summaryrefslogtreecommitdiff
path: root/tex/context/base/l-unicode.lua
blob: f4480e93c40698876d340da36f80bb9061afab20 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
if not modules then modules = { } end modules ['l-unicode'] = {
    version   = 1.001,
    comment   = "companion to luat-lib.mkiv",
    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
    copyright = "PRAGMA ADE / ConTeXt Development Team",
    license   = "see context related readme files"
}

if not unicode then

    unicode = { utf8 = { } }

    local floor, char = math.floor, string.char

    function unicode.utf8.utfchar(n)
        if n < 0x80 then
            return char(n)
        elseif n < 0x800 then
            return char(
                0xC0 + floor(n/0x40),
                0x80 + (n % 0x40)
            )
        elseif n < 0x10000 then
            return char(
                0xE0 + floor(n/0x1000),
                0x80 + (floor(n/0x40) % 0x40),
                0x80 + (n % 0x40)
            )
        elseif n < 0x40000 then
            return char(
                0xF0 + floor(n/0x40000),
                0x80 + floor(n/0x1000),
                0x80 + (floor(n/0x40) % 0x40),
                0x80 + (n % 0x40)
            )
        else
         -- return char(
         --     0xF1 + floor(n/0x1000000),
         --     0x80 + floor(n/0x40000),
         --     0x80 + floor(n/0x1000),
         --     0x80 + (floor(n/0x40) % 0x40),
         --     0x80 + (n % 0x40)
         -- )
            return "?"
        end
    end

end

local unicode = unicode

utf = utf or unicode.utf8

local concat = table.concat
local utfchar, utfbyte, utfgsub = utf.char, utf.byte, utf.gsub
local char, byte, find, bytepairs, utfvalues, format = string.char, string.byte, string.find, string.bytepairs, string.utfvalues, string.format
local type = type

local utfsplitlines = string.utfsplitlines

-- 0  EF BB BF      UTF-8
-- 1  FF FE         UTF-16-little-endian
-- 2  FE FF         UTF-16-big-endian
-- 3  FF FE 00 00   UTF-32-little-endian
-- 4  00 00 FE FF   UTF-32-big-endian

unicode.utfname = {
    [0] = 'utf-8',
    [1] = 'utf-16-le',
    [2] = 'utf-16-be',
    [3] = 'utf-32-le',
    [4] = 'utf-32-be'
}

-- \000 fails in <= 5.0 but is valid in >=5.1 where %z is depricated

function unicode.utftype(f)
    local str = f:read(4)
    if not str then
        f:seek('set')
        return 0
 -- elseif find(str,"^%z%z\254\255") then            -- depricated
 -- elseif find(str,"^\000\000\254\255") then        -- not permitted and bugged
    elseif find(str,"\000\000\254\255",1,true) then  -- seems to work okay (TH)
        return 4
 -- elseif find(str,"^\255\254%z%z") then            -- depricated
 -- elseif find(str,"^\255\254\000\000") then        -- not permitted and bugged
    elseif find(str,"\255\254\000\000",1,true) then  -- seems to work okay (TH)
        return 3
    elseif find(str,"^\254\255") then
        f:seek('set',2)
        return 2
    elseif find(str,"^\255\254") then
        f:seek('set',2)
        return 1
    elseif find(str,"^\239\187\191") then
        f:seek('set',3)
        return 0
    else
        f:seek('set')
        return 0
    end
end

--~ function unicode.utf16_to_utf8(str, endian) -- maybe a gsub is faster or an lpeg
--~     local result, tmp, n, m, p, r, t = { }, { }, 0, 0, 0, 0, 0 -- we reuse tmp
--~     -- lf | cr | crlf / (cr:13, lf:10)
--~     local function doit() -- inline this
--~         if n == 10 then
--~             if p ~= 13 then
--~                 if t > 0 then
--~                     r = r + 1
--~                     result[r] = concat(tmp,"",1,t)
--~                     t = 0
--~                 end
--~                 p = 0
--~             end
--~         elseif n == 13 then
--~             if t > 0 then
--~                 r = r + 1
--~                 result[r] = concat(tmp,"",1,t)
--~                 t = 0
--~             end
--~             p = n
--~         else
--~             t = t + 1
--~             tmp[t] = utfchar(n)
--~             p = 0
--~         end
--~     end
--~     for l,r in bytepairs(str) do
--~         if r then
--~             if endian then -- maybe make two loops
--~                 n = 256*l + r
--~             else
--~                 n = 256*r + l
--~             end
--~             if m > 0 then
--~                 n = (m-0xD800)*0x400 + (n-0xDC00) + 0x10000
--~                 m = 0
--~                 doit()
--~             elseif n >= 0xD800 and n <= 0xDBFF then
--~                 m = n
--~             else
--~                 doit()
--~             end
--~         end
--~     end
--~     if t > 0 then
--~         r = r + 1
--~         result[r] = concat(tmp,"",1,t) -- we reused tmp, hence t
--~     end
--~     return result
--~ end

--~ function unicode.utf32_to_utf8(str, endian)
--~     local result, tmp, n, m, p, r, t = { }, { }, 0, -1, 0, 0, 0
--~     -- lf | cr | crlf / (cr:13, lf:10)
--~     local function doit() -- inline this
--~         if n == 10 then
--~             if p ~= 13 then
--~                 if t > 0 then
--~                     r = r + 1
--~                     result[r] = concat(tmp,"",1,t)
--~                     t = 0
--~                 end
--~                 p = 0
--~             end
--~         elseif n == 13 then
--~             if t > 0 then
--~                 r = r + 1
--~                 result[r] = concat(tmp,"",1,t)
--~                 t = 0
--~             end
--~             p = n
--~         else
--~             t = t + 1
--~             tmp[t] = utfchar(n)
--~             p = 0
--~         end
--~     end
--~     for a,b in bytepairs(str) do
--~         if a and b then
--~             if m < 0 then
--~                 if endian then -- maybe make two loops
--~                     m = 256*256*256*a + 256*256*b
--~                 else
--~                     m = 256*b + a
--~                 end
--~             else
--~                 if endian then -- maybe make two loops
--~                     n = m + 256*a + b
--~                 else
--~                     n = m + 256*256*256*b + 256*256*a
--~                 end
--~                 m = -1
--~                 doit()
--~             end
--~         else
--~             break
--~         end
--~     end
--~     if #tmp > 0 then
--~         r = r + 1
--~         result[r] = concat(tmp,"",1,t) -- we reused tmp, hence t
--~     end
--~     return result
--~ end

local function utf16_to_utf8_be(t)
    if type(t) == "string" then
        t = utfsplitlines(str)
    end
    local result = { } -- we reuse result
    for i=1,#t do
        local r, more = 0, 0
        for left, right in bytepairs(t[i]) do
            if right then
                local now = 256*left + right
                if more > 0 then
                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                    more = 0
                    r = r + 1
                    result[r] = utfchar(now)
                elseif now >= 0xD800 and now <= 0xDBFF then
                    more = now
                else
                    r = r + 1
                    result[r] = utfchar(now)
                end
            end
        end
        t[i] = concat(result,"",1,r) -- we reused tmp, hence t
    end
    return t
end

local function utf16_to_utf8_le(t)
    if type(t) == "string" then
        t = utfsplitlines(str)
    end
    local result = { } -- we reuse result
    for i=1,#t do
        local r, more = 0, 0
        for left, right in bytepairs(t[i]) do
            if right then
                local now = 256*right + left
                if more > 0 then
                    now = (more-0xD800)*0x400 + (now-0xDC00) + 0x10000 -- the 0x10000 smells wrong
                    more = 0
                    r = r + 1
                    result[r] = utfchar(now)
                elseif now >= 0xD800 and now <= 0xDBFF then
                    more = now
                else
                    r = r + 1
                    result[r] = utfchar(now)
                end
            end
        end
        t[i] = concat(result,"",1,r) -- we reused tmp, hence t
    end
    return t
end

local function utf32_to_utf8_be(t)
    if type(t) == "string" then
        t = utfsplitlines(t)
    end
    local result = { } -- we reuse result
    for i=1,#t do
        local r, more = 0, -1
        for a,b in bytepairs(t[i]) do
            if a and b then
                if more < 0 then
                    more = 256*256*256*a + 256*256*b
                else
                    r = r + 1
                    result[t] = utfchar(more + 256*a + b)
                    more = -1
                end
            else
                break
            end
        end
        t[i] = concat(result,"",1,r)
    end
    return t
end

local function utf32_to_utf8_le(t)
    if type(t) == "string" then
        t = utfsplitlines(t)
    end
    local result = { } -- we reuse result
    for i=1,#t do
        local r, more = 0, -1
        for a,b in bytepairs(t[i]) do
            if a and b then
                if more < 0 then
                    more = 256*b + a
                else
                    r = r + 1
                    result[t] = utfchar(more + 256*256*256*b + 256*256*a)
                    more = -1
                end
            else
                break
            end
        end
        t[i] = concat(result,"",1,r)
    end
    return t
end

unicode.utf32_to_utf8_be = utf32_to_utf8_be
unicode.utf32_to_utf8_le = utf32_to_utf8_le
unicode.utf16_to_utf8_be = utf16_to_utf8_be
unicode.utf16_to_utf8_le = utf16_to_utf8_le

function unicode.utf8_to_utf8(t)
    return type(t) == "string" and utfsplitlines(t) or t
end

function unicode.utf16_to_utf8(t,endian)
    return endian and utf16_to_utf8_be(t) or utf16_to_utf8_le(t) or t
end

function unicode.utf32_to_utf8(t,endian)
    return endian and utf32_to_utf8_be(t) or utf32_to_utf8_le(t) or t
end

local function little(c)
    local b = byte(c)
    if b < 0x10000 then
        return char(b%256,b/256)
    else
        b = b - 0x10000
        local b1, b2 = b/1024 + 0xD800, b%1024 + 0xDC00
        return char(b1%256,b1/256,b2%256,b2/256)
    end
end

local function big(c)
    local b = byte(c)
    if b < 0x10000 then
        return char(b/256,b%256)
    else
        b = b - 0x10000
        local b1, b2 = b/1024 + 0xD800, b%1024 + 0xDC00
        return char(b1/256,b1%256,b2/256,b2%256)
    end
end

function unicode.utf8_to_utf16(str,littleendian)
    if littleendian then
        return char(255,254) .. utfgsub(str,".",little)
    else
        return char(254,255) .. utfgsub(str,".",big)
    end
end

function unicode.utfcodes(str)
    local t, n = { }, 0
    for u in utfvalues(str) do
        n = n + 1
        t[n] = format("0x%04X",u)
    end
    return concat(t,separator or " ")
end

function unicode.ustring(s)
    return format("U+%05X",type(s) == "number" and s or utfbyte(s))
end

function unicode.xstring(s)
    return format("0x%05X",type(s) == "number" and s or utfbyte(s))
end

--~ print(unicode.utfcodes(str))

local lpegmatch = lpeg.match
local patterns = lpeg.patterns
local utftype = patterns.utftype

function unicode.filetype(data)
    return data and lpegmatch(utftype,data) or "unknown"
end

local toentities = lpeg.Cs (
    (
        patterns.utf8one
            + (
                patterns.utf8two
              + patterns.utf8three
              + patterns.utf8four
            ) / function(s) local b = utfbyte(s) if b < 127 then return s else return format("&#%X;",b) end end
    )^0
)

patterns.toentities = toentities

function utf.toentities(str)
    return lpegmatch(toentities,str)
end

--~ local utfchr = { } -- 60K -> 2.638 M extra mem but currently not called that often (on latin)
--~
--~ setmetatable(utfchr, { __index = function(t,k) local v = utfchar(k) t[k] = v return v end } )
--~
--~ collectgarbage("collect")
--~ local u = collectgarbage("count")*1024
--~ local t = os.clock()
--~ for i=1,1000 do
--~     for i=1,600 do
--~         local a = utfchr[i]
--~     end
--~ end
--~ print(os.clock()-t,collectgarbage("count")*1024-u)

--~ collectgarbage("collect")
--~ local t = os.clock()
--~ for i=1,1000 do
--~     for i=1,600 do
--~         local a = utfchar(i)
--~     end
--~ end
--~ print(os.clock()-t,collectgarbage("count")*1024-u)

--~ local byte = string.byte
--~ local utfchar = utf.char
--~ local lpegmatch = lpeg.match, lpeg.P, lpeg.C, lpeg.R, lpeg.Cs

local P, C, R, Cs = lpeg.P, lpeg.C, lpeg.R, lpeg.Cs

local one  = P(1)
local two  = C(1) * C(1)
local four = C(R(utfchar(0xD8),utfchar(0xFF))) * C(1) * C(1) * C(1)

-- actually one of them is already utf ... sort of useless this one

local pattern = P("\254\255") * Cs( (
                    four  / function(a,b,c,d)
                                local ab = 0xFF * byte(a) + byte(b)
                                local cd = 0xFF * byte(c) + byte(d)
                                return utfchar((ab-0xD800)*0x400 + (cd-0xDC00) + 0x10000)
                            end
                  + two   / function(a,b)
                                return utfchar(byte(a)*256 + byte(b))
                            end
                  + one
                )^1 )
              + P("\255\254") * Cs( (
                    four  / function(b,a,d,c)
                                local ab = 0xFF * byte(a) + byte(b)
                                local cd = 0xFF * byte(c) + byte(d)
                                return utfchar((ab-0xD800)*0x400 + (cd-0xDC00) + 0x10000)
                            end
                  + two   / function(b,a)
                                return utfchar(byte(a)*256 + byte(b))
                            end
                  + one
                )^1 )

function string.toutf(s)
    return lpegmatch(pattern,s) or s -- todo: utf32
end

local validatedutf = Cs (
    (
        patterns.utf8one
      + patterns.utf8two
      + patterns.utf8three
      + patterns.utf8four
      + P(1) / "�"
    )^0
)

patterns.validatedutf = validatedutf

function string.validutf(str)
    return lpegmatch(validatedutf,str)
end