04x},", start, end)) endif let start = n - if p[0] =~ '\.\.' - let end = ('0x' . substitute(p[0], '.*\.\.', '', '')) + 0 - else - let end = n - endif endif + let end = n_last endif endif endfor @@ -235,8 +242,8 @@ func! BuildAmbiguousTable() " New buffer to put the result in. new - file ambiguous - call setline(1, " static struct interval ambiguous[] =") + 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 @@ -276,5 +283,8 @@ edit http://www.unicode.org/Public/UNIDA " Parse each line, create a list of lists. call ParseWidthProps() -" Build the ambiguous table. -call BuildAmbiguousTable() +" Build the double width table. +call BuildWidthTable('[WF]', 'doublewidth') + +" Build the ambiguous width table. +call BuildWidthTable('A', 'ambiguous') diff --git a/src/mbyte.c b/src/mbyte.c --- a/src/mbyte.c +++ b/src/mbyte.c @@ -1200,6 +1200,49 @@ intable(table, size, c) utf_char2cells(c) int c; { + /* Sorted list of non-overlapping intervals of East Asian double width + * characters, generated with ../runtime/tools/unicode.vim. */ + static struct interval doublewidth[] = + { + {0x1100, 0x115f}, + {0x11a3, 0x11a7}, + {0x11fa, 0x11ff}, + {0x2329, 0x232a}, + {0x2e80, 0x2e99}, + {0x2e9b, 0x2ef3}, + {0x2f00, 0x2fd5}, + {0x2ff0, 0x2ffb}, + {0x3000, 0x3029}, + {0x3030, 0x303e}, + {0x3041, 0x3096}, + {0x309b, 0x30ff}, + {0x3105, 0x312d}, + {0x3131, 0x318e}, + {0x3190, 0x31b7}, + {0x31c0, 0x31e3}, + {0x31f0, 0x321e}, + {0x3220, 0x3247}, + {0x3250, 0x32fe}, + {0x3300, 0x4dbf}, + {0x4e00, 0xa48c}, + {0xa490, 0xa4c6}, + {0xa960, 0xa97c}, + {0xac00, 0xd7a3}, + {0xd7b0, 0xd7c6}, + {0xd7cb, 0xd7fb}, + {0xf900, 0xfaff}, + {0xfe10, 0xfe19}, + {0xfe30, 0xfe52}, + {0xfe54, 0xfe66}, + {0xfe68, 0xfe6b}, + {0xff01, 0xff60}, + {0xffe0, 0xffe6}, + {0x1f200, 0x1f200}, + {0x1f210, 0x1f231}, + {0x1f240, 0x1f248}, + {0x20000, 0x2fffd}, + {0x30000, 0x3fffd} + }; /* Sorted list of non-overlapping intervals of East Asian Ambiguous * characters, generated with ../runtime/tools/unicode.vim. */ static struct interval ambiguous[] = @@ -1403,20 +1446,7 @@ utf_char2cells(c) #else if (!utf_printable(c)) return 6; /* unprintable, displays */ - if (c >= 0x1100 - && (c <= 0x115f /* Hangul Jamo */ - || c == 0x2329 - || c == 0x232a - || (c >= 0x2e80 && c <= 0xa4cf - && c != 0x303f) /* CJK ... Yi */ - || (c >= 0xac00 && c <= 0xd7a3) /* Hangul Syllables */ - || (c >= 0xf900 && c <= 0xfaff) /* CJK Compatibility - Ideographs */ - || (c >= 0xfe30 && c <= 0xfe6f) /* CJK Compatibility Forms */ - || (c >= 0xff00 && c <= 0xff60) /* Fullwidth Forms */ - || (c >= 0xffe0 && c <= 0xffe6) - || (c >= 0x20000 && c <= 0x2fffd) - || (c >= 0x30000 && c <= 0x3fffd))) + if (intable(doublewidth, sizeof(doublewidth), c)) return 2; #endif } diff --git a/src/version.c b/src/version.c --- a/src/version.c +++ b/src/version.c @@ -682,6 +682,8 @@ static char *(features[]) = static int included_patches[] = { /* Add new patch number below this line */ /**/ + 348, +/**/ 347, /**/ 346,