dparse.lexer 281/867(32%) line coverage

      
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
250
260
270
280
290
300
310
320
330
340
350
360
370
380
390
400
410
420
430
440
450
460
470
480
490
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
670
680
690
700
710
720
730
740
750
760
770
780
790
800
810
820
830
840
850
860
870
880
890
900
910
920
930
940
950
960
970
980
990
1000
1010
1020
1030
1040
1050
1060
1070
1080
1090
1100
1110
1120
1130
1140
1150
1160
1170
1180
1190
1200
1210
1220
1230
1240
1250
1260
1270
1280
1290
1300
1310
1320
1330
1340
1350
1360
1370
1380
1390
1400
1410
1420
1430
1440
1450
1460
1470
1480
1490
1500
1510
1520
1530
1540
1550
1560
1570
1580
1590
1600
1610
1620
1630
1640
1650
1660
1670
1680
1690
1700
1710
1720
1730
1740
1750
1760
1770
1780
1790
1800
1810
1820
1830
1840
1850
1860
1870
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
2110
2120
2130
2140
2150
2160
2170
2180
2190
2200
2210
2220
2230
2240
2250
2260
2270
2280
2290
2300
2310
2320
2330
2340
2350
2360
2370
2380
2390
2400
2410
2420
2430
2440
2450
2460
2470
2480
2490
2500
2510
2520
2530
2540
2550
2560
2570
2580
2590
2600
2610
2620
2630
2640
2650
2660
2670
2680
2690
2700
2710
2720
2730
2740
2750
2760
2770
2780
2790
2800
2810
2820
2830
2840
2850
2860
2870
2880
2890
2900
2910
2920
2930
2940
2950
2960
2970
2980
2990
3000
3010
3020
3030
3040
3050
3060
3070
3080
3090
3100
3110
3120
3130
3140
3150
3160
3170
3180
3190
3200
3210
3220
3230
3240
3250
3260
3270
3280
3290
3300
3310
3320
3330
3340
3350
3360
3370
3380
3390
3400
3410
3420
3430
3440
3450
3460
3470
3480
3490
3500
3510
3520
3530
3540
3550
3560
3570
3580
3590
3600
3610
3620
3630
3640
3650
3660
3670
3680
3690
3700
3710
3720
3730
3740
3750
3760
3770
3780
3790
3800
3810
3820
3830
3840
3850
3860
3870
3880
3890
3900
3910
3920
3930
3940
3950
3960
3970
3980
3990
4000
4010
4020
4030
4040
4050
4060
4070
4080
4090
4100
4110
4120
4130
4140
4150
4160
4170
4180
4190
4200
4210
4220
4230
4240
4250
4260
4270
4280
4290
4300
4310
4320
4330
4340
4350
4360
4370
4380
4390
4400
4410
4420
4430
4440
4450
4460
4470
4480
4490
4500
4510
4520
4530
4540
4550
4560
4570
4580
4590
4600
4610
4620
4630
4640
4650
4660
4670
4680
4690
4700
4710
4720
4730
4740
4750
4760
4770
4780
4790
4800
4810
4820
4830
4840
4850
4860
4870
4880
4890
4900
4910
4920
4930
4940
4950
4960
4970
4980
4990
5000
5010
5020
5030
5040
5050
5060
5070
5080
5090
5100
5110
5120
5130
5140
5150
5160
5170
5180
5190
5200
5210
5220
5230
5240
5250
5260
5270
5280
5290
5300
5310
5320
5330
5340
5350
5360
5370
5380
5390
5400
5410
5420
5430
5440
5450
5460
5470
5480
5490
5500
5510
5520
5530
5540
5550
5560
5570
5580
5590
5600
5610
5620
5630
5640
5650
5660
5670
5680
5690
5700
5711182
5720
5730
5740
5750
5760
5770
5780
5790
5800
5810
5820
583203
584203
585203
586203
5870
5880
5890
5900
5910
5920
5930
5940
5950
5960
5970
5980
5990
6000
6010
602979
603979
604979
6050
6060
6070
6080
6090
6100
6110
6120
6130
6140
61555372
6160
61727647
6180
6190
6200
62127686
6220
6230
6240
6250
6260
6270
6280
6290
6300
6310
6329011
6339011
6349011
6359011
6369011
63755
63855
63955
64055
64155
64255
6430
6440
6450
6460
6470
6480
6490
6500
6510
6520
6530
6540
6550
6560
65718620
65818620
6590
6609066
66118633
66218633
66318633
6640
6650
6660
6670
6680
6691102
6700
671551
672551
6730
6740
6750
6760
6770
6780
6790
6800
6810
6820
6830
6840
6850
686551
687551
688551
6890
6900
6910
6920
6930
6940
6950
6960
6970
6980
6990
7000
7010
7020
7030
7040
7050
7060
7070
7080
7090
7100
7110
7120
7130
7140
7150
7160
7170
7180
7190
7200
7210
7220
7230
7240
7250
7260
7270
7280
7290
7300
7310
7320
7330
7340
7350
7360
7370
7380
7390
7400
7410
7420
7430
7440
7450
7460
7470
7480
7490
7500
7510
7520
7530
7540
7550
7560
7570
7580
7590
7600
7610
7620
7630
7640
7650
7660
7670
7680
7690
7700
7710
7720
7730
7740
7750
7760
7770
7780
7790
7800
7810
7820
7830
7840
7850
7860
7870
7880
7890
7900
7910
7920
7930
7940
7950
7960
7970
7980
7990
8000
8010
8020
8030
8040
8050
8060
8070
8080
8090
8100
8110
8120
8130
8140
8150
8160
8170
8180
8190
8200
8210
8220
8230
8240
8250
8260
8270
8280
8290
8300
8310
832723
8330
8340
8350
8360
8370
8381274
8391274
8401274
8410
8420
8430
8440
8450
8462548
8470
8482548
8490
85021510
8511274
8520
8530
8542548
8550
8561273
8571273
8581273
8590
8600
8611
8620
8630
8640
8651274
8660
8670
8680
8690
8700
8710
8720
8730
8740
8750
8760
8770
8780
8790
8800
8810
8820
8830
8840
8850
8860
8870
88819
88957
89011
8910
8920
8930
8940
8950
8968
8970
8988
8998
90016
90116
90216
9038
90416
9050
9060
9070
9080
9098
9100
9110
9120
9130
9140
9150
9160
9170
9180
9190
9201255
9211255
9220
9230
9241274
9250
9260
9270
9280
9290
9300
9310
9320
9330
9340
9350
9360
9370
9380
9390
9400
9410
9420
9430
9440
9450
9460
9470
9480
9490
9500
9510
9520
9530
9540
9550
9560
9570
9580
9590
9600
9610
9620
9630
9640
9650
9660
9670
9680
9690
9700
9710
9720
9730
9740
9750
9760
9770
9780
9790
9800
9810
9820
9830
9840
9850
9860
9870
9880
9890
9900
9910
9920
9930
9940
9950
9960
9970
9980
9990
10000
10010
10020
10030
10040
10050
10060
10070
10080
10090
10100
10110
10120
10130
10140
10150
10160
10170
10180
10190
10200
10210
10220
10230
10240
10250
10260
10270
10280
10290
10300
10310
10320
10330
10340
10350
10360
10370
10380
10390
10400
10410
10420
10430
10440
10450
10460
10470
10480
10490
10500
10510
10520
10530
10540
10550
10560
10570
10580
10590
10600
10610
10620
10630
10640
106510
106610
106710
10680
10690
10700
107120
107210
10730
10740
107510
10760
107710
107820
10790
108010
108110
10820
10830
10840
10850
10860
108710
108810
10890
10900
10910
10920
10930
10940
10950
1096276
1097276
1098793
10990
11000
11010
11021586
11030
1104793
11050
11060
11070
11081586
1109276
1110517
11110
1112276
1113276
11140
11150
11160
11170
11180
11190
11200
112112
112212
112312
112412
1125448
11260
11270
11280
1129436
11300
1131218
11320
11330
11340
1135218
11360
113724
113848
11390
114012
114112
11420
11430
1144194
11450
11460
11470
11480
11490
11500
11510
11520
11530
1154194
11550
115612
115712
11580
11590
11600
11610
11620
11630
11640
11651648
11662475
11670
11682475
11690
11700
11710
11720
11730
11740
11750
11764950
11770
11782452
11790
11800
11810
11822475
11830
11841648
11851648
11860
1187827
11880
118951
11900
11910
1192776
11930
11941648
11951648
11961648
11970
11980
11990
12000
12010
12020
1203170
1204170
1205170
12060
1207170
1208300
12090
1210300
12110
12120
12130
12140
12150
12160
12170
1218600
12190
1220300
12210
12220
12230
1224300
12250
1226170
1227170
12280
12290
1230130
12310
12320
12330
12340
12350
12360
12370
12380
12390
12400
12410
12420
12430
12440
12450
12460
12470
12480
12490
12500
12510
12520
12530
12540
12550
12560
12570
12580
12590
1260170
1261170
12620
12630
12640
12650
12660
12671818
12680
12690
12700
12710
12720
12730
12741818
12750
12760
12770
12780
12795454
12800
12810
12820
12830
12840
12850
12860
12870
12880
12890
12900
12910
12920
12930
12940
12950
12960
12970
12980
12990
13000
13010
13020
13030
13040
13050
13060
13070
13080
13090
13100
13110
13120
13130
13140
13150
13160
13170
13180
13190
13200
13210
13220
13230
13240
13250
13260
13270
13280
13290
13300
13310
13320
13330
13340
13350
13360
13370
13380
13390
13400
13410
13420
13430
13440
13450
13460
13470
13480
13490
13500
13510
13520
13530
13540
13550
13560
13570
13580
13590
13600
13610
13620
13630
13640
13650
13660
13670
13680
13690
13700
13710
13720
13730
13740
13750
13760
13770
13780
13790
13800
13810
13820
13830
13840
13850
13860
13870
13880
13890
13900
13910
13920
13930
13940
13950
13960
13970
13980
13990
14000
14010
14020
14030
14040
14050
14060
14070
14080
14090
14100
14110
14120
14130
14140
14150
14160
14170
14180
14190
14200
14210
14220
14230
14240
14250
14260
14270
14280
14290
14300
14310
14320
14330
14340
14350
14360
14370
14380
14390
14400
14410
14420
14430
14440
14450
14460
14470
14480
14490
14500
14510
14520
14530
14540
14550
14560
14570
14580
14590
14600
14610
14620
14630
14640
14650
14660
14670
14680
14690
14700
14710
14720
14730
14740
14750
14760
14770
14780
14790
14800
14810
14820
14830
14840
14850
14860
14870
14880
14890
14900
14910
149251
149351
14940
14950
14960
14970
149851
14990
15000
15012
15022
15032
15042
15052
15062
150751
150851
150951
151051
151151
151251
15130
15140
15150
15160
15170
15180
15190
15200
15210
15220
15230
15240
15250
15260
15270
15280
15290
15300
15310
15320
15330
15340
15350
15360
15370
15380
15390
15400
15410
15420
15430
15440
15450
15460
15470
15480
15490
15500
15510
15520
15530
15540
15550
15560
15570
15580
15590
15600
15610
15620
15630
15640
15650
15660
15670
15680
15690
15700
15710
15720
15730
15740
15750
15760
15770
15780
15790
15800
15810
15820
15830
15840
15850
15860
15870
15880
15890
15900
15910
15920
15930
15940
15950
15960
15970
15980
15990
16000
16010
16020
16030
16040
16050
16060
16070
16080
16090
16100
161151
16120
16130
16140
16150
16160
161782
161882
16190
162082
16210
162282
16230
16240
16250
16260
16270
162882
16290
16300
16310
16320
16330
163482
16350
1636164
16370
163882
163982
16400
16410
16420
16430
16440
16450
16460
16470
16480
16490
16500
16510
16520
165313267
16540
16550
16560
16570
165813843
16590
16600
16610
166227686
16630
166413776
16650
166613776
166713776
16680
16690
167013843
167113267
16720
1673576
16740
167513267
16760
16770
16780
16790
16800
16810
16824588
16830
16840
16850
16860
16870
16884588
16890
16900
16910
16920
16930
16940
16950
16960
16970
16980
16990
17000
17010
17020
17030
17044588
17054588
17064588
17074588
17080
17090
17100
17110
17120
17130
17140
17150
17160
17170
17180
17190
17200
17210
17220
17230
17240
17250
17260
17270
17280
17290
17300
17310
17320
17330
173432304
17350
17360
17370
17380
173932304
17400
174132304
174232304
17430
17440
17450
17460
17470
17480
17490
17500
17510
17520
17530
17540
17550
17560
17570
17580
17590
176032304
176132304
176214739
176317565
176417565
17650
17660
17670
17680
17690
17700
17710
17720
17730
17740
17750
17760
17770
17780
17790
17800
17810
17820
17830
17840
17850
17860
17870
17880
17890
17900
17910
17920
17930
17940
17950
17960
17970
17980
17990
18000
18010
18020
18030
18040
18050
18060
18070
18080
18090
18100
18110
18120
18130
18140
18150
18160
18170
18180
18190
18200
18210
18220
18230
18240
18250
18260
18270
18280
18290
18300
18310
18320
18330
18340
18350
18360
18370
18380
18390
18400
18410
18420
18430
18440
18450
18460
18470
18480
18490
18500
18510
18520
18530
18540
18550
18560
18570
18580
18590
18600
18610
18620
18630
18640
18650
18660
18670
18680
18690
18700
18710
18720
18730
18740
18750
18760
18770
18780
18790
18800
18810
18820
18830
18840
18850
18860
18870
18880
18890
18900
18910
18920
18930
18940
18950
18960
18970
18980
18990
19000
19010
19020
19030
19040
19050
19060
19070
19080
19090
19100
19110
19120
19130
19140
19150
19160
19170
19180
19190
19200
19210
19220
19230
19240
19250
19260
19270
19280
19290
19300
19310
19320
19330
19340
19350
19360
19370
19380
19390
19400
19410
19420
19430
19440
19450
19460
19470
19480
19490
19500
19510
19520
19530
19540
19550
19560
19570
19580
19590
19600
19610
19620
19630
19640
19650
19660
19670
19680
19690
19700
19710
19720
19730
19740
19750
19760
19770
19780
19790
19800
19810
19820
19830
19840
19850
19860
19870
19880
19890
19900
19910
19920
19930
19940
19950
19960
19970
19980
19990
20000
20010
20020
20030
20040
20050
20060
20070
20080
20090
20100
20110
20120
20130
20140
20150
20160
20170
20180
20190
20200
20210
20220
20230
20240
20250
20260
20270
20280
20290
20300
20310
20320
20330
20340
20350
20360
20370
20380
20390
20400
204113
20420
20430
20440
20450
20460
204713
204813
204939
20500
20510
20520
20530
20540
20550
20560
20570
20580
205913
20600
20610
20620
20630
206413
206526
20660
206713
206813
206913
20700
2071159783
20720
207353248
207456893
20750
20763645
20770
20783645
20793645
20803645
20810
20820
208313
208413
208513
20860
20870
20880
20890
20900
20910
20920
209370744
20940
209535372
20960
20970
20980
20990
21000
21010
21020
21030
21040
21050
21060
21070
21080
21090
21100
21110
21120
21130
21140
211535372
211635372
211735372
211835372
211931727
21203645
21213645
21223645
21230
21240
21253645
21263645
21273645
21283645
21293645
21303645
21313645
21323645
21333645
21340
21350
21360
21370
21380
213935372
214035372
214136261
21420
214364343
214431727
2145889
21460
21473645
21480
21490
21500
21510
21520
215335372
215435372
21550
21560
21570
215835372
215935372
216035372
216168618
21620
216333246
21640
21650
21660
216733246
216833246
216933246
217033246
217133246
217233246
21730
217435372
21750
21765541
21775541
21785541
217910337
218010337
218110337
218230913
218330913
218430913
218530913
21864459
21874459
21880
218935372
219035372
219135372
219235372
21930
21940
21950
21960
21970
21983645
21990
22000
22010
22023645
22030
22040
22050
22063645
22073645
22087290
22090
22103632
22113632
22123632
22133632
22140
22153632
22167264
22170
22180
22190
22200
222113
222213
222313
222413
222526
22260
22270
22280
22290
22300
22310
22320
22330
22340
22350
22360
22370
22380
22390
22400
22410
22420
22430
22440
22450
22460
22470
22480
22490
22500
22510
22520
22530
22540
22550
22560
22570
22580
22590
22600
22610
22620
22630
22640
22650
22660
22670
22680
22690
22700
22710
22720
22730
22740
22750
22760
22770
22780
22790
22800
22810
22820
22830
22840
22850
22860
22870
22880
22890
22900
22910
22920
22930
22940
22950
22960
22970
22980
22990
23000
23010
23020
23030
23040
23050
23060
23070
23080
23090
23100
23110
23120
23130
23140
23150
23160
23170
23180
23190
23200
23210
23220
23230
23240
23250
23260
23270
23280
23290
23300
23310
23320
23330
23340
23350
23360
23370
23380
23390
23400
23410
23420
23430
23440
23450
23460
234731420
234831420
23490
235027647
23510
23523773
23530
23540
23550
23560
23570
23580
23590
23600
23610
23620
23630
23640
23650
23660
23670
23680
23690
23700
23710
23720
23730
23740
237515049
23760
23770
23780
237915049
238015049
23810
23820
23830
23840
23850
23860
23870
23880
23890
23900
23910
23920
23930
23940
23950
23960
23970
23980
23990
24000
24010
24020
24030
24040
24050
24060
24070
24080
24090
24100
24110
24120
24130
24140
24150
24160
24170
24180
module dparse.lexer; import std.typecons; import std.typetuple; import std.array; import std.algorithm; import std.range; import std.experimental.lexer; import std.traits; import core.cpuid : sse42; version (D_InlineAsm_X86_64) { version (Windows) {} else version = iasm64NotWindows; } /// Operators private enum operators = [ ",", ".", "..", "...", "/", "/=", "!", "!<", "!<=", "!<>", "!<>=", "!=", "!>", "!>=", "$", "%", "%=", "&", "&&", "&=", "(", ")", "*", "*=", "+", "++", "+=", "-", "--", "-=", ":", ";", "<", "<<", "<<=", "<=", "<>", "<>=", "=", "==", "=>", ">", ">=", ">>", ">>=", ">>>", ">>>=", "?", "@", "[", "]", "^", "^=", "^^", "^^=", "{", "|", "|=", "||", "}", "~", "~=" ]; /// Kewords private enum keywords = [ "abstract", "alias", "align", "asm", "assert", "auto", "bool", "break", "byte", "case", "cast", "catch", "cdouble", "cent", "cfloat", "char", "class", "const", "continue", "creal", "dchar", "debug", "default", "delegate", "delete", "deprecated", "do", "double", "else", "enum", "export", "extern", "false", "final", "finally", "float", "for", "foreach", "foreach_reverse", "function", "goto", "idouble", "if", "ifloat", "immutable", "import", "in", "inout", "int", "interface", "invariant", "ireal", "is", "lazy", "long", "macro", "mixin", "module", "new", "nothrow", "null", "out", "override", "package", "pragma", "private", "protected", "public", "pure", "real", "ref", "return", "scope", "shared", "short", "static", "struct", "super", "switch", "synchronized", "template", "this", "throw", "true", "try", "typedef", "typeid", "typeof", "ubyte", "ucent", "uint", "ulong", "union", "unittest", "ushort", "version", "void", "volatile", "wchar", "while", "with", "__DATE__", "__EOF__", "__FILE__", "__FILE_FULL_PATH__", "__FUNCTION__", "__gshared", "__LINE__", "__MODULE__", "__parameters", "__PRETTY_FUNCTION__", "__TIME__", "__TIMESTAMP__", "__traits", "__vector", "__VENDOR__", "__VERSION__" ]; /// Other tokens private enum dynamicTokens = [ "specialTokenSequence", "comment", "identifier", "scriptLine", "whitespace", "doubleLiteral", "floatLiteral", "idoubleLiteral", "ifloatLiteral", "intLiteral", "longLiteral", "realLiteral", "irealLiteral", "uintLiteral", "ulongLiteral", "characterLiteral", "dstringLiteral", "stringLiteral", "wstringLiteral" ]; private enum pseudoTokenHandlers = [ "\"", "lexStringLiteral", "`", "lexWysiwygString", "//", "lexSlashSlashComment", "/*", "lexSlashStarComment", "/+", "lexSlashPlusComment", ".", "lexDot", "'", "lexCharacterLiteral", "0", "lexNumber", "1", "lexDecimal", "2", "lexDecimal", "3", "lexDecimal", "4", "lexDecimal", "5", "lexDecimal", "6", "lexDecimal", "7", "lexDecimal", "8", "lexDecimal", "9", "lexDecimal", "q\"", "lexDelimitedString", "q{", "lexTokenString", "r\"", "lexWysiwygString", "x\"", "lexHexString", " ", "lexWhitespace", "\t", "lexWhitespace", "\r", "lexWhitespace", "\n", "lexWhitespace", "\v", "lexWhitespace", "\f", "lexWhitespace", "\u2028", "lexLongNewline", "\u2029", "lexLongNewline", "#!", "lexScriptLine", "#line", "lexSpecialTokenSequence" ]; /// Token ID type for the D lexer. public alias IdType = TokenIdType!(operators, dynamicTokens, keywords); /** * Function used for converting an IdType to a string. * * Examples: * --- * IdType c = tok!"case"; * assert (str(c) == "case"); * --- */ public alias str = tokenStringRepresentation!(IdType, operators, dynamicTokens, keywords); /** * Template used to refer to D token types. * * See the $(B operators), $(B keywords), and $(B dynamicTokens) enums for * values that can be passed to this template. * Example: * --- * import dparse.lexer; * IdType t = tok!"floatLiteral"; * --- */ public template tok(string token) { alias tok = TokenId!(IdType, operators, dynamicTokens, keywords, token); } private enum extraFields = q{ string comment; string trailingComment; int opCmp(size_t i) const pure nothrow @safe { if (index < i) return -1; if (index > i) return 1; return 0; } int opCmp(ref const typeof(this) other) const pure nothrow @safe { return opCmp(other.index); } }; /// The token type in the D lexer public alias Token = std.experimental.lexer.TokenStructure!(IdType, extraFields); /** * Configure whitespace handling */ public enum WhitespaceBehavior : ubyte { include = 0b0000_0000, skip = 0b0000_0001, } /** * Configure string lexing behavior */ public enum StringBehavior : ubyte { /// Do not include quote characters, process escape sequences compiler = 0b0000_0000, /// Opening quotes, closing quotes, and string suffixes are included in the /// string token includeQuoteChars = 0b0000_0001, /// String escape sequences are not replaced notEscaped = 0b0000_0010, /// Not modified at all. Useful for formatters or highlighters source = includeQuoteChars | notEscaped } public enum CommentBehavior : bool { intern = true, noIntern = false } /** * Lexer configuration struct */ public struct LexerConfig { string fileName; StringBehavior stringBehavior; WhitespaceBehavior whitespaceBehavior; CommentBehavior commentBehavior = CommentBehavior.intern; } /** * Basic type token types. */ public alias BasicTypes = AliasSeq!(tok!"int", tok!"bool", tok!"byte", tok!"cdouble", tok!"cent", tok!"cfloat", tok!"char", tok!"creal", tok!"dchar", tok!"double", tok!"float", tok!"idouble", tok!"ifloat", tok!"ireal", tok!"long", tok!"real", tok!"short", tok!"ubyte", tok!"ucent", tok!"uint", tok!"ulong", tok!"ushort", tok!"void", tok!"wchar"); /** * Returns: true if the given ID is for a basic type. */ public bool isBasicType(IdType type) nothrow pure @safe @nogc { switch (type) { foreach (T; BasicTypes) { case T: return true; } default: return false; } } /** * Number literal token types. */ public alias NumberLiterals = AliasSeq!(tok!"doubleLiteral", tok!"floatLiteral", tok!"idoubleLiteral", tok!"ifloatLiteral", tok!"intLiteral", tok!"longLiteral", tok!"realLiteral", tok!"irealLiteral", tok!"uintLiteral", tok!"ulongLiteral"); /** * Returns: true if the given ID type is for a number literal. */ public bool isNumberLiteral(IdType type) nothrow pure @safe @nogc { switch (type) { foreach (T; NumberLiterals) { case T: return true; } default: return false; } } /** * Number literal token types. */ public alias IntegerLiterals = AliasSeq!(tok!"intLiteral", tok!"longLiteral", tok!"uintLiteral", tok!"ulongLiteral"); /** * Returns: true if the given ID type is for a integer literal. */ public bool isIntegerLiteral(IdType type) nothrow pure @safe @nogc { switch (type) { foreach (T; IntegerLiterals) { case T: return true; } default: return false; } } /** * Operator token types. */ public alias Operators = AliasSeq!(tok!",", tok!".", tok!"..", tok!"...", tok!"/", tok!"/=", tok!"!", tok!"!<", tok!"!<=", tok!"!<>", tok!"!<>=", tok!"!=", tok!"!>", tok!"!>=", tok!"$", tok!"%", tok!"%=", tok!"&", tok!"&&", tok!"&=", tok!"(", tok!")", tok!"*", tok!"*=", tok!"+", tok!"++", tok!"+=", tok!"-", tok!"--", tok!"-=", tok!":", tok!";", tok!"<", tok!"<<", tok!"<<=", tok!"<=", tok!"<>", tok!"<>=", tok!"=", tok!"==", tok!"=>", tok!">", tok!">=", tok!">>", tok!">>=", tok!">>>", tok!">>>=", tok!"?", tok!"@", tok!"[", tok!"]", tok!"^", tok!"^=", tok!"^^", tok!"^^=", tok!"{", tok!"|", tok!"|=", tok!"||", tok!"}", tok!"~", tok!"~="); /** * Returns: true if the given ID type is for an operator. */ public bool isOperator(IdType type) nothrow pure @safe @nogc { switch (type) { foreach (T; Operators) { case T: return true; } default: return false; } } /** * Keyword token types. */ public alias Keywords = AliasSeq!(tok!"abstract", tok!"alias", tok!"align", tok!"asm", tok!"assert", tok!"auto", tok!"break", tok!"case", tok!"cast", tok!"catch", tok!"class", tok!"const", tok!"continue", tok!"debug", tok!"default", tok!"delegate", tok!"delete", tok!"deprecated", tok!"do", tok!"else", tok!"enum", tok!"export", tok!"extern", tok!"false", tok!"final", tok!"finally", tok!"for", tok!"foreach", tok!"foreach_reverse", tok!"function", tok!"goto", tok!"if", tok!"immutable", tok!"import", tok!"in", tok!"inout", tok!"interface", tok!"invariant", tok!"is", tok!"lazy", tok!"macro", tok!"mixin", tok!"module", tok!"new", tok!"nothrow", tok!"null", tok!"out", tok!"override", tok!"package", tok!"pragma", tok!"private", tok!"protected", tok!"public", tok!"pure", tok!"ref", tok!"return", tok!"scope", tok!"shared", tok!"static", tok!"struct", tok!"super", tok!"switch", tok!"synchronized", tok!"template", tok!"this", tok!"throw", tok!"true", tok!"try", tok!"typedef", tok!"typeid", tok!"typeof", tok!"union", tok!"unittest", tok!"version", tok!"volatile", tok!"while", tok!"with", tok!"__DATE__", tok!"__EOF__", tok!"__FILE__", tok!"__FILE_FULL_PATH__", tok!"__FUNCTION__", tok!"__gshared", tok!"__LINE__", tok!"__MODULE__", tok!"__parameters", tok!"__PRETTY_FUNCTION__", tok!"__TIME__", tok!"__TIMESTAMP__", tok!"__traits", tok!"__vector", tok!"__VENDOR__", tok!"__VERSION__"); /** * Returns: true if the given ID type is for a keyword. */ public bool isKeyword(IdType type) pure nothrow @safe @nogc { switch (type) { foreach (T; Keywords) { case T: return true; } default: return false; } } /** * String literal token types */ public alias StringLiterals = AliasSeq!(tok!"dstringLiteral", tok!"stringLiteral", tok!"wstringLiteral"); /** * Returns: true if the given ID type is for a string literal. */ public bool isStringLiteral(IdType type) pure nothrow @safe @nogc { switch (type) { foreach (T; StringLiterals) { case T: return true; } default: return false; } } /** * Protection token types. */ public alias Protections = AliasSeq!(tok!"export", tok!"package", tok!"private", tok!"public", tok!"protected"); /** * Returns: true if the given ID type is for a protection attribute. */ public bool isProtection(IdType type) pure nothrow @safe @nogc { switch (type) { foreach (T; Protections) { case T: return true; } default: return false; } } public alias SpecialTokens = AliasSeq!(tok!"__DATE__", tok!"__TIME__", tok!"__TIMESTAMP__", tok!"__VENDOR__", tok!"__VERSION__", tok!"__FILE__", tok!"__FILE_FULL_PATH__", tok!"__LINE__", tok!"__MODULE__", tok!"__FUNCTION__", tok!"__PRETTY_FUNCTION__"); public bool isSpecialToken(IdType type) pure nothrow @safe @nogc { switch (type) { foreach (T; SpecialTokens) { case T: return true; } default: return false; } } public alias Literals = AliasSeq!(StringLiterals, NumberLiterals, tok!"characterLiteral", SpecialTokens, tok!"true", tok!"false", tok!"null", tok!"$"); public bool isLiteral(IdType type) pure nothrow @safe @nogc { switch (type) { foreach (T; Literals) { case T: return true; } default: return false; } } /** * Returns: an array of tokens lexed from the given source code to the output range. All * whitespace tokens are skipped and comments are attached to the token nearest * to them. */ const(Token)[] getTokensForParser(R)(R sourceCode, LexerConfig config, StringCache* cache) if (is(Unqual!(ElementEncodingType!R) : ubyte) && isDynamicArray!R) { enum CommentType : ubyte { notDoc, line, block } static CommentType commentType(string comment) pure nothrow @safe { if (comment.length < 3) return CommentType.notDoc; if (comment[0 ..3] == "///") return CommentType.line; if (comment[0 ..3] == "/++" || comment[0 ..3] == "/**") return CommentType.block; return CommentType.notDoc; } config.whitespaceBehavior = WhitespaceBehavior.skip; config.commentBehavior = CommentBehavior.noIntern; auto leadingCommentAppender = appender!(char[])(); leadingCommentAppender.reserve(1024); auto trailingCommentAppender = appender!(char[])(); trailingCommentAppender.reserve(1024); bool hadDdoc; string empty = cache.intern(""); auto output = appender!(typeof(return))(); auto lexer = DLexer(sourceCode, config, cache); size_t tokenCount; loop: while (!lexer.empty) switch (lexer.front.type) { case tok!"specialTokenSequence": case tok!"whitespace": lexer.popFront(); break; case tok!"comment": final switch (commentType(lexer.front.text)) { case CommentType.block: case CommentType.line: if (tokenCount > 0 && lexer.front.line == output.data[tokenCount - 1].line) { if (!trailingCommentAppender.data.empty) trailingCommentAppender.put('\n'); unDecorateComment(lexer.front.text, trailingCommentAppender); hadDdoc = true; } else { if (!leadingCommentAppender.data.empty) leadingCommentAppender.put('\n'); unDecorateComment(lexer.front.text, leadingCommentAppender); hadDdoc = true; } lexer.popFront(); break; case CommentType.notDoc: lexer.popFront(); break; } break; case tok!"__EOF__": if (!trailingCommentAppender.data.empty) (cast() output.data[$ - 1].trailingComment) = cache.intern(cast(string) trailingCommentAppender.data); break loop; default: Token t = lexer.front; lexer.popFront(); tokenCount++; if (!output.data.empty && !trailingCommentAppender.data.empty) { (cast() output.data[$ - 1].trailingComment) = cache.intern(cast(string) trailingCommentAppender.data); hadDdoc = false; } t.comment = leadingCommentAppender.data.length > 0 ? cache.intern(cast(string) leadingCommentAppender.data) : (hadDdoc ? empty : null); leadingCommentAppender.clear(); trailingCommentAppender.clear(); hadDdoc = false; output.put(t); break; } return output.data; } /** * The D lexer struct. */ public struct DLexer { mixin Lexer!(Token, lexIdentifier, isSeparating, operators, dynamicTokens, keywords, pseudoTokenHandlers); /// @disable this(); /** * Params: * range = the bytes that compose the source code that will be lexed. * config = the lexer configuration to use. * cache = the string interning cache for de-duplicating identifiers and * other token text. * haveSSE42 = Parse streaming SIMD Extensions 4.2 in inline assembly */ this(R)(R range, const LexerConfig config, StringCache* cache, bool haveSSE42 = sse42()) pure nothrow @safe if (is(Unqual!(ElementEncodingType!R) : ubyte) && isDynamicArray!R) { this.haveSSE42 = haveSSE42; auto r = (range.length >= 3 && range[0] == 0xef && range[1] == 0xbb && range[2] == 0xbf) ? range[3 .. $] : range; this.range = LexerRange(cast(const(ubyte)[]) r); this.config = config; this.cache = cache; popFront(); } /// public void popFront()() pure nothrow @safe { do _popFront(); while (config.whitespaceBehavior == WhitespaceBehavior.skip && _front.type == tok!"whitespace"); } private pure nothrow @safe: bool isWhitespace() { switch (range.bytes[range.index]) { case ' ': case '\r': case '\n': case '\t': case '\v': case '\f': return true; case 0xe2: auto peek = range.peek(2); return peek.length == 2 && peek[0] == 0x80 && (peek[1] == 0xa8 || peek[1] == 0xa9); default: return false; } } void popFrontWhitespaceAware() { switch (range.bytes[range.index]) { case '\r': range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '\n') { range.popFront(); range.incrementLine(); } else range.incrementLine(); return; case '\n': range.popFront(); range.incrementLine(); return; case 0xe2: auto lookahead = range.peek(3); if (lookahead.length == 3 && lookahead[1] == 0x80 && (lookahead[2] == 0xa8 || lookahead[2] == 0xa9)) { range.index+=3; range.column+=3; range.incrementLine(); return; } else { range.popFront(); return; } default: range.popFront(); return; } } void lexWhitespace(ref Token token) @trusted { mixin (tokenStart); loop: do { version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { skip!(true, '\t', ' ', '\v', '\f')(range.bytes.ptr + range.index, &range.index, &range.column); } } switch (range.bytes[range.index]) { case '\r': range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '\n') { range.popFront(); } range.column = 1; range.line += 1; break; case '\n': range.popFront(); range.column = 1; range.line += 1; break; case ' ': case '\t': case '\v': case '\f': range.popFront(); break; case 0xe2: if (range.index + 2 >= range.bytes.length) break loop; if (range.bytes[range.index + 1] != 0x80) break loop; if (range.bytes[range.index + 2] == 0xa8 || range.bytes[range.index + 2] == 0xa9) { range.index += 3; range.column += 3; range.column = 1; range.line += 1; break; } break loop; default: break loop; } } while (!(range.index >= range.bytes.length)); string text = config.whitespaceBehavior == WhitespaceBehavior.include ? cache.intern(range.slice(mark)) : ""; token = Token(tok!"whitespace", text, line, column, index); } void lexNumber(ref Token token) { mixin (tokenStart); if (range.bytes[range.index] == '0' && range.index + 1 < range.bytes.length) { immutable ahead = range.bytes[range.index + 1]; switch (ahead) { case 'x': case 'X': range.index += 2; range.column += 2; lexHex(token, mark, line, column, index); return; case 'b': case 'B': range.index += 2; range.column += 2; lexBinary(token, mark, line, column, index); return; default: lexDecimal(token, mark, line, column, index); return; } } else lexDecimal(token, mark, line, column, index); } void lexHex(ref Token token) { mixin (tokenStart); lexHex(token, mark, line, column, index); } void lexHex(ref Token token, size_t mark, size_t line, size_t column, size_t index) @trusted { IdType type = tok!"intLiteral"; bool foundDot; hexLoop: while (!(range.index >= range.bytes.length)) { switch (range.bytes[range.index]) { case 'a': .. case 'f': case 'A': .. case 'F': case '0': .. case '9': case '_': version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { immutable ulong i = rangeMatch!(false, '0', '9', 'a', 'f', 'A', 'F', '_', '_') (range.bytes.ptr + range.index); range.column += i; range.index += i; } else range.popFront(); } else range.popFront(); break; case 'u': case 'U': lexIntSuffix(type); break hexLoop; case 'i': if (foundDot) lexFloatSuffix(type); break hexLoop; case 'L': if (foundDot) lexFloatSuffix(type); else lexIntSuffix(type); break hexLoop; case 'p': case 'P': lexExponent(type); break hexLoop; case '.': if (foundDot || !(range.index + 1 < range.bytes.length) || range.peekAt(1) == '.') break hexLoop; else { // The following bit of silliness tries to tell the // difference between "int dot identifier" and // "double identifier". if (range.index + 1 < range.bytes.length) { switch (range.peekAt(1)) { case '0': .. case '9': case 'A': .. case 'F': case 'a': .. case 'f': goto doubleLiteral; default: break hexLoop; } } else { doubleLiteral: range.popFront(); foundDot = true; type = tok!"doubleLiteral"; } } break; default: break hexLoop; } } token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexBinary(ref Token token) { mixin (tokenStart); return lexBinary(token, mark, line, column, index); } void lexBinary(ref Token token, size_t mark, size_t line, size_t column, size_t index) @trusted { IdType type = tok!"intLiteral"; binaryLoop: while (!(range.index >= range.bytes.length)) { switch (range.bytes[range.index]) { case '0': case '1': case '_': version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { immutable ulong i = rangeMatch!(false, '0', '1', '_', '_')( range.bytes.ptr + range.index); range.column += i; range.index += i; } else range.popFront(); } else range.popFront(); break; case 'u': case 'U': case 'L': lexIntSuffix(type); break binaryLoop; default: break binaryLoop; } } token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexDecimal(ref Token token) { mixin (tokenStart); lexDecimal(token, mark, line, column, index); } void lexDecimal(ref Token token, size_t mark, size_t line, size_t column, size_t index) @trusted { bool foundDot = range.bytes[range.index] == '.'; IdType type = tok!"intLiteral"; if (foundDot) { range.popFront(); type = tok!"doubleLiteral"; } decimalLoop: while (!(range.index >= range.bytes.length)) { switch (range.bytes[range.index]) { case '0': .. case '9': case '_': version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { immutable ulong i = rangeMatch!(false, '0', '9', '_', '_')(range.bytes.ptr + range.index); range.column += i; range.index += i; } else range.popFront(); } else range.popFront(); break; case 'u': case 'U': if (!foundDot) lexIntSuffix(type); break decimalLoop; case 'i': lexFloatSuffix(type); break decimalLoop; case 'L': if (foundDot) lexFloatSuffix(type); else lexIntSuffix(type); break decimalLoop; case 'f': case 'F': lexFloatSuffix(type); break decimalLoop; case 'e': case 'E': lexExponent(type); break decimalLoop; case '.': if (foundDot || !(range.index + 1 < range.bytes.length) || range.peekAt(1) == '.') break decimalLoop; else { // The following bit of silliness tries to tell the // difference between "int dot identifier" and // "double identifier". if (range.index + 1 < range.bytes.length) { immutable ch = range.peekAt(1); if (ch <= 0x2f || (ch >= '0' && ch <= '9') || (ch >= ':' && ch <= '@') || (ch >= '[' && ch <= '^') || (ch >= '{' && ch <= '~') || ch == '`' || ch == '_') { goto doubleLiteral; } else break decimalLoop; } else { doubleLiteral: range.popFront(); foundDot = true; type = tok!"doubleLiteral"; } } break; default: break decimalLoop; } } token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexIntSuffix(ref IdType type) pure nothrow @safe { bool secondPass; if (range.bytes[range.index] == 'u' || range.bytes[range.index] == 'U') { U: if (type == tok!"intLiteral") type = tok!"uintLiteral"; else type = tok!"ulongLiteral"; range.popFront(); if (secondPass) return; if (range.index < range.bytes.length && (range.bytes[range.index] == 'L' || range.bytes[range.index] == 'l')) goto L; goto I; } if (range.bytes[range.index] == 'L' || range.bytes[range.index] == 'l') { L: if (type == tok!"uintLiteral") type = tok!"ulongLiteral"; else type = tok!"longLiteral"; range.popFront(); if (range.index < range.bytes.length && (range.bytes[range.index] == 'U' || range.bytes[range.index] == 'u')) { secondPass = true; goto U; } goto I; } I: if (range.index < range.bytes.length && range.bytes[range.index] == 'i') { warning("Complex number literals are deprecated"); range.popFront(); if (type == tok!"longLiteral" || type == tok!"ulongLiteral") type = tok!"idoubleLiteral"; else type = tok!"ifloatLiteral"; } } void lexFloatSuffix(ref IdType type) pure nothrow @safe { switch (range.bytes[range.index]) { case 'L': range.popFront(); type = tok!"doubleLiteral"; break; case 'f': case 'F': range.popFront(); type = tok!"floatLiteral"; break; default: break; } if (range.index < range.bytes.length && range.bytes[range.index] == 'i') { warning("Complex number literals are deprecated"); range.popFront(); if (type == tok!"floatLiteral") type = tok!"ifloatLiteral"; else type = tok!"idoubleLiteral"; } } void lexExponent(ref IdType type) pure nothrow @safe { range.popFront(); bool foundSign = false; bool foundDigit = false; while (range.index < range.bytes.length) { switch (range.bytes[range.index]) { case '-': case '+': if (foundSign) { if (!foundDigit) error("Expected an exponent"); return; } foundSign = true; range.popFront(); break; case '0': .. case '9': case '_': foundDigit = true; range.popFront(); break; case 'L': case 'f': case 'F': case 'i': lexFloatSuffix(type); return; default: if (!foundDigit) error("Expected an exponent"); return; } } } void lexScriptLine(ref Token token) { mixin (tokenStart); while (!(range.index >= range.bytes.length) && !isNewline) { range.popFront(); } token = Token(tok!"scriptLine", cache.intern(range.slice(mark)), line, column, index); } void lexSpecialTokenSequence(ref Token token) { mixin (tokenStart); while (!(range.index >= range.bytes.length) && !isNewline) { range.popFront(); } token = Token(tok!"specialTokenSequence", cache.intern(range.slice(mark)), line, column, index); } void lexSlashStarComment(ref Token token) @trusted { mixin (tokenStart); IdType type = tok!"comment"; range.popFrontN(2); while (range.index < range.bytes.length) { version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) skip!(false, '\r', '\n', '/', '*', 0xe2)(range.bytes.ptr + range.index, &range.index, &range.column); } if (range.bytes[range.index] == '*') { range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '/') { range.popFront(); break; } } else popFrontWhitespaceAware(); } if (config.commentBehavior == CommentBehavior.intern) token = Token(type, cache.intern(range.slice(mark)), line, column, index); else token = Token(type, cast(string) range.slice(mark), line, column, index); } void lexSlashSlashComment(ref Token token) @trusted { mixin (tokenStart); IdType type = tok!"comment"; range.popFrontN(2); while (range.index < range.bytes.length) { version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { skip!(false, '\r', '\n', 0xe2)(range.bytes.ptr + range.index, &range.index, &range.column); } } if (range.bytes[range.index] == '\r' || range.bytes[range.index] == '\n') break; range.popFront(); } if (config.commentBehavior == CommentBehavior.intern) token = Token(type, cache.intern(range.slice(mark)), line, column, index); else token = Token(type, cast(string) range.slice(mark), line, column, index); } void lexSlashPlusComment(ref Token token) @trusted { mixin (tokenStart); IdType type = tok!"comment"; range.index += 2; range.column += 2; int depth = 1; while (depth > 0 && !(range.index >= range.bytes.length)) { version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { skip!(false, '+', '/', '\\', '\r', '\n', 0xe2)(range.bytes.ptr + range.index, &range.index, &range.column); } } if (range.bytes[range.index] == '+') { range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '/') { range.popFront(); depth--; } } else if (range.bytes[range.index] == '/') { range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '+') { range.popFront(); depth++; } } else popFrontWhitespaceAware(); } if (config.commentBehavior == CommentBehavior.intern) token = Token(type, cache.intern(range.slice(mark)), line, column, index); else token = Token(type, cast(string) range.slice(mark), line, column, index); } void lexStringLiteral(ref Token token) @trusted { mixin (tokenStart); range.popFront(); while (true) { if (range.index >= range.bytes.length) { error("Error: unterminated string literal"); token = Token(tok!""); return; } version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { skip!(false, '"', '\\', '\r', '\n', 0xe2)(range.bytes.ptr + range.index, &range.index, &range.column); } } if (range.bytes[range.index] == '"') { range.popFront(); break; } else if (range.bytes[range.index] == '\\') { lexEscapeSequence(); } else popFrontWhitespaceAware(); } IdType type = tok!"stringLiteral"; lexStringSuffix(type); token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexWysiwygString(ref Token token) @trusted { mixin (tokenStart); IdType type = tok!"stringLiteral"; immutable bool backtick = range.bytes[range.index] == '`'; if (backtick) { range.popFront(); while (true) { if (range.index >= range.bytes.length) { error("Error: unterminated string literal"); token = Token(tok!""); return; } version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { skip!(false, '\r', '\n', 0xe2, '`')(range.bytes.ptr + range.index, &range.index, &range.column); } } if (range.bytes[range.index] == '`') { range.popFront(); break; } else popFrontWhitespaceAware(); } } else { range.popFront(); if (range.index >= range.bytes.length) { error("Error: unterminated string literal"); token = Token(tok!""); return; } range.popFront(); while (true) { if (range.index >= range.bytes.length) { error("Error: unterminated string literal"); token = Token(tok!""); return; } else if (range.bytes[range.index] == '"') { range.popFront(); break; } else popFrontWhitespaceAware(); } } lexStringSuffix(type); token = Token(type, cache.intern(range.slice(mark)), line, column, index); } private ubyte lexStringSuffix(ref IdType type) pure nothrow @safe { if (range.index >= range.bytes.length) { type = tok!"stringLiteral"; return 0; } else { switch (range.bytes[range.index]) { case 'w': range.popFront(); type = tok!"wstringLiteral"; return 'w'; case 'd': range.popFront(); type = tok!"dstringLiteral"; return 'd'; case 'c': range.popFront(); type = tok!"stringLiteral"; return 'c'; default: type = tok!"stringLiteral"; return 0; } } } void lexDelimitedString(ref Token token) { mixin (tokenStart); range.index += 2; range.column += 2; ubyte open; ubyte close; switch (range.bytes[range.index]) { case '<': open = '<'; close = '>'; range.popFront(); lexNormalDelimitedString(token, mark, line, column, index, open, close); break; case '{': open = '{'; close = '}'; range.popFront(); lexNormalDelimitedString(token, mark, line, column, index, open, close); break; case '[': open = '['; close = ']'; range.popFront(); lexNormalDelimitedString(token, mark, line, column, index, open, close); break; case '(': open = '('; close = ')'; range.popFront(); lexNormalDelimitedString(token, mark, line, column, index, open, close); break; default: lexHeredocString(token, mark, line, column, index); break; } } void lexNormalDelimitedString(ref Token token, size_t mark, size_t line, size_t column, size_t index, ubyte open, ubyte close) { int depth = 1; while (!(range.index >= range.bytes.length) && depth > 0) { if (range.bytes[range.index] == open) { depth++; range.popFront(); } else if (range.bytes[range.index] == close) { depth--; range.popFront(); if (depth <= 0) { if (range.bytes[range.index] == '"') { range.popFront(); } else { error("Error: `\"` expected to end delimited string literal"); token = Token(tok!""); return; } } } else popFrontWhitespaceAware(); } IdType type = tok!"stringLiteral"; lexStringSuffix(type); token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexHeredocString(ref Token token, size_t mark, size_t line, size_t column, size_t index) { Token ident; lexIdentifier(ident); if (isNewline()) popFrontWhitespaceAware(); else error("Newline expected"); while (!(range.index >= range.bytes.length)) { if (isNewline()) { popFrontWhitespaceAware(); if (!range.canPeek(ident.text.length)) { error(ident.text ~ " expected"); break; } if (range.peek(ident.text.length - 1) == ident.text) { range.popFrontN(ident.text.length); break; } } else { range.popFront(); } } if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '"') { range.popFront(); } else error("`\"` expected"); IdType type = tok!"stringLiteral"; lexStringSuffix(type); token = Token(type, cache.intern(range.slice(mark)), line, column, index); } void lexTokenString(ref Token token) { mixin (tokenStart); assert (range.bytes[range.index] == 'q'); range.popFront(); assert (range.bytes[range.index] == '{'); range.popFront(); auto app = appender!string(); app.put("q{"); int depth = 1; immutable WhitespaceBehavior oldWhitespace = config.whitespaceBehavior; immutable StringBehavior oldString = config.stringBehavior; config.whitespaceBehavior = WhitespaceBehavior.include; config.stringBehavior = StringBehavior.source; scope (exit) { config.whitespaceBehavior = oldWhitespace; config.stringBehavior = oldString; } advance(_front); while (depth > 0 && !empty) { auto t = front(); if (t.text is null) app.put(str(t.type)); else app.put(t.text); if (t.type == tok!"}") { depth--; if (depth > 0) popFront(); } else if (t.type == tok!"{") { depth++; popFront(); } else popFront(); } IdType type = tok!"stringLiteral"; auto b = lexStringSuffix(type); if (b != 0) app.put(b); token = Token(type, cache.intern(cast(const(ubyte)[]) app.data), line, column, index); } void lexHexString(ref Token token) { mixin (tokenStart); range.index += 2; range.column += 2; loop: while (true) { if (range.index >= range.bytes.length) { error("Error: unterminated hex string literal"); token = Token(tok!""); return; } else if (isWhitespace()) popFrontWhitespaceAware(); else switch (range.bytes[range.index]) { case '0': .. case '9': case 'A': .. case 'F': case 'a': .. case 'f': range.popFront(); break; case '"': range.popFront(); break loop; default: error("Error: invalid character in hex string"); token = Token(tok!""); return; } } IdType type = tok!"stringLiteral"; lexStringSuffix(type); token = Token(type, cache.intern(range.slice(mark)), line, column, index); } bool lexEscapeSequence() { range.popFront(); if (range.index >= range.bytes.length) { error("Error: non-terminated character escape sequence."); return false; } switch (range.bytes[range.index]) { case '\'': case '"': case '?': case '\\': case 'a': case 'b': case 'f': case 'n': case 'r': case 't': case 'v': range.popFront(); break; case 'x': range.popFront(); foreach (i; 0 .. 2) { if (range.index >= range.bytes.length) { error("Error: 2 hex digits expected."); return false; } switch (range.bytes[range.index]) { case '0': .. case '9': case 'a': .. case 'f': case 'A': .. case 'F': range.popFront(); break; default: error("Error: 2 hex digits expected."); return false; } } break; case '0': if (!(range.index + 1 < range.bytes.length) || ((range.index + 1 < range.bytes.length) && range.peekAt(1) == '\'')) { range.popFront(); break; } goto case; case '1': .. case '7': for (size_t i = 0; i < 3 && !(range.index >= range.bytes.length) && range.bytes[range.index] >= '0' && range.bytes[range.index] <= '7'; i++) range.popFront(); break; case 'u': range.popFront(); foreach (i; 0 .. 4) { if (range.index >= range.bytes.length) { error("Error: at least 4 hex digits expected."); return false; } switch (range.bytes[range.index]) { case '0': .. case '9': case 'a': .. case 'f': case 'A': .. case 'F': range.popFront(); break; default: error("Error: at least 4 hex digits expected."); return false; } } break; case 'U': range.popFront(); foreach (i; 0 .. 8) { if (range.index >= range.bytes.length) { error("Error: at least 8 hex digits expected."); return false; } switch (range.bytes[range.index]) { case '0': .. case '9': case 'a': .. case 'f': case 'A': .. case 'F': range.popFront(); break; default: error("Error: at least 8 hex digits expected."); return false; } } break; default: while (true) { if (range.index >= range.bytes.length) { error("Error: non-terminated character escape sequence."); return false; } if (range.bytes[range.index] == ';') { range.popFront(); break; } else { range.popFront(); } } } return true; } void lexCharacterLiteral(ref Token token) { mixin (tokenStart); range.popFront(); if (range.empty) goto err; if (range.bytes[range.index] == '\\') lexEscapeSequence(); else if (range.bytes[range.index] == '\'') { range.popFront(); token = Token(tok!"characterLiteral", cache.intern(range.slice(mark)), line, column, index); } else if (range.bytes[range.index] & 0x80) { while (range.bytes[range.index] & 0x80) range.popFront(); } else popFrontWhitespaceAware(); if (range.index < range.bytes.length && range.bytes[range.index] == '\'') { range.popFront(); token = Token(tok!"characterLiteral", cache.intern(range.slice(mark)), line, column, index); } else { err: error("Error: Expected `'` to end character literal"); token = Token(tok!""); } } void lexIdentifier(ref Token token) @trusted { mixin (tokenStart); if (isSeparating(0)) { error("Invalid identifier"); range.popFront(); } while (true) { version (iasm64NotWindows) { if (haveSSE42 && range.index + 16 < range.bytes.length) { immutable ulong i = rangeMatch!(false, 'a', 'z', 'A', 'Z', '_', '_') (range.bytes.ptr + range.index); range.column += i; range.index += i; } } if (isSeparating(0)) break; else range.popFront(); } token = Token(tok!"identifier", cache.intern(range.slice(mark)), line, column, index); } void lexDot(ref Token token) { mixin (tokenStart); if (!(range.index + 1 < range.bytes.length)) { range.popFront(); token = Token(tok!".", null, line, column, index); return; } switch (range.peekAt(1)) { case '0': .. case '9': lexNumber(token); return; case '.': range.popFront(); range.popFront(); if (!(range.index >= range.bytes.length) && range.bytes[range.index] == '.') { range.popFront(); token = Token(tok!"...", null, line, column, index); } else token = Token(tok!"..", null, line, column, index); return; default: range.popFront(); token = Token(tok!".", null, line, column, index); return; } } void lexLongNewline(ref Token token) @nogc { mixin (tokenStart); range.popFront(); range.popFront(); range.popFront(); range.incrementLine(); string text = config.whitespaceBehavior == WhitespaceBehavior.include ? cache.intern(range.slice(mark)) : ""; token = Token(tok!"whitespace", text, line, column, index); } bool isNewline() @nogc { if (range.bytes[range.index] == '\n') return true; if (range.bytes[range.index] == '\r') return true; return (range.bytes[range.index] & 0x80) && (range.index + 2 < range.bytes.length) && (range.peek(2) == "\u2028" || range.peek(2) == "\u2029"); } bool isSeparating(size_t offset) @nogc { enum : ubyte { n, y, m // no, yes, maybe } if (range.index + offset >= range.bytes.length) return true; auto c = range.bytes[range.index + offset]; static immutable ubyte[256] LOOKUP_TABLE = [ y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, y, n, n, n, n, n, n, n, n, n, n, y, y, y, y, y, y, y, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, y, y, y, y, n, y, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, n, y, y, y, y, y, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m, m ]; immutable ubyte result = LOOKUP_TABLE[c]; if (result == n) return false; if (result == y) return true; if (result == m) { auto r = range; range.popFrontN(offset); return (r.canPeek(2) && (r.peek(2) == "\u2028" || r.peek(2) == "\u2029")); } assert (false); } enum tokenStart = q{ size_t index = range.index; size_t column = range.column; size_t line = range.line; auto mark = range.mark(); }; void error(string message) { messages ~= Message(range.line, range.column, message, true); } void warning(string message) { messages ~= Message(range.line, range.column, message, false); assert (messages.length > 0); } static struct Message { size_t line; size_t column; string message; bool isError; } Message[] messages; StringCache* cache; LexerConfig config; bool haveSSE42; } /// copy from phobos b/c we need to build on older versions of dmd /// Returns : the next power of two from a given value private static size_t nextPow2(size_t value) { import core.bitop : bsr; return 1 << bsr(value) + 1; } /** * Creates a token range from the given source code. Creates a default lexer * configuration and a GC-managed string cache. */ public auto byToken(R)(R range) if (is(Unqual!(ElementEncodingType!R) : ubyte) && isDynamicArray!R) { uint bc = cast(uint)((range.length > 2^^31UL) ? 2^^31 : nextPow2(1 + range.length / 32)); LexerConfig config; StringCache* cache = new StringCache(bc); return DLexer(range, config, cache); } /** * Creates a token range from the given source code. Uses the given string * cache. */ public auto byToken(R)(R range, StringCache* cache) if (is(Unqual!(ElementEncodingType!R) : ubyte) && isDynamicArray!R) { LexerConfig config; return DLexer(range, config, cache); } /** * Creates a token range from the given source code. Uses the provided lexer * configuration and string cache. */ public auto byToken(R)(R range, const LexerConfig config, StringCache* cache) if (is(Unqual!(ElementEncodingType!R) : ubyte) && isDynamicArray!R) { return DLexer(range, config, cache); } /** * Removes "decoration" such as leading whitespace, leading + and * characters, * and places the result into the given output range */ public void unDecorateComment(T)(string comment, auto ref T outputRange) if (isOutputRange!(T, string)) in { assert (comment.length >= 3); } body { import std.string : lineSplitter, stripRight; static void adjustBeginningAndEnd(string s, ref size_t a, ref size_t b) pure nothrow @nogc @safe { immutable char c = s[1]; while (a < b && s[a] == c) a++; while (b > a && s[b] == c) b--; b++; } string leadingChars; size_t i = 3; size_t j; bool hasOutput = false; bool lastWasBlank = false; switch (comment[0 .. 3]) { case "///": j = comment.length; foreach (line; lineSplitter(comment)) { auto l = line[3 .. $]; if (leadingChars.empty) { size_t k = 0; while (k < l.length && (l[k] == ' ' || l[k] == '\t')) k++; leadingChars = l[0 .. k]; } immutable string stripped = l.stripRight(); if (hasOutput) outputRange.put('\n'); else hasOutput = true; if (stripped.length >= leadingChars.length && stripped.startsWith(leadingChars)) outputRange.put(stripped[leadingChars.length .. $]); else outputRange.put(stripped); } break; case "/++": case "/**": if (comment.length == 3) { comment = ""; goto default; } j = comment.length - 2; // Skip beginning and ending stars and plusses adjustBeginningAndEnd(comment, i, j); foreach (line; lineSplitter(comment[i .. j])) { immutable string stripped = line.stripRight(); if (leadingChars.empty) { size_t k = 0; while (k < line.length && (line[k] == ' ' || line[k] == '\t')) k++; if (k < line.length && line[k] == comment[1]) { k++; while (k < line.length && (line[k] == ' ' || line[k] == '\t')) k++; } if (k == stripped.length) continue; leadingChars = line[0 .. k]; } if (stripped.startsWith(leadingChars)) { if (stripped.length > leadingChars.length) { if (hasOutput) outputRange.put('\n'); hasOutput = true; if (lastWasBlank) outputRange.put('\n'); lastWasBlank = false; outputRange.put(stripped[leadingChars.length .. $]); } } else if (hasOutput && stripped.length == leadingChars.stripRight().length) lastWasBlank = true; else if (!stripped.empty && !leadingChars.startsWith(stripped)) { if (hasOutput) outputRange.put('\n'); hasOutput = true; if (lastWasBlank) outputRange.put('\n'); lastWasBlank = false; outputRange.put(stripped); } else lastWasBlank = false; } break; default: outputRange.put(comment); break; } } /// unittest { import std.array:array, appender; import std.stdio:stderr; stderr.writeln("Running unittest for unDecorateComment..."); string[] inputs = [ "/***************\n*******************/", "/***************\n *\n ******************/", "/**\n*/", "/** */", "/***/", "/** abcde */", "/// abcde\n/// abcde", "/**\n * stuff\n */", "/**\n *\n * stuff\n */", "/**\n *\n * stuff\n *\n */", "/**\n *\n * stuff\n *\n*/", "/**\n * abcde\n * abcde \n */", "/**\n * abcde\n *\n * abcde\n */", ]; string[] outputs = [ "", "", "", "", "", "abcde", "abcde\nabcde", "stuff", "stuff", "stuff", "stuff", "abcde\n abcde", "abcde\n\nabcde" ]; assert(inputs.length == outputs.length); foreach (pair; zip(inputs, outputs)) { foreach (b; [true, false]) { auto app = appender!string(); unDecorateComment(b ? pair[0] : pair[0].replace("*", "+"), app); assert(pair[1] == app.data, "[[" ~ pair[0] ~ "]] => [[" ~ app.data ~ "]]"); } } stderr.writeln("Unittest for unDecorateComment passed."); } /** * The string cache is used for string interning. * * It will only store a single copy of any string that it is asked to hold. * Interned strings can be compared for equality by comparing their $(B .ptr) * field. * * Default and postbilt constructors are disabled. When a StringCache goes out * of scope, the memory held by it is freed. * * See_also: $(LINK http://en.wikipedia.org/wiki/String_interning) */ struct StringCache { public pure nothrow @nogc: @disable this(); @disable this(this); /** * Params: bucketCount = the initial number of buckets. Must be a * power of two */ this(size_t bucketCount) nothrow @trusted @nogc in { import core.bitop : popcnt; static if (size_t.sizeof == 8) { immutable low = popcnt(cast(uint) bucketCount); immutable high = popcnt(cast(uint) (bucketCount >> 32)); assert ((low == 0 && high == 1) || (low == 1 && high == 0)); } else { static assert (size_t.sizeof == 4); assert (popcnt(cast(uint) bucketCount) == 1); } } body { buckets = (cast(Node**) calloc((Node*).sizeof, bucketCount))[0 .. bucketCount]; } ~this() { Block* current = rootBlock; while (current !is null) { Block* prev = current; current = current.next; free(cast(void*) prev); } foreach (nodePointer; buckets) { Node* currentNode = nodePointer; while (currentNode !is null) { if (currentNode.mallocated) free(currentNode.str.ptr); Node* prev = currentNode; currentNode = currentNode.next; free(prev); } } rootBlock = null; free(buckets.ptr); buckets = null; } /** * Caches a string. */ string intern(const(ubyte)[] str) @safe { if (str is null || str.length == 0) return ""; return _intern(str); } /** * ditto */ string intern(string str) @trusted { return intern(cast(ubyte[]) str); } /** * The default bucket count for the string cache. */ static enum defaultBucketCount = 4096; private: string _intern(const(ubyte)[] bytes) @trusted { immutable uint hash = hashBytes(bytes); immutable size_t index = hash & (buckets.length - 1); Node* s = find(bytes, hash); if (s !is null) return cast(string) s.str; ubyte[] mem = void; bool mallocated = bytes.length > BIG_STRING; if (mallocated) mem = (cast(ubyte*) malloc(bytes.length))[0 .. bytes.length]; else mem = allocate(bytes.length); mem[] = bytes[]; Node* node = cast(Node*) malloc(Node.sizeof); node.str = mem; node.hash = hash; node.next = buckets[index]; node.mallocated = mallocated; buckets[index] = node; return cast(string) mem; } Node* find(const(ubyte)[] bytes, uint hash) @trusted { import std.algorithm : equal; immutable size_t index = hash & (buckets.length - 1); Node* node = buckets[index]; while (node !is null) { if (node.hash == hash && bytes == cast(ubyte[]) node.str) return node; node = node.next; } return node; } static uint hashBytes(const(ubyte)[] data) pure nothrow @trusted @nogc in { assert (data !is null); assert (data.length > 0); } body { immutable uint m = 0x5bd1e995; immutable int r = 24; uint h = cast(uint) data.length; while (data.length >= 4) { uint k = (cast(ubyte) data[3]) << 24 | (cast(ubyte) data[2]) << 16 | (cast(ubyte) data[1]) << 8 | (cast(ubyte) data[0]); k *= m; k ^= k >> r; k *= m; h *= m; h ^= k; data = data[4 .. $]; } switch (data.length & 3) { case 3: h ^= data[2] << 16; goto case; case 2: h ^= data[1] << 8; goto case; case 1: h ^= data[0]; h *= m; break; default: break; } h ^= h >> 13; h *= m; h ^= h >> 15; return h; } ubyte[] allocate(size_t numBytes) pure nothrow @trusted @nogc in { assert (numBytes != 0); } out (result) { assert (result.length == numBytes); } body { Block* r = rootBlock; size_t i = 0; while (i <= 3 && r !is null) { immutable size_t available = r.bytes.length; immutable size_t oldUsed = r.used; immutable size_t newUsed = oldUsed + numBytes; if (newUsed <= available) { r.used = newUsed; return r.bytes[oldUsed .. newUsed]; } i++; r = r.next; } Block* b = cast(Block*) calloc(Block.sizeof, 1); b.used = numBytes; b.next = rootBlock; rootBlock = b; return b.bytes[0 .. numBytes]; } static struct Node { ubyte[] str = void; Node* next = void; uint hash = void; bool mallocated = void; } static struct Block { Block* next; size_t used; enum BLOCK_CAPACITY = BLOCK_SIZE - size_t.sizeof - (void*).sizeof; ubyte[BLOCK_CAPACITY] bytes; } static assert (BLOCK_SIZE == Block.sizeof); enum BLOCK_SIZE = 1024 * 16; // If a string would take up more than 1/4 of a block, allocate it outside // of the block. enum BIG_STRING = BLOCK_SIZE / 4; Node*[] buckets; Block* rootBlock; } private extern(C) void* calloc(size_t, size_t) nothrow pure @nogc @trusted; private extern(C) void* malloc(size_t) nothrow pure @nogc @trusted; private extern(C) void free(void*) nothrow pure @nogc @trusted; unittest { auto source = cast(ubyte[]) q{ import std.stdio;}c; auto tokens = getTokensForParser(source, LexerConfig(), new StringCache(StringCache.defaultBucketCount)); assert (tokens.map!"a.type"().equal([tok!"import", tok!"identifier", tok!".", tok!"identifier", tok!";"])); } /// Test \x char sequence unittest { auto toks = (string s) => byToken(cast(ubyte[])s); // valid immutable hex = ['0','1','2','3','4','5','6','7','8','9','a','b','c','d','e','f','A','B','C','D','E','F']; auto source = ""; foreach (h1; hex) foreach (h2; hex) source ~= "'\\x" ~ h1 ~ h2 ~ "'"; assert (toks(source).filter!(t => t.type != tok!"characterLiteral").empty); // invalid assert (toks(`'\x'`).messages[0] == DLexer.Message(1,4,"Error: 2 hex digits expected.",true)); assert (toks(`'\x_'`).messages[0] == DLexer.Message(1,4,"Error: 2 hex digits expected.",true)); assert (toks(`'\xA'`).messages[0] == DLexer.Message(1,5,"Error: 2 hex digits expected.",true)); assert (toks(`'\xAY'`).messages[0] == DLexer.Message(1,5,"Error: 2 hex digits expected.",true)); assert (toks(`'\xXX'`).messages[0] == DLexer.Message(1,4,"Error: 2 hex digits expected.",true)); } version (iasm64NotWindows) { /** * Returns: */ ushort newlineMask(const ubyte*) pure nothrow @trusted @nogc { asm pure nothrow @nogc { naked; movdqu XMM1, [RDI]; mov RAX, 3; mov RDX, 16; mov R8, 0x0d0d0d0d0d0d0d0dL; movq XMM2, R8; shufpd XMM2, XMM2, 0; pcmpeqb XMM2, XMM1; mov R9, 0x0a0a0a0a0a0a0a0aL; movq XMM3, R9; shufpd XMM3, XMM3, 0; pcmpeqb XMM3, XMM1; mov R10, 0xe280a8L; movq XMM4, R10; pcmpestrm XMM4, XMM1, 0b01001100; movdqa XMM4, XMM0; mov R11, 0xe280a9L; movq XMM5, R11; pcmpestrm XMM5, XMM1, 0b01001100; movdqa XMM5, XMM0; mov RCX, 0x0a0d; dec RAX; movq XMM6, RCX; pcmpestrm XMM6, XMM1, 0b01001100; movdqa XMM6, XMM0; movdqa XMM7, XMM6; pslldq XMM7, 1; movdqa XMM0, XMM4; por XMM0, XMM5; por XMM7, XMM6; movdqa XMM1, XMM2; por XMM1, XMM3; pxor XMM7, XMM1; por XMM7, XMM0; por XMM7, XMM6; pmovmskb RAX, XMM7; and RAX, 0b0011_1111_1111_1111; ret; } } /** * Skips between 0 and 16 bytes that match (or do not match) one of the * given $(B chars). */ void skip(bool matching, chars...)(const ubyte*, ulong*, ulong*) pure nothrow @trusted @nogc if (chars.length <= 8) { enum constant = ByteCombine!chars; enum charsLength = chars.length; static if (matching) enum flags = 0b0001_0000; else enum flags = 0b0000_0000; asm pure nothrow @nogc { naked; movdqu XMM1, [RDX]; mov R10, constant; movq XMM2, R10; mov RAX, charsLength; mov RDX, 16; pcmpestri XMM2, XMM1, flags; add [RSI], RCX; add [RDI], RCX; ret; } } /** * Returns: the number of bytes starting at the given location that match * (or do not match if $(B invert) is true) the byte ranges in $(B chars). */ ulong rangeMatch(bool invert, chars...)(const ubyte*) pure nothrow @trusted @nogc { static assert (chars.length % 2 == 0); enum constant = ByteCombine!chars; static if (invert) enum rangeMatchFlags = 0b0000_0100; else enum rangeMatchFlags = 0b0001_0100; enum charsLength = chars.length; asm pure nothrow @nogc { naked; movdqu XMM1, [RDI]; mov R10, constant; movq XMM2, R10; mov RAX, charsLength; mov RDX, 16; pcmpestri XMM2, XMM1, rangeMatchFlags; mov RAX, RCX; ret; } } template ByteCombine(c...) { static assert (c.length <= 8); static if (c.length > 1) enum ulong ByteCombine = c[0] | (ByteCombine!(c[1..$]) << 8); else enum ulong ByteCombine = c[0]; } } unittest { import core.exception : RangeError; import std.exception : assertNotThrown; static immutable src1 = "/++"; static immutable src2 = "/**"; LexerConfig cf; StringCache ca = StringCache(16); assertNotThrown!RangeError(getTokensForParser(src1, cf, &ca)); assertNotThrown!RangeError(getTokensForParser(src2, cf, &ca)); }