summaryrefslogtreecommitdiff
path: root/context/data/scite/lexers/scite-context-lexer.lua
blob: 8d06b4923a1b1616c072d20f861a03d4808fbb10 (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
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
local info = {
    version   = 1.324,
    comment   = "basics for scintilla lpeg lexer for context/metafun",
    author    = "Hans Hagen, PRAGMA-ADE, Hasselt NL",
    copyright = "PRAGMA ADE / ConTeXt Development Team",
    license   = "see context related readme files",
    comment   = "contains copyrighted code from mitchell.att.foicica.com",

}

-- todo: move all code here
-- todo: explore adapted dll ... properties + init

-- The fold and lex functions are copied and patched from original code by Mitchell (see
-- lexer.lua). All errors are mine.
--
-- Starting with SciTE version 3.20 there is an issue with coloring. As we still lack
-- a connection with scite itself (properties as well as printing to the log pane) we
-- cannot trace this (on windows). As far as I can see, there are no fundamental
-- changes in lexer.lua or LexLPeg.cxx so it must be in scintilla itself. So for the
-- moment I stick to 3.10. Indicators are: no lexing of 'next' and 'goto <label>' in the
-- Lua lexer and no brace highlighting either. Interesting is that it does work ok in
-- the cld lexer (so the Lua code is okay). Also the fact that char-def.lua lexes fast
-- is a signal that the lexer quits somewhere halfway.
--
-- After checking 3.24 and adapting to the new lexer tables things are okay again. So,
-- this version assumes 3.24 or higher. In 3.24 we have a different token result, i.e. no
-- longer a { tag, pattern } but just two return values. I didn't check other changes but
-- will do that when I run into issues.
--
-- I've considered making a whole copy and patch the other functions too as we need
-- an extra nesting model. However, I don't want to maintain too much. An unfortunate
-- change in 3.03 is that no longer a script can be specified. This means that instead
-- of loading the extensions via the properties file, we now need to load them in our
-- own lexers, unless of course we replace lexer.lua completely (which adds another
-- installation issue).
--
-- Another change has been that _LEXERHOME is no longer available. It looks like more and
-- more functionality gets dropped so maybe at some point we need to ship our own dll/so
-- files. For instance, I'd like to have access to the current filename and other scite
-- properties. For instance, we could cache some info with each file, if only we had
-- knowledge of what file we're dealing with.
--
-- For huge files folding can be pretty slow and I do have some large ones that I keep
-- open all the time. Loading is normally no ussue, unless one has remembered the status
-- and the cursor is at the last line of a 200K line file. Optimizing the fold function
-- brought down loading of char-def.lua from 14 sec => 8 sec. Replacing the word_match
-- function and optimizing the lex function gained another 2+ seconds. A 6 second load
-- is quite ok for me. The changed lexer table structure (no subtables) brings loading
-- down to a few seconds.
--
-- When the lexer path is copied to the textadept lexer path, and the theme definition to
-- theme path (as lexer.lua), the lexer works there as well. When I have time and motive
-- I will make a proper setup file to tune the look and feel a bit and associate suffixes
-- with the context lexer. The textadept editor has a nice style tracing option but lacks
-- the tabs for selecting files that scite has. It also has no integrated run that pipes
-- to the log pane (I wonder if it could borrow code from the console2 project). Interesting
-- is that the jit version of textadept crashes on lexing large files (and does not feel
-- faster either).
--
-- Function load(lexer_name) starts with _M.WHITESPACE = lexer_name..'_whitespace' which
-- means that we need to have it frozen at the moment we load another lexer. Because spacing
-- is used to revert to a parent lexer we need to make sure that we load children as late
-- as possible in order not to get the wrong whitespace trigger. This took me quite a while
-- to figure out (not being that familiar with the internals). The lex and fold functions
-- have been optimized. It is a pitty that there is no proper print available. Another thing
-- needed is a default style in ourown theme style definition, as otherwise we get wrong
-- nested lexers, especially if they are larger than a view. This is the hardest part of
-- getting things right.
--
-- Eventually it might be safer to copy the other methods from lexer.lua here as well so
-- that we have no dependencies, apart from the c library (for which at some point the api
-- will be stable I hope).
--
-- It's a pitty that there is no scintillua library for the OSX version of scite. Even
-- better would be to have the scintillua library as integral part of scite as that way I
-- could use OSX alongside windows and linux (depending on needs). Also nice would be to
-- have a proper interface to scite then because currently the lexer is rather isolated and the
-- lua version does not provide all standard libraries. It would also be good to have lpeg
-- support in the regular scite lua extension (currently you need to pick it up from someplace
-- else).

local lpeg = require 'lpeg'

local R, P, S, C, V, Cp, Cs, Ct, Cmt, Cc, Cf, Cg, Carg = lpeg.R, lpeg.P, lpeg.S, lpeg.C, lpeg.V, lpeg.Cp, lpeg.Cs, lpeg.Ct, lpeg.Cmt, lpeg.Cc, lpeg.Cf, lpeg.Cg, lpeg.Carg
local lpegmatch = lpeg.match
local find, gmatch, match, lower, upper, gsub = string.find, string.gmatch, string.match, string.lower, string.upper, string.gsub
local concat = table.concat
local global = _G
local type, next, setmetatable, rawset = type, next, setmetatable, rawset

if lexer then
    -- in recent c++ code the lexername and loading is hard coded
elseif _LEXERHOME then
    dofile(_LEXERHOME .. '/lexer.lua') -- pre 3.03 situation
else
    dofile('lexer.lua') -- whatever
end

local LEXER = lexer

lexer.context    = lexer.context or { }
local context    = lexer.context

context.patterns = context.patterns or { }
local patterns   = context.patterns

lexer._CONTEXTEXTENSIONS = true

local locations = {
 -- lexer.context.path,
   "data", -- optional data directory
   "..",   -- regular scite directory
}

local function collect(name)
--  local definitions = loadfile(name .. ".luc") or loadfile(name .. ".lua")
    local okay, definitions = pcall(function () return require(name) end)
    if okay then
        if type(definitions) == "function" then
            definitions = definitions()
        end
        if type(definitions) == "table" then
            return definitions
        end
    end
end

function context.loaddefinitions(name)
    for i=1,#locations do
        local data = collect(locations[i] .. "/" .. name)
        if data then
            return data
        end
    end
end

-- maybe more efficient:

function context.word_match(words,word_chars,case_insensitive)
    local chars = '%w_' -- maybe just "" when word_chars
    if word_chars then
        chars = '^([' .. chars .. gsub(word_chars,'([%^%]%-])', '%%%1') ..']+)'
    else
        chars = '^([' .. chars ..']+)'
    end
    if case_insensitive then
        local word_list = { }
        for i=1,#words do
            word_list[lower(words[i])] = true
        end
        return P(function(input, index)
            local s, e, word = find(input,chars,index)
            return word and word_list[lower(word)] and e + 1 or nil
        end)
    else
        local word_list = { }
        for i=1,#words do
            word_list[words[i]] = true
        end
        return P(function(input, index)
            local s, e, word = find(input,chars,index)
            return word and word_list[word] and e + 1 or nil
        end)
    end
end

local idtoken = R("az","AZ","\127\255","__")
local digit   = R("09")
local sign    = S("+-")
local period  = P(".")
local space   = S(" \n\r\t\f\v")

patterns.idtoken  = idtoken

patterns.digit    = digit
patterns.sign     = sign
patterns.period   = period

patterns.cardinal = digit^1
patterns.integer  = sign^-1 * digit^1

patterns.real     =
    sign^-1 * (                    -- at most one
        digit^1 * period * digit^0 -- 10.0 10.
      + digit^0 * period * digit^1 -- 0.10 .10
      + digit^1                    -- 10
   )

patterns.restofline = (1-S("\n\r"))^1
patterns.space      = space
patterns.spacing    = space^1
patterns.nospacing  = (1-space)^1
patterns.anything   = P(1)

local endof = S("\n\r\f")

patterns.startofline = P(function(input,index)
    return (index == 1 or lpegmatch(endof,input,index-1)) and index
end)

function context.exact_match(words,word_chars,case_insensitive)
    local characters = concat(words)
    local pattern -- the concat catches _ etc
    if word_chars == true or word_chars == false or word_chars == nil then
        word_chars = ""
    end
    if type(word_chars) == "string" then
        pattern = S(characters) + idtoken
        if case_insensitive then
            pattern = pattern + S(upper(characters)) + S(lower(characters))
        end
        if word_chars ~= "" then
            pattern = pattern + S(word_chars)
        end
    elseif word_chars then
        pattern = word_chars
    end
    if case_insensitive then
        local list = { }
        for i=1,#words do
            list[lower(words[i])] = true
        end
        return Cmt(pattern^1, function(_,i,s)
            return list[lower(s)] -- and i or nil
        end)
    else
        local list = { }
        for i=1,#words do
            list[words[i]] = true
        end
        return Cmt(pattern^1, function(_,i,s)
            return list[s] -- and i or nil
        end)
    end
end

-- spell checking (we can only load lua files)
--
-- return {
--     min = 3,
--     max = 40,
--     n = 12345,
--     words = {
--         ["someword"]    = "someword",
--         ["anotherword"] = "Anotherword",
--     },
-- }

local lists = { }

function context.setwordlist(tag,limit) -- returns hash (lowercase keys and original values)
    if not tag or tag == "" then
        return false, 3
    end
    local list = lists[tag]
    if not list then
        list = context.loaddefinitions("spell-" .. tag)
        if not list or type(list) ~= "table" then
            list = { words = false, min = 3 }
        else
            list.words = list.words or false
            list.min   = list.min or 3
        end
        lists[tag] = list
    end
    return list.words, list.min
end

patterns.wordtoken   = R("az","AZ","\127\255")
patterns.wordpattern = patterns.wordtoken^3 -- todo: if limit and #s < limit then

-- -- pre 3.24:
--
-- function context.checkedword(validwords,validminimum,s,i) -- ,limit
--     if not validwords then -- or #s < validminimum then
--         return true, { "text", i } -- { "default", i }
--     else
--         -- keys are lower
--         local word = validwords[s]
--         if word == s then
--             return true, { "okay", i } -- exact match
--         elseif word then
--             return true, { "warning", i } -- case issue
--         else
--             local word = validwords[lower(s)]
--             if word == s then
--                 return true, { "okay", i } -- exact match
--             elseif word then
--                 return true, { "warning", i } -- case issue
--             elseif upper(s) == s then
--                 return true, { "warning", i } -- probably a logo or acronym
--             else
--                 return true, { "error", i }
--             end
--         end
--     end
-- end

function context.checkedword(validwords,validminimum,s,i) -- ,limit
    if not validwords then -- or #s < validminimum then
        return true, "text", i -- { "default", i }
    else
        -- keys are lower
        local word = validwords[s]
        if word == s then
            return true, "okay", i -- exact match
        elseif word then
            return true, "warning", i -- case issue
        else
            local word = validwords[lower(s)]
            if word == s then
                return true, "okay", i -- exact match
            elseif word then
                return true, "warning", i -- case issue
            elseif upper(s) == s then
                return true, "warning", i -- probably a logo or acronym
            else
                return true, "error", i
            end
        end
    end
end

function context.styleofword(validwords,validminimum,s) -- ,limit
    if not validwords or #s < validminimum then
        return "text"
    else
        -- keys are lower
        local word = validwords[s]
        if word == s then
            return "okay" -- exact match
        elseif word then
            return "warning" -- case issue
        else
            local word = validwords[lower(s)]
            if word == s then
                return "okay" -- exact match
            elseif word then
                return "warning" -- case issue
            elseif upper(s) == s then
                return "warning" -- probably a logo or acronym
            else
                return "error"
            end
        end
    end
end

-- overloaded functions

local h_table, b_table, n_table = { }, { }, { }

setmetatable(h_table, { __index = function(t,level) local v = { level, LEXER.FOLD_HEADER } t[level] = v return v end })
setmetatable(b_table, { __index = function(t,level) local v = { level, LEXER.FOLD_BLANK  } t[level] = v return v end })
setmetatable(n_table, { __index = function(t,level) local v = { level                    } t[level] = v return v end })

-- -- todo: move the local functions outside (see below) .. old variant < 3.24
--
-- local newline = P("\r\n") + S("\r\n")
-- local p_yes   = Cp() * Cs((1-newline)^1) * newline^-1
-- local p_nop   = newline
--
-- local function fold_by_parsing(text,start_pos,start_line,start_level,lexer)
--     local foldsymbols = lexer._foldsymbols
--     if not foldsymbols then
--         return { }
--     end
--     local patterns = foldsymbols._patterns
--     if not patterns then
--         return { }
--     end
--     local nofpatterns = #patterns
--     if nofpatterns == 0 then
--         return { }
--     end
--     local folds = { }
--     local line_num = start_line
--     local prev_level = start_level
--     local current_level = prev_level
--     local validmatches = foldsymbols._validmatches
--     if not validmatches then
--         validmatches = { }
--         for symbol, matches in next, foldsymbols do -- whatever = { start = 1, stop = -1 }
--             if not find(symbol,"^_") then -- brrr
--                 for s, _ in next, matches do
--                     validmatches[s] = true
--                 end
--             end
--         end
--         foldsymbols._validmatches = validmatches
--     end
--     -- of course we could instead build a nice lpeg checker .. something for
--     -- a rainy day with a stack of new cd's at hand
--     local function action_y(pos,line)
--         for i=1,nofpatterns do
--             for s, m in gmatch(line,patterns[i]) do
--                 if validmatches[m] then
--                     local symbols = foldsymbols[get_style_at(start_pos + pos + s - 1)]
--                     if symbols then
--                         local action = symbols[m]
--                         if action then
--                             if type(action) == 'number' then -- we could store this in validmatches if there was only one symbol category
--                                 current_level = current_level + action
--                             else
--                                 current_level = current_level + action(text,pos,line,s,m)
--                             end
--                             if current_level < FOLD_BASE then
--                                 current_level = FOLD_BASE
--                             end
--                         end
--                     end
--                 end
--             end
--         end
--         if current_level > prev_level then
--             folds[line_num] = h_table[prev_level] -- { prev_level, FOLD_HEADER }
--         else
--             folds[line_num] = n_table[prev_level] -- { prev_level }
--         end
--         prev_level = current_level
--         line_num = line_num + 1
--     end
--     local function action_n()
--         folds[line_num] = b_table[prev_level] -- { prev_level, FOLD_BLANK }
--         line_num = line_num + 1
--     end
--     if lexer._reset_parser then
--         lexer._reset_parser()
--     end
--     local lpegpattern = (p_yes/action_y + p_nop/action_n)^0 -- not too efficient but indirect function calls are neither but
--     lpegmatch(lpegpattern,text)                             -- keys are not pressed that fast ... large files are slow anyway
--     return folds
-- end

-- The 3.24 variant; no longer subtable optimization is needed:

local newline = P("\r\n") + S("\r\n")
local p_yes   = Cp() * Cs((1-newline)^1) * newline^-1
local p_nop   = newline

local folders = { }

local function fold_by_parsing(text,start_pos,start_line,start_level,lexer)
    local folder = folders[lexer]
    if not folder then
        --
        local pattern, folds, text, start_pos, line_num, prev_level, current_level
        --
        local fold_symbols = lexer._foldsymbols
        local fold_pattern = lexer._foldpattern -- use lpeg instead (context extension)
        --
local FOLD_BASE   = LEXER.FOLD_BASE   or SC_FOLDLEVELBASE
local FOLD_HEADER = LEXER.FOLD_HEADER or SC_FOLDLEVELHEADERFLAG
local FOLD_BLANK  = LEXER.FOLD_BLANK  or SC_FOLDLEVELWHITEFLAG
local get_style_at = LEXER.get_style_at or GetStyleAt
        --
        if fold_pattern then
            -- if no functions are found then we could have a faster one

         -- fold_pattern = Cp() * C(fold_pattern) * Carg(1) / function(s,match,pos)
         --     local symbols = fold_symbols[get_style_at(start_pos + pos + s - 1)]
         --     local l = symbols and symbols[match]
         --     if l then
         --         local t = type(l)
         --         if t == 'number' then
         --             current_level = current_level + l
         --         elseif t == 'function' then
         --             current_level = current_level + l(text, pos, line, s, match)
         --         end
         --     end
         -- end
         -- fold_pattern = (fold_pattern + P(1))^0
         -- local action_y = function(pos,line)
         --     lpegmatch(fold_pattern,line,1,pos)
         --     folds[line_num] = prev_level
         --     if current_level > prev_level then
         --         folds[line_num] = prev_level + FOLD_HEADER
         --     end
         --     if current_level < FOLD_BASE then
         --         current_level = FOLD_BASE
         --     end
         --     prev_level = current_level
         --     line_num = line_num + 1
         -- end
         -- local action_n = function()
         --     folds[line_num] = prev_level + FOLD_BLANK
         --     line_num = line_num + 1
         -- end
         -- pattern = (p_yes/action_y + p_nop/action_n)^0

            fold_pattern = Cp() * C(fold_pattern) / function(s,match)
                local symbols = fold_symbols[get_style_at(start_pos + s)]
                if symbols then
                    local l = symbols[match]
                    if l then
                        current_level = current_level + l
                    end
                end
            end
            local action_y = function()
                folds[line_num] = prev_level
                if current_level > prev_level then
                    folds[line_num] = prev_level + FOLD_HEADER
                end
                if current_level < FOLD_BASE then
                    current_level = FOLD_BASE
                end
                prev_level = current_level
                line_num = line_num + 1
            end
            local action_n = function()
                folds[line_num] = prev_level + FOLD_BLANK
                line_num = line_num + 1
            end
            pattern = ((fold_pattern + (1-newline))^1 * newline / action_y + newline/action_n)^0

         else
            -- the traditional one but a bit optimized
            local fold_symbols_patterns = fold_symbols._patterns
            local action_y = function(pos,line)
                for j = 1, #fold_symbols_patterns do
                    for s, match in gmatch(line,fold_symbols_patterns[j]) do -- '()('..patterns[i]..')'
                        local symbols = fold_symbols[get_style_at(start_pos + pos + s - 1)]
                        local l = symbols and symbols[match]
                        local t = type(l)
                        if t == 'number' then
                            current_level = current_level + l
                        elseif t == 'function' then
                            current_level = current_level + l(text, pos, line, s, match)
                        end
                    end
                end
                folds[line_num] = prev_level
                if current_level > prev_level then
                    folds[line_num] = prev_level + FOLD_HEADER
                end
                if current_level < FOLD_BASE then
                    current_level = FOLD_BASE
                end
                prev_level = current_level
                line_num = line_num + 1
            end
            local action_n = function()
                folds[line_num] = prev_level + FOLD_BLANK
                line_num = line_num + 1
            end
            pattern = (p_yes/action_y + p_nop/action_n)^0
        end
        --
        local reset_parser = lexer._reset_parser
        --
        folder = function(_text_,_start_pos_,_start_line_,_start_level_)
            if reset_parser then
                reset_parser()
            end
            folds         = { }
            text          = _text_
            start_pos     = _start_pos_
            line_num      = _start_line_
            prev_level    = _start_level_
            current_level = prev_level
            lpegmatch(pattern,text)
--             return folds
local t = folds
folds = nil
return t -- so folds can be collected
        end
        folders[lexer] = folder
    end
    return folder(text,start_pos,start_line,start_level,lexer)
end

-- local function fold_by_indentation(text,start_pos,start_line,start_level)
--     local folds = { }
--     local current_line = start_line
--     local prev_level = start_level
--     for line in gmatch(text,'[\t ]*(.-)\r?\n') do
--         if line ~= "" then
--             local current_level = FOLD_BASE + get_indent_amount(current_line)
--             if current_level > prev_level then -- next level
--                 local i = current_line - 1
--                 while true do
--                     local f = folds[i]
--                     if f and f[2] == FOLD_BLANK then
--                         i = i - 1
--                     else
--                         break
--                     end
--                 end
--                 local f = folds[i]
--                 if f then
--                     f[2] = FOLD_HEADER
--                 end -- low indent
--                 folds[current_line] = n_table[current_level] -- { current_level } -- high indent
--             elseif current_level < prev_level then -- prev level
--                 local f = folds[current_line - 1]
--                 if f then
--                     f[1] = prev_level -- high indent
--                 end
--                 folds[current_line] = n_table[current_level] -- { current_level } -- low indent
--             else -- same level
--                 folds[current_line] = n_table[prev_level] -- { prev_level }
--             end
--             prev_level = current_level
--         else
--             folds[current_line] = b_table[prev_level] -- { prev_level, FOLD_BLANK }
--         end
--         current_line = current_line + 1
--     end
--     return folds
-- end

-- local function fold_by_indentation(text,start_pos,start_line,start_level)
--     local folds = { }
--     local current_line = start_line
--     local prev_level = start_level
--     for line in gmatch(text,'[\t ]*(.-)\r?\n') do
--         if line ~= '' then
--             local current_level = FOLD_BASE + get_indent_amount(current_line)
--             if current_level > prev_level then -- next level
--                 local i = current_line - 1
--                 local f
--                 while true do
--                     f = folds[i]
--                     if not f then
--                         break
--                     elseif f[2] == FOLD_BLANK then
--                         i = i - 1
--                     else
--                         f[2] = FOLD_HEADER -- low indent
--                         break
--                     end
--                 end
--                 folds[current_line] = { current_level } -- high indent
--             elseif current_level < prev_level then -- prev level
--                 local f = folds[current_line - 1]
--                 if f then
--                     f[1] = prev_level -- high indent
--                 end
--                 folds[current_line] = { current_level } -- low indent
--             else -- same level
--                 folds[current_line] = { prev_level }
--             end
--             prev_level = current_level
--         else
--             folds[current_line] = { prev_level, FOLD_BLANK }
--         end
--         current_line = current_line + 1
--     end
--     for line, level in next, folds do
--         folds[line] = level[1] + (level[2] or 0)
--     end
--     return folds
-- end

local folds, current_line, prev_level

local function action_y()
local FOLD_BASE   = LEXER.FOLD_BASE   or SC_FOLDLEVELBASE
local FOLD_HEADER = LEXER.FOLD_HEADER or SC_FOLDLEVELHEADERFLAG
local FOLD_BLANK  = LEXER.FOLD_BLANK  or SC_FOLDLEVELWHITEFLAG
local get_indent_amount = LEXER.get_indent_amount or GetIndentAmount
    local current_level = FOLD_BASE + get_indent_amount(current_line)
    if current_level > prev_level then -- next level
        local i = current_line - 1
        local f
        while true do
            f = folds[i]
            if not f then
                break
            elseif f[2] == FOLD_BLANK then
                i = i - 1
            else
                f[2] = FOLD_HEADER -- low indent
                break
            end
        end
        folds[current_line] = { current_level } -- high indent
    elseif current_level < prev_level then -- prev level
        local f = folds[current_line - 1]
        if f then
            f[1] = prev_level -- high indent
        end
        folds[current_line] = { current_level } -- low indent
    else -- same level
        folds[current_line] = { prev_level }
    end
    prev_level = current_level
    current_line = current_line + 1
end

local function action_n()
    folds[current_line] = { prev_level, LEXER.FOLD_BLANK }
    current_line = current_line + 1
end

local pattern = ( S("\t ")^0 * ( (1-S("\n\r"))^1 / action_y + P(true) / action_n) * newline )^0

local function fold_by_indentation(text,start_pos,start_line,start_level)
    -- initialize
    folds        = { }
    current_line = start_line
    prev_level   = start_level
    -- define
    -- -- not here .. pattern binds and local functions are not frozen
    -- analyze
    lpegmatch(pattern,text)
    -- flatten
    for line, level in next, folds do
        folds[line] = level[1] + (level[2] or 0)
    end
    -- done
--     return folds
local t = folds
folds = nil
return t -- so folds can be collected
end

local function fold_by_line(text,start_pos,start_line,start_level)
    local folds = { }
    -- can also be lpeg'd
    for _ in gmatch(text,".-\r?\n") do
        folds[start_line] = n_table[start_level] -- { start_level }
        start_line = start_line + 1
    end
    return folds
end

local threshold_by_lexer       =  512 * 1024 -- we don't know the filesize yet
local threshold_by_parsing     =  512 * 1024 -- we don't know the filesize yet
local threshold_by_indentation =  512 * 1024 -- we don't know the filesize yet
local threshold_by_line        =  512 * 1024 -- we don't know the filesize yet

function context.fold(text,start_pos,start_line,start_level) -- hm, we had size thresholds .. where did they go
    if text == '' then
        return { }
    end
    local lexer = global._LEXER
    local fold_by_lexer = lexer._fold
    local fold_by_symbols = lexer._foldsymbols
    local filesize = 0 -- we don't know that
local get_property = LEXER.get_property or GetProperty
    if fold_by_lexer then
        if filesize <= threshold_by_lexer then
            return fold_by_lexer(text,start_pos,start_line,start_level,lexer)
        end
    elseif fold_by_symbols then -- and get_property('fold.by.parsing',1) > 0 then
        if filesize <= threshold_by_parsing then
            return fold_by_parsing(text,start_pos,start_line,start_level,lexer)
        end
    elseif get_property('fold.by.indentation',1) > 0 then
        if filesize <= threshold_by_indentation then
            return fold_by_indentation(text,start_pos,start_line,start_level,lexer)
        end
    elseif get_property('fold.by.line',1) > 0 then
        if filesize <= threshold_by_line then
            return fold_by_line(text,start_pos,start_line,start_level,lexer)
        end
    end
    return { }
end

-- The following code is mostly unchanged:

local function add_rule(lexer, id, rule)
    if not lexer._RULES then
        lexer._RULES = {}
        lexer._RULEORDER = {}
    end
    lexer._RULES[id] = rule
    lexer._RULEORDER[#lexer._RULEORDER + 1] = id
end

local function add_style(lexer, token_name, style)
    local len = lexer._STYLES.len
    if len == 32 then
        len = len + 8
    end
    if len >= 128 then
        print('Too many styles defined (128 MAX)')
    end
    lexer._TOKENS[token_name] = len
    lexer._STYLES[len] = style
    lexer._STYLES.len = len + 1
end

local function join_tokens(lexer)
    local patterns, order = lexer._RULES, lexer._RULEORDER
    local token_rule = patterns[order[1]]
    for i=2,#order do
        token_rule = token_rule + patterns[order[i]]
    end
    lexer._TOKENRULE = token_rule
    return lexer._TOKENRULE
end

local function add_lexer(grammar, lexer, token_rule)
    local token_rule = join_tokens(lexer)
    local lexer_name = lexer._NAME
    local children = lexer._CHILDREN
    for i=1,#children do
        local child = children[i]
        if child._CHILDREN then
            add_lexer(grammar, child)
        end
        local child_name = child._NAME
        local rules = child._EMBEDDEDRULES[lexer_name]
        local rules_token_rule = grammar['__'..child_name] or rules.token_rule
        grammar[child_name] = (-rules.end_rule * rules_token_rule)^0 * rules.end_rule^-1 * V(lexer_name)
        local embedded_child = '_' .. child_name
        grammar[embedded_child] = rules.start_rule * (-rules.end_rule * rules_token_rule)^0 * rules.end_rule^-1
        token_rule = V(embedded_child) + token_rule
    end
    grammar['__' .. lexer_name] = token_rule
    grammar[lexer_name] = token_rule^0
end

local function build_grammar(lexer, initial_rule)
    local children = lexer._CHILDREN
    if children then
        local lexer_name = lexer._NAME
        if not initial_rule then
            initial_rule = lexer_name
        end
        local grammar = { initial_rule }
        add_lexer(grammar, lexer)
        lexer._INITIALRULE = initial_rule
        lexer._GRAMMAR = Ct(P(grammar))
    else
        lexer._GRAMMAR = Ct(join_tokens(lexer)^0)
    end
end

-- so far. We need these local functions in the next one.
--
-- Before 3.24 we had tokens[..] = { category, position }, now it's a two values.

local lineparsers = { }

function context.lex(text,init_style)
    local lexer = global._LEXER
    local grammar = lexer._GRAMMAR
    if not grammar then
        return { }
    elseif lexer._LEXBYLINE then -- we could keep token
        local tokens = { }
        local offset = 0
        local noftokens = 0
     -- -- pre 3.24
     --
     -- for line in gmatch(text,'[^\r\n]*\r?\n?') do -- could be an lpeg
     --     local line_tokens = lpegmatch(grammar,line)
     --     if line_tokens then
     --         for i=1,#line_tokens do
     --             local token = line_tokens[i]
     --             token[2] = token[2] + offset
     --             noftokens = noftokens + 1
     --             tokens[noftokens] = token
     --         end
     --     end
     --     offset = offset + #line
     --     if noftokens > 0 and tokens[noftokens][2] ~= offset then
     --         noftokens = noftokens + 1
     --         tokens[noftokens] = { 'default', offset + 1 }
     --     end
     -- end

     -- for line in gmatch(text,'[^\r\n]*\r?\n?') do
     --     local line_tokens = lpegmatch(grammar,line)
     --     if line_tokens then
     --         for i=1,#line_tokens,2 do
     --             noftokens = noftokens + 1
     --             tokens[noftokens] = line_tokens[i]
     --             noftokens = noftokens + 1
     --             tokens[noftokens] = line_tokens[i + 1] + offset
     --         end
     --     end
     --     offset = offset + #line
     --     if noftokens > 0 and tokens[noftokens] ~= offset then
     --         noftokens = noftokens + 1
     --         tokens[noftokens] = 'default'
     --         noftokens = noftokens + 1
     --         tokens[noftokens] = offset + 1
     --     end
     -- end

        local lineparser = lineparsers[lexer]
        if not lineparser then -- probably a cmt is more efficient
            lineparser = C((1-newline)^0 * newline) / function(line)
                local length = #line
                local line_tokens = length > 0 and lpegmatch(grammar,line)
                if line_tokens then
                    for i=1,#line_tokens,2 do
                        noftokens = noftokens + 1
                        tokens[noftokens] = line_tokens[i]
                        noftokens = noftokens + 1
                        tokens[noftokens] = line_tokens[i + 1] + offset
                    end
                end
                offset = offset + length
                if noftokens > 0 and tokens[noftokens] ~= offset then
                    noftokens = noftokens + 1
                    tokens[noftokens] = 'default'
                    noftokens = noftokens + 1
                    tokens[noftokens] = offset + 1
                end
            end
            lineparser = lineparser^0
            lineparsers[lexer] = lineparser
        end
        lpegmatch(lineparser,text)
        return tokens

    elseif lexer._CHILDREN then
        -- as we cannot print, tracing is not possible ... this might change as we can as well
        -- generate them all in one go (sharing as much as possible)
        local hash = lexer._HASH -- hm, was _hash
        if not hash then
            hash = { }
            lexer._HASH = hash
        end
        grammar = hash[init_style]
        if grammar then
            lexer._GRAMMAR = grammar
        else
            for style, style_num in next, lexer._TOKENS do
                if style_num == init_style then
                    -- the name of the lexers is filtered from the whitespace
                    -- specification
                    local lexer_name = match(style,'^(.+)_whitespace') or lexer._NAME
                    if lexer._INITIALRULE ~= lexer_name then
                        grammar = hash[lexer_name]
                        if not grammar then
                            build_grammar(lexer,lexer_name)
                            grammar = lexer._GRAMMAR
                            hash[lexer_name] = grammar
                        end
                    end
                    break
                end
            end
            grammar = grammar or lexer._GRAMMAR
            hash[init_style] = grammar
        end
        return lpegmatch(grammar,text)
    else
        return lpegmatch(grammar,text)
    end
end

-- todo: keywords: one lookup and multiple matches

-- function context.token(name, patt)
--     return Ct(patt * Cc(name) * Cp())
-- end
--
-- -- hm, changed in 3.24 .. no longer a table

function context.token(name, patt)
    return patt * Cc(name) * Cp()
end

lexer.fold        = context.fold
lexer.lex         = context.lex
lexer.token       = context.token
lexer.exact_match = context.exact_match

-- helper .. alas ... the lexer's lua instance is rather crippled .. not even
-- math is part of it

local floor = math and math.floor
local char  = string.char

if not floor then

    floor = function(n)
        return tonumber(string.format("%d",n))
    end

    math = math or { }

    math.floor = floor

end

local function 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

context.utfchar = utfchar

-- a helper from l-lpeg:

local gmatch = string.gmatch

local function make(t)
    local p
    for k, v in next, t do
        if not p then
            if next(v) then
                p = P(k) * make(v)
            else
                p = P(k)
            end
        else
            if next(v) then
                p = p + P(k) * make(v)
            else
                p = p + P(k)
            end
        end
    end
    return p
end

function lpeg.utfchartabletopattern(list)
    local tree = { }
    for i=1,#list do
        local t = tree
        for c in gmatch(list[i],".") do
            if not t[c] then
                t[c] = { }
            end
            t = t[c]
        end
    end
    return make(tree)
end

-- patterns.invisibles =
--     P(utfchar(0x00A0)) -- nbsp
--   + P(utfchar(0x2000)) -- enquad
--   + P(utfchar(0x2001)) -- emquad
--   + P(utfchar(0x2002)) -- enspace
--   + P(utfchar(0x2003)) -- emspace
--   + P(utfchar(0x2004)) -- threeperemspace
--   + P(utfchar(0x2005)) -- fourperemspace
--   + P(utfchar(0x2006)) -- sixperemspace
--   + P(utfchar(0x2007)) -- figurespace
--   + P(utfchar(0x2008)) -- punctuationspace
--   + P(utfchar(0x2009)) -- breakablethinspace
--   + P(utfchar(0x200A)) -- hairspace
--   + P(utfchar(0x200B)) -- zerowidthspace
--   + P(utfchar(0x202F)) -- narrownobreakspace
--   + P(utfchar(0x205F)) -- math thinspace

patterns.invisibles = lpeg.utfchartabletopattern {
    utfchar(0x00A0), -- nbsp
    utfchar(0x2000), -- enquad
    utfchar(0x2001), -- emquad
    utfchar(0x2002), -- enspace
    utfchar(0x2003), -- emspace
    utfchar(0x2004), -- threeperemspace
    utfchar(0x2005), -- fourperemspace
    utfchar(0x2006), -- sixperemspace
    utfchar(0x2007), -- figurespace
    utfchar(0x2008), -- punctuationspace
    utfchar(0x2009), -- breakablethinspace
    utfchar(0x200A), -- hairspace
    utfchar(0x200B), -- zerowidthspace
    utfchar(0x202F), -- narrownobreakspace
    utfchar(0x205F), -- math thinspace
}

-- now we can make:

patterns.iwordtoken   = patterns.wordtoken - patterns.invisibles
patterns.iwordpattern = patterns.iwordtoken^3

-- require("themes/scite-context-theme")

-- In order to deal with some bug in additional styles (I have no cue what is
-- wrong, but additional styles get ignored and clash somehow) I just copy the
-- original lexer code ... see original for comments.

return LEXER