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
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
|
/* inline.h
*
* Copyright (C) 2012 by Larry Wall and others
*
* You may distribute under the terms of either the GNU General Public
* License or the Artistic License, as specified in the README file.
*
* This file contains tables and code adapted from
* https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which requires this
* copyright notice:
Copyright (c) 2008-2009 Bjoern Hoehrmann <bjoern@hoehrmann.de>
Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the "Software"), to deal in
the Software without restriction, including without limitation the rights to
use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies
of the Software, and to permit persons to whom the Software is furnished to do
so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
*
* This file is a home for static inline functions that cannot go in other
* header files, because they depend on proto.h (included after most other
* headers) or struct definitions.
*
* Each section names the header file that the functions "belong" to.
*/
/* ------------------------------- av.h ------------------------------- */
/*
=for apidoc_section $AV
=for apidoc av_count
Returns the number of elements in the array C<av>. This is the true length of
the array, including any undefined elements. It is always the same as
S<C<av_top_index(av) + 1>>.
=cut
*/
PERL_STATIC_INLINE Size_t
Perl_av_count(pTHX_ AV *av)
{
PERL_ARGS_ASSERT_AV_COUNT;
assert(SvTYPE(av) == SVt_PVAV);
return AvFILL(av) + 1;
}
/* ------------------------------- av.c ------------------------------- */
/*
=for apidoc av_store_simple
This is a cut-down version of av_store that assumes that the array is
very straightforward - no magic, not readonly, and AvREAL - and that
C<key> is not negative. This function MUST NOT be used in situations
where any of those assumptions may not hold.
Stores an SV in an array. The array index is specified as C<key>. It
can be dereferenced to get the C<SV*> that was stored there (= C<val>)).
Note that the caller is responsible for suitably incrementing the reference
count of C<val> before the call.
Approximate Perl equivalent: C<splice(@myarray, $key, 1, $val)>.
=cut
*/
PERL_STATIC_INLINE SV**
Perl_av_store_simple(pTHX_ AV *av, SSize_t key, SV *val)
{
SV** ary;
PERL_ARGS_ASSERT_AV_STORE_SIMPLE;
assert(SvTYPE(av) == SVt_PVAV);
assert(!SvMAGICAL(av));
assert(!SvREADONLY(av));
assert(AvREAL(av));
assert(key > -1);
ary = AvARRAY(av);
if (AvFILLp(av) < key) {
if (key > AvMAX(av)) {
av_extend(av,key);
ary = AvARRAY(av);
}
AvFILLp(av) = key;
} else
SvREFCNT_dec(ary[key]);
ary[key] = val;
return &ary[key];
}
/*
=for apidoc av_fetch_simple
This is a cut-down version of av_fetch that assumes that the array is
very straightforward - no magic, not readonly, and AvREAL - and that
C<key> is not negative. This function MUST NOT be used in situations
where any of those assumptions may not hold.
Returns the SV at the specified index in the array. The C<key> is the
index. If lval is true, you are guaranteed to get a real SV back (in case
it wasn't real before), which you can then modify. Check that the return
value is non-null before dereferencing it to a C<SV*>.
The rough perl equivalent is C<$myarray[$key]>.
=cut
*/
PERL_STATIC_INLINE SV**
Perl_av_fetch_simple(pTHX_ AV *av, SSize_t key, I32 lval)
{
PERL_ARGS_ASSERT_AV_FETCH_SIMPLE;
assert(SvTYPE(av) == SVt_PVAV);
assert(!SvMAGICAL(av));
assert(!SvREADONLY(av));
assert(AvREAL(av));
assert(key > -1);
if ( (key > AvFILLp(av)) || !AvARRAY(av)[key]) {
return lval ? av_store_simple(av,key,newSV(0)) : NULL;
} else {
return &AvARRAY(av)[key];
}
}
/* ------------------------------- cv.h ------------------------------- */
/*
=for apidoc_section $CV
=for apidoc CvGV
Returns the GV associated with the CV C<sv>, reifying it if necessary.
=cut
*/
PERL_STATIC_INLINE GV *
Perl_CvGV(pTHX_ CV *sv)
{
PERL_ARGS_ASSERT_CVGV;
return CvNAMED(sv)
? Perl_cvgv_from_hek(aTHX_ sv)
: ((XPVCV*)MUTABLE_PTR(SvANY(sv)))->xcv_gv_u.xcv_gv;
}
PERL_STATIC_INLINE I32 *
Perl_CvDEPTH(const CV * const sv)
{
PERL_ARGS_ASSERT_CVDEPTH;
assert(SvTYPE(sv) == SVt_PVCV || SvTYPE(sv) == SVt_PVFM);
return &((XPVCV*)SvANY(sv))->xcv_depth;
}
/*
CvPROTO returns the prototype as stored, which is not necessarily what
the interpreter should be using. Specifically, the interpreter assumes
that spaces have been stripped, which has been the case if the prototype
was added by toke.c, but is generally not the case if it was added elsewhere.
Since we can't enforce the spacelessness at assignment time, this routine
provides a temporary copy at parse time with spaces removed.
I<orig> is the start of the original buffer, I<len> is the length of the
prototype and will be updated when this returns.
*/
#ifdef PERL_CORE
PERL_STATIC_INLINE char *
S_strip_spaces(pTHX_ const char * orig, STRLEN * const len)
{
SV * tmpsv;
char * tmps;
tmpsv = newSVpvn_flags(orig, *len, SVs_TEMP);
tmps = SvPVX(tmpsv);
while ((*len)--) {
if (!isSPACE(*orig))
*tmps++ = *orig;
orig++;
}
*tmps = '\0';
*len = tmps - SvPVX(tmpsv);
return SvPVX(tmpsv);
}
#endif
/* ------------------------------- mg.h ------------------------------- */
#if defined(PERL_CORE) || defined(PERL_EXT)
/* assumes get-magic and stringification have already occurred */
PERL_STATIC_INLINE STRLEN
S_MgBYTEPOS(pTHX_ MAGIC *mg, SV *sv, const char *s, STRLEN len)
{
assert(mg->mg_type == PERL_MAGIC_regex_global);
assert(mg->mg_len != -1);
if (mg->mg_flags & MGf_BYTES || !DO_UTF8(sv))
return (STRLEN)mg->mg_len;
else {
const STRLEN pos = (STRLEN)mg->mg_len;
/* Without this check, we may read past the end of the buffer: */
if (pos > sv_or_pv_len_utf8(sv, s, len)) return len+1;
return sv_or_pv_pos_u2b(sv, s, pos, NULL);
}
}
#endif
/* ------------------------------- pad.h ------------------------------ */
#if defined(PERL_IN_PAD_C) || defined(PERL_IN_OP_C)
PERL_STATIC_INLINE bool
S_PadnameIN_SCOPE(const PADNAME * const pn, const U32 seq)
{
PERL_ARGS_ASSERT_PADNAMEIN_SCOPE;
/* is seq within the range _LOW to _HIGH ?
* This is complicated by the fact that PL_cop_seqmax
* may have wrapped around at some point */
if (COP_SEQ_RANGE_LOW(pn) == PERL_PADSEQ_INTRO)
return FALSE; /* not yet introduced */
if (COP_SEQ_RANGE_HIGH(pn) == PERL_PADSEQ_INTRO) {
/* in compiling scope */
if (
(seq > COP_SEQ_RANGE_LOW(pn))
? (seq - COP_SEQ_RANGE_LOW(pn) < (U32_MAX >> 1))
: (COP_SEQ_RANGE_LOW(pn) - seq > (U32_MAX >> 1))
)
return TRUE;
}
else if (
(COP_SEQ_RANGE_LOW(pn) > COP_SEQ_RANGE_HIGH(pn))
?
( seq > COP_SEQ_RANGE_LOW(pn)
|| seq <= COP_SEQ_RANGE_HIGH(pn))
: ( seq > COP_SEQ_RANGE_LOW(pn)
&& seq <= COP_SEQ_RANGE_HIGH(pn))
)
return TRUE;
return FALSE;
}
#endif
/* ------------------------------- pp.h ------------------------------- */
PERL_STATIC_INLINE I32
Perl_TOPMARK(pTHX)
{
DEBUG_s(DEBUG_v(PerlIO_printf(Perl_debug_log,
"MARK top %p %" IVdf "\n",
PL_markstack_ptr,
(IV)*PL_markstack_ptr)));
return *PL_markstack_ptr;
}
PERL_STATIC_INLINE I32
Perl_POPMARK(pTHX)
{
DEBUG_s(DEBUG_v(PerlIO_printf(Perl_debug_log,
"MARK pop %p %" IVdf "\n",
(PL_markstack_ptr-1),
(IV)*(PL_markstack_ptr-1))));
assert((PL_markstack_ptr > PL_markstack) || !"MARK underflow");
return *PL_markstack_ptr--;
}
/* ----------------------------- regexp.h ----------------------------- */
/* PVLVs need to act as a superset of all scalar types - they are basically
* PVMGs with a few extra fields.
* REGEXPs are first class scalars, but have many fields that can't be copied
* into a PVLV body.
*
* Hence we take a different approach - instead of a copy, PVLVs store a pointer
* back to the original body. To avoid increasing the size of PVLVs just for the
* rare case of REGEXP assignment, this pointer is stored in the memory usually
* used for SvLEN(). Hence the check for SVt_PVLV below, and the ? : ternary to
* read the pointer from the two possible locations. The macro SvLEN() wraps the
* access to the union's member xpvlenu_len, but there is no equivalent macro
* for wrapping the union's member xpvlenu_rx, hence the direct reference here.
*
* See commit df6b4bd56551f2d3 for more details. */
PERL_STATIC_INLINE struct regexp *
Perl_ReANY(const REGEXP * const re)
{
XPV* const p = (XPV*)SvANY(re);
PERL_ARGS_ASSERT_REANY;
assert(isREGEXP(re));
return SvTYPE(re) == SVt_PVLV ? p->xpv_len_u.xpvlenu_rx
: (struct regexp *)p;
}
/* ------------------------------- sv.h ------------------------------- */
PERL_STATIC_INLINE bool
Perl_SvTRUE(pTHX_ SV *sv)
{
PERL_ARGS_ASSERT_SVTRUE;
if (UNLIKELY(sv == NULL))
return FALSE;
SvGETMAGIC(sv);
return SvTRUE_nomg_NN(sv);
}
PERL_STATIC_INLINE bool
Perl_SvTRUE_nomg(pTHX_ SV *sv)
{
PERL_ARGS_ASSERT_SVTRUE_NOMG;
if (UNLIKELY(sv == NULL))
return FALSE;
return SvTRUE_nomg_NN(sv);
}
PERL_STATIC_INLINE bool
Perl_SvTRUE_NN(pTHX_ SV *sv)
{
PERL_ARGS_ASSERT_SVTRUE_NN;
SvGETMAGIC(sv);
return SvTRUE_nomg_NN(sv);
}
PERL_STATIC_INLINE bool
Perl_SvTRUE_common(pTHX_ SV * sv, const bool sv_2bool_is_fallback)
{
PERL_ARGS_ASSERT_SVTRUE_COMMON;
if (UNLIKELY(SvIMMORTAL_INTERP(sv)))
return SvIMMORTAL_TRUE(sv);
if (! SvOK(sv))
return FALSE;
if (SvPOK(sv))
return SvPVXtrue(sv);
if (SvIOK(sv))
return SvIVX(sv) != 0; /* casts to bool */
if (SvROK(sv) && !(SvOBJECT(SvRV(sv)) && HvAMAGIC(SvSTASH(SvRV(sv)))))
return TRUE;
if (sv_2bool_is_fallback)
return sv_2bool_nomg(sv);
return isGV_with_GP(sv);
}
PERL_STATIC_INLINE SV *
Perl_SvREFCNT_inc(SV *sv)
{
if (LIKELY(sv != NULL))
SvREFCNT(sv)++;
return sv;
}
PERL_STATIC_INLINE SV *
Perl_SvREFCNT_inc_NN(SV *sv)
{
PERL_ARGS_ASSERT_SVREFCNT_INC_NN;
SvREFCNT(sv)++;
return sv;
}
PERL_STATIC_INLINE void
Perl_SvREFCNT_inc_void(SV *sv)
{
if (LIKELY(sv != NULL))
SvREFCNT(sv)++;
}
PERL_STATIC_INLINE void
Perl_SvREFCNT_dec(pTHX_ SV *sv)
{
if (LIKELY(sv != NULL)) {
U32 rc = SvREFCNT(sv);
if (LIKELY(rc > 1))
SvREFCNT(sv) = rc - 1;
else
Perl_sv_free2(aTHX_ sv, rc);
}
}
PERL_STATIC_INLINE void
Perl_SvREFCNT_dec_NN(pTHX_ SV *sv)
{
U32 rc = SvREFCNT(sv);
PERL_ARGS_ASSERT_SVREFCNT_DEC_NN;
if (LIKELY(rc > 1))
SvREFCNT(sv) = rc - 1;
else
Perl_sv_free2(aTHX_ sv, rc);
}
PERL_STATIC_INLINE void
Perl_SvAMAGIC_on(SV *sv)
{
PERL_ARGS_ASSERT_SVAMAGIC_ON;
assert(SvROK(sv));
if (SvOBJECT(SvRV(sv))) HvAMAGIC_on(SvSTASH(SvRV(sv)));
}
PERL_STATIC_INLINE void
Perl_SvAMAGIC_off(SV *sv)
{
PERL_ARGS_ASSERT_SVAMAGIC_OFF;
if (SvROK(sv) && SvOBJECT(SvRV(sv)))
HvAMAGIC_off(SvSTASH(SvRV(sv)));
}
PERL_STATIC_INLINE U32
Perl_SvPADSTALE_on(SV *sv)
{
assert(!(SvFLAGS(sv) & SVs_PADTMP));
return SvFLAGS(sv) |= SVs_PADSTALE;
}
PERL_STATIC_INLINE U32
Perl_SvPADSTALE_off(SV *sv)
{
assert(!(SvFLAGS(sv) & SVs_PADTMP));
return SvFLAGS(sv) &= ~SVs_PADSTALE;
}
#if defined(PERL_CORE) || defined (PERL_EXT)
PERL_STATIC_INLINE STRLEN
S_sv_or_pv_pos_u2b(pTHX_ SV *sv, const char *pv, STRLEN pos, STRLEN *lenp)
{
PERL_ARGS_ASSERT_SV_OR_PV_POS_U2B;
if (SvGAMAGIC(sv)) {
U8 *hopped = utf8_hop((U8 *)pv, pos);
if (lenp) *lenp = (STRLEN)(utf8_hop(hopped, *lenp) - hopped);
return (STRLEN)(hopped - (U8 *)pv);
}
return sv_pos_u2b_flags(sv,pos,lenp,SV_CONST_RETURN);
}
#endif
/* ------------------------------- utf8.h ------------------------------- */
/*
=for apidoc_section $unicode
*/
PERL_STATIC_INLINE void
Perl_append_utf8_from_native_byte(const U8 byte, U8** dest)
{
/* Takes an input 'byte' (Latin1 or EBCDIC) and appends it to the UTF-8
* encoded string at '*dest', updating '*dest' to include it */
PERL_ARGS_ASSERT_APPEND_UTF8_FROM_NATIVE_BYTE;
if (NATIVE_BYTE_IS_INVARIANT(byte))
*((*dest)++) = byte;
else {
*((*dest)++) = UTF8_EIGHT_BIT_HI(byte);
*((*dest)++) = UTF8_EIGHT_BIT_LO(byte);
}
}
/*
=for apidoc valid_utf8_to_uvchr
Like C<L<perlapi/utf8_to_uvchr_buf>>, but should only be called when it is
known that the next character in the input UTF-8 string C<s> is well-formed
(I<e.g.>, it passes C<L<perlapi/isUTF8_CHAR>>. Surrogates, non-character code
points, and non-Unicode code points are allowed.
=cut
*/
PERL_STATIC_INLINE UV
Perl_valid_utf8_to_uvchr(const U8 *s, STRLEN *retlen)
{
const UV expectlen = UTF8SKIP(s);
const U8* send = s + expectlen;
UV uv = *s;
PERL_ARGS_ASSERT_VALID_UTF8_TO_UVCHR;
if (retlen) {
*retlen = expectlen;
}
/* An invariant is trivially returned */
if (expectlen == 1) {
return uv;
}
/* Remove the leading bits that indicate the number of bytes, leaving just
* the bits that are part of the value */
uv = NATIVE_UTF8_TO_I8(uv) & UTF_START_MASK(expectlen);
/* Now, loop through the remaining bytes, accumulating each into the
* working total as we go. (I khw tried unrolling the loop for up to 4
* bytes, but there was no performance improvement) */
for (++s; s < send; s++) {
uv = UTF8_ACCUMULATE(uv, *s);
}
return UNI_TO_NATIVE(uv);
}
/*
=for apidoc is_utf8_invariant_string
Returns TRUE if the first C<len> bytes of the string C<s> are the same
regardless of the UTF-8 encoding of the string (or UTF-EBCDIC encoding on
EBCDIC machines); otherwise it returns FALSE. That is, it returns TRUE if they
are UTF-8 invariant. On ASCII-ish machines, all the ASCII characters and only
the ASCII characters fit this definition. On EBCDIC machines, the ASCII-range
characters are invariant, but so also are the C1 controls.
If C<len> is 0, it will be calculated using C<strlen(s)>, (which means if you
use this option, that C<s> can't have embedded C<NUL> characters and has to
have a terminating C<NUL> byte).
See also
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.
=cut
*/
#define is_utf8_invariant_string(s, len) \
is_utf8_invariant_string_loc(s, len, NULL)
/*
=for apidoc is_utf8_invariant_string_loc
Like C<L</is_utf8_invariant_string>> but upon failure, stores the location of
the first UTF-8 variant character in the C<ep> pointer; if all characters are
UTF-8 invariant, this function does not change the contents of C<*ep>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_invariant_string_loc(const U8* const s, STRLEN len, const U8 ** ep)
{
const U8* send;
const U8* x = s;
PERL_ARGS_ASSERT_IS_UTF8_INVARIANT_STRING_LOC;
if (len == 0) {
len = strlen((const char *)s);
}
send = s + len;
/* This looks like 0x010101... */
# define PERL_COUNT_MULTIPLIER (~ (UINTMAX_C(0)) / 0xFF)
/* This looks like 0x808080... */
# define PERL_VARIANTS_WORD_MASK (PERL_COUNT_MULTIPLIER * 0x80)
# define PERL_WORDSIZE sizeof(PERL_UINTMAX_T)
# define PERL_WORD_BOUNDARY_MASK (PERL_WORDSIZE - 1)
/* Evaluates to 0 if 'x' is at a word boundary; otherwise evaluates to 1, by
* or'ing together the lowest bits of 'x'. Hopefully the final term gets
* optimized out completely on a 32-bit system, and its mask gets optimized out
* on a 64-bit system */
# define PERL_IS_SUBWORD_ADDR(x) (1 & ( PTR2nat(x) \
| ( PTR2nat(x) >> 1) \
| ( ( (PTR2nat(x) \
& PERL_WORD_BOUNDARY_MASK) >> 2))))
#ifndef EBCDIC
/* Do the word-at-a-time iff there is at least one usable full word. That
* means that after advancing to a word boundary, there still is at least a
* full word left. The number of bytes needed to advance is 'wordsize -
* offset' unless offset is 0. */
if ((STRLEN) (send - x) >= PERL_WORDSIZE
/* This term is wordsize if subword; 0 if not */
+ PERL_WORDSIZE * PERL_IS_SUBWORD_ADDR(x)
/* 'offset' */
- (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK))
{
/* Process per-byte until reach word boundary. XXX This loop could be
* eliminated if we knew that this platform had fast unaligned reads */
while (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK) {
if (! UTF8_IS_INVARIANT(*x)) {
if (ep) {
*ep = x;
}
return FALSE;
}
x++;
}
/* Here, we know we have at least one full word to process. Process
* per-word as long as we have at least a full word left */
do {
if ((* (PERL_UINTMAX_T *) x) & PERL_VARIANTS_WORD_MASK) {
/* Found a variant. Just return if caller doesn't want its
* exact position */
if (! ep) {
return FALSE;
}
# if BYTEORDER == 0x1234 || BYTEORDER == 0x12345678 \
|| BYTEORDER == 0x4321 || BYTEORDER == 0x87654321
*ep = x + variant_byte_number(* (PERL_UINTMAX_T *) x);
assert(*ep >= s && *ep < send);
return FALSE;
# else /* If weird byte order, drop into next loop to do byte-at-a-time
checks. */
break;
# endif
}
x += PERL_WORDSIZE;
} while (x + PERL_WORDSIZE <= send);
}
#endif /* End of ! EBCDIC */
/* Process per-byte */
while (x < send) {
if (! UTF8_IS_INVARIANT(*x)) {
if (ep) {
*ep = x;
}
return FALSE;
}
x++;
}
return TRUE;
}
/* See if the platform has builtins for finding the most/least significant bit,
* and which one is right for using on 32 and 64 bit operands */
#if (__has_builtin(__builtin_clz) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == INTSIZE
# define PERL_CLZ_32 __builtin_clz
# endif
# if defined(U64TYPE) && U64SIZE == INTSIZE
# define PERL_CLZ_64 __builtin_clz
# endif
#endif
#if (__has_builtin(__builtin_ctz) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == INTSIZE
# define PERL_CTZ_32 __builtin_ctz
# endif
# if defined(U64TYPE) && U64SIZE == INTSIZE
# define PERL_CTZ_64 __builtin_ctz
# endif
#endif
#if (__has_builtin(__builtin_clzl) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == LONGSIZE && ! defined(PERL_CLZ_32)
# define PERL_CLZ_32 __builtin_clzl
# endif
# if defined(U64TYPE) && U64SIZE == LONGSIZE && ! defined(PERL_CLZ_64)
# define PERL_CLZ_64 __builtin_clzl
# endif
#endif
#if (__has_builtin(__builtin_ctzl) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == LONGSIZE && ! defined(PERL_CTZ_32)
# define PERL_CTZ_32 __builtin_ctzl
# endif
# if defined(U64TYPE) && U64SIZE == LONGSIZE && ! defined(PERL_CTZ_64)
# define PERL_CTZ_64 __builtin_ctzl
# endif
#endif
#if (__has_builtin(__builtin_clzll) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == LONGLONGSIZE && ! defined(PERL_CLZ_32)
# define PERL_CLZ_32 __builtin_clzll
# endif
# if defined(U64TYPE) && U64SIZE == LONGLONGSIZE && ! defined(PERL_CLZ_64)
# define PERL_CLZ_64 __builtin_clzll
# endif
#endif
#if (__has_builtin(__builtin_ctzll) || PERL_GCC_VERSION_GE(3,4,0))
# if U32SIZE == LONGLONGSIZE && ! defined(PERL_CTZ_32)
# define PERL_CTZ_32 __builtin_ctzll
# endif
# if defined(U64TYPE) && U64SIZE == LONGLONGSIZE && ! defined(PERL_CTZ_64)
# define PERL_CTZ_64 __builtin_ctzll
# endif
#endif
#if defined(_MSC_VER)
# include <intrin.h>
# pragma intrinsic(_BitScanForward)
# pragma intrinsic(_BitScanReverse)
# ifdef _WIN64
# pragma intrinsic(_BitScanForward64)
# pragma intrinsic(_BitScanReverse64)
# endif
#endif
/* The reason there are not checks to see if ffs() and ffsl() are available for
* determining the lsb, is because these don't improve on the deBruijn method
* fallback, which is just a branchless integer multiply, array element
* retrieval, and shift. The others, even if the function call overhead is
* optimized out, have to cope with the possibility of the input being all
* zeroes, and almost certainly will have conditionals for this eventuality.
* khw, at the time of this commit, looked at the source for both gcc and clang
* to verify this. (gcc used a method inferior to deBruijn.) */
/* Below are functions to find the first, last, or only set bit in a word. On
* platforms with 64-bit capability, there is a pair for each operation; the
* first taking a 64 bit operand, and the second a 32 bit one. The logic is
* the same in each pair, so the second is stripped of most comments. */
#ifdef U64TYPE /* HAS_QUAD not usable outside the core */
PERL_STATIC_INLINE unsigned
Perl_lsbit_pos64(U64 word)
{
/* Find the position (0..63) of the least significant set bit in the input
* word */
ASSUME(word != 0);
/* If we can determine that the platform has a usable fast method to get
* this info, use that */
# if defined(PERL_CTZ_64)
# define PERL_HAS_FAST_GET_LSB_POS64
return (unsigned) PERL_CTZ_64(word);
# elif U64SIZE == 8 && defined(_WIN64) && defined(_MSC_VER)
# define PERL_HAS_FAST_GET_LSB_POS64
{
unsigned long index;
_BitScanForward64(&index, word);
return (unsigned)index;
}
# else
/* Here, we didn't find a fast method for finding the lsb. Fall back to
* making the lsb the only set bit in the word, and use our function that
* works on words with a single bit set.
*
* Isolate the lsb;
* https://stackoverflow.com/questions/757059/position-of-least-significant-bit-that-is-set
*
* The word will look like this, with a rightmost set bit in position 's':
* ('x's are don't cares, and 'y's are their complements)
* s
* x..x100..00
* y..y011..11 Complement
* y..y100..00 Add 1
* 0..0100..00 And with the original
*
* (Yes, complementing and adding 1 is just taking the negative on 2's
* complement machines, but not on 1's complement ones, and some compilers
* complain about negating an unsigned.)
*/
return single_1bit_pos64(word & (~word + 1));
# endif
}
# define lsbit_pos_uintmax_(word) lsbit_pos64(word)
#else /* ! QUAD */
# define lsbit_pos_uintmax_(word) lsbit_pos32(word)
#endif
PERL_STATIC_INLINE unsigned /* Like above for 32 bit word */
Perl_lsbit_pos32(U32 word)
{
/* Find the position (0..31) of the least significant set bit in the input
* word */
ASSUME(word != 0);
#if defined(PERL_CTZ_32)
# define PERL_HAS_FAST_GET_LSB_POS32
return (unsigned) PERL_CTZ_32(word);
#elif U32SIZE == 4 && defined(_MSC_VER)
# define PERL_HAS_FAST_GET_LSB_POS32
{
unsigned long index;
_BitScanForward(&index, word);
return (unsigned)index;
}
#else
return single_1bit_pos32(word & (~word + 1));
#endif
}
/* Convert the leading zeros count to the bit position of the first set bit.
* This just subtracts from the highest position, 31 or 63. But some compilers
* don't optimize this optimally, and so a bit of bit twiddling encourages them
* to do the right thing. It turns out that subtracting a smaller non-negative
* number 'x' from 2**n-1 for any n is the same as taking the exclusive-or of
* the two numbers. To see why, first note that the sum of any number, x, and
* its complement, x', is all ones. So all ones minus x is x'. Then note that
* the xor of x and all ones is x'. */
#define LZC_TO_MSBIT_POS_(size, lzc) ((size##SIZE * CHARBITS - 1) ^ (lzc))
#ifdef U64TYPE /* HAS_QUAD not usable outside the core */
PERL_STATIC_INLINE unsigned
Perl_msbit_pos64(U64 word)
{
/* Find the position (0..63) of the most significant set bit in the input
* word */
ASSUME(word != 0);
/* If we can determine that the platform has a usable fast method to get
* this, use that */
# if defined(PERL_CLZ_64)
# define PERL_HAS_FAST_GET_MSB_POS64
return (unsigned) LZC_TO_MSBIT_POS_(U64, PERL_CLZ_64(word));
# elif U64SIZE == 8 && defined(_WIN64) && defined(_MSC_VER)
# define PERL_HAS_FAST_GET_MSB_POS64
{
unsigned long index;
_BitScanReverse64(&index, word);
return (unsigned)index;
}
# else
/* Here, we didn't find a fast method for finding the msb. Fall back to
* making the msb the only set bit in the word, and use our function that
* works on words with a single bit set.
*
* Isolate the msb; http://codeforces.com/blog/entry/10330
*
* Only the most significant set bit matters. Or'ing word with its right
* shift of 1 makes that bit and the next one to its right both 1.
* Repeating that with the right shift of 2 makes for 4 1-bits in a row.
* ... We end with the msb and all to the right being 1. */
word |= (word >> 1);
word |= (word >> 2);
word |= (word >> 4);
word |= (word >> 8);
word |= (word >> 16);
word |= (word >> 32);
/* Then subtracting the right shift by 1 clears all but the left-most of
* the 1 bits, which is our desired result */
word -= (word >> 1);
/* Now we have a single bit set */
return single_1bit_pos64(word);
# endif
}
# define msbit_pos_uintmax_(word) msbit_pos64(word)
#else /* ! QUAD */
# define msbit_pos_uintmax_(word) msbit_pos32(word)
#endif
PERL_STATIC_INLINE unsigned
Perl_msbit_pos32(U32 word)
{
/* Find the position (0..31) of the most significant set bit in the input
* word */
ASSUME(word != 0);
#if defined(PERL_CLZ_32)
# define PERL_HAS_FAST_GET_MSB_POS32
return (unsigned) LZC_TO_MSBIT_POS_(U32, PERL_CLZ_32(word));
#elif U32SIZE == 4 && defined(_MSC_VER)
# define PERL_HAS_FAST_GET_MSB_POS32
{
unsigned long index;
_BitScanReverse(&index, word);
return (unsigned)index;
}
#else
word |= (word >> 1);
word |= (word >> 2);
word |= (word >> 4);
word |= (word >> 8);
word |= (word >> 16);
word -= (word >> 1);
return single_1bit_pos32(word);
#endif
}
#if UVSIZE == U64SIZE
# define msbit_pos(word) msbit_pos64(word)
# define lsbit_pos(word) lsbit_pos64(word)
#elif UVSIZE == U32SIZE
# define msbit_pos(word) msbit_pos32(word)
# define lsbit_pos(word) lsbit_pos32(word)
#endif
#ifdef U64TYPE /* HAS_QUAD not usable outside the core */
PERL_STATIC_INLINE unsigned
Perl_single_1bit_pos64(U64 word)
{
/* Given a 64-bit word known to contain all zero bits except one 1 bit,
* find and return the 1's position: 0..63 */
# ifdef PERL_CORE /* macro not exported */
ASSUME(isPOWER_OF_2(word));
# else
ASSUME(word && (word & (word-1)) == 0);
# endif
/* The only set bit is both the most and least significant bit. If we have
* a fast way of finding either one, use that.
*
* It may appear at first glance that those functions call this one, but
* they don't if the corresponding #define is set */
# ifdef PERL_HAS_FAST_GET_MSB_POS64
return msbit_pos64(word);
# elif defined(PERL_HAS_FAST_GET_LSB_POS64)
return lsbit_pos64(word);
# else
/* The position of the only set bit in a word can be quickly calculated
* using deBruijn sequences. See for example
* https://en.wikipedia.org/wiki/De_Bruijn_sequence */
return PL_deBruijn_bitpos_tab64[(word * PERL_deBruijnMagic64_)
>> PERL_deBruijnShift64_];
# endif
}
#endif
PERL_STATIC_INLINE unsigned
Perl_single_1bit_pos32(U32 word)
{
/* Given a 32-bit word known to contain all zero bits except one 1 bit,
* find and return the 1's position: 0..31 */
#ifdef PERL_CORE /* macro not exported */
ASSUME(isPOWER_OF_2(word));
#else
ASSUME(word && (word & (word-1)) == 0);
#endif
#ifdef PERL_HAS_FAST_GET_MSB_POS32
return msbit_pos32(word);
#elif defined(PERL_HAS_FAST_GET_LSB_POS32)
return lsbit_pos32(word);
/* Unlikely, but possible for the platform to have a wider fast operation but
* not a narrower one. But easy enough to handle the case by widening the
* parameter size. (Going the other way, emulating 64 bit by two 32 bit ops
* would be slower than the deBruijn method.) */
#elif defined(PERL_HAS_FAST_GET_MSB_POS64)
return msbit_pos64(word);
#elif defined(PERL_HAS_FAST_GET_LSB_POS64)
return lsbit_pos64(word);
#else
return PL_deBruijn_bitpos_tab32[(word * PERL_deBruijnMagic32_)
>> PERL_deBruijnShift32_];
#endif
}
#ifndef EBCDIC
PERL_STATIC_INLINE unsigned int
Perl_variant_byte_number(PERL_UINTMAX_T word)
{
/* This returns the position in a word (0..7) of the first variant byte in
* it. This is a helper function. Note that there are no branches */
/* Get just the msb bits of each byte */
word &= PERL_VARIANTS_WORD_MASK;
/* This should only be called if we know there is a variant byte in the
* word */
assert(word);
# if BYTEORDER == 0x1234 || BYTEORDER == 0x12345678
/* Bytes are stored like
* Byte8 ... Byte2 Byte1
* 63..56...15...8 7...0
* so getting the lsb of the whole modified word is getting the msb of the
* first byte that has its msb set */
word = lsbit_pos_uintmax_(word);
/* Here, word contains the position 7,15,23,...55,63 of that bit. Convert
* to 0..7 */
return (unsigned int) ((word + 1) >> 3) - 1;
# elif BYTEORDER == 0x4321 || BYTEORDER == 0x87654321
/* Bytes are stored like
* Byte1 Byte2 ... Byte8
* 63..56 55..47 ... 7...0
* so getting the msb of the whole modified word is getting the msb of the
* first byte that has its msb set */
word = msbit_pos_uintmax_(word);
/* Here, word contains the position 63,55,...,23,15,7 of that bit. Convert
* to 0..7 */
word = ((word + 1) >> 3) - 1;
/* And invert the result because of the reversed byte order on this
* platform */
word = CHARBITS - word - 1;
return (unsigned int) word;
# else
# error Unexpected byte order
# endif
}
#endif
#if defined(PERL_CORE) || defined(PERL_EXT)
/*
=for apidoc variant_under_utf8_count
This function looks at the sequence of bytes between C<s> and C<e>, which are
assumed to be encoded in ASCII/Latin1, and returns how many of them would
change should the string be translated into UTF-8. Due to the nature of UTF-8,
each of these would occupy two bytes instead of the single one in the input
string. Thus, this function returns the precise number of bytes the string
would expand by when translated to UTF-8.
Unlike most of the other functions that have C<utf8> in their name, the input
to this function is NOT a UTF-8-encoded string. The function name is slightly
I<odd> to emphasize this.
This function is internal to Perl because khw thinks that any XS code that
would want this is probably operating too close to the internals. Presenting a
valid use case could change that.
See also
C<L<perlapi/is_utf8_invariant_string>>
and
C<L<perlapi/is_utf8_invariant_string_loc>>,
=cut
*/
PERL_STATIC_INLINE Size_t
S_variant_under_utf8_count(const U8* const s, const U8* const e)
{
const U8* x = s;
Size_t count = 0;
PERL_ARGS_ASSERT_VARIANT_UNDER_UTF8_COUNT;
# ifndef EBCDIC
/* Test if the string is long enough to use word-at-a-time. (Logic is the
* same as for is_utf8_invariant_string()) */
if ((STRLEN) (e - x) >= PERL_WORDSIZE
+ PERL_WORDSIZE * PERL_IS_SUBWORD_ADDR(x)
- (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK))
{
/* Process per-byte until reach word boundary. XXX This loop could be
* eliminated if we knew that this platform had fast unaligned reads */
while (PTR2nat(x) & PERL_WORD_BOUNDARY_MASK) {
count += ! UTF8_IS_INVARIANT(*x++);
}
/* Process per-word as long as we have at least a full word left */
do { /* Commit 03c1e4ab1d6ee9062fb3f94b0ba31db6698724b1 contains an
explanation of how this works */
PERL_UINTMAX_T increment
= ((((* (PERL_UINTMAX_T *) x) & PERL_VARIANTS_WORD_MASK) >> 7)
* PERL_COUNT_MULTIPLIER)
>> ((PERL_WORDSIZE - 1) * CHARBITS);
count += (Size_t) increment;
x += PERL_WORDSIZE;
} while (x + PERL_WORDSIZE <= e);
}
# endif
/* Process per-byte */
while (x < e) {
if (! UTF8_IS_INVARIANT(*x)) {
count++;
}
x++;
}
return count;
}
#endif
#ifndef PERL_IN_REGEXEC_C /* Keep these around for that file */
# undef PERL_WORDSIZE
# undef PERL_COUNT_MULTIPLIER
# undef PERL_WORD_BOUNDARY_MASK
# undef PERL_VARIANTS_WORD_MASK
#endif
/*
=for apidoc is_utf8_string
Returns TRUE if the first C<len> bytes of string C<s> form a valid
Perl-extended-UTF-8 string; returns FALSE otherwise. If C<len> is 0, it will
be calculated using C<strlen(s)> (which means if you use this option, that C<s>
can't have embedded C<NUL> characters and has to have a terminating C<NUL>
byte). Note that all characters being ASCII constitute 'a valid UTF-8 string'.
This function considers Perl's extended UTF-8 to be valid. That means that
code points above Unicode, surrogates, and non-character code points are
considered valid by this function. Use C<L</is_strict_utf8_string>>,
C<L</is_c9strict_utf8_string>>, or C<L</is_utf8_string_flags>> to restrict what
code points are considered valid.
See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
=cut
*/
#define is_utf8_string(s, len) is_utf8_string_loclen(s, len, NULL, NULL)
#if defined(PERL_CORE) || defined (PERL_EXT)
/*
=for apidoc is_utf8_non_invariant_string
Returns TRUE if L<perlapi/is_utf8_invariant_string> returns FALSE for the first
C<len> bytes of the string C<s>, but they are, nonetheless, legal Perl-extended
UTF-8; otherwise returns FALSE.
A TRUE return means that at least one code point represented by the sequence
either is a wide character not representable as a single byte, or the
representation differs depending on whether the sequence is encoded in UTF-8 or
not.
See also
C<L<perlapi/is_utf8_invariant_string>>,
C<L<perlapi/is_utf8_string>>
=cut
This is commonly used to determine if a SV's UTF-8 flag should be turned on.
It generally needn't be if its string is entirely UTF-8 invariant, and it
shouldn't be if it otherwise contains invalid UTF-8.
It is an internal function because khw thinks that XS code shouldn't be working
at this low a level. A valid use case could change that.
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_non_invariant_string(const U8* const s, STRLEN len)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_NON_INVARIANT_STRING;
if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
return FALSE;
}
return is_utf8_string(first_variant, len - (first_variant - s));
}
#endif
/*
=for apidoc is_strict_utf8_string
Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8-encoded string that is fully interchangeable by any application using
Unicode rules; otherwise it returns FALSE. If C<len> is 0, it will be
calculated using C<strlen(s)> (which means if you use this option, that C<s>
can't have embedded C<NUL> characters and has to have a terminating C<NUL>
byte). Note that all characters being ASCII constitute 'a valid UTF-8 string'.
This function returns FALSE for strings containing any
code points above the Unicode max of 0x10FFFF, surrogate code points, or
non-character code points.
See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.
=cut
*/
#define is_strict_utf8_string(s, len) is_strict_utf8_string_loclen(s, len, NULL, NULL)
/*
=for apidoc is_c9strict_utf8_string
Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8-encoded string that conforms to
L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>;
otherwise it returns FALSE. If C<len> is 0, it will be calculated using
C<strlen(s)> (which means if you use this option, that C<s> can't have embedded
C<NUL> characters and has to have a terminating C<NUL> byte). Note that all
characters being ASCII constitute 'a valid UTF-8 string'.
This function returns FALSE for strings containing any code points above the
Unicode max of 0x10FFFF or surrogate code points, but accepts non-character
code points per
L<Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.
See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_flags>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.
=cut
*/
#define is_c9strict_utf8_string(s, len) is_c9strict_utf8_string_loclen(s, len, NULL, 0)
/*
=for apidoc is_utf8_string_flags
Returns TRUE if the first C<len> bytes of string C<s> form a valid
UTF-8 string, subject to the restrictions imposed by C<flags>;
returns FALSE otherwise. If C<len> is 0, it will be calculated
using C<strlen(s)> (which means if you use this option, that C<s> can't have
embedded C<NUL> characters and has to have a terminating C<NUL> byte). Note
that all characters being ASCII constitute 'a valid UTF-8 string'.
If C<flags> is 0, this gives the same results as C<L</is_utf8_string>>; if
C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
as C<L</is_strict_utf8_string>>; and if C<flags> is
C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives the same results as
C<L</is_c9strict_utf8_string>>. Otherwise C<flags> may be any
combination of the C<UTF8_DISALLOW_I<foo>> flags understood by
C<L</utf8n_to_uvchr>>, with the same meanings.
See also
C<L</is_utf8_invariant_string>>,
C<L</is_utf8_invariant_string_loc>>,
C<L</is_utf8_string>>,
C<L</is_utf8_string_loc>>,
C<L</is_utf8_string_loc_flags>>,
C<L</is_utf8_string_loclen>>,
C<L</is_utf8_string_loclen_flags>>,
C<L</is_utf8_fixed_width_buf_flags>>,
C<L</is_utf8_fixed_width_buf_loc_flags>>,
C<L</is_utf8_fixed_width_buf_loclen_flags>>,
C<L</is_strict_utf8_string>>,
C<L</is_strict_utf8_string_loc>>,
C<L</is_strict_utf8_string_loclen>>,
C<L</is_c9strict_utf8_string>>,
C<L</is_c9strict_utf8_string_loc>>,
and
C<L</is_c9strict_utf8_string_loclen>>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_string_flags(const U8 *s, STRLEN len, const U32 flags)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
|UTF8_DISALLOW_PERL_EXTENDED)));
if (len == 0) {
len = strlen((const char *)s);
}
if (flags == 0) {
return is_utf8_string(s, len);
}
if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
== UTF8_DISALLOW_ILLEGAL_INTERCHANGE)
{
return is_strict_utf8_string(s, len);
}
if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
== UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE)
{
return is_c9strict_utf8_string(s, len);
}
if (! is_utf8_invariant_string_loc(s, len, &first_variant)) {
const U8* const send = s + len;
const U8* x = first_variant;
while (x < send) {
STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
if (UNLIKELY(! cur_len)) {
return FALSE;
}
x += cur_len;
}
}
return TRUE;
}
/*
=for apidoc is_utf8_string_loc
Like C<L</is_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.
See also C<L</is_utf8_string_loclen>>.
=cut
*/
#define is_utf8_string_loc(s, len, ep) is_utf8_string_loclen(s, len, ep, 0)
/*
=for apidoc is_utf8_string_loclen
Like C<L</is_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.
See also C<L</is_utf8_string_loc>>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN;
if (len == 0) {
len = strlen((const char *) s);
}
if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
if (el)
*el = len;
if (ep) {
*ep = s + len;
}
return TRUE;
}
{
const U8* const send = s + len;
const U8* x = first_variant;
STRLEN outlen = first_variant - s;
while (x < send) {
const STRLEN cur_len = isUTF8_CHAR(x, send);
if (UNLIKELY(! cur_len)) {
break;
}
x += cur_len;
outlen++;
}
if (el)
*el = outlen;
if (ep) {
*ep = x;
}
return (x == send);
}
}
/* The perl core arranges to never call the DFA below without there being at
* least one byte available to look at. This allows the DFA to use a do {}
* while loop which means that calling it with a UTF-8 invariant has a single
* conditional, same as the calling code checking for invariance ahead of time.
* And having the calling code remove that conditional speeds up by that
* conditional, the case where it wasn't invariant. So there's no reason to
* check before caling this.
*
* But we don't know this for non-core calls, so have to retain the check for
* them. */
#ifdef PERL_CORE
# define PERL_NON_CORE_CHECK_EMPTY(s,e) assert((e) > (s))
#else
# define PERL_NON_CORE_CHECK_EMPTY(s,e) if ((e) <= (s)) return FALSE
#endif
/*
* DFA for checking input is valid UTF-8 syntax.
*
* This uses adaptations of the table and algorithm given in
* https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
* documentation of the original version. A copyright notice for the original
* version is given at the beginning of this file. The Perl adapations are
* documented at the definition of PL_extended_utf8_dfa_tab[].
*
* This dfa is fast. There are three exit conditions:
* 1) a well-formed code point, acceptable to the table
* 2) the beginning bytes of an incomplete character, whose completion might
* or might not be acceptable
* 3) unacceptable to the table. Some of the adaptations have certain,
* hopefully less likely to occur, legal inputs be unacceptable to the
* table, so these must be sorted out afterwards.
*
* This macro is a complete implementation of the code executing the DFA. It
* is passed the input sequence bounds and the table to use, and what to do
* for each of the exit conditions. There are three canned actions, likely to
* be the ones you want:
* DFA_RETURN_SUCCESS_
* DFA_RETURN_FAILURE_
* DFA_GOTO_TEASE_APART_FF_
*
* You pass a parameter giving the action to take for each of the three
* possible exit conditions:
*
* 'accept_action' This is executed when the DFA accepts the input.
* DFA_RETURN_SUCCESS_ is the most likely candidate.
* 'reject_action' This is executed when the DFA rejects the input.
* DFA_RETURN_FAILURE_ is a candidate, or 'goto label' where
* you have written code to distinguish the rejecting state
* results. Because it happens in several places, and
* involves #ifdefs, the special action
* DFA_GOTO_TEASE_APART_FF_ is what you want with
* PL_extended_utf8_dfa_tab. On platforms without
* EXTRA_LONG_UTF8, there is no need to tease anything apart,
* so this evaluates to DFA_RETURN_FAILURE_; otherwise you
* need to have a label 'tease_apart_FF' that it will transfer
* to.
* 'incomplete_char_action' This is executed when the DFA ran off the end
* before accepting or rejecting the input.
* DFA_RETURN_FAILURE_ is the likely action, but you could
* have a 'goto', or NOOP. In the latter case the DFA drops
* off the end, and you place your code to handle this case
* immediately after it.
*/
#define DFA_RETURN_SUCCESS_ return s - s0
#define DFA_RETURN_FAILURE_ return 0
#ifdef HAS_EXTRA_LONG_UTF8
# define DFA_TEASE_APART_FF_ goto tease_apart_FF
#else
# define DFA_TEASE_APART_FF_ DFA_RETURN_FAILURE_
#endif
#define PERL_IS_UTF8_CHAR_DFA(s0, e, dfa_tab, \
accept_action, \
reject_action, \
incomplete_char_action) \
STMT_START { \
const U8 * s = s0; \
UV state = 0; \
\
PERL_NON_CORE_CHECK_EMPTY(s,e); \
\
do { \
state = dfa_tab[256 + state + dfa_tab[*s]]; \
s++; \
\
if (state == 0) { /* Accepting state */ \
accept_action; \
} \
\
if (UNLIKELY(state == 1)) { /* Rejecting state */ \
reject_action; \
} \
} while (s < e); \
\
/* Here, dropped out of loop before end-of-char */ \
incomplete_char_action; \
} STMT_END
/*
=for apidoc isUTF8_CHAR
Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
that represents some code point; otherwise it evaluates to 0. If non-zero, the
value gives how many bytes starting at C<s> comprise the code point's
representation. Any bytes remaining before C<e>, but beyond the ones needed to
form the first code point in C<s>, are not examined.
The code point can be any that will fit in an IV on this machine, using Perl's
extension to official UTF-8 to represent those higher than the Unicode maximum
of 0x10FFFF. That means that this macro is used to efficiently decide if the
next few bytes in C<s> is legal UTF-8 for a single character.
Use C<L</isSTRICT_UTF8_CHAR>> to restrict the acceptable code points to those
defined by Unicode to be fully interchangeable across applications;
C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
code points; and C<L</isUTF8_CHAR_flags>> for a more customized definition.
Use C<L</is_utf8_string>>, C<L</is_utf8_string_loc>>, and
C<L</is_utf8_string_loclen>> to check entire strings.
Note also that a UTF-8 "invariant" character (i.e. ASCII on non-EBCDIC
machines) is a valid UTF-8 character.
=cut
This uses an adaptation of the table and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version. A copyright notice for the original
version is given at the beginning of this file. The Perl adapation is
documented at the definition of PL_extended_utf8_dfa_tab[].
*/
PERL_STATIC_INLINE Size_t
Perl_isUTF8_CHAR(const U8 * const s0, const U8 * const e)
{
PERL_ARGS_ASSERT_ISUTF8_CHAR;
PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
DFA_RETURN_SUCCESS_,
DFA_TEASE_APART_FF_,
DFA_RETURN_FAILURE_);
/* Here, we didn't return success, but dropped out of the loop. In the
* case of PL_extended_utf8_dfa_tab, this means the input is either
* malformed, or the start byte was FF on a platform that the dfa doesn't
* handle FF's. Call a helper function. */
#ifdef HAS_EXTRA_LONG_UTF8
tease_apart_FF:
/* In the case of PL_extended_utf8_dfa_tab, getting here means the input is
* either malformed, or was for the largest possible start byte, which we
* now check, not inline */
if (*s0 != I8_TO_NATIVE_UTF8(0xFF)) {
return 0;
}
return is_utf8_FF_helper_(s0, e,
FALSE /* require full, not partial char */
);
#endif
}
/*
=for apidoc isSTRICT_UTF8_CHAR
Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
Unicode code point completely acceptable for open interchange between all
applications; otherwise it evaluates to 0. If non-zero, the value gives how
many bytes starting at C<s> comprise the code point's representation. Any
bytes remaining before C<e>, but beyond the ones needed to form the first code
point in C<s>, are not examined.
The largest acceptable code point is the Unicode maximum 0x10FFFF, and must not
be a surrogate nor a non-character code point. Thus this excludes any code
point from Perl's extended UTF-8.
This is used to efficiently decide if the next few bytes in C<s> is
legal Unicode-acceptable UTF-8 for a single character.
Use C<L</isC9_STRICT_UTF8_CHAR>> to use the L<Unicode Corrigendum
#9|http://www.unicode.org/versions/corrigendum9.html> definition of allowable
code points; C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8;
and C<L</isUTF8_CHAR_flags>> for a more customized definition.
Use C<L</is_strict_utf8_string>>, C<L</is_strict_utf8_string_loc>>, and
C<L</is_strict_utf8_string_loclen>> to check entire strings.
=cut
This uses an adaptation of the tables and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version. A copyright notice for the original
version is given at the beginning of this file. The Perl adapation is
documented at the definition of strict_extended_utf8_dfa_tab[].
*/
PERL_STATIC_INLINE Size_t
Perl_isSTRICT_UTF8_CHAR(const U8 * const s0, const U8 * const e)
{
PERL_ARGS_ASSERT_ISSTRICT_UTF8_CHAR;
PERL_IS_UTF8_CHAR_DFA(s0, e, PL_strict_utf8_dfa_tab,
DFA_RETURN_SUCCESS_,
goto check_hanguls,
DFA_RETURN_FAILURE_);
check_hanguls:
/* Here, we didn't return success, but dropped out of the loop. In the
* case of PL_strict_utf8_dfa_tab, this means the input is either
* malformed, or was for certain Hanguls; handle them specially */
/* The dfa above drops out for incomplete or illegal inputs, and certain
* legal Hanguls; check and return accordingly */
return is_HANGUL_ED_utf8_safe(s0, e);
}
/*
=for apidoc isC9_STRICT_UTF8_CHAR
Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8 that represents some
Unicode non-surrogate code point; otherwise it evaluates to 0. If non-zero,
the value gives how many bytes starting at C<s> comprise the code point's
representation. Any bytes remaining before C<e>, but beyond the ones needed to
form the first code point in C<s>, are not examined.
The largest acceptable code point is the Unicode maximum 0x10FFFF. This
differs from C<L</isSTRICT_UTF8_CHAR>> only in that it accepts non-character
code points. This corresponds to
L<Unicode Corrigendum #9|http://www.unicode.org/versions/corrigendum9.html>.
which said that non-character code points are merely discouraged rather than
completely forbidden in open interchange. See
L<perlunicode/Noncharacter code points>.
Use C<L</isUTF8_CHAR>> to check for Perl's extended UTF-8; and
C<L</isUTF8_CHAR_flags>> for a more customized definition.
Use C<L</is_c9strict_utf8_string>>, C<L</is_c9strict_utf8_string_loc>>, and
C<L</is_c9strict_utf8_string_loclen>> to check entire strings.
=cut
This uses an adaptation of the tables and algorithm given in
https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides comprehensive
documentation of the original version. A copyright notice for the original
version is given at the beginning of this file. The Perl adapation is
documented at the definition of PL_c9_utf8_dfa_tab[].
*/
PERL_STATIC_INLINE Size_t
Perl_isC9_STRICT_UTF8_CHAR(const U8 * const s0, const U8 * const e)
{
PERL_ARGS_ASSERT_ISC9_STRICT_UTF8_CHAR;
PERL_IS_UTF8_CHAR_DFA(s0, e, PL_c9_utf8_dfa_tab,
DFA_RETURN_SUCCESS_,
DFA_RETURN_FAILURE_,
DFA_RETURN_FAILURE_);
}
/*
=for apidoc is_strict_utf8_string_loc
Like C<L</is_strict_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.
See also C<L</is_strict_utf8_string_loclen>>.
=cut
*/
#define is_strict_utf8_string_loc(s, len, ep) \
is_strict_utf8_string_loclen(s, len, ep, 0)
/*
=for apidoc is_strict_utf8_string_loclen
Like C<L</is_strict_utf8_string>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.
See also C<L</is_strict_utf8_string_loc>>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_STRICT_UTF8_STRING_LOCLEN;
if (len == 0) {
len = strlen((const char *) s);
}
if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
if (el)
*el = len;
if (ep) {
*ep = s + len;
}
return TRUE;
}
{
const U8* const send = s + len;
const U8* x = first_variant;
STRLEN outlen = first_variant - s;
while (x < send) {
const STRLEN cur_len = isSTRICT_UTF8_CHAR(x, send);
if (UNLIKELY(! cur_len)) {
break;
}
x += cur_len;
outlen++;
}
if (el)
*el = outlen;
if (ep) {
*ep = x;
}
return (x == send);
}
}
/*
=for apidoc is_c9strict_utf8_string_loc
Like C<L</is_c9strict_utf8_string>> but stores the location of the failure (in
the case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.
See also C<L</is_c9strict_utf8_string_loclen>>.
=cut
*/
#define is_c9strict_utf8_string_loc(s, len, ep) \
is_c9strict_utf8_string_loclen(s, len, ep, 0)
/*
=for apidoc is_c9strict_utf8_string_loclen
Like C<L</is_c9strict_utf8_string>> but stores the location of the failure (in
the case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8 encoded
characters in the C<el> pointer.
See also C<L</is_c9strict_utf8_string_loc>>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_c9strict_utf8_string_loclen(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_C9STRICT_UTF8_STRING_LOCLEN;
if (len == 0) {
len = strlen((const char *) s);
}
if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
if (el)
*el = len;
if (ep) {
*ep = s + len;
}
return TRUE;
}
{
const U8* const send = s + len;
const U8* x = first_variant;
STRLEN outlen = first_variant - s;
while (x < send) {
const STRLEN cur_len = isC9_STRICT_UTF8_CHAR(x, send);
if (UNLIKELY(! cur_len)) {
break;
}
x += cur_len;
outlen++;
}
if (el)
*el = outlen;
if (ep) {
*ep = x;
}
return (x == send);
}
}
/*
=for apidoc is_utf8_string_loc_flags
Like C<L</is_utf8_string_flags>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer.
See also C<L</is_utf8_string_loclen_flags>>.
=cut
*/
#define is_utf8_string_loc_flags(s, len, ep, flags) \
is_utf8_string_loclen_flags(s, len, ep, 0, flags)
/* The above 3 actual functions could have been moved into the more general one
* just below, and made #defines that call it with the right 'flags'. They are
* currently kept separate to increase their chances of getting inlined */
/*
=for apidoc is_utf8_string_loclen_flags
Like C<L</is_utf8_string_flags>> but stores the location of the failure (in the
case of "utf8ness failure") or the location C<s>+C<len> (in the case of
"utf8ness success") in the C<ep> pointer, and the number of UTF-8
encoded characters in the C<el> pointer.
See also C<L</is_utf8_string_loc_flags>>.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_string_loclen_flags(const U8 *s, STRLEN len, const U8 **ep, STRLEN *el, const U32 flags)
{
const U8 * first_variant;
PERL_ARGS_ASSERT_IS_UTF8_STRING_LOCLEN_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
|UTF8_DISALLOW_PERL_EXTENDED)));
if (len == 0) {
len = strlen((const char *) s);
}
if (flags == 0) {
return is_utf8_string_loclen(s, len, ep, el);
}
if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
== UTF8_DISALLOW_ILLEGAL_INTERCHANGE)
{
return is_strict_utf8_string_loclen(s, len, ep, el);
}
if ((flags & ~UTF8_DISALLOW_PERL_EXTENDED)
== UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE)
{
return is_c9strict_utf8_string_loclen(s, len, ep, el);
}
if (is_utf8_invariant_string_loc(s, len, &first_variant)) {
if (el)
*el = len;
if (ep) {
*ep = s + len;
}
return TRUE;
}
{
const U8* send = s + len;
const U8* x = first_variant;
STRLEN outlen = first_variant - s;
while (x < send) {
const STRLEN cur_len = isUTF8_CHAR_flags(x, send, flags);
if (UNLIKELY(! cur_len)) {
break;
}
x += cur_len;
outlen++;
}
if (el)
*el = outlen;
if (ep) {
*ep = x;
}
return (x == send);
}
}
/*
=for apidoc utf8_distance
Returns the number of UTF-8 characters between the UTF-8 pointers C<a>
and C<b>.
WARNING: use only if you *know* that the pointers point inside the
same UTF-8 buffer.
=cut
*/
PERL_STATIC_INLINE IV
Perl_utf8_distance(pTHX_ const U8 *a, const U8 *b)
{
PERL_ARGS_ASSERT_UTF8_DISTANCE;
return (a < b) ? -1 * (IV) utf8_length(a, b) : (IV) utf8_length(b, a);
}
/*
=for apidoc utf8_hop
Return the UTF-8 pointer C<s> displaced by C<off> characters, either
forward or backward.
WARNING: do not use the following unless you *know* C<off> is within
the UTF-8 data pointed to by C<s> *and* that on entry C<s> is aligned
on the first byte of character or just after the last byte of a character.
=cut
*/
PERL_STATIC_INLINE U8 *
Perl_utf8_hop(const U8 *s, SSize_t off)
{
PERL_ARGS_ASSERT_UTF8_HOP;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
if (off >= 0) {
while (off--)
s += UTF8SKIP(s);
}
else {
while (off++) {
s--;
while (UTF8_IS_CONTINUATION(*s))
s--;
}
}
GCC_DIAG_IGNORE(-Wcast-qual)
return (U8 *)s;
GCC_DIAG_RESTORE
}
/*
=for apidoc utf8_hop_forward
Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
forward.
C<off> must be non-negative.
C<s> must be before or equal to C<end>.
When moving forward it will not move beyond C<end>.
Will not exceed this limit even if the string is not valid "UTF-8".
=cut
*/
PERL_STATIC_INLINE U8 *
Perl_utf8_hop_forward(const U8 *s, SSize_t off, const U8 *end)
{
PERL_ARGS_ASSERT_UTF8_HOP_FORWARD;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
assert(s <= end);
assert(off >= 0);
while (off--) {
STRLEN skip = UTF8SKIP(s);
if ((STRLEN)(end - s) <= skip) {
GCC_DIAG_IGNORE(-Wcast-qual)
return (U8 *)end;
GCC_DIAG_RESTORE
}
s += skip;
}
GCC_DIAG_IGNORE(-Wcast-qual)
return (U8 *)s;
GCC_DIAG_RESTORE
}
/*
=for apidoc utf8_hop_back
Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
backward.
C<off> must be non-positive.
C<s> must be after or equal to C<start>.
When moving backward it will not move before C<start>.
Will not exceed this limit even if the string is not valid "UTF-8".
=cut
*/
PERL_STATIC_INLINE U8 *
Perl_utf8_hop_back(const U8 *s, SSize_t off, const U8 *start)
{
PERL_ARGS_ASSERT_UTF8_HOP_BACK;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
assert(start <= s);
assert(off <= 0);
while (off++ && s > start) {
do {
s--;
} while (UTF8_IS_CONTINUATION(*s) && s > start);
}
GCC_DIAG_IGNORE(-Wcast-qual)
return (U8 *)s;
GCC_DIAG_RESTORE
}
/*
=for apidoc utf8_hop_safe
Return the UTF-8 pointer C<s> displaced by up to C<off> characters,
either forward or backward.
When moving backward it will not move before C<start>.
When moving forward it will not move beyond C<end>.
Will not exceed those limits even if the string is not valid "UTF-8".
=cut
*/
PERL_STATIC_INLINE U8 *
Perl_utf8_hop_safe(const U8 *s, SSize_t off, const U8 *start, const U8 *end)
{
PERL_ARGS_ASSERT_UTF8_HOP_SAFE;
/* Note: cannot use UTF8_IS_...() too eagerly here since e.g
* the bitops (especially ~) can create illegal UTF-8.
* In other words: in Perl UTF-8 is not just for Unicode. */
assert(start <= s && s <= end);
if (off >= 0) {
return utf8_hop_forward(s, off, end);
}
else {
return utf8_hop_back(s, off, start);
}
}
/*
=for apidoc isUTF8_CHAR_flags
Evaluates to non-zero if the first few bytes of the string starting at C<s> and
looking no further than S<C<e - 1>> are well-formed UTF-8, as extended by Perl,
that represents some code point, subject to the restrictions given by C<flags>;
otherwise it evaluates to 0. If non-zero, the value gives how many bytes
starting at C<s> comprise the code point's representation. Any bytes remaining
before C<e>, but beyond the ones needed to form the first code point in C<s>,
are not examined.
If C<flags> is 0, this gives the same results as C<L</isUTF8_CHAR>>;
if C<flags> is C<UTF8_DISALLOW_ILLEGAL_INTERCHANGE>, this gives the same results
as C<L</isSTRICT_UTF8_CHAR>>;
and if C<flags> is C<UTF8_DISALLOW_ILLEGAL_C9_INTERCHANGE>, this gives
the same results as C<L</isC9_STRICT_UTF8_CHAR>>.
Otherwise C<flags> may be any combination of the C<UTF8_DISALLOW_I<foo>> flags
understood by C<L</utf8n_to_uvchr>>, with the same meanings.
The three alternative macros are for the most commonly needed validations; they
are likely to run somewhat faster than this more general one, as they can be
inlined into your code.
Use L</is_utf8_string_flags>, L</is_utf8_string_loc_flags>, and
L</is_utf8_string_loclen_flags> to check entire strings.
=cut
*/
PERL_STATIC_INLINE STRLEN
Perl_isUTF8_CHAR_flags(const U8 * const s0, const U8 * const e, const U32 flags)
{
PERL_ARGS_ASSERT_ISUTF8_CHAR_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
|UTF8_DISALLOW_PERL_EXTENDED)));
PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
goto check_success,
DFA_TEASE_APART_FF_,
DFA_RETURN_FAILURE_);
check_success:
return is_utf8_char_helper_(s0, e, flags);
#ifdef HAS_EXTRA_LONG_UTF8
tease_apart_FF:
/* In the case of PL_extended_utf8_dfa_tab, getting here means the input is
* either malformed, or was for the largest possible start byte, which
* indicates perl extended UTF-8, well above the Unicode maximum */
if ( *s0 != I8_TO_NATIVE_UTF8(0xFF)
|| (flags & (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_PERL_EXTENDED)))
{
return 0;
}
/* Otherwise examine the sequence not inline */
return is_utf8_FF_helper_(s0, e,
FALSE /* require full, not partial char */
);
#endif
}
/*
=for apidoc is_utf8_valid_partial_char
Returns 0 if the sequence of bytes starting at C<s> and looking no further than
S<C<e - 1>> is the UTF-8 encoding, as extended by Perl, for one or more code
points. Otherwise, it returns 1 if there exists at least one non-empty
sequence of bytes that when appended to sequence C<s>, starting at position
C<e> causes the entire sequence to be the well-formed UTF-8 of some code point;
otherwise returns 0.
In other words this returns TRUE if C<s> points to a partial UTF-8-encoded code
point.
This is useful when a fixed-length buffer is being tested for being well-formed
UTF-8, but the final few bytes in it don't comprise a full character; that is,
it is split somewhere in the middle of the final code point's UTF-8
representation. (Presumably when the buffer is refreshed with the next chunk
of data, the new first bytes will complete the partial code point.) This
function is used to verify that the final bytes in the current buffer are in
fact the legal beginning of some code point, so that if they aren't, the
failure can be signalled without having to wait for the next read.
=cut
*/
#define is_utf8_valid_partial_char(s, e) \
is_utf8_valid_partial_char_flags(s, e, 0)
/*
=for apidoc is_utf8_valid_partial_char_flags
Like C<L</is_utf8_valid_partial_char>>, it returns a boolean giving whether
or not the input is a valid UTF-8 encoded partial character, but it takes an
extra parameter, C<flags>, which can further restrict which code points are
considered valid.
If C<flags> is 0, this behaves identically to
C<L</is_utf8_valid_partial_char>>. Otherwise C<flags> can be any combination
of the C<UTF8_DISALLOW_I<foo>> flags accepted by C<L</utf8n_to_uvchr>>. If
there is any sequence of bytes that can complete the input partial character in
such a way that a non-prohibited character is formed, the function returns
TRUE; otherwise FALSE. Non character code points cannot be determined based on
partial character input. But many of the other possible excluded types can be
determined from just the first one or two bytes.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_valid_partial_char_flags(const U8 * const s0, const U8 * const e, const U32 flags)
{
PERL_ARGS_ASSERT_IS_UTF8_VALID_PARTIAL_CHAR_FLAGS;
assert(0 == (flags & ~(UTF8_DISALLOW_ILLEGAL_INTERCHANGE
|UTF8_DISALLOW_PERL_EXTENDED)));
PERL_IS_UTF8_CHAR_DFA(s0, e, PL_extended_utf8_dfa_tab,
DFA_RETURN_FAILURE_,
DFA_TEASE_APART_FF_,
NOOP);
/* The NOOP above causes the DFA to drop down here iff the input was a
* partial character. flags=0 => can return TRUE immediately; otherwise we
* need to check (not inline) if the partial character is the beginning of
* a disallowed one */
if (flags == 0) {
return TRUE;
}
return cBOOL(is_utf8_char_helper_(s0, e, flags));
#ifdef HAS_EXTRA_LONG_UTF8
tease_apart_FF:
/* Getting here means the input is either malformed, or, in the case of
* PL_extended_utf8_dfa_tab, was for the largest possible start byte. The
* latter case has to be extended UTF-8, so can fail immediately if that is
* forbidden */
if ( *s0 != I8_TO_NATIVE_UTF8(0xFF)
|| (flags & (UTF8_DISALLOW_SUPER|UTF8_DISALLOW_PERL_EXTENDED)))
{
return 0;
}
return is_utf8_FF_helper_(s0, e,
TRUE /* Require to be a partial character */
);
#endif
}
/*
=for apidoc is_utf8_fixed_width_buf_flags
Returns TRUE if the fixed-width buffer starting at C<s> with length C<len>
is entirely valid UTF-8, subject to the restrictions given by C<flags>;
otherwise it returns FALSE.
If C<flags> is 0, any well-formed UTF-8, as extended by Perl, is accepted
without restriction. If the final few bytes of the buffer do not form a
complete code point, this will return TRUE anyway, provided that
C<L</is_utf8_valid_partial_char_flags>> returns TRUE for them.
If C<flags> in non-zero, it can be any combination of the
C<UTF8_DISALLOW_I<foo>> flags accepted by C<L</utf8n_to_uvchr>>, and with the
same meanings.
This function differs from C<L</is_utf8_string_flags>> only in that the latter
returns FALSE if the final few bytes of the string don't form a complete code
point.
=cut
*/
#define is_utf8_fixed_width_buf_flags(s, len, flags) \
is_utf8_fixed_width_buf_loclen_flags(s, len, 0, 0, flags)
/*
=for apidoc is_utf8_fixed_width_buf_loc_flags
Like C<L</is_utf8_fixed_width_buf_flags>> but stores the location of the
failure in the C<ep> pointer. If the function returns TRUE, C<*ep> will point
to the beginning of any partial character at the end of the buffer; if there is
no partial character C<*ep> will contain C<s>+C<len>.
See also C<L</is_utf8_fixed_width_buf_loclen_flags>>.
=cut
*/
#define is_utf8_fixed_width_buf_loc_flags(s, len, loc, flags) \
is_utf8_fixed_width_buf_loclen_flags(s, len, loc, 0, flags)
/*
=for apidoc is_utf8_fixed_width_buf_loclen_flags
Like C<L</is_utf8_fixed_width_buf_loc_flags>> but stores the number of
complete, valid characters found in the C<el> pointer.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_utf8_fixed_width_buf_loclen_flags(const U8 * const s,
STRLEN len,
const U8 **ep,
STRLEN *el,
const U32 flags)
{
const U8 * maybe_partial;
PERL_ARGS_ASSERT_IS_UTF8_FIXED_WIDTH_BUF_LOCLEN_FLAGS;
if (! ep) {
ep = &maybe_partial;
}
/* If it's entirely valid, return that; otherwise see if the only error is
* that the final few bytes are for a partial character */
return is_utf8_string_loclen_flags(s, len, ep, el, flags)
|| is_utf8_valid_partial_char_flags(*ep, s + len, flags);
}
PERL_STATIC_INLINE UV
Perl_utf8n_to_uvchr_msgs(const U8 *s,
STRLEN curlen,
STRLEN *retlen,
const U32 flags,
U32 * errors,
AV ** msgs)
{
/* This is the inlined portion of utf8n_to_uvchr_msgs. It handles the
* simple cases, and, if necessary calls a helper function to deal with the
* more complex ones. Almost all well-formed non-problematic code points
* are considered simple, so that it's unlikely that the helper function
* will need to be called.
*
* This is an adaptation of the tables and algorithm given in
* https://bjoern.hoehrmann.de/utf-8/decoder/dfa/, which provides
* comprehensive documentation of the original version. A copyright notice
* for the original version is given at the beginning of this file. The
* Perl adapation is documented at the definition of PL_strict_utf8_dfa_tab[].
*/
const U8 * const s0 = s;
const U8 * send = s0 + curlen;
UV uv = 0; /* The 0 silences some stupid compilers */
UV state = 0;
PERL_ARGS_ASSERT_UTF8N_TO_UVCHR_MSGS;
/* This dfa is fast. If it accepts the input, it was for a well-formed,
* non-problematic code point, which can be returned immediately.
* Otherwise we call a helper function to figure out the more complicated
* cases. */
while (s < send && LIKELY(state != 1)) {
UV type = PL_strict_utf8_dfa_tab[*s];
uv = (state == 0)
? ((0xff >> type) & NATIVE_UTF8_TO_I8(*s))
: UTF8_ACCUMULATE(uv, *s);
state = PL_strict_utf8_dfa_tab[256 + state + type];
if (state != 0) {
s++;
continue;
}
if (retlen) {
*retlen = s - s0 + 1;
}
if (errors) {
*errors = 0;
}
if (msgs) {
*msgs = NULL;
}
return UNI_TO_NATIVE(uv);
}
/* Here is potentially problematic. Use the full mechanism */
return _utf8n_to_uvchr_msgs_helper(s0, curlen, retlen, flags, errors, msgs);
}
PERL_STATIC_INLINE UV
Perl_utf8_to_uvchr_buf_helper(pTHX_ const U8 *s, const U8 *send, STRLEN *retlen)
{
PERL_ARGS_ASSERT_UTF8_TO_UVCHR_BUF_HELPER;
assert(s < send);
if (! ckWARN_d(WARN_UTF8)) {
/* EMPTY is not really allowed, and asserts on debugging builds. But
* on non-debugging we have to deal with it, and this causes it to
* return the REPLACEMENT CHARACTER, as the documentation indicates */
return utf8n_to_uvchr(s, send - s, retlen,
(UTF8_ALLOW_ANY | UTF8_ALLOW_EMPTY));
}
else {
UV ret = utf8n_to_uvchr(s, send - s, retlen, 0);
if (retlen && ret == 0 && *s != '\0') {
*retlen = (STRLEN) -1;
}
return ret;
}
}
/* ------------------------------- perl.h ----------------------------- */
/*
=for apidoc_section $utility
=for apidoc is_safe_syscall
Test that the given C<pv> (with length C<len>) doesn't contain any internal
C<NUL> characters.
If it does, set C<errno> to C<ENOENT>, optionally warn using the C<syscalls>
category, and return FALSE.
Return TRUE if the name is safe.
C<what> and C<op_name> are used in any warning.
Used by the C<IS_SAFE_SYSCALL()> macro.
=cut
*/
PERL_STATIC_INLINE bool
Perl_is_safe_syscall(pTHX_ const char *pv, STRLEN len, const char *what, const char *op_name)
{
/* While the Windows CE API provides only UCS-16 (or UTF-16) APIs
* perl itself uses xce*() functions which accept 8-bit strings.
*/
PERL_ARGS_ASSERT_IS_SAFE_SYSCALL;
if (len > 1) {
char *null_at;
if (UNLIKELY((null_at = (char *)memchr(pv, 0, len-1)) != NULL)) {
SETERRNO(ENOENT, LIB_INVARG);
Perl_ck_warner(aTHX_ packWARN(WARN_SYSCALLS),
"Invalid \\0 character in %s for %s: %s\\0%s",
what, op_name, pv, null_at+1);
return FALSE;
}
}
return TRUE;
}
/*
Return true if the supplied filename has a newline character
immediately before the first (hopefully only) NUL.
My original look at this incorrectly used the len from SvPV(), but
that's incorrect, since we allow for a NUL in pv[len-1].
So instead, strlen() and work from there.
This allow for the user reading a filename, forgetting to chomp it,
then calling:
open my $foo, "$file\0";
*/
#ifdef PERL_CORE
PERL_STATIC_INLINE bool
S_should_warn_nl(const char *pv)
{
STRLEN len;
PERL_ARGS_ASSERT_SHOULD_WARN_NL;
len = strlen(pv);
return len > 0 && pv[len-1] == '\n';
}
#endif
#if defined(PERL_IN_PP_C) || defined(PERL_IN_PP_HOT_C)
PERL_STATIC_INLINE bool
S_lossless_NV_to_IV(const NV nv, IV *ivp)
{
/* This function determines if the input NV 'nv' may be converted without
* loss of data to an IV. If not, it returns FALSE taking no other action.
* But if it is possible, it does the conversion, returning TRUE, and
* storing the converted result in '*ivp' */
PERL_ARGS_ASSERT_LOSSLESS_NV_TO_IV;
# if defined(NAN_COMPARE_BROKEN) && defined(Perl_isnan)
/* Normally any comparison with a NaN returns false; if we can't rely
* on that behaviour, check explicitly */
if (UNLIKELY(Perl_isnan(nv))) {
return FALSE;
}
# endif
/* Written this way so that with an always-false NaN comparison we
* return false */
if (!(LIKELY(nv >= (NV) IV_MIN) && LIKELY(nv < IV_MAX_P1))) {
return FALSE;
}
if ((IV) nv != nv) {
return FALSE;
}
*ivp = (IV) nv;
return TRUE;
}
#endif
/* ------------------ pp.c, regcomp.c, toke.c, universal.c ------------ */
#if defined(PERL_IN_PP_C) || defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_TOKE_C) || defined(PERL_IN_UNIVERSAL_C)
#define MAX_CHARSET_NAME_LENGTH 2
PERL_STATIC_INLINE const char *
S_get_regex_charset_name(const U32 flags, STRLEN* const lenp)
{
PERL_ARGS_ASSERT_GET_REGEX_CHARSET_NAME;
/* Returns a string that corresponds to the name of the regex character set
* given by 'flags', and *lenp is set the length of that string, which
* cannot exceed MAX_CHARSET_NAME_LENGTH characters */
*lenp = 1;
switch (get_regex_charset(flags)) {
case REGEX_DEPENDS_CHARSET: return DEPENDS_PAT_MODS;
case REGEX_LOCALE_CHARSET: return LOCALE_PAT_MODS;
case REGEX_UNICODE_CHARSET: return UNICODE_PAT_MODS;
case REGEX_ASCII_RESTRICTED_CHARSET: return ASCII_RESTRICT_PAT_MODS;
case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
*lenp = 2;
return ASCII_MORE_RESTRICT_PAT_MODS;
}
/* The NOT_REACHED; hides an assert() which has a rather complex
* definition in perl.h. */
NOT_REACHED; /* NOTREACHED */
return "?"; /* Unknown */
}
#endif
/*
Return false if any get magic is on the SV other than taint magic.
*/
PERL_STATIC_INLINE bool
Perl_sv_only_taint_gmagic(SV *sv)
{
MAGIC *mg = SvMAGIC(sv);
PERL_ARGS_ASSERT_SV_ONLY_TAINT_GMAGIC;
while (mg) {
if (mg->mg_type != PERL_MAGIC_taint
&& !(mg->mg_flags & MGf_GSKIP)
&& mg->mg_virtual->svt_get) {
return FALSE;
}
mg = mg->mg_moremagic;
}
return TRUE;
}
/* ------------------ cop.h ------------------------------------------- */
/* implement GIMME_V() macro */
PERL_STATIC_INLINE U8
Perl_gimme_V(pTHX)
{
I32 cxix;
U8 gimme = (PL_op->op_flags & OPf_WANT);
if (gimme)
return gimme;
cxix = PL_curstackinfo->si_cxsubix;
if (cxix < 0)
return PL_curstackinfo->si_type == PERLSI_SORT ? G_SCALAR: G_VOID;
assert(cxstack[cxix].blk_gimme & G_WANT);
return (cxstack[cxix].blk_gimme & G_WANT);
}
/* Enter a block. Push a new base context and return its address. */
PERL_STATIC_INLINE PERL_CONTEXT *
Perl_cx_pushblock(pTHX_ U8 type, U8 gimme, SV** sp, I32 saveix)
{
PERL_CONTEXT * cx;
PERL_ARGS_ASSERT_CX_PUSHBLOCK;
CXINC;
cx = CX_CUR();
cx->cx_type = type;
cx->blk_gimme = gimme;
cx->blk_oldsaveix = saveix;
cx->blk_oldsp = (I32)(sp - PL_stack_base);
cx->blk_oldcop = PL_curcop;
cx->blk_oldmarksp = (I32)(PL_markstack_ptr - PL_markstack);
cx->blk_oldscopesp = PL_scopestack_ix;
cx->blk_oldpm = PL_curpm;
cx->blk_old_tmpsfloor = PL_tmps_floor;
PL_tmps_floor = PL_tmps_ix;
CX_DEBUG(cx, "PUSH");
return cx;
}
/* Exit a block (RETURN and LAST). */
PERL_STATIC_INLINE void
Perl_cx_popblock(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_POPBLOCK;
CX_DEBUG(cx, "POP");
/* these 3 are common to cx_popblock and cx_topblock */
PL_markstack_ptr = PL_markstack + cx->blk_oldmarksp;
PL_scopestack_ix = cx->blk_oldscopesp;
PL_curpm = cx->blk_oldpm;
/* LEAVE_SCOPE() should have made this true. /(?{})/ cheats
* and leaves a CX entry lying around for repeated use, so
* skip for multicall */ \
assert( (CxTYPE(cx) == CXt_SUB && CxMULTICALL(cx))
|| PL_savestack_ix == cx->blk_oldsaveix);
PL_curcop = cx->blk_oldcop;
PL_tmps_floor = cx->blk_old_tmpsfloor;
}
/* Continue a block elsewhere (e.g. NEXT, REDO, GOTO).
* Whereas cx_popblock() restores the state to the point just before
* cx_pushblock() was called, cx_topblock() restores it to the point just
* *after* cx_pushblock() was called. */
PERL_STATIC_INLINE void
Perl_cx_topblock(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_TOPBLOCK;
CX_DEBUG(cx, "TOP");
/* these 3 are common to cx_popblock and cx_topblock */
PL_markstack_ptr = PL_markstack + cx->blk_oldmarksp;
PL_scopestack_ix = cx->blk_oldscopesp;
PL_curpm = cx->blk_oldpm;
PL_stack_sp = PL_stack_base + cx->blk_oldsp;
}
PERL_STATIC_INLINE void
Perl_cx_pushsub(pTHX_ PERL_CONTEXT *cx, CV *cv, OP *retop, bool hasargs)
{
U8 phlags = CX_PUSHSUB_GET_LVALUE_MASK(Perl_was_lvalue_sub);
PERL_ARGS_ASSERT_CX_PUSHSUB;
PERL_DTRACE_PROBE_ENTRY(cv);
cx->blk_sub.old_cxsubix = PL_curstackinfo->si_cxsubix;
PL_curstackinfo->si_cxsubix = cx - PL_curstackinfo->si_cxstack;
cx->blk_sub.cv = cv;
cx->blk_sub.olddepth = CvDEPTH(cv);
cx->blk_sub.prevcomppad = PL_comppad;
cx->cx_type |= (hasargs) ? CXp_HASARGS : 0;
cx->blk_sub.retop = retop;
SvREFCNT_inc_simple_void_NN(cv);
cx->blk_u16 = PL_op->op_private & (phlags|OPpDEREF);
}
/* subsets of cx_popsub() */
PERL_STATIC_INLINE void
Perl_cx_popsub_common(pTHX_ PERL_CONTEXT *cx)
{
CV *cv;
PERL_ARGS_ASSERT_CX_POPSUB_COMMON;
assert(CxTYPE(cx) == CXt_SUB);
PL_comppad = cx->blk_sub.prevcomppad;
PL_curpad = LIKELY(PL_comppad) ? AvARRAY(PL_comppad) : NULL;
cv = cx->blk_sub.cv;
CvDEPTH(cv) = cx->blk_sub.olddepth;
cx->blk_sub.cv = NULL;
SvREFCNT_dec(cv);
PL_curstackinfo->si_cxsubix = cx->blk_sub.old_cxsubix;
}
/* handle the @_ part of leaving a sub */
PERL_STATIC_INLINE void
Perl_cx_popsub_args(pTHX_ PERL_CONTEXT *cx)
{
AV *av;
PERL_ARGS_ASSERT_CX_POPSUB_ARGS;
assert(CxTYPE(cx) == CXt_SUB);
assert(AvARRAY(MUTABLE_AV(
PadlistARRAY(CvPADLIST(cx->blk_sub.cv))[
CvDEPTH(cx->blk_sub.cv)])) == PL_curpad);
CX_POP_SAVEARRAY(cx);
av = MUTABLE_AV(PAD_SVl(0));
if (UNLIKELY(AvREAL(av)))
/* abandon @_ if it got reified */
clear_defarray(av, 0);
else {
CLEAR_ARGARRAY(av);
}
}
PERL_STATIC_INLINE void
Perl_cx_popsub(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_POPSUB;
assert(CxTYPE(cx) == CXt_SUB);
PERL_DTRACE_PROBE_RETURN(cx->blk_sub.cv);
if (CxHASARGS(cx))
cx_popsub_args(cx);
cx_popsub_common(cx);
}
PERL_STATIC_INLINE void
Perl_cx_pushformat(pTHX_ PERL_CONTEXT *cx, CV *cv, OP *retop, GV *gv)
{
PERL_ARGS_ASSERT_CX_PUSHFORMAT;
cx->blk_format.old_cxsubix = PL_curstackinfo->si_cxsubix;
PL_curstackinfo->si_cxsubix= cx - PL_curstackinfo->si_cxstack;
cx->blk_format.cv = cv;
cx->blk_format.retop = retop;
cx->blk_format.gv = gv;
cx->blk_format.dfoutgv = PL_defoutgv;
cx->blk_format.prevcomppad = PL_comppad;
cx->blk_u16 = 0;
SvREFCNT_inc_simple_void_NN(cv);
CvDEPTH(cv)++;
SvREFCNT_inc_void(cx->blk_format.dfoutgv);
}
PERL_STATIC_INLINE void
Perl_cx_popformat(pTHX_ PERL_CONTEXT *cx)
{
CV *cv;
GV *dfout;
PERL_ARGS_ASSERT_CX_POPFORMAT;
assert(CxTYPE(cx) == CXt_FORMAT);
dfout = cx->blk_format.dfoutgv;
setdefout(dfout);
cx->blk_format.dfoutgv = NULL;
SvREFCNT_dec_NN(dfout);
PL_comppad = cx->blk_format.prevcomppad;
PL_curpad = LIKELY(PL_comppad) ? AvARRAY(PL_comppad) : NULL;
cv = cx->blk_format.cv;
cx->blk_format.cv = NULL;
--CvDEPTH(cv);
SvREFCNT_dec_NN(cv);
PL_curstackinfo->si_cxsubix = cx->blk_format.old_cxsubix;
}
PERL_STATIC_INLINE void
Perl_push_evalortry_common(pTHX_ PERL_CONTEXT *cx, OP *retop, SV *namesv)
{
cx->blk_eval.retop = retop;
cx->blk_eval.old_namesv = namesv;
cx->blk_eval.old_eval_root = PL_eval_root;
cx->blk_eval.cur_text = PL_parser ? PL_parser->linestr : NULL;
cx->blk_eval.cv = NULL; /* later set by doeval_compile() */
cx->blk_eval.cur_top_env = PL_top_env;
assert(!(PL_in_eval & ~ 0x3F));
assert(!(PL_op->op_type & ~0x1FF));
cx->blk_u16 = (PL_in_eval & 0x3F) | ((U16)PL_op->op_type << 7);
}
PERL_STATIC_INLINE void
Perl_cx_pusheval(pTHX_ PERL_CONTEXT *cx, OP *retop, SV *namesv)
{
PERL_ARGS_ASSERT_CX_PUSHEVAL;
Perl_push_evalortry_common(aTHX_ cx, retop, namesv);
cx->blk_eval.old_cxsubix = PL_curstackinfo->si_cxsubix;
PL_curstackinfo->si_cxsubix = cx - PL_curstackinfo->si_cxstack;
}
PERL_STATIC_INLINE void
Perl_cx_pushtry(pTHX_ PERL_CONTEXT *cx, OP *retop)
{
PERL_ARGS_ASSERT_CX_PUSHTRY;
Perl_push_evalortry_common(aTHX_ cx, retop, NULL);
/* Don't actually change it, just store the current value so it's restored
* by the common popeval */
cx->blk_eval.old_cxsubix = PL_curstackinfo->si_cxsubix;
}
PERL_STATIC_INLINE void
Perl_cx_popeval(pTHX_ PERL_CONTEXT *cx)
{
SV *sv;
PERL_ARGS_ASSERT_CX_POPEVAL;
assert(CxTYPE(cx) == CXt_EVAL);
PL_in_eval = CxOLD_IN_EVAL(cx);
assert(!(PL_in_eval & 0xc0));
PL_eval_root = cx->blk_eval.old_eval_root;
sv = cx->blk_eval.cur_text;
if (sv && CxEVAL_TXT_REFCNTED(cx)) {
cx->blk_eval.cur_text = NULL;
SvREFCNT_dec_NN(sv);
}
sv = cx->blk_eval.old_namesv;
if (sv) {
cx->blk_eval.old_namesv = NULL;
SvREFCNT_dec_NN(sv);
}
PL_curstackinfo->si_cxsubix = cx->blk_eval.old_cxsubix;
}
/* push a plain loop, i.e.
* { block }
* while (cond) { block }
* for (init;cond;continue) { block }
* This loop can be last/redo'ed etc.
*/
PERL_STATIC_INLINE void
Perl_cx_pushloop_plain(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_PUSHLOOP_PLAIN;
cx->blk_loop.my_op = cLOOP;
}
/* push a true for loop, i.e.
* for var (list) { block }
*/
PERL_STATIC_INLINE void
Perl_cx_pushloop_for(pTHX_ PERL_CONTEXT *cx, void *itervarp, SV* itersave)
{
PERL_ARGS_ASSERT_CX_PUSHLOOP_FOR;
/* this one line is common with cx_pushloop_plain */
cx->blk_loop.my_op = cLOOP;
cx->blk_loop.itervar_u.svp = (SV**)itervarp;
cx->blk_loop.itersave = itersave;
#ifdef USE_ITHREADS
cx->blk_loop.oldcomppad = PL_comppad;
#endif
}
/* pop all loop types, including plain */
PERL_STATIC_INLINE void
Perl_cx_poploop(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_POPLOOP;
assert(CxTYPE_is_LOOP(cx));
if ( CxTYPE(cx) == CXt_LOOP_ARY
|| CxTYPE(cx) == CXt_LOOP_LAZYSV)
{
/* Free ary or cur. This assumes that state_u.ary.ary
* aligns with state_u.lazysv.cur. See cx_dup() */
SV *sv = cx->blk_loop.state_u.lazysv.cur;
cx->blk_loop.state_u.lazysv.cur = NULL;
SvREFCNT_dec_NN(sv);
if (CxTYPE(cx) == CXt_LOOP_LAZYSV) {
sv = cx->blk_loop.state_u.lazysv.end;
cx->blk_loop.state_u.lazysv.end = NULL;
SvREFCNT_dec_NN(sv);
}
}
if (cx->cx_type & (CXp_FOR_PAD|CXp_FOR_GV)) {
SV *cursv;
SV **svp = (cx)->blk_loop.itervar_u.svp;
if ((cx->cx_type & CXp_FOR_GV))
svp = &GvSV((GV*)svp);
cursv = *svp;
*svp = cx->blk_loop.itersave;
cx->blk_loop.itersave = NULL;
SvREFCNT_dec(cursv);
}
}
PERL_STATIC_INLINE void
Perl_cx_pushwhen(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_PUSHWHEN;
cx->blk_givwhen.leave_op = cLOGOP->op_other;
}
PERL_STATIC_INLINE void
Perl_cx_popwhen(pTHX_ PERL_CONTEXT *cx)
{
PERL_ARGS_ASSERT_CX_POPWHEN;
assert(CxTYPE(cx) == CXt_WHEN);
PERL_UNUSED_ARG(cx);
PERL_UNUSED_CONTEXT;
/* currently NOOP */
}
PERL_STATIC_INLINE void
Perl_cx_pushgiven(pTHX_ PERL_CONTEXT *cx, SV *orig_defsv)
{
PERL_ARGS_ASSERT_CX_PUSHGIVEN;
cx->blk_givwhen.leave_op = cLOGOP->op_other;
cx->blk_givwhen.defsv_save = orig_defsv;
}
PERL_STATIC_INLINE void
Perl_cx_popgiven(pTHX_ PERL_CONTEXT *cx)
{
SV *sv;
PERL_ARGS_ASSERT_CX_POPGIVEN;
assert(CxTYPE(cx) == CXt_GIVEN);
sv = GvSV(PL_defgv);
GvSV(PL_defgv) = cx->blk_givwhen.defsv_save;
cx->blk_givwhen.defsv_save = NULL;
SvREFCNT_dec(sv);
}
/* ------------------ util.h ------------------------------------------- */
/*
=for apidoc_section $string
=for apidoc foldEQ
Returns true if the leading C<len> bytes of the strings C<s1> and C<s2> are the
same
case-insensitively; false otherwise. Uppercase and lowercase ASCII range bytes
match themselves and their opposite case counterparts. Non-cased and non-ASCII
range bytes match only themselves.
=cut
*/
PERL_STATIC_INLINE I32
Perl_foldEQ(const char *s1, const char *s2, I32 len)
{
const U8 *a = (const U8 *)s1;
const U8 *b = (const U8 *)s2;
PERL_ARGS_ASSERT_FOLDEQ;
assert(len >= 0);
while (len--) {
if (*a != *b && *a != PL_fold[*b])
return 0;
a++,b++;
}
return 1;
}
PERL_STATIC_INLINE I32
Perl_foldEQ_latin1(const char *s1, const char *s2, I32 len)
{
/* Compare non-UTF-8 using Unicode (Latin1) semantics. Works on all folds
* representable without UTF-8, except for LATIN_SMALL_LETTER_SHARP_S, and
* does not check for this. Nor does it check that the strings each have
* at least 'len' characters. */
const U8 *a = (const U8 *)s1;
const U8 *b = (const U8 *)s2;
PERL_ARGS_ASSERT_FOLDEQ_LATIN1;
assert(len >= 0);
while (len--) {
if (*a != *b && *a != PL_fold_latin1[*b]) {
return 0;
}
a++, b++;
}
return 1;
}
/*
=for apidoc_section $locale
=for apidoc foldEQ_locale
Returns true if the leading C<len> bytes of the strings C<s1> and C<s2> are the
same case-insensitively in the current locale; false otherwise.
=cut
*/
PERL_STATIC_INLINE I32
Perl_foldEQ_locale(const char *s1, const char *s2, I32 len)
{
const U8 *a = (const U8 *)s1;
const U8 *b = (const U8 *)s2;
PERL_ARGS_ASSERT_FOLDEQ_LOCALE;
assert(len >= 0);
while (len--) {
if (*a != *b && *a != PL_fold_locale[*b])
return 0;
a++,b++;
}
return 1;
}
/*
=for apidoc_section $string
=for apidoc my_strnlen
The C library C<strnlen> if available, or a Perl implementation of it.
C<my_strnlen()> computes the length of the string, up to C<maxlen>
characters. It will never attempt to address more than C<maxlen>
characters, making it suitable for use with strings that are not
guaranteed to be NUL-terminated.
=cut
Description stolen from http://man.openbsd.org/strnlen.3,
implementation stolen from PostgreSQL.
*/
#ifndef HAS_STRNLEN
PERL_STATIC_INLINE Size_t
Perl_my_strnlen(const char *str, Size_t maxlen)
{
const char *end = (char *) memchr(str, '\0', maxlen);
PERL_ARGS_ASSERT_MY_STRNLEN;
if (end == NULL) return maxlen;
return end - str;
}
#endif
#if ! defined (HAS_MEMRCHR) && (defined(PERL_CORE) || defined(PERL_EXT))
PERL_STATIC_INLINE void *
S_my_memrchr(const char * s, const char c, const STRLEN len)
{
/* memrchr(), since many platforms lack it */
const char * t = s + len - 1;
PERL_ARGS_ASSERT_MY_MEMRCHR;
while (t >= s) {
if (*t == c) {
return (void *) t;
}
t--;
}
return NULL;
}
#endif
PERL_STATIC_INLINE char *
Perl_mortal_getenv(const char * str)
{
/* This implements a (mostly) thread-safe, sequential-call-safe getenv().
*
* It's (mostly) thread-safe because it uses a mutex to prevent other
* threads (that look at this mutex) from destroying the result before this
* routine has a chance to copy the result to a place that won't be
* destroyed before the caller gets a chance to handle it. That place is a
* mortal SV. khw chose this over SAVEFREEPV because he is under the
* impression that the SV will hang around longer under more circumstances
*
* The reason it isn't completely thread-safe is that other code could
* simply not pay attention to the mutex. All of the Perl core uses the
* mutex, but it is possible for code from, say XS, to not use this mutex,
* defeating the safety.
*
* getenv() returns, in some implementations, a pointer to a spot in the
* **environ array, which could be invalidated at any time by this or
* another thread changing the environment. Other implementations copy the
* **environ value to a static buffer, returning a pointer to that. That
* buffer might or might not be invalidated by a getenv() call in another
* thread. If it does get zapped, we need an exclusive lock. Otherwise,
* many getenv() calls can safely be running simultaneously, so a
* many-reader (but no simultaneous writers) lock is ok. There is a
* Configure probe to see if another thread destroys the buffer, and the
* mutex is defined accordingly.
*
* But in all cases, using the mutex prevents these problems, as long as
* all code uses the same mutex..
*
* A complication is that this can be called during phases where the
* mortalization process isn't available. These are in interpreter
* destruction or early in construction. khw believes that at these times
* there shouldn't be anything else going on, so plain getenv is safe AS
* LONG AS the caller acts on the return before calling it again. */
char * ret;
dTHX;
PERL_ARGS_ASSERT_MORTAL_GETENV;
/* Can't mortalize without stacks. khw believes that no other threads
* should be running, so no need to lock things, and this may be during a
* phase when locking isn't even available */
if (UNLIKELY(PL_scopestack_ix == 0)) {
return getenv(str);
}
#ifdef PERL_MEM_LOG
/* A major complication arises under PERL_MEM_LOG. When that is active,
* every memory allocation may result in logging, depending on the value of
* ENV{PERL_MEM_LOG} at the moment. That means, as we create the SV for
* saving ENV{foo}'s value (but before saving it), the logging code will
* call us recursively to find out what ENV{PERL_MEM_LOG} is. Without some
* care that could lead to: 1) infinite recursion; or 2) deadlock (trying to
* lock a boolean mutex recursively); 3) destroying the getenv() static
* buffer; or 4) destroying the temporary created by this for the copy
* causes a log entry to be made which could cause a new temporary to be
* created, which will need to be destroyed at some point, leading to an
* infinite loop.
*
* The solution adopted here (after some gnashing of teeth) is to detect
* the recursive calls and calls from the logger, and treat them specially.
* Let's say we want to do getenv("foo"). We first find
* getenv(PERL_MEM_LOG) and save it to a fixed-length per-interpreter
* variable, so no temporary is required. Then we do getenv(foo}, and in
* the process of creating a temporary to save it, this function will be
* called recursively to do a getenv(PERL_MEM_LOG). On the recursed call,
* we detect that it is such a call and return our saved value instead of
* locking and doing a new getenv(). This solves all of problems 1), 2),
* and 3). Because all the getenv()s are done while the mutex is locked,
* the state cannot have changed. To solve 4), we don't create a temporary
* when this is called from the logging code. That code disposes of the
* return value while the mutex is still locked.
*
* The value of getenv(PERL_MEM_LOG) can be anything, but only initial
* digits and 3 particular letters are significant; the rest are ignored by
* the memory logging code. Thus the per-interpreter variable only needs
* to be large enough to save the significant information, the size of
* which is known at compile time. The first byte is extra, reserved for
* flags for our use. To protect against overflowing, only the reserved
* byte, as many digits as don't overflow, and the three letters are
* stored.
*
* The reserved byte has two bits:
* 0x1 if set indicates that if we get here, it is a recursive call of
* getenv()
* 0x2 if set indicates that the call is from the logging code.
*
* If the flag indicates this is a recursive call, just return the stored
* value of PL_mem_log; An empty value gets turned into NULL. */
if (strEQ(str, "PERL_MEM_LOG") && PL_mem_log[0] & 0x1) {
if (PL_mem_log[1] == '\0') {
return NULL;
} else {
return PL_mem_log + 1;
}
}
#endif
GETENV_LOCK;
#ifdef PERL_MEM_LOG
/* Here we are in a critical section. As explained above, we do our own
* getenv(PERL_MEM_LOG), saving the result safely. */
ret = getenv("PERL_MEM_LOG");
if (ret == NULL) { /* No logging active */
/* Return that immediately if called from the logging code */
if (PL_mem_log[0] & 0x2) {
GETENV_UNLOCK;
return NULL;
}
PL_mem_log[1] = '\0';
}
else {
char *mem_log_meat = PL_mem_log + 1; /* first byte reserved */
/* There is nothing to prevent the value of PERL_MEM_LOG from being an
* extremely long string. But we want only a few characters from it.
* PL_mem_log has been made large enough to hold just the ones we need.
* First the file descriptor. */
if (isDIGIT(*ret)) {
const char * s = ret;
if (UNLIKELY(*s == '0')) {
/* Reduce multiple leading zeros to a single one. This is to
* allow the caller to change what to do with leading zeros. */
*mem_log_meat++ = '0';
s++;
while (*s == '0') {
s++;
}
}
/* If the input overflows, copy just enough for the result to also
* overflow, plus 1 to make sure */
while (isDIGIT(*s) && s < ret + TYPE_DIGITS(UV) + 1) {
*mem_log_meat++ = *s++;
}
}
/* Then each of the three significant characters */
if (strchr(ret, 'm')) {
*mem_log_meat++ = 'm';
}
if (strchr(ret, 's')) {
*mem_log_meat++ = 's';
}
if (strchr(ret, 't')) {
*mem_log_meat++ = 't';
}
*mem_log_meat = '\0';
assert(mem_log_meat < PL_mem_log + sizeof(PL_mem_log));
}
/* If we are being called from the logger, it only needs the significant
* portion of PERL_MEM_LOG, and doesn't need a safe copy */
if (PL_mem_log[0] & 0x2) {
assert(strEQ(str, "PERL_MEM_LOG"));
GETENV_UNLOCK;
return PL_mem_log + 1;
}
/* Here is a generic getenv(). This could be a getenv("PERL_MEM_LOG") that
* is coming from other than the logging code, so it should be treated the
* same as any other getenv(), returning the full value, not just the
* significant part, and having its value saved. Set the flag that
* indicates any call to this routine will be a recursion from here */
PL_mem_log[0] = 0x1;
#endif
/* Now get the value of the real desired variable, and save a copy */
ret = getenv(str);
if (ret != NULL) {
ret = SvPVX(sv_2mortal(newSVpv(ret, 0)));
}
GETENV_UNLOCK;
#ifdef PERL_MEM_LOG
/* Clear the buffer */
Zero(PL_mem_log, sizeof(PL_mem_log), char);
#endif
return ret;
}
PERL_STATIC_INLINE bool
Perl_sv_isbool(pTHX_ const SV *sv)
{
return SvIOK(sv) && SvPOK(sv) && SvIsCOW_static(sv) &&
(SvPVX_const(sv) == PL_Yes || SvPVX_const(sv) == PL_No);
}
#ifdef USE_ITHREADS
PERL_STATIC_INLINE AV *
Perl_cop_file_avn(pTHX_ const COP *cop) {
PERL_ARGS_ASSERT_COP_FILE_AVN;
const char *file = CopFILE(cop);
if (file) {
GV *gv = gv_fetchfile_flags(file, strlen(file), GVF_NOADD);
if (gv) {
return GvAVn(gv);
}
else
return NULL;
}
else
return NULL;
}
#endif
/*
* ex: set ts=8 sts=4 sw=4 et:
*/
|