This file is indexed.

/usr/share/perl5/HTML/Element.pm is in libhtml-tree-perl 4.2-1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

   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
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
package HTML::Element;

=head1 NAME

HTML::Element - Class for objects that represent HTML elements

=head1 VERSION

Version 4.2

=cut

use strict;
use Carp           ();
use HTML::Entities ();
use HTML::Tagset   ();
use integer;    # vroom vroom!

use vars qw( $VERSION );
$VERSION = 4.2;

# This contorls encoding entities on output.
# When set entities won't be re-encoded.
# Defaulting off because parser defaults to unencoding entities
our $encoded_content = 0;

use vars qw($html_uc $Debug $ID_COUNTER %list_type_to_sub);

=head1 SYNOPSIS

    use HTML::Element;
    $a = HTML::Element->new('a', href => 'http://www.perl.com/');
    $a->push_content("The Perl Homepage");

    $tag = $a->tag;
    print "$tag starts out as:",  $a->starttag, "\n";
    print "$tag ends as:",  $a->endtag, "\n";
    print "$tag\'s href attribute is: ", $a->attr('href'), "\n";

    $links_r = $a->extract_links();
    print "Hey, I found ", scalar(@$links_r), " links.\n";

    print "And that, as HTML, is: ", $a->as_HTML, "\n";
    $a = $a->delete;

=head1 DESCRIPTION

(This class is part of the L<HTML::Tree|HTML::Tree> dist.)

Objects of the HTML::Element class can be used to represent elements
of HTML document trees.  These objects have attributes, notably attributes that
designates each element's parent and content.  The content is an array
of text segments and other HTML::Element objects.  A tree with HTML::Element
objects as nodes can represent the syntax tree for a HTML document.

=head1 HOW WE REPRESENT TREES

Consider this HTML document:

  <html lang='en-US'>
    <head>
      <title>Stuff</title>
      <meta name='author' content='Jojo'>
    </head>
    <body>
     <h1>I like potatoes!</h1>
    </body>
  </html>

Building a syntax tree out of it makes a tree-structure in memory
that could be diagrammed as:

                     html (lang='en-US')
                      / \
                    /     \
                  /         \
                head        body
               /\               \
             /    \               \
           /        \               \
         title     meta              h1
          |       (name='author',     |
       "Stuff"    content='Jojo')    "I like potatoes"

This is the traditional way to diagram a tree, with the "root" at the
top, and it's this kind of diagram that people have in mind when they
say, for example, that "the meta element is under the head element
instead of under the body element".  (The same is also said with
"inside" instead of "under" -- the use of "inside" makes more sense
when you're looking at the HTML source.)

Another way to represent the above tree is with indenting:

  html (attributes: lang='en-US')
    head
      title
        "Stuff"
      meta (attributes: name='author' content='Jojo')
    body
      h1
        "I like potatoes"

Incidentally, diagramming with indenting works much better for very
large trees, and is easier for a program to generate.  The C<< $tree->dump >>
method uses indentation just that way.

However you diagram the tree, it's stored the same in memory -- it's a
network of objects, each of which has attributes like so:

  element #1:  _tag: 'html'
               _parent: none
               _content: [element #2, element #5]
               lang: 'en-US'

  element #2:  _tag: 'head'
               _parent: element #1
               _content: [element #3, element #4]

  element #3:  _tag: 'title'
               _parent: element #2
               _content: [text segment "Stuff"]

  element #4   _tag: 'meta'
               _parent: element #2
               _content: none
               name: author
               content: Jojo

  element #5   _tag: 'body'
               _parent: element #1
               _content: [element #6]

  element #6   _tag: 'h1'
               _parent: element #5
               _content: [text segment "I like potatoes"]

The "treeness" of the tree-structure that these elements comprise is
not an aspect of any particular object, but is emergent from the
relatedness attributes (_parent and _content) of these element-objects
and from how you use them to get from element to element.

While you could access the content of a tree by writing code that says
"access the 'src' attribute of the root's I<first> child's I<seventh>
child's I<third> child", you're more likely to have to scan the contents
of a tree, looking for whatever nodes, or kinds of nodes, you want to
do something with.  The most straightforward way to look over a tree
is to "traverse" it; an HTML::Element method (C<< $h->traverse >>) is
provided for this purpose; and several other HTML::Element methods are
based on it.

(For everything you ever wanted to know about trees, and then some,
see Niklaus Wirth's I<Algorithms + Data Structures = Programs> or
Donald Knuth's I<The Art of Computer Programming, Volume 1>.)

=cut

$Debug = 0 unless defined $Debug;

=head2 Version

Why is this a sub?

=cut

sub Version { $VERSION; }

my $nillio = [];

*HTML::Element::emptyElement   = \%HTML::Tagset::emptyElement;      # legacy
*HTML::Element::optionalEndTag = \%HTML::Tagset::optionalEndTag;    # legacy
*HTML::Element::linkElements   = \%HTML::Tagset::linkElements;      # legacy
*HTML::Element::boolean_attr   = \%HTML::Tagset::boolean_attr;      # legacy
*HTML::Element::canTighten     = \%HTML::Tagset::canTighten;        # legacy

# Constants for signalling back to the traverser:
my $travsignal_package = __PACKAGE__ . '::_travsignal';
my ( $ABORT, $PRUNE, $PRUNE_SOFTLY, $OK, $PRUNE_UP )
    = map { my $x = $_; bless \$x, $travsignal_package; }
    qw(
    ABORT  PRUNE   PRUNE_SOFTLY   OK   PRUNE_UP
);

=head2 ABORT OK PRUNE PRUNE_SOFTLY PRUNE_UP

Constants for signalling back to the traverser

=cut

## Comments from Father Chrysostomos RT #58880
## The sole purpose for empty parentheses after a sub name is to make it
## parse as a 0-ary (nihilary?) function. I.e., ABORT+1 should parse as
## ABORT()+1, not ABORT(+1). The parentheses also tell perl that it can
### be inlined.
##Deparse is really useful for demonstrating this:
##$ perl -MO=Deparse,-p -e 'sub ABORT {7} print ABORT+8'
# Vs
# perl -MO=Deparse,-p -e 'sub ABORT() {7} print ABORT+8'
#
# With the parentheses, it not only makes it parse as a term.
# It even resolves the constant at compile-time, making the code run faster.

## no critic
sub ABORT ()        {$ABORT}
sub PRUNE ()        {$PRUNE}
sub PRUNE_SOFTLY () {$PRUNE_SOFTLY}
sub OK ()           {$OK}
sub PRUNE_UP ()     {$PRUNE_UP}
## use critic

$html_uc = 0;

# set to 1 if you want tag and attribute names from starttag and endtag
#  to be uc'd

# regexs for XML names
# http://www.w3.org/TR/2006/REC-xml11-20060816/NT-NameStartChar
my $START_CHAR
    = qr/(?:\:|[A-Z]|_|[a-z]|[\x{C0}-\x{D6}]|[\x{D8}-\x{F6}]|[\x{F8}-\x{2FF}]|[\x{370}-\x{37D}]|[\x{37F}-\x{1FFF}]|[\x{200C}-\x{200D}]|[\x{2070}-\x{218F}]|[\x{2C00}-\x{2FEF}]|[\x{3001}-\x{D7FF}]|[\x{F900}-\x{FDCF}]|[\x{FDF0}-\x{FFFD}]|[\x{10000}-\x{EFFFF}])/;

# http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-NameChar
my $NAME_CHAR
    = qr/(?:$START_CHAR|-|\.|[0-9]|\x{B7}|[\x{0300}-\x{036F}]|[\x{203F}-\x{2040}])/;

# Elements that does not have corresponding end tags (i.e. are empty)

#==========================================================================

=head1 BASIC METHODS

=head2 $h = HTML::Element->new('tag', 'attrname' => 'value', ... )

This constructor method returns a new HTML::Element object.  The tag
name is a required argument; it will be forced to lowercase.
Optionally, you can specify other initial attributes at object
creation time.

=cut

#
# An HTML::Element is represented by blessed hash reference, much like
# Tree::DAG_Node objects.  Key-names not starting with '_' are reserved
# for the SGML attributes of the element.
# The following special keys are used:
#
#    '_tag':    The tag name (i.e., the generic identifier)
#    '_parent': A reference to the HTML::Element above (when forming a tree)
#    '_pos':    The current position (a reference to a HTML::Element) is
#               where inserts will be placed (look at the insert_element
#               method)  If not set, the implicit value is the object itself.
#    '_content': A ref to an array of nodes under this.
#                It might not be set.
#
# Example: <img src="gisle.jpg" alt="Gisle's photo"> is represented like this:
#
#  bless {
#     _tag => 'img',
#     src  => 'gisle.jpg',
#     alt  => "Gisle's photo",
#  }, 'HTML::Element';
#

sub new {
    my $class = shift;
    $class = ref($class) || $class;

    my $tag = shift;
    Carp::croak("No tagname") unless defined $tag and length $tag;
    Carp::croak "\"$tag\" isn't a good tag name!"
        if $tag =~ m/[<>\/\x00-\x20]/;    # minimal sanity, certainly!
    my $self = bless { _tag => scalar( $class->_fold_case($tag) ) }, $class;
    my ( $attr, $val );
    while ( ( $attr, $val ) = splice( @_, 0, 2 ) ) {
## RT #42209 why does this default to the attribute name and not remain unset or the empty string?
        $val = $attr unless defined $val;
        $self->{ $class->_fold_case($attr) } = $val;
    }
    if ( $tag eq 'html' ) {
        $self->{'_pos'} = undef;
    }
    return $self;
}

=head2 $h->attr('attr') or $h->attr('attr', 'value')

Returns (optionally sets) the value of the given attribute of $h.  The
attribute name (but not the value, if provided) is forced to
lowercase.  If trying to read the value of an attribute not present
for this element, the return value is undef.
If setting a new value, the old value of that attribute is
returned.

If methods are provided for accessing an attribute (like C<< $h->tag >> for
"_tag", C<< $h->content_list >>, etc. below), use those instead of calling
attr C<< $h->attr >>, whether for reading or setting.

Note that setting an attribute to C<undef> (as opposed to "", the empty
string) actually deletes the attribute.

=cut

sub attr {
    my $self = shift;
    my $attr = scalar( $self->_fold_case(shift) );
    if (@_) {    # set
        if ( defined $_[0] ) {
            my $old = $self->{$attr};
            $self->{$attr} = $_[0];
            return $old;
        }
        else {    # delete, actually
            return delete $self->{$attr};
        }
    }
    else {        # get
        return $self->{$attr};
    }
}

=head2 $h->tag() or $h->tag('tagname')

Returns (optionally sets) the tag name (also known as the generic
identifier) for the element $h.  In setting, the tag name is always
converted to lower case.

There are four kinds of "pseudo-elements" that show up as
HTML::Element objects:

=over

=item Comment pseudo-elements

These are element objects with a C<$h-E<gt>tag> value of "~comment",
and the content of the comment is stored in the "text" attribute
(C<$h-E<gt>attr("text")>).  For example, parsing this code with
HTML::TreeBuilder...

  <!-- I like Pie.
     Pie is good
  -->

produces an HTML::Element object with these attributes:

  "_tag",
  "~comment",
  "text",
  " I like Pie.\n     Pie is good\n  "

=item Declaration pseudo-elements

Declarations (rarely encountered) are represented as HTML::Element
objects with a tag name of "~declaration", and content in the "text"
attribute.  For example, this:

  <!DOCTYPE foo>

produces an element whose attributes include:

  "_tag", "~declaration", "text", "DOCTYPE foo"

=item Processing instruction pseudo-elements

PIs (rarely encountered) are represented as HTML::Element objects with
a tag name of "~pi", and content in the "text" attribute.  For
example, this:

  <?stuff foo?>

produces an element whose attributes include:

  "_tag", "~pi", "text", "stuff foo?"

(assuming a recent version of HTML::Parser)

=item ~literal pseudo-elements

These objects are not currently produced by HTML::TreeBuilder, but can
be used to represent a "super-literal" -- i.e., a literal you want to
be immune from escaping.  (Yes, I just made that term up.)

That is, this is useful if you want to insert code into a tree that
you plan to dump out with C<as_HTML>, where you want, for some reason,
to suppress C<as_HTML>'s normal behavior of amp-quoting text segments.

For example, this:

  my $literal = HTML::Element->new('~literal',
    'text' => 'x < 4 & y > 7'
  );
  my $span = HTML::Element->new('span');
  $span->push_content($literal);
  print $span->as_HTML;

prints this:

  <span>x < 4 & y > 7</span>

Whereas this:

  my $span = HTML::Element->new('span');
  $span->push_content('x < 4 & y > 7');
    # normal text segment
  print $span->as_HTML;

prints this:

  <span>x &lt; 4 &amp; y &gt; 7</span>

Unless you're inserting lots of pre-cooked code into existing trees,
and dumping them out again, it's not likely that you'll find
C<~literal> pseudo-elements useful.

=back

=cut

sub tag {
    my $self = shift;
    if (@_) {    # set
        $self->{'_tag'} = $self->_fold_case( $_[0] );
    }
    else {       # get
        $self->{'_tag'};
    }
}

=head2 $h->parent() or $h->parent($new_parent)

Returns (optionally sets) the parent (aka "container") for this element.
The parent should either be undef, or should be another element.

You B<should not> use this to directly set the parent of an element.
Instead use any of the other methods under "Structure-Modifying
Methods", below.

Note that not($h->parent) is a simple test for whether $h is the
root of its subtree.

=cut

sub parent {
    my $self = shift;
    if (@_) {    # set
        Carp::croak "an element can't be made its own parent"
            if defined $_[0] and ref $_[0] and $self eq $_[0];    # sanity
        $self->{'_parent'} = $_[0];
    }
    else {
        $self->{'_parent'};                                       # get
    }
}

=head2 $h->content_list()

Returns a list of the child nodes of this element -- i.e., what
nodes (elements or text segments) are inside/under this element. (Note
that this may be an empty list.)

In a scalar context, this returns the count of the items,
as you may expect.

=cut

sub content_list {
    return wantarray
        ? @{ shift->{'_content'} || return () }
        : scalar @{ shift->{'_content'} || return 0 };
}

=head2 $h->content()

This somewhat deprecated method returns the content of this element;
but unlike content_list, this returns either undef (which you should
understand to mean no content), or a I<reference to the array> of
content items, each of which is either a text segment (a string, i.e.,
a defined non-reference scalar value), or an HTML::Element object.
Note that even if an arrayref is returned, it may be a reference to an
empty array.

While older code should feel free to continue to use C<< $h->content >>,
new code should use C<< $h->content_list >> in almost all conceivable
cases.  It is my experience that in most cases this leads to simpler
code anyway, since it means one can say:

    @children = $h->content_list;

instead of the inelegant:

    @children = @{$h->content || []};

If you do use C<< $h->content >> (or C<< $h->content_array_ref >>), you should not
use the reference returned by it (assuming it returned a reference,
and not undef) to directly set or change the content of an element or
text segment!  Instead use L<content_refs_list> or any of the other
methods under "Structure-Modifying Methods", below.

=cut

# a read-only method!  can't say $h->content( [] )!
sub content {
    return shift->{'_content'};
}

=head2 $h->content_array_ref()

This is like C<content> (with all its caveats and deprecations) except
that it is guaranteed to return an array reference.  That is, if the
given node has no C<_content> attribute, the C<content> method would
return that undef, but C<content_array_ref> would set the given node's
C<_content> value to C<[]> (a reference to a new, empty array), and
return that.

=cut

sub content_array_ref {
    return shift->{'_content'} ||= [];
}

=head2 $h->content_refs_list

This returns a list of scalar references to each element of C<$h>'s
content list.  This is useful in case you want to in-place edit any
large text segments without having to get a copy of the current value
of that segment value, modify that copy, then use the
C<splice_content> to replace the old with the new.  Instead, here you
can in-place edit:

    foreach my $item_r ($h->content_refs_list) {
        next if ref $$item_r;
        $$item_r =~ s/honour/honor/g;
    }

You I<could> currently achieve the same affect with:

    foreach my $item (@{ $h->content_array_ref }) {
        # deprecated!
        next if ref $item;
        $item =~ s/honour/honor/g;
    }

...except that using the return value of C<< $h->content >> or
C<< $h->content_array_ref >> to do that is deprecated, and just might stop
working in the future.

=cut

sub content_refs_list {
    return \( @{ shift->{'_content'} || return () } );
}

=head2 $h->implicit() or $h->implicit($bool)

Returns (optionally sets) the "_implicit" attribute.  This attribute is
a flag that's used for indicating that the element was not originally
present in the source, but was added to the parse tree (by
HTML::TreeBuilder, for example) in order to conform to the rules of
HTML structure.

=cut

sub implicit {
    return shift->attr( '_implicit', @_ );
}

=head2 $h->pos() or $h->pos($element)

Returns (and optionally sets) the "_pos" (for "current I<pos>ition")
pointer of C<$h>.  This attribute is a pointer used during some
parsing operations, whose value is whatever HTML::Element element
at or under C<$h> is currently "open", where C<< $h->insert_element(NEW) >>
will actually insert a new element.

(This has nothing to do with the Perl function called "pos", for
controlling where regular expression matching starts.)

If you set C<< $h->pos($element) >>, be sure that C<$element> is
either C<$h>, or an element under C<$h>.

If you've been modifying the tree under C<$h> and are no longer
sure C<< $h->pos >> is valid, you can enforce validity with:

    $h->pos(undef) unless $h->pos->is_inside($h);

=cut

sub pos {
    my $self = shift;
    my $pos  = $self->{'_pos'};
    if (@_) {    # set
        my $parm = shift;
        if ( defined $parm and $parm ne $self ) {
            $self->{'_pos'} = $parm;    # means that element
        }
        else {
            $self->{'_pos'} = undef;    # means $self
        }
    }
    return $pos if defined($pos);
    return $self;
}

=head2 $h->all_attr()

Returns all this element's attributes and values, as key-value pairs.
This will include any "internal" attributes (i.e., ones not present
in the original element, and which will not be represented if/when you
call C<< $h->as_HTML >>).  Internal attributes are distinguished by the fact
that the first character of their key (not value! key!) is an
underscore ("_").

Example output of C<< $h->all_attr() >> :
C<'_parent', >I<[object_value]>C< , '_tag', 'em', 'lang', 'en-US',
'_content', >I<[array-ref value]>.

=head2 $h->all_attr_names()

Like all_attr, but only returns the names of the attributes.

Example output of C<< $h->all_attr_names() >> :
C<'_parent', '_tag', 'lang', '_content', >.

=cut

sub all_attr {
    return %{ $_[0] };

    # Yes, trivial.  But no other way for the user to do the same
    #  without breaking encapsulation.
    # And if our object representation changes, this method's behavior
    #  should stay the same.
}

sub all_attr_names {
    return keys %{ $_[0] };
}

=head2 $h->all_external_attr()

Like C<all_attr>, except that internal attributes are not present.

=head2 $h->all_external_attr_names()

Like C<all_external_attr_names>, except that internal attributes' names
are not present.

=cut

sub all_external_attr {
    my $self = $_[0];
    return map( ( length($_) && substr( $_, 0, 1 ) eq '_' )
        ? ()
        : ( $_, $self->{$_} ),
        keys %$self );
}

sub all_external_attr_names {
    return grep !( length($_) && substr( $_, 0, 1 ) eq '_' ), keys %{ $_[0] };
}

=head2 $h->id() or $h->id($string)

Returns (optionally sets to C<$string>) the "id" attribute.
C<< $h->id(undef) >> deletes the "id" attribute.

=cut

sub id {
    if ( @_ == 1 ) {
        return $_[0]{'id'};
    }
    elsif ( @_ == 2 ) {
        if ( defined $_[1] ) {
            return $_[0]{'id'} = $_[1];
        }
        else {
            return delete $_[0]{'id'};
        }
    }
    else {
        Carp::croak '$node->id can\'t take ' . scalar(@_) . ' parameters!';
    }
}

=head2 $h->idf() or $h->idf($string)

Just like the C<id> method, except that if you call C<< $h->idf() >> and
no "id" attribute is defined for this element, then it's set to a
likely-to-be-unique value, and returned.  (The "f" is for "force".)

=cut

sub _gensym {
    unless ( defined $ID_COUNTER ) {

        # start it out...
        $ID_COUNTER = sprintf( '%04x', rand(0x1000) );
        $ID_COUNTER =~ tr<0-9a-f><J-NP-Z>;    # yes, skip letter "oh"
        $ID_COUNTER .= '00000';
    }
    ++$ID_COUNTER;
}

sub idf {
    my $nparms = scalar @_;

    if ( $nparms == 1 ) {
        my $x;
        if ( defined( $x = $_[0]{'id'} ) and length $x ) {
            return $x;
        }
        else {
            return $_[0]{'id'} = _gensym();
        }
    }
    if ( $nparms == 2 ) {
        if ( defined $_[1] ) {
            return $_[0]{'id'} = $_[1];
        }
        else {
            return delete $_[0]{'id'};
        }
    }
    Carp::croak '$node->idf can\'t take ' . scalar(@_) . ' parameters!';
}

=head1 STRUCTURE-MODIFYING METHODS

These methods are provided for modifying the content of trees
by adding or changing nodes as parents or children of other nodes.

=head2 $h->push_content($element_or_text, ...)

Adds the specified items to the I<end> of the content list of the
element C<$h>.  The items of content to be added should each be either a
text segment (a string), an HTML::Element object, or an arrayref.
Arrayrefs are fed thru C<< $h->new_from_lol(that_arrayref) >> to
convert them into elements, before being added to the content
list of C<$h>.  This means you can say things concise things like:

  $body->push_content(
    ['br'],
    ['ul',
      map ['li', $_], qw(Peaches Apples Pears Mangos)
    ]
  );

See C<new_from_lol> method's documentation, far below, for more
explanation.

The push_content method will try to consolidate adjacent text segments
while adding to the content list.  That's to say, if $h's content_list is

  ('foo bar ', $some_node, 'baz!')

and you call

   $h->push_content('quack?');

then the resulting content list will be this:

  ('foo bar ', $some_node, 'baz!quack?')

and not this:

  ('foo bar ', $some_node, 'baz!', 'quack?')

If that latter is what you want, you'll have to override the
feature of consolidating text by using splice_content,
as in:

  $h->splice_content(scalar($h->content_list),0,'quack?');

Similarly, if you wanted to add 'Skronk' to the beginning of
the content list, calling this:

   $h->unshift_content('Skronk');

then the resulting content list will be this:

  ('Skronkfoo bar ', $some_node, 'baz!')

and not this:

  ('Skronk', 'foo bar ', $some_node, 'baz!')

What you'd to do get the latter is:

  $h->splice_content(0,0,'Skronk');

=cut

sub push_content {
    my $self = shift;
    return $self unless @_;

    my $content = ( $self->{'_content'} ||= [] );
    for (@_) {
        if ( ref($_) eq 'ARRAY' ) {

            # magically call new_from_lol
            push @$content, $self->new_from_lol($_);
            $content->[-1]->{'_parent'} = $self;
        }
        elsif ( ref($_) ) {    # insert an element
            $_->detach if $_->{'_parent'};
            $_->{'_parent'} = $self;
            push( @$content, $_ );
        }
        else {                 # insert text segment
            if ( @$content && !ref $content->[-1] ) {

                # last content element is also text segment -- append
                $content->[-1] .= $_;
            }
            else {
                push( @$content, $_ );
            }
        }
    }
    return $self;
}

=head2 $h->unshift_content($element_or_text, ...)

Just like C<push_content>, but adds to the I<beginning> of the $h
element's content list.

The items of content to be added should each be
either a text segment (a string), an HTML::Element object, or
an arrayref (which is fed thru C<new_from_lol>).

The unshift_content method will try to consolidate adjacent text segments
while adding to the content list.  See above for a discussion of this.

=cut

sub unshift_content {
    my $self = shift;
    return $self unless @_;

    my $content = ( $self->{'_content'} ||= [] );
    for ( reverse @_ ) {    # so they get added in the order specified
        if ( ref($_) eq 'ARRAY' ) {

            # magically call new_from_lol
            unshift @$content, $self->new_from_lol($_);
            $content->[0]->{'_parent'} = $self;
        }
        elsif ( ref $_ ) {    # insert an element
            $_->detach if $_->{'_parent'};
            $_->{'_parent'} = $self;
            unshift( @$content, $_ );
        }
        else {                # insert text segment
            if ( @$content && !ref $content->[0] ) {

                # last content element is also text segment -- prepend
                $content->[0] = $_ . $content->[0];
            }
            else {
                unshift( @$content, $_ );
            }
        }
    }
    return $self;
}

# Cf.  splice ARRAY,OFFSET,LENGTH,LIST

=head2 $h->splice_content($offset, $length, $element_or_text, ...)

Detaches the elements from $h's list of content-nodes, starting at
$offset and continuing for $length items, replacing them with the
elements of the following list, if any.  Returns the elements (if any)
removed from the content-list.  If $offset is negative, then it starts
that far from the end of the array, just like Perl's normal C<splice>
function.  If $length and the following list is omitted, removes
everything from $offset onward.

The items of content to be added (if any) should each be either a text
segment (a string), an arrayref (which is fed thru C<new_from_lol>),
or an HTML::Element object that's not already
a child of $h.

=cut

sub splice_content {
    my ( $self, $offset, $length, @to_add ) = @_;
    Carp::croak "splice_content requires at least one argument"
        if @_ < 2;    # at least $h->splice_content($offset);
    return $self unless @_;

    my $content = ( $self->{'_content'} ||= [] );

    # prep the list

    my @out;
    if ( @_ > 2 ) {    # self, offset, length, ...
        foreach my $n (@to_add) {
            if ( ref($n) eq 'ARRAY' ) {
                $n = $self->new_from_lol($n);
                $n->{'_parent'} = $self;
            }
            elsif ( ref($n) ) {
                $n->detach;
                $n->{'_parent'} = $self;
            }
        }
        @out = splice @$content, $offset, $length, @to_add;
    }
    else {    #  self, offset
        @out = splice @$content, $offset;
    }
    foreach my $n (@out) {
        $n->{'_parent'} = undef if ref $n;
    }
    return @out;
}

=head2 $h->detach()

This unlinks $h from its parent, by setting its 'parent' attribute to
undef, and by removing it from the content list of its parent (if it
had one).  The return value is the parent that was detached from (or
undef, if $h had no parent to start with).  Note that neither $h nor
its parent are explicitly destroyed.

=cut

sub detach {
    my $self = $_[0];
    return unless ( my $parent = $self->{'_parent'} );
    $self->{'_parent'} = undef;
    my $cohort = $parent->{'_content'} || return $parent;
    @$cohort = grep { not( ref($_) and $_ eq $self ) } @$cohort;

    # filter $self out, if parent has any evident content

    return $parent;
}

=head2 $h->detach_content()

This unlinks all of $h's children from $h, and returns them.
Note that these are not explicitly destroyed; for that, you
can just use $h->delete_content.

=cut

sub detach_content {
    my $c = $_[0]->{'_content'} || return ();    # in case of no content
    for (@$c) {
        $_->{'_parent'} = undef if ref $_;
    }
    return splice @$c;
}

=head2 $h->replace_with( $element_or_text, ... ) 

This replaces C<$h> in its parent's content list with the nodes
specified.  The element C<$h> (which by then may have no parent)
is returned.  This causes a fatal error if C<$h> has no parent.
The list of nodes to insert may contain C<$h>, but at most once.
Aside from that possible exception, the nodes to insert should not
already be children of C<$h>'s parent.

Also, note that this method does not destroy C<$h> -- use
C<< $h->replace_with(...)->delete >> if you need that.

=cut

sub replace_with {
    my ( $self, @replacers ) = @_;
    Carp::croak "the target node has no parent"
        unless my ($parent) = $self->{'_parent'};

    my $parent_content = $parent->{'_content'};
    Carp::croak "the target node's parent has no content!?"
        unless $parent_content and @$parent_content;

    my $replacers_contains_self;
    for (@replacers) {
        if ( !ref $_ ) {

            # noop
        }
        elsif ( $_ eq $self ) {

            # noop, but check that it's there just once.
            Carp::croak "Replacement list contains several copies of target!"
                if $replacers_contains_self++;
        }
        elsif ( $_ eq $parent ) {
            Carp::croak "Can't replace an item with its parent!";
        }
        elsif ( ref($_) eq 'ARRAY' ) {
            $_ = $self->new_from_lol($_);
            $_->{'_parent'} = $parent;
        }
        else {
            $_->detach;
            $_->{'_parent'} = $parent;

            # each of these are necessary
        }
    }    # for @replacers
    @$parent_content = map { ( ref($_) and $_ eq $self ) ? @replacers : $_ }
        @$parent_content;

    $self->{'_parent'} = undef unless $replacers_contains_self;

    # if replacers does contain self, then the parent attribute is fine as-is

    return $self;
}

=head2 $h->preinsert($element_or_text...)

Inserts the given nodes right BEFORE C<$h> in C<$h>'s parent's
content list.  This causes a fatal error if C<$h> has no parent.
None of the given nodes should be C<$h> or other children of C<$h>.
Returns C<$h>.

=cut

sub preinsert {
    my $self = shift;
    return $self unless @_;
    return $self->replace_with( @_, $self );
}

=head2 $h->postinsert($element_or_text...)

Inserts the given nodes right AFTER C<$h> in C<$h>'s parent's content
list.  This causes a fatal error if C<$h> has no parent.  None of
the given nodes should be C<$h> or other children of C<$h>.  Returns
C<$h>.

=cut

sub postinsert {
    my $self = shift;
    return $self unless @_;
    return $self->replace_with( $self, @_ );
}

=head2 $h->replace_with_content()

This replaces C<$h> in its parent's content list with its own content.
The element C<$h> (which by then has no parent or content of its own) is
returned.  This causes a fatal error if C<$h> has no parent.  Also, note
that this does not destroy C<$h> -- use
C<< $h->replace_with_content->delete >> if you need that.

=cut

sub replace_with_content {
    my $self = $_[0];
    Carp::croak "the target node has no parent"
        unless my ($parent) = $self->{'_parent'};

    my $parent_content = $parent->{'_content'};
    Carp::croak "the target node's parent has no content!?"
        unless $parent_content and @$parent_content;

    my $content_r = $self->{'_content'} || [];
    @$parent_content = map { ( ref($_) and $_ eq $self ) ? @$content_r : $_ }
        @$parent_content;

    $self->{'_parent'} = undef;    # detach $self from its parent

    # Update parentage link, removing from $self's content list
    for ( splice @$content_r ) { $_->{'_parent'} = $parent if ref $_ }

    return $self;                  # note: doesn't destroy it.
}

=head2 $h->delete_content()

Clears the content of C<$h>, calling C<< $h->delete >> for each content
element.  Compare with C<< $h->detach_content >>.

Returns C<$h>.

=cut

sub delete_content {
    for (
        splice @{
            delete( $_[0]->{'_content'} )

                # Deleting it here (while holding its value, for the moment)
                #  will keep calls to detach() from trying to uselessly filter
                #  the list (as they won't be able to see it once it's been
                #  deleted)
                || return ( $_[0] )    # in case of no content
        },
        0

        # the splice is so we can null the array too, just in case
        # something somewhere holds a ref to it
        )
    {
        $_->delete if ref $_;
    }
    $_[0];
}

=head2 $h->delete() destroy destroy_content

Detaches this element from its parent (if it has one) and explicitly
destroys the element and all its descendants.  The return value is
undef.

Perl uses garbage collection based on reference counting; when no
references to a data structure exist, it's implicitly destroyed --
i.e., when no value anywhere points to a given object anymore, Perl
knows it can free up the memory that the now-unused object occupies.

But this fails with HTML::Element trees, because a parent element
always holds references to its children, and its children elements
hold references to the parent, so no element ever looks like it's
I<not> in use.  So, to destroy those elements, you need to call
C<< $h->delete >> on the parent.

=cut

# two handy aliases
sub destroy         { shift->delete(@_) }
sub destroy_content { shift->delete_content(@_) }

sub delete {
    my $self = $_[0];
    $self->delete_content    # recurse down
        if $self->{'_content'} && @{ $self->{'_content'} };

    $self->detach if $self->{'_parent'} and $self->{'_parent'}{'_content'};

    # not the typical case

    %$self = ();             # null out the whole object on the way out
    return;
}

=head2 $h->clone()

Returns a copy of the element (whose children are clones (recursively)
of the original's children, if any).

The returned element is parentless.  Any '_pos' attributes present in the
source element/tree will be absent in the copy.  For that and other reasons,
the clone of an HTML::TreeBuilder object that's in mid-parse (i.e, the head
of a tree that HTML::TreeBuilder is elaborating) cannot (currently) be used
to continue the parse.

You are free to clone HTML::TreeBuilder trees, just as long as:
1) they're done being parsed, or 2) you don't expect to resume parsing
into the clone.  (You can continue parsing into the original; it is
never affected.)

=cut

sub clone {

    #print "Cloning $_[0]\n";
    my $it = shift;
    Carp::croak "clone() can be called only as an object method"
        unless ref $it;
    Carp::croak "clone() takes no arguments" if @_;

    my $new = bless {%$it}, ref($it);    # COPY!!! HOOBOY!
    delete @$new{ '_content', '_parent', '_pos', '_head', '_body' };

    # clone any contents
    if ( $it->{'_content'} and @{ $it->{'_content'} } ) {
        $new->{'_content'}
            = [ ref($it)->clone_list( @{ $it->{'_content'} } ) ];
        for ( @{ $new->{'_content'} } ) {
            $_->{'_parent'} = $new if ref $_;
        }
    }

    return $new;
}

=head2 HTML::Element->clone_list(...nodes...)

Returns a list consisting of a copy of each node given.
Text segments are simply copied; elements are cloned by
calling $it->clone on each of them.

Note that this must be called as a class method, not as an instance
method.  C<clone_list> will croak if called as an instance method.
You can also call it like so:

    ref($h)->clone_list(...nodes...)

=cut

sub clone_list {
    Carp::croak "clone_list can be called only as a class method"
        if ref shift @_;

    # all that does is get me here
    return map {
        ref($_)
            ? $_->clone    # copy by method
            : $_           # copy by evaluation
    } @_;
}

=head2 $h->normalize_content

Normalizes the content of C<$h> -- i.e., concatenates any adjacent
text nodes.  (Any undefined text segments are turned into empty-strings.)
Note that this does not recurse into C<$h>'s descendants.

=cut

sub normalize_content {
    my $start = $_[0];
    my $c;
    return
        unless $c = $start->{'_content'} and ref $c and @$c;   # nothing to do
        # TODO: if we start having text elements, deal with catenating those too?
    my @stretches = (undef);    # start with a barrier

    # I suppose this could be rewritten to treat stretches as it goes, instead
    #  of at the end.  But feh.

    # Scan:
    for ( my $i = 0; $i < @$c; ++$i ) {
        if ( defined $c->[$i] and ref $c->[$i] ) {    # not a text segment
            if ( $stretches[0] ) {

                # put in a barrier
                if ( $stretches[0][1] == 1 ) {

                    #print "Nixing stretch at ", $i-1, "\n";
                    undef $stretches[0]; # nix the previous one-node "stretch"
                }
                else {

                    #print "End of stretch at ", $i-1, "\n";
                    unshift @stretches, undef;
                }
            }

            # else no need for a barrier
        }
        else {                           # text segment
            $c->[$i] = '' unless defined $c->[$i];
            if ( $stretches[0] ) {
                ++$stretches[0][1];      # increase length
            }
            else {

                #print "New stretch at $i\n";
                unshift @stretches, [ $i, 1 ];    # start and length
            }
        }
    }

    # Now combine.  Note that @stretches is in reverse order, so the indexes
    # still make sense as we work our way thru (i.e., backwards thru $c).
    foreach my $s (@stretches) {
        if ( $s and $s->[1] > 1 ) {

            #print "Stretch at ", $s->[0], " for ", $s->[1], "\n";
            $c->[ $s->[0] ]
                .= join( '', splice( @$c, $s->[0] + 1, $s->[1] - 1 ) )

                # append the subsequent ones onto the first one.
        }
    }
    return;
}

=head2 $h->delete_ignorable_whitespace()

This traverses under C<$h> and deletes any text segments that are ignorable
whitespace.  You should not use this if C<$h> under a 'pre' element.

=cut

sub delete_ignorable_whitespace {

    # This doesn't delete all sorts of whitespace that won't actually
    #  be used in rendering, tho -- that's up to the rendering application.
    # For example:
    #   <input type='text' name='foo'>
    #     [some whitespace]
    #   <input type='text' name='bar'>
    # The WS between the two elements /will/ get used by the renderer.
    # But here:
    #   <input type='hidden' name='foo' value='1'>
    #     [some whitespace]
    #   <input type='text' name='bar' value='2'>
    # the WS between them won't be rendered in any way, presumably.

    #my $Debug = 4;
    die "delete_ignorable_whitespace can be called only as an object method"
        unless ref $_[0];

    print "About to tighten up...\n" if $Debug > 2;
    my (@to_do) = ( $_[0] );    # Start off.
    my ( $i, $sibs, $ptag, $this );    # scratch for the loop...
    while (@to_do) {
        if (   ( $ptag = ( $this = shift @to_do )->{'_tag'} ) eq 'pre'
            or $ptag eq 'textarea'
            or $HTML::Tagset::isCDATA_Parent{$ptag} )
        {

            # block the traversal under those
            print "Blocking traversal under $ptag\n" if $Debug;
            next;
        }
        next unless ( $sibs = $this->{'_content'} and @$sibs );
        for ( $i = $#$sibs; $i >= 0; --$i ) {   # work backwards thru the list
            if ( ref $sibs->[$i] ) {
                unshift @to_do, $sibs->[$i];

                # yes, this happens in pre order -- we're going backwards
                # thru this sibling list.  I doubt it actually matters, tho.
                next;
            }
            next if $sibs->[$i] =~ m<[^\n\r\f\t ]>s;   # it's /all/ whitespace

            print "Under $ptag whose canTighten ",
                "value is ", 0 + $HTML::Element::canTighten{$ptag}, ".\n"
                if $Debug > 3;

            # It's all whitespace...

            if ( $i == 0 ) {
                if ( @$sibs == 1 ) {                   # I'm an only child
                    next unless $HTML::Element::canTighten{$ptag};    # parent
                }
                else {    # I'm leftmost of many
                          # if either my parent or sib are eligible, I'm good.
                    next
                        unless $HTML::Element::canTighten{$ptag}    # parent
                            or (ref $sibs->[1]
                                and $HTML::Element::canTighten{ $sibs->[1]
                                        {'_tag'} }    # right sib
                            );
                }
            }
            elsif ( $i == $#$sibs ) {                 # I'm rightmost of many
                    # if either my parent or sib are eligible, I'm good.
                next
                    unless $HTML::Element::canTighten{$ptag}    # parent
                        or (ref $sibs->[ $i - 1 ]
                            and $HTML::Element::canTighten{ $sibs->[ $i - 1 ]
                                    {'_tag'} }                  # left sib
                        );
            }
            else {    # I'm the piggy in the middle
                      # My parent doesn't matter -- it all depends on my sibs
                next
                    unless ref $sibs->[ $i - 1 ]
                        or ref $sibs->[ $i + 1 ];

                # if NEITHER sib is a node, quit

                next if

                    # bailout condition: if BOTH are INeligible nodes
                    #  (as opposed to being text, or being eligible nodes)
                    ref $sibs->[ $i - 1 ]
                        and ref $sibs->[ $i + 1 ]
                        and !$HTML::Element::canTighten{ $sibs->[ $i - 1 ]
                                {'_tag'} }    # left sib
                        and !$HTML::Element::canTighten{ $sibs->[ $i + 1 ]
                                {'_tag'} }    # right sib
                ;
            }

       # Unknown tags aren't in canTighten and so AREN'T subject to tightening

            print "  delendum: child $i of $ptag\n" if $Debug > 3;
            splice @$sibs, $i, 1;
        }

        # end of the loop-over-children
    }

    # end of the while loop.

    return;
}

=head2 $h->insert_element($element, $implicit)

Inserts (via push_content) a new element under the element at
C<< $h->pos() >>.  Then updates C<< $h->pos() >> to point to the inserted
element, unless $element is a prototypically empty element like
"br", "hr", "img", etc.  The new C<< $h->pos() >> is returned.  This
method is useful only if your particular tree task involves setting
C<< $h->pos() >>.

=cut

sub insert_element {
    my ( $self, $tag, $implicit ) = @_;
    return $self->pos() unless $tag;    # noop if nothing to insert

    my $e;
    if ( ref $tag ) {
        $e   = $tag;
        $tag = $e->tag;
    }
    else {    # just a tag name -- so make the element
        $e = $self->element_class->new($tag);
        ++( $self->{'_element_count'} ) if exists $self->{'_element_count'};

        # undocumented.  see TreeBuilder.
    }

    $e->{'_implicit'} = 1 if $implicit;

    my $pos = $self->{'_pos'};
    $pos = $self unless defined $pos;

    $pos->push_content($e);

    $self->{'_pos'} = $pos = $e
        unless $self->_empty_element_map->{$tag} || $e->{'_empty_element'};

    $pos;
}

#==========================================================================
# Some things to override in XML::Element

sub _empty_element_map {
    \%HTML::Element::emptyElement;
}

sub _fold_case_LC {
    if (wantarray) {
        shift;
        map lc($_), @_;
    }
    else {
        return lc( $_[1] );
    }
}

sub _fold_case_NOT {
    if (wantarray) {
        shift;
        @_;
    }
    else {
        return $_[1];
    }
}

*_fold_case = \&_fold_case_LC;

#==========================================================================

=head1 DUMPING METHODS

=head2 $h->dump()

=head2 $h->dump(*FH)  ; # or *FH{IO} or $fh_obj

Prints the element and all its children to STDOUT (or to a specified
filehandle), in a format useful
only for debugging.  The structure of the document is shown by
indentation (no end tags).

=cut

sub dump {
    my ( $self, $fh, $depth ) = @_;
    $fh    = *STDOUT{IO} unless defined $fh;
    $depth = 0           unless defined $depth;
    print $fh "  " x $depth, $self->starttag, " \@", $self->address,
        $self->{'_implicit'} ? " (IMPLICIT)\n" : "\n";
    for ( @{ $self->{'_content'} } ) {
        if ( ref $_ ) {    # element
            $_->dump( $fh, $depth + 1 );    # recurse
        }
        else {                              # text node
            print $fh "  " x ( $depth + 1 );
            if ( length($_) > 65 or m<[\x00-\x1F]> ) {

                # it needs prettyin' up somehow or other
                my $x
                    = ( length($_) <= 65 )
                    ? $_
                    : ( substr( $_, 0, 65 ) . '...' );
                $x =~ s<([\x00-\x1F])>
                     <'\\x'.(unpack("H2",$1))>eg;
                print $fh qq{"$x"\n};
            }
            else {
                print $fh qq{"$_"\n};
            }
        }
    }
}

=head2 $h->as_HTML() or $h->as_HTML($entities)

=head2 or $h->as_HTML($entities, $indent_char)

=head2 or $h->as_HTML($entities, $indent_char, \%optional_end_tags)

Returns a string representing in HTML the element and its
descendants.  The optional argument C<$entities> specifies a string of
the entities to encode.  For compatibility with previous versions,
specify C<'E<lt>E<gt>&'> here.  If omitted or undef, I<all> unsafe
characters are encoded as HTML entities.  See L<HTML::Entities> for
details.  If passed an empty string, no entities are encoded.

If $indent_char is specified and defined, the HTML to be output is
intented, using the string you specify (which you probably should
set to "\t", or some number of spaces, if you specify it).

If C<\%optional_end_tags> is specified and defined, it should be
a reference to a hash that holds a true value for every tag name
whose end tag is optional.  Defaults to
C<\%HTML::Element::optionalEndTag>, which is an alias to 
C<%HTML::Tagset::optionalEndTag>, which, at time of writing, contains
true values for C<p, li, dt, dd>.  A useful value to pass is an empty
hashref, C<{}>, which means that no end-tags are optional for this dump.
Otherwise, possibly consider copying C<%HTML::Tagset::optionalEndTag> to a 
hash of your own, adding or deleting values as you like, and passing
a reference to that hash.

=cut

sub as_HTML {
    my ( $self, $entities, $indent, $omissible_map ) = @_;

    #my $indent_on = defined($indent) && length($indent);
    my @html = ();

    $omissible_map ||= \%HTML::Element::optionalEndTag;
    my $empty_element_map = $self->_empty_element_map;

    my $last_tag_tightenable    = 0;
    my $this_tag_tightenable    = 0;
    my $nonindentable_ancestors = 0;    # count of nonindentible tags over us.

    my ( $tag, $node, $start, $depth ); # per-iteration scratch

    if ( defined($indent) && length($indent) ) {
        $self->traverse(
            sub {
                ( $node, $start, $depth ) = @_;
                if ( ref $node ) {      # it's an element

                    # detect bogus classes. RT #35948, #61673
                    $node->can('starttag')
                        or Carp::confess( "Object of class "
                            . ref($node)
                            . " cannot be processed by HTML::Element" );

                    $tag = $node->{'_tag'};

                    if ($start) {       # on the way in
                        if ((   $this_tag_tightenable
                                = $HTML::Element::canTighten{$tag}
                            )
                            and !$nonindentable_ancestors
                            and $last_tag_tightenable
                            )
                        {
                            push
                                @html,
                                "\n",
                                $indent x $depth,
                                $node->starttag($entities),
                                ;
                        }
                        else {
                            push( @html, $node->starttag($entities) );
                        }
                        $last_tag_tightenable = $this_tag_tightenable;

                        ++$nonindentable_ancestors
                            if $tag eq 'pre'
                                or $HTML::Tagset::isCDATA_Parent{$tag};

                    }
                    elsif (
                        not(   $empty_element_map->{$tag}
                            or $omissible_map->{$tag} )
                        )
                    {

                        # on the way out
                        if (   $tag eq 'pre'
                            or $HTML::Tagset::isCDATA_Parent{$tag} )
                        {
                            --$nonindentable_ancestors;
                            $last_tag_tightenable
                                = $HTML::Element::canTighten{$tag};
                            push @html, $node->endtag;

                        }
                        else {    # general case
                            if ((   $this_tag_tightenable
                                    = $HTML::Element::canTighten{$tag}
                                )
                                and !$nonindentable_ancestors
                                and $last_tag_tightenable
                                )
                            {
                                push
                                    @html,
                                    "\n",
                                    $indent x $depth,
                                    $node->endtag,
                                    ;
                            }
                            else {
                                push @html, $node->endtag;
                            }
                            $last_tag_tightenable = $this_tag_tightenable;

                           #print "$tag tightenable: $this_tag_tightenable\n";
                        }
                    }
                }
                else {    # it's a text segment

                    $last_tag_tightenable = 0;    # I guess this is right
                    HTML::Entities::encode_entities( $node, $entities )

                        # That does magic things if $entities is undef.
                        unless (
                        ( defined($entities) && !length($entities) )

                        # If there's no entity to encode, don't call it
                        || $HTML::Tagset::isCDATA_Parent{ $_[3]{'_tag'} }

                        # To keep from amp-escaping children of script et al.
                        # That doesn't deal with descendants; but then, CDATA
                        #  parents shouldn't /have/ descendants other than a
                        #  text children (or comments?)
                        || $encoded_content
                        );
                    if ($nonindentable_ancestors) {
                        push @html, $node;    # say no go
                    }
                    else {
                        if ($last_tag_tightenable) {
                            $node =~ s<[\n\r\f\t ]+>< >s;

                            #$node =~ s< $><>s;
                            $node =~ s<^ ><>s;
                            push
                                @html,
                                "\n",
                                $indent x $depth,
                                $node,

           #Text::Wrap::wrap($indent x $depth, $indent x $depth, "\n" . $node)
                                ;
                        }
                        else {
                            push
                                @html,
                                $node,

                                #Text::Wrap::wrap('', $indent x $depth, $node)
                                ;
                        }
                    }
                }
                1;    # keep traversing
            }
        );            # End of parms to traverse()
    }
    else {            # no indenting -- much simpler code
        $self->traverse(
            sub {
                ( $node, $start ) = @_;
                if ( ref $node ) {

                    # detect bogus classes. RT #35948
                    $node->isa( $self->element_class )
                        or Carp::confess( "Object of class "
                            . ref($node)
                            . " cannot be processed by HTML::Element" );

                    $tag = $node->{'_tag'};
                    if ($start) {    # on the way in
                        push( @html, $node->starttag($entities) );
                    }
                    elsif (
                        not(   $empty_element_map->{$tag}
                            or $omissible_map->{$tag} )
                        )
                    {

                        # on the way out
                        push( @html, $node->endtag );
                    }
                }
                else {

                    # simple text content
                    HTML::Entities::encode_entities( $node, $entities )

                        # That does magic things if $entities is undef.
                        unless (
                        ( defined($entities) && !length($entities) )

                        # If there's no entity to encode, don't call it
                        || $HTML::Tagset::isCDATA_Parent{ $_[3]{'_tag'} }

                        # To keep from amp-escaping children of script et al.
                        # That doesn't deal with descendants; but then, CDATA
                        #  parents shouldn't /have/ descendants other than a
                        #  text children (or comments?)
                        || $encoded_content
                        );
                    push( @html, $node );
                }
                1;    # keep traversing
            }
        );            # End of parms to traverse()
    }

    if ( $self->{_store_declarations} && defined $self->{_decl} ) {
        unshift @html, sprintf "<!%s>\n", $self->{_decl}->{text};
    }

    return join( '', @html );
}

=head2 $h->as_text()

=head2 $h->as_text(skip_dels => 1, extra_chars => '\xA0')

Returns a string consisting of only the text parts of the element's
descendants.

Text under 'script' or 'style' elements is never included in what's
returned.  If C<skip_dels> is true, then text content under "del"
nodes is not included in what's returned.

=head2 $h->as_trimmed_text(...) as_text_trimmed

This is just like as_text(...) except that leading and trailing
whitespace is deleted, and any internal whitespace is collapsed.

This will not remove hard spaces, unicode spaces, or any other
non ASCII white space unless you supplye the extra characters as
a string argument. e.g. $h->as_trimmed_text(extra_chars => '\xA0')

=cut

sub as_text {

    # Yet another iteratively implemented traverser
    my ( $this, %options ) = @_;
    my $skip_dels = $options{'skip_dels'} || 0;
    my (@pile) = ($this);
    my $tag;
    my $text = '';
    while (@pile) {
        if ( !defined( $pile[0] ) ) {    # undef!
                                         # no-op
        }
        elsif ( !ref( $pile[0] ) ) {     # text bit!  save it!
            $text .= shift @pile;
        }
        else {                           # it's a ref -- traverse under it
            unshift @pile, @{ $this->{'_content'} || $nillio }
                unless ( $tag = ( $this = shift @pile )->{'_tag'} ) eq 'style'
                or $tag eq 'script'
                or ( $skip_dels and $tag eq 'del' );
        }
    }
    return $text;
}

# extra_chars added for RT #26436
sub as_trimmed_text {
    my ( $this, %options ) = @_;
    my $text = $this->as_text(%options);
    my $extra_chars = $options{'extra_chars'} || '';

    $text =~ s/[\n\r\f\t$extra_chars ]+$//s;
    $text =~ s/^[\n\r\f\t$extra_chars ]+//s;
    $text =~ s/[\n\r\f\t$extra_chars ]+/ /g;
    return $text;
}

sub as_text_trimmed { shift->as_trimmed_text(@_) }   # alias, because I forget

=head2 $h->as_XML()

Returns a string representing in XML the element and its descendants.

The XML is not indented.

=cut

# TODO: make it wrap, if not indent?

sub as_XML {

    # based an as_HTML
    my ($self) = @_;

    #my $indent_on = defined($indent) && length($indent);
    my @xml               = ();
    my $empty_element_map = $self->_empty_element_map;

    my ( $tag, $node, $start );    # per-iteration scratch
    $self->traverse(
        sub {
            ( $node, $start ) = @_;
            if ( ref $node ) {     # it's an element
                $tag = $node->{'_tag'};
                if ($start) {      # on the way in

                    foreach my $attr ( $node->all_attr_names() ) {
                        Carp::croak(
                            "$tag has an invalid attribute name '$attr'")
                            unless ( $attr eq '/' || $self->_valid_name($attr) );
                    }

                    if ( $empty_element_map->{$tag}
                        and !@{ $node->{'_content'} || $nillio } )
                    {
                        push( @xml, $node->starttag_XML( undef, 1 ) );
                    }
                    else {
                        push( @xml, $node->starttag_XML(undef) );
                    }
                }
                else {    # on the way out
                    unless ( $empty_element_map->{$tag}
                        and !@{ $node->{'_content'} || $nillio } )
                    {
                        push( @xml, $node->endtag_XML() );
                    }     # otherwise it will have been an <... /> tag.
                }
            }
            else {        # it's just text
                _xml_escape($node);
                push( @xml, $node );
            }
            1;            # keep traversing
        }
    );

    join( '', @xml, "\n" );
}

sub _xml_escape {

# DESTRUCTIVE (a.k.a. "in-place")
# Five required escapes: http://www.w3.org/TR/2006/REC-xml11-20060816/#syntax
# We allow & if it's part of a valid escape already: http://www.w3.org/TR/2006/REC-xml11-20060816/#sec-references
    foreach my $x (@_) {

        # In strings with no encoded entities all & should be encoded.
        if ($encoded_content) {
            $x
                =~ s/&(?!                 # An ampersand that isn't followed by...
                (\#\d+; |                 # A hash mark, digits and semicolon, or
                \#x[\da-f]+; |            # A hash mark, "x", hex digits and semicolon, or
                $START_CHAR$NAME_CHAR+; ) # A valid unicode entity name and semicolon
           )/&amp;/gx;    # Needs to be escaped to amp
        }
        else {
            $x =~ s/&/&amp;/g;
        }

        # simple character escapes
        $x =~ s/</&lt;/g;
        $x =~ s/>/&gt;/g;
        $x =~ s/"/&quot;/g;
        $x =~ s/'/&apos;/g;
    }
    return;
}

=head2 $h->as_Lisp_form()

Returns a string representing the element and its descendants as a
Lisp form.  Unsafe characters are encoded as octal escapes.

The Lisp form is indented, and contains external ("href", etc.)  as
well as internal attributes ("_tag", "_content", "_implicit", etc.),
except for "_parent", which is omitted.

Current example output for a given element:

  ("_tag" "img" "border" "0" "src" "pie.png" "usemap" "#main.map")

=cut

# NOTES:
#
# It's been suggested that attribute names be made :-keywords:
#   (:_tag "img" :border 0 :src "pie.png" :usemap "#main.map")
# However, it seems that Scheme has no such data type as :-keywords.
# So, for the moment at least, I tend toward simplicity, uniformity,
#  and universality, where everything a string or a list.

sub as_Lisp_form {
    my @out;

    my $sub;
    my $depth = 0;
    my ( @list, $val );
    $sub = sub {    # Recursor
        my $self = $_[0];
        @list = ( '_tag', $self->{'_tag'} );
        @list = () unless defined $list[-1];    # unlikely

        for ( sort keys %$self ) {              # predictable ordering
            next
                if $_ eq '_content'
                    or $_ eq '_tag'
                    or $_ eq '_parent'
                    or $_ eq '/';

            # Leave the other private attributes, I guess.
            push @list, $_, $val
                if defined( $val = $self->{$_} );    # and !ref $val;
        }

        for (@list) {

            # octal-escape it
            s<([^\x20\x21\x23\x27-\x5B\x5D-\x7E])>
         <sprintf('\\%03o',ord($1))>eg;
            $_ = qq{"$_"};
        }
        push @out, ( '  ' x $depth ) . '(' . join ' ', splice @list;
        if ( @{ $self->{'_content'} || $nillio } ) {
            $out[-1] .= " \"_content\" (\n";
            ++$depth;
            foreach my $c ( @{ $self->{'_content'} } ) {
                if ( ref($c) ) {

                    # an element -- recurse
                    $sub->($c);
                }
                else {

                    # a text segment -- stick it in and octal-escape it
                    push @out, $c;
                    $out[-1] =~ s<([^\x20\x21\x23\x27-\x5B\x5D-\x7E])>
             <sprintf('\\%03o',ord($1))>eg;

                    # And quote and indent it.
                    $out[-1] .= "\"\n";
                    $out[-1] = ( '  ' x $depth ) . '"' . $out[-1];
                }
            }
            --$depth;
            substr( $out[-1], -1 )
                = "))\n";    # end of _content and of the element
        }
        else {
            $out[-1] .= ")\n";
        }
        return;
    };

    $sub->( $_[0] );
    undef $sub;
    return join '', @out;
}

=head2 format

Formats text output. Defaults to HTML::FormatText.

Takes a second argument that is a reference to a formatter.

=cut

sub format {
    my ( $self, $formatter ) = @_;
    unless ( defined $formatter ) {
        require HTML::FormatText;
        $formatter = HTML::FormatText->new();
    }
    $formatter->format($self);
}

=head2 $h->starttag() or $h->starttag($entities)

Returns a string representing the complete start tag for the element.
I.e., leading "<", tag name, attributes, and trailing ">".
All values are surrounded with
double-quotes, and appropriate characters are encoded.  If C<$entities>
is omitted or undef, I<all> unsafe characters are encoded as HTML
entities.  See L<HTML::Entities> for details.  If you specify some
value for C<$entities>, remember to include the double-quote character in
it.  (Previous versions of this module would basically behave as if
C<'&"E<gt>'> were specified for C<$entities>.)  If C<$entities> is
an empty string, no entity is escaped.

=cut

sub starttag {
    my ( $self, $entities ) = @_;

    my $name = $self->{'_tag'};

    return $self->{'text'}              if $name eq '~literal';
    return "<!" . $self->{'text'} . ">" if $name eq '~declaration';
    return "<?" . $self->{'text'} . ">" if $name eq '~pi';

    if ( $name eq '~comment' ) {
        if ( ref( $self->{'text'} || '' ) eq 'ARRAY' ) {

            # Does this ever get used?  And is this right?
            return
                "<!"
                . join( ' ', map( "--$_--", @{ $self->{'text'} } ) ) . ">";
        }
        else {
            return "<!--" . $self->{'text'} . "-->";
        }
    }

    my $tag = $html_uc ? "<\U$name" : "<\L$name";
    my $val;
    for ( sort keys %$self ) {    # predictable ordering
        next if !length $_ or m/^_/s or $_ eq '/';
        $val = $self->{$_};
        next if !defined $val;    # or ref $val;
        if ($_ eq $val &&         # if attribute is boolean, for this element
            exists( $HTML::Element::boolean_attr{$name} )
            && (ref( $HTML::Element::boolean_attr{$name} )
                ? $HTML::Element::boolean_attr{$name}{$_}
                : $HTML::Element::boolean_attr{$name} eq $_
            )
            )
        {
            $tag .= $html_uc ? " \U$_" : " \L$_";
        }
        else {                    # non-boolean attribute

            if ( ref $val eq 'HTML::Element'
                and $val->{_tag} eq '~literal' )
            {
                $val = $val->{text};
            }
            else {
                HTML::Entities::encode_entities( $val, $entities )
                    unless (
                    defined($entities) && !length($entities)
                    || $encoded_content

                    );
            }

            $val = qq{"$val"};
            $tag .= $html_uc ? qq{ \U$_\E=$val} : qq{ \L$_\E=$val};
        }
    }    # for keys
    if ( scalar $self->content_list == 0
        && $self->_empty_element_map->{ $self->tag } )
    {
        return $tag . " />";
    }
    else {
        return $tag . ">";
    }
}

=head2 starttag_XML

Returns a string representing the complete start tag for the element.

=cut

sub starttag_XML {
    my ($self) = @_;

    # and a third parameter to signal emptiness?

    my $name = $self->{'_tag'};

    return $self->{'text'}               if $name eq '~literal';
    return '<!' . $self->{'text'} . '>'  if $name eq '~declaration';
    return "<?" . $self->{'text'} . "?>" if $name eq '~pi';

    if ( $name eq '~comment' ) {
        if ( ref( $self->{'text'} || '' ) eq 'ARRAY' ) {

            # Does this ever get used?  And is this right?
            $name = join( ' ', @{ $self->{'text'} } );
        }
        else {
            $name = $self->{'text'};
        }
        $name =~ s/--/-&#45;/g;    # can't have double --'s in XML comments
        return "<!-- $name -->";
    }

    my $tag = "<$name";
    my $val;
    for ( sort keys %$self ) {     # predictable ordering
        next if !length $_ or m/^_/s or $_ eq '/';

        # Hm -- what to do if val is undef?
        # I suppose that shouldn't ever happen.
        next if !defined( $val = $self->{$_} );    # or ref $val;
        _xml_escape($val);
        $tag .= qq{ $_="$val"};
    }
    @_ == 3 ? "$tag />" : "$tag>";
}

=head2 $h->endtag() || endtag_XML

Returns a string representing the complete end tag for this element.
I.e., "</", tag name, and ">".

=cut

sub endtag {
    $html_uc ? "</\U$_[0]->{'_tag'}>" : "</\L$_[0]->{'_tag'}>";
}

# TODO: document?
sub endtag_XML {
    "</$_[0]->{'_tag'}>";
}

#==========================================================================
# This, ladies and germs, is an iterative implementation of a
# recursive algorithm.  DON'T TRY THIS AT HOME.
# Basically, the algorithm says:
#
# To traverse:
#   1: pre-order visit this node
#   2: traverse any children of this node
#   3: post-order visit this node, unless it's a text segment,
#       or a prototypically empty node (like "br", etc.)
# Add to that the consideration of the callbacks' return values,
# so you can block visitation of the children, or siblings, or
# abort the whole excursion, etc.
#
# So, why all this hassle with making the code iterative?
# It makes for real speed, because it eliminates the whole
# hassle of Perl having to allocate scratch space for each
# instance of the recursive sub.  Since the algorithm
# is basically simple (and not all recursive ones are!) and
# has few necessary lexicals (basically just the current node's
# content list, and the current position in it), it was relatively
# straightforward to store that information not as the frame
# of a sub, but as a stack, i.e., a simple Perl array (well, two
# of them, actually: one for content-listrefs, one for indexes of
# current position in each of those).

my $NIL = [];

sub traverse {
    my ( $start, $callback, $ignore_text ) = @_;

    Carp::croak "traverse can be called only as an object method"
        unless ref $start;

    Carp::croak('must provide a callback for traverse()!')
        unless defined $callback and ref $callback;

    # Elementary type-checking:
    my ( $c_pre, $c_post );
    if ( UNIVERSAL::isa( $callback, 'CODE' ) ) {
        $c_pre = $c_post = $callback;
    }
    elsif ( UNIVERSAL::isa( $callback, 'ARRAY' ) ) {
        ( $c_pre, $c_post ) = @$callback;
        Carp::croak(
            "pre-order callback \"$c_pre\" is true but not a coderef!")
            if $c_pre and not UNIVERSAL::isa( $c_pre, 'CODE' );
        Carp::croak(
            "pre-order callback \"$c_post\" is true but not a coderef!")
            if $c_post and not UNIVERSAL::isa( $c_post, 'CODE' );
        return $start unless $c_pre or $c_post;

        # otherwise there'd be nothing to actually do!
    }
    else {
        Carp::croak("$callback is not a known kind of reference")
            unless ref($callback);
    }

    my $empty_element_map = $start->_empty_element_map;

    my (@C) = [$start];    # a stack containing lists of children
    my (@I) = (-1);        # initial value must be -1 for each list
         # a stack of indexes to current position in corresponding lists in @C
         # In each of these, 0 is the active point

    # scratch:
    my ($rv,           # return value of callback
        $this,         # current node
        $content_r,    # child list of $this
    );

    # THE BIG LOOP
    while (@C) {

        # Move to next item in this frame
        if ( !defined( $I[0] ) or ++$I[0] >= @{ $C[0] } ) {

            # We either went off the end of this list, or aborted the list
            # So call the post-order callback:
            if (    $c_post
                and defined $I[0]
                and @C > 1

                # to keep the next line from autovivifying
                and defined( $this = $C[1][ $I[1] ] )    # sanity, and
                     # suppress callbacks on exiting the fictional top frame
                and ref($this)    # sanity
                and not(
                    $this->{'_empty_element'}
                    || ( $empty_element_map->{ $this->{'_tag'} || '' }
                        && !@{ $this->{'_content'} } )    # RT #49932
                )    # things that don't get post-order callbacks
                )
            {
                shift @I;
                shift @C;

                #print "Post! at depth", scalar(@I), "\n";
                $rv = $c_post->(

                    #map $_, # copy to avoid any messiness
                    $this,     # 0: this
                    0,         # 1: startflag (0 for post-order call)
                    @I - 1,    # 2: depth
                );

                if ( defined($rv) and ref($rv) eq $travsignal_package ) {
                    $rv = $$rv;    #deref
                    if ( $rv eq 'ABORT' ) {
                        last;      # end of this excursion!
                    }
                    elsif ( $rv eq 'PRUNE' ) {

                        # NOOP on post!!
                    }
                    elsif ( $rv eq 'PRUNE_SOFTLY' ) {

                        # NOOP on post!!
                    }
                    elsif ( $rv eq 'OK' ) {

                        # noop
                    }
                    elsif ( $rv eq 'PRUNE_UP' ) {
                        $I[0] = undef;
                    }
                    else {
                        die "Unknown travsignal $rv\n";

                        # should never happen
                    }
                }
            }
            else {
                shift @I;
                shift @C;
            }
            next;
        }

        $this = $C[0][ $I[0] ];

        if ($c_pre) {
            if ( defined $this and ref $this ) {    # element
                $rv = $c_pre->(

                    #map $_, # copy to avoid any messiness
                    $this,     # 0: this
                    1,         # 1: startflag (1 for pre-order call)
                    @I - 1,    # 2: depth
                );
            }
            else {             # text segment
                next if $ignore_text;
                $rv = $c_pre->(

                    #map $_, # copy to avoid any messiness
                    $this,           # 0: this
                    1,               # 1: startflag (1 for pre-order call)
                    @I - 1,          # 2: depth
                    $C[1][ $I[1] ],  # 3: parent
                                     # And there will always be a $C[1], since
                             #  we can't start traversing at a text node
                    $I[0]    # 4: index of self in parent's content list
                );
            }
            if ( not $rv ) {    # returned false.  Same as PRUNE.
                next;           # prune
            }
            elsif ( ref($rv) eq $travsignal_package ) {
                $rv = $$rv;     # deref
                if ( $rv eq 'ABORT' ) {
                    last;       # end of this excursion!
                }
                elsif ( $rv eq 'PRUNE' ) {
                    next;
                }
                elsif ( $rv eq 'PRUNE_SOFTLY' ) {
                    if (ref($this)
                        and not( $this->{'_empty_element'}
                            || $empty_element_map->{ $this->{'_tag'} || '' } )
                        )
                    {

             # push a dummy empty content list just to trigger a post callback
                        unshift @I, -1;
                        unshift @C, $NIL;
                    }
                    next;
                }
                elsif ( $rv eq 'OK' ) {

                    # noop
                }
                elsif ( $rv eq 'PRUNE_UP' ) {
                    $I[0] = undef;
                    next;

                    # equivalent of last'ing out of the current child list.

            # Used to have PRUNE_UP_SOFTLY and ABORT_SOFTLY here, but the code
            # for these was seriously upsetting, served no particularly clear
            # purpose, and could not, I think, be easily implemented with a
            # recursive routine.  All bad things!
                }
                else {
                    die "Unknown travsignal $rv\n";

                    # should never happen
                }
            }

            # else fall thru to meaning same as \'OK'.
        }

        # end of pre-order calling

        # Now queue up content list for the current element...
        if (ref $this
            and not(    # ...except for those which...
                not( $content_r = $this->{'_content'} and @$content_r )

                # ...have empty content lists...
                and $this->{'_empty_element'}
                || $empty_element_map->{ $this->{'_tag'} || '' }

                # ...and that don't get post-order callbacks
            )
            )
        {
            unshift @I, -1;
            unshift @C, $content_r || $NIL;

            #print $this->{'_tag'}, " ($this) adds content_r ", $C[0], "\n";
        }
    }
    return $start;
}

=head1 SECONDARY STRUCTURAL METHODS

These methods all involve some structural aspect of the tree;
either they report some aspect of the tree's structure, or they involve
traversal down the tree, or walking up the tree.

=head2 $h->is_inside('tag', ...) or $h->is_inside($element, ...)

Returns true if the $h element is, or is contained anywhere inside an
element that is any of the ones listed, or whose tag name is any of
the tag names listed.

=cut

sub is_inside {
    my $self = shift;
    return unless @_;    # if no items specified, I guess this is right.

    my $current = $self;

    # the loop starts by looking at the given element
    while ( defined $current and ref $current ) {
        for (@_) {
            if (ref) {    # element
                return 1 if $_ eq $current;
            }
            else {        # tag name
                return 1 if $_ eq $current->{'_tag'};
            }
        }
        $current = $current->{'_parent'};
    }
    0;
}

=head2 $h->is_empty()

Returns true if $h has no content, i.e., has no elements or text
segments under it.  In other words, this returns true if $h is a leaf
node, AKA a terminal node.  Do not confuse this sense of "empty" with
another sense that it can have in SGML/HTML/XML terminology, which
means that the element in question is of the type (like HTML's "hr",
"br", "img", etc.) that I<can't> have any content.

That is, a particular "p" element may happen to have no content, so
$that_p_element->is_empty will be true -- even though the prototypical
"p" element isn't "empty" (not in the way that the prototypical "hr"
element is).

If you think this might make for potentially confusing code, consider
simply using the clearer exact equivalent:  not($h->content_list)

=cut

sub is_empty {
    my $self = shift;
    !$self->{'_content'} || !@{ $self->{'_content'} };
}

=head2 $h->pindex()

Return the index of the element in its parent's contents array, such
that $h would equal

  $h->parent->content->[$h->pindex]
  or
  ($h->parent->content_list)[$h->pindex]

assuming $h isn't root.  If the element $h is root, then
$h->pindex returns undef.

=cut

sub pindex {
    my $self = shift;

    my $parent = $self->{'_parent'}    || return;
    my $pc     = $parent->{'_content'} || return;
    for ( my $i = 0; $i < @$pc; ++$i ) {
        return $i if ref $pc->[$i] and $pc->[$i] eq $self;
    }
    return;    # we shouldn't ever get here
}

#--------------------------------------------------------------------------

=head2 $h->left()

In scalar context: returns the node that's the immediate left sibling
of $h.  If $h is the leftmost (or only) child of its parent (or has no
parent), then this returns undef.

In list context: returns all the nodes that're the left siblings of $h
(starting with the leftmost).  If $h is the leftmost (or only) child
of its parent (or has no parent), then this returns empty-list.

(See also $h->preinsert(LIST).)

=cut

sub left {
    Carp::croak "left() is supposed to be an object method"
        unless ref $_[0];
    my $pc = ( $_[0]->{'_parent'} || return )->{'_content'}
        || die "parent is childless?";

    die "parent is childless" unless @$pc;
    return if @$pc == 1;    # I'm an only child

    if (wantarray) {
        my @out;
        foreach my $j (@$pc) {
            return @out if ref $j and $j eq $_[0];
            push @out, $j;
        }
    }
    else {
        for ( my $i = 0; $i < @$pc; ++$i ) {
            return $i ? $pc->[ $i - 1 ] : undef
                if ref $pc->[$i] and $pc->[$i] eq $_[0];
        }
    }

    die "I'm not in my parent's content list?";
    return;
}

=head2 $h->right()

In scalar context: returns the node that's the immediate right sibling
of $h.  If $h is the rightmost (or only) child of its parent (or has
no parent), then this returns undef.

In list context: returns all the nodes that're the right siblings of
$h, starting with the leftmost.  If $h is the rightmost (or only) child
of its parent (or has no parent), then this returns empty-list.

(See also $h->postinsert(LIST).)

=cut

sub right {
    Carp::croak "right() is supposed to be an object method"
        unless ref $_[0];
    my $pc = ( $_[0]->{'_parent'} || return )->{'_content'}
        || die "parent is childless?";

    die "parent is childless" unless @$pc;
    return if @$pc == 1;    # I'm an only child

    if (wantarray) {
        my ( @out, $seen );
        foreach my $j (@$pc) {
            if ($seen) {
                push @out, $j;
            }
            else {
                $seen = 1 if ref $j and $j eq $_[0];
            }
        }
        die "I'm not in my parent's content list?" unless $seen;
        return @out;
    }
    else {
        for ( my $i = 0; $i < @$pc; ++$i ) {
            return +( $i == $#$pc ) ? undef : $pc->[ $i + 1 ]
                if ref $pc->[$i] and $pc->[$i] eq $_[0];
        }
        die "I'm not in my parent's content list?";
        return;
    }
}

#--------------------------------------------------------------------------

=head2 $h->address()

Returns a string representing the location of this node in the tree.
The address consists of numbers joined by a '.', starting with '0',
and followed by the pindexes of the nodes in the tree that are
ancestors of $h, starting from the top.

So if the way to get to a node starting at the root is to go to child
2 of the root, then child 10 of that, and then child 0 of that, and
then you're there -- then that node's address is "0.2.10.0".

As a bit of a special case, the address of the root is simply "0".

I forsee this being used mainly for debugging, but you may
find your own uses for it.

=head2 $h->address($address)

This returns the node (whether element or text-segment) at
the given address in the tree that $h is a part of.  (That is,
the address is resolved starting from $h->root.)

If there is no node at the given address, this returns undef.

You can specify "relative addressing" (i.e., that indexing is supposed
to start from $h and not from $h->root) by having the address start
with a period -- e.g., $h->address(".3.2") will look at child 3 of $h,
and child 2 of that.

=cut

sub address {
    if ( @_ == 1 ) {    # report-address form
        return join(
            '.',
            reverse(    # so it starts at the top
                map( $_->pindex() || '0',    # so that root's undef -> '0'
                    $_[0],                   # self and...
                    $_[0]->lineage )
            )
        );
    }
    else {                                   # get-node-at-address
        my @stack = split( /\./, $_[1] );
        my $here;

        if ( @stack and !length $stack[0] ) {    # relative addressing
            $here = $_[0];
            shift @stack;
        }
        else {                                   # absolute addressing
            return unless 0 == shift @stack;   # to pop the initial 0-for-root
            $here = $_[0]->root;
        }

        while (@stack) {
            return
                unless $here->{'_content'}
                    and @{ $here->{'_content'} } > $stack[0];

            # make sure the index isn't too high
            $here = $here->{'_content'}[ shift @stack ];
            return if @stack and not ref $here;

            # we hit a text node when we expected a non-terminal element node
        }

        return $here;
    }
}

=head2 $h->depth()

Returns a number expressing C<$h>'s depth within its tree, i.e., how many
steps away it is from the root.  If C<$h> has no parent (i.e., is root),
its depth is 0.

=cut

sub depth {
    my $here  = $_[0];
    my $depth = 0;
    while ( defined( $here = $here->{'_parent'} ) and ref($here) ) {
        ++$depth;
    }
    return $depth;
}

=head2 $h->root()

Returns the element that's the top of C<$h>'s tree.  If C<$h> is
root, this just returns C<$h>.  (If you want to test whether C<$h>
I<is> the root, instead of asking what its root is, just test
C<< not($h->parent) >>.)

=cut

sub root {
    my $here = my $root = shift;
    while ( defined( $here = $here->{'_parent'} ) and ref($here) ) {
        $root = $here;
    }
    return $root;
}

=head2 $h->lineage()

Returns the list of C<$h>'s ancestors, starting with its parent,
and then that parent's parent, and so on, up to the root.  If C<$h>
is root, this returns an empty list.

If you simply want a count of the number of elements in C<$h>'s lineage,
use $h->depth.

=cut

sub lineage {
    my $here = shift;
    my @lineage;
    while ( defined( $here = $here->{'_parent'} ) and ref($here) ) {
        push @lineage, $here;
    }
    return @lineage;
}

=head2 $h->lineage_tag_names()

Returns the list of the tag names of $h's ancestors, starting
with its parent, and that parent's parent, and so on, up to the
root.  If $h is root, this returns an empty list.
Example output: C<('em', 'td', 'tr', 'table', 'body', 'html')>

=cut

sub lineage_tag_names {
    my $here = my $start = shift;
    my @lineage_names;
    while ( defined( $here = $here->{'_parent'} ) and ref($here) ) {
        push @lineage_names, $here->{'_tag'};
    }
    return @lineage_names;
}

=head2 $h->descendants()

In list context, returns the list of all $h's descendant elements,
listed in pre-order (i.e., an element appears before its
content-elements).  Text segments DO NOT appear in the list.
In scalar context, returns a count of all such elements.

=head2 $h->descendents()

This is just an alias to the C<descendants> method.

=cut

sub descendents { shift->descendants(@_) }

sub descendants {
    my $start = shift;
    if (wantarray) {
        my @descendants;
        $start->traverse(
            [    # pre-order sub only
                sub {
                    push( @descendants, $_[0] );
                    return 1;
                },
                undef    # no post
            ],
            1,           # ignore text
        );
        shift @descendants;    # so $self doesn't appear in the list
        return @descendants;
    }
    else {                     # just returns a scalar
        my $descendants = -1;    # to offset $self being counted
        $start->traverse(
            [                    # pre-order sub only
                sub {
                    ++$descendants;
                    return 1;
                },
                undef            # no post
            ],
            1,                   # ignore text
        );
        return $descendants;
    }
}

=head2 $h->find_by_tag_name('tag', ...)

In list context, returns a list of elements at or under $h that have
any of the specified tag names.  In scalar context, returns the first
(in pre-order traversal of the tree) such element found, or undef if
none.

=head2 $h->find('tag', ...)

This is just an alias to C<find_by_tag_name>.  (There was once
going to be a whole find_* family of methods, but then look_down
filled that niche, so there turned out not to be much reason for the
verboseness of the name "find_by_tag_name".)

=cut

sub find { shift->find_by_tag_name(@_) }

# yup, a handy alias

sub find_by_tag_name {
    my (@pile) = shift(@_);    # start out the to-do stack for the traverser
    Carp::croak "find_by_tag_name can be called only as an object method"
        unless ref $pile[0];
    return () unless @_;
    my (@tags) = $pile[0]->_fold_case(@_);
    my ( @matching, $this, $this_tag );
    while (@pile) {
        $this_tag = ( $this = shift @pile )->{'_tag'};
        foreach my $t (@tags) {
            if ( $t eq $this_tag ) {
                if (wantarray) {
                    push @matching, $this;
                    last;
                }
                else {
                    return $this;
                }
            }
        }
        unshift @pile, grep ref($_), @{ $this->{'_content'} || next };
    }
    return @matching if wantarray;
    return;
}

=head2 $h->find_by_attribute('attribute', 'value')

In a list context, returns a list of elements at or under $h that have
the specified attribute, and have the given value for that attribute.
In a scalar context, returns the first (in pre-order traversal of the
tree) such element found, or undef if none.

This method is B<deprecated> in favor of the more expressive
C<look_down> method, which new code should use instead.

=cut

sub find_by_attribute {

    # We could limit this to non-internal attributes, but hey.
    my ( $self, $attribute, $value ) = @_;
    Carp::croak "Attribute must be a defined value!"
        unless defined $attribute;
    $attribute = $self->_fold_case($attribute);

    my @matching;
    my $wantarray = wantarray;
    my $quit;
    $self->traverse(
        [    # pre-order only
            sub {
                if ( exists $_[0]{$attribute}
                    and $_[0]{$attribute} eq $value )
                {
                    push @matching, $_[0];
                    return HTML::Element::ABORT
                        unless $wantarray;    # only take the first
                }
                1;                            # keep traversing
            },
            undef                             # no post
        ],
        1,                                    # yes, ignore text nodes.
    );

    if ($wantarray) {
        return @matching;
    }
    else {
        return unless @matching;
        return $matching[0];
    }
}

#--------------------------------------------------------------------------

=head2 $h->look_down( ...criteria... )

This starts at $h and looks thru its element descendants (in
pre-order), looking for elements matching the criteria you specify.
In list context, returns all elements that match all the given
criteria; in scalar context, returns the first such element (or undef,
if nothing matched).

There are three kinds of criteria you can specify:

=over

=item (attr_name, attr_value)

This means you're looking for an element with that value for that
attribute.  Example: C<"alt", "pix!">.  Consider that you can search
on internal attribute values too: C<"_tag", "p">.

=item (attr_name, qr/.../)

This means you're looking for an element whose value for that
attribute matches the specified Regexp object.

=item a coderef

This means you're looking for elements where coderef->(each_element)
returns true.  Example:

  my @wide_pix_images
    = $h->look_down(
                    "_tag", "img",
                    "alt", "pix!",
                    sub { $_[0]->attr('width') > 350 }
                   );

=back

Note that C<(attr_name, attr_value)> and C<(attr_name, qr/.../)>
criteria are almost always faster than coderef
criteria, so should presumably be put before them in your list of
criteria.  That is, in the example above, the sub ref is called only
for elements that have already passed the criteria of having a "_tag"
attribute with value "img", and an "alt" attribute with value "pix!".
If the coderef were first, it would be called on every element, and
I<then> what elements pass that criterion (i.e., elements for which
the coderef returned true) would be checked for their "_tag" and "alt"
attributes.

Note that comparison of string attribute-values against the string
value in C<(attr_name, attr_value)> is case-INsensitive!  A criterion
of C<('align', 'right')> I<will> match an element whose "align" value
is "RIGHT", or "right" or "rIGhT", etc.

Note also that C<look_down> considers "" (empty-string) and undef to
be different things, in attribute values.  So this:

  $h->look_down("alt", "")

will find elements I<with> an "alt" attribute, but where the value for
the "alt" attribute is "".  But this:

  $h->look_down("alt", undef)

is the same as:

  $h->look_down(sub { !defined($_[0]->attr('alt')) } )

That is, it finds elements that do not have an "alt" attribute at all
(or that do have an "alt" attribute, but with a value of undef --
which is not normally possible).

Note that when you give several criteria, this is taken to mean you're
looking for elements that match I<all> your criterion, not just I<any>
of them.  In other words, there is an implicit "and", not an "or".  So
if you wanted to express that you wanted to find elements with a
"name" attribute with the value "foo" I<or> with an "id" attribute
with the value "baz", you'd have to do it like:

  @them = $h->look_down(
    sub {
      # the lcs are to fold case
      lc($_[0]->attr('name')) eq 'foo'
      or lc($_[0]->attr('id')) eq 'baz'
    }
  );

Coderef criteria are more expressive than C<(attr_name, attr_value)>
and C<(attr_name, qr/.../)>
criteria, and all C<(attr_name, attr_value)>
and C<(attr_name, qr/.../)>
criteria could be
expressed in terms of coderefs.  However, C<(attr_name, attr_value)>
and C<(attr_name, qr/.../)>
criteria are a convenient shorthand.  (In fact, C<look_down> itself is
basically "shorthand" too, since anything you can do with C<look_down>
you could do by traversing the tree, either with the C<traverse>
method or with a routine of your own.  However, C<look_down> often
makes for very concise and clear code.)

=cut

sub look_down {
    ref( $_[0] ) or Carp::croak "look_down works only as an object method";

    my @criteria;
    for ( my $i = 1; $i < @_; ) {
        Carp::croak "Can't use undef as an attribute name"
            unless defined $_[$i];
        if ( ref $_[$i] ) {
            Carp::croak "A " . ref( $_[$i] ) . " value is not a criterion"
                unless ref $_[$i] eq 'CODE';
            push @criteria, $_[ $i++ ];
        }
        else {
            Carp::croak "param list to look_down ends in a key!" if $i == $#_;
            push @criteria, [
                scalar( $_[0]->_fold_case( $_[$i] ) ),
                defined( $_[ $i + 1 ] )
                ? ( ( ref $_[ $i + 1 ] ? $_[ $i + 1 ] : lc( $_[ $i + 1 ] ) ),
                    ref( $_[ $i + 1 ] )
                    )

                    # yes, leave that LC!
                : undef
            ];
            $i += 2;
        }
    }
    Carp::croak "No criteria?" unless @criteria;

    my (@pile) = ( $_[0] );
    my ( @matching, $val, $this );
Node:
    while ( defined( $this = shift @pile ) ) {

        # Yet another traverser implemented with merely iterative code.
        foreach my $c (@criteria) {
            if ( ref($c) eq 'CODE' ) {
                next Node unless $c->($this);    # jump to the continue block
            }
            else {                               # it's an attr-value pair
                next Node                        # jump to the continue block
                    if                           # two values are unequal if:
                        ( defined( $val = $this->{ $c->[0] } ) )
                    ? (     !defined $c->[ 1
                                ]    # actual is def, critval is undef => fail
                                     # allow regex matching
                                     # allow regex matching
                                or (
                                  $c->[2] eq 'Regexp'
                                ? $val !~ $c->[1]
                                : ( ref $val ne $c->[2]

                                        # have unequal ref values => fail
                                        or lc($val) ne lc( $c->[1] )

                                       # have unequal lc string values => fail
                                )
                                )
                        )
                    : (     defined $c->[1]
                        )    # actual is undef, critval is def => fail
            }
        }

        # We make it this far only if all the criteria passed.
        return $this unless wantarray;
        push @matching, $this;
    }
    continue {
        unshift @pile, grep ref($_), @{ $this->{'_content'} || $nillio };
    }
    return @matching if wantarray;
    return;
}

=head2 $h->look_up( ...criteria... )

This is identical to $h->look_down, except that whereas $h->look_down
basically scans over the list:

   ($h, $h->descendants)

$h->look_up instead scans over the list

   ($h, $h->lineage)

So, for example, this returns all ancestors of $h (possibly including
$h itself) that are "td" elements with an "align" attribute with a
value of "right" (or "RIGHT", etc.):

   $h->look_up("_tag", "td", "align", "right");

=cut

sub look_up {
    ref( $_[0] ) or Carp::croak "look_up works only as an object method";

    my @criteria;
    for ( my $i = 1; $i < @_; ) {
        Carp::croak "Can't use undef as an attribute name"
            unless defined $_[$i];
        if ( ref $_[$i] ) {
            Carp::croak "A " . ref( $_[$i] ) . " value is not a criterion"
                unless ref $_[$i] eq 'CODE';
            push @criteria, $_[ $i++ ];
        }
        else {
            Carp::croak "param list to look_up ends in a key!" if $i == $#_;
            push @criteria, [
                scalar( $_[0]->_fold_case( $_[$i] ) ),
                defined( $_[ $i + 1 ] )
                ? ( ( ref $_[ $i + 1 ] ? $_[ $i + 1 ] : lc( $_[ $i + 1 ] ) ),
                    ref( $_[ $i + 1 ] )
                    )
                : undef    # Yes, leave that LC!
            ];
            $i += 2;
        }
    }
    Carp::croak "No criteria?" unless @criteria;

    my ( @matching, $val );
    my $this = $_[0];
Node:
    while (1) {

       # You'll notice that the code here is almost the same as for look_down.
        foreach my $c (@criteria) {
            if ( ref($c) eq 'CODE' ) {
                next Node unless $c->($this);    # jump to the continue block
            }
            else {                               # it's an attr-value pair
                next Node                        # jump to the continue block
                    if                           # two values are unequal if:
                        ( defined( $val = $this->{ $c->[0] } ) )
                    ? (     !defined $c->[ 1
                                ]    # actual is def, critval is undef => fail
                                or (
                                  $c->[2] eq 'Regexp'
                                ? $val !~ $c->[1]
                                : ( ref $val ne $c->[2]

                                        # have unequal ref values => fail
                                        or lc($val) ne $c->[1]

                                       # have unequal lc string values => fail
                                )
                                )
                        )
                    : (     defined $c->[1]
                        )    # actual is undef, critval is def => fail
            }
        }

        # We make it this far only if all the criteria passed.
        return $this unless wantarray;
        push @matching, $this;
    }
    continue {
        last unless defined( $this = $this->{'_parent'} ) and ref $this;
    }

    return @matching if wantarray;
    return;
}

#--------------------------------------------------------------------------

=head2 $h->traverse(...options...)

Lengthy discussion of HTML::Element's unnecessary and confusing
C<traverse> method has been moved to a separate file:
L<HTML::Element::traverse>

=head2 $h->attr_get_i('attribute')

In list context, returns a list consisting of the values of the given
attribute for $self and for all its ancestors starting from $self and
working its way up.  Nodes with no such attribute are skipped.
("attr_get_i" stands for "attribute get, with inheritance".)
In scalar context, returns the first such value, or undef if none.

Consider a document consisting of:

   <html lang='i-klingon'>
     <head><title>Pati Pata</title></head>
     <body>
       <h1 lang='la'>Stuff</h1>
       <p lang='es-MX' align='center'>
         Foo bar baz <cite>Quux</cite>.
       </p>
       <p>Hooboy.</p>
     </body>
   </html>

If $h is the "cite" element, $h->attr_get_i("lang") in list context
will return the list ('es-MX', 'i-klingon').  In scalar context, it
will return the value 'es-MX'.

If you call with multiple attribute names...

=head2 $h->attr_get_i('a1', 'a2', 'a3')

...in list context, this will return a list consisting of
the values of these attributes which exist in $self and its ancestors.
In scalar context, this returns the first value (i.e., the value of
the first existing attribute from the first element that has
any of the attributes listed).  So, in the above example,

  $h->attr_get_i('lang', 'align');

will return:

   ('es-MX', 'center', 'i-klingon') # in list context
  or
   'es-MX' # in scalar context.

But note that this:

 $h->attr_get_i('align', 'lang');

will return:

   ('center', 'es-MX', 'i-klingon') # in list context
  or
   'center' # in scalar context.

=cut

sub attr_get_i {
    if ( @_ > 2 ) {
        my $self = shift;
        Carp::croak "No attribute names can be undef!"
            if grep !defined($_), @_;
        my @attributes = $self->_fold_case(@_);
        if (wantarray) {
            my @out;
            foreach my $x ( $self, $self->lineage ) {
                push @out,
                    map { exists( $x->{$_} ) ? $x->{$_} : () } @attributes;
            }
            return @out;
        }
        else {
            foreach my $x ( $self, $self->lineage ) {
                foreach my $attribute (@attributes) {
                    return $x->{$attribute}
                        if exists $x->{$attribute};    # found
                }
            }
            return;                                    # never found
        }
    }
    else {

        # Single-attribute search.  Simpler, most common, so optimize
        #  for the most common case
        Carp::croak "Attribute name must be a defined value!"
            unless defined $_[1];
        my $self      = $_[0];
        my $attribute = $self->_fold_case( $_[1] );
        if (wantarray) {                               # list context
            return
                map { exists( $_->{$attribute} ) ? $_->{$attribute} : () }
                $self, $self->lineage;
        }
        else {                                         # scalar context
            foreach my $x ( $self, $self->lineage ) {
                return $x->{$attribute} if exists $x->{$attribute};    # found
            }
            return;    # never found
        }
    }
}

=head2 $h->tagname_map()

Scans across C<$h> and all its descendants, and makes a hash (a
reference to which is returned) where each entry consists of a key
that's a tag name, and a value that's a reference to a list to all
elements that have that tag name.  I.e., this method returns:

   {
     # Across $h and all descendants...
     'a'   => [ ...list of all 'a'   elements... ],
     'em'  => [ ...list of all 'em'  elements... ],
     'img' => [ ...list of all 'img' elements... ],
   }

(There are entries in the hash for only those tagnames that occur
at/under C<$h> -- so if there's no "img" elements, there'll be no
"img" entry in the hashr(ref) returned.)

Example usage:

    my $map_r = $h->tagname_map();
    my @heading_tags = sort grep m/^h\d$/s, keys %$map_r;
    if(@heading_tags) {
      print "Heading levels used: @heading_tags\n";
    } else {
      print "No headings.\n"
    }

=cut

sub tagname_map {
    my (@pile) = $_[0];    # start out the to-do stack for the traverser
    Carp::croak "find_by_tag_name can be called only as an object method"
        unless ref $pile[0];
    my ( %map, $this_tag, $this );
    while (@pile) {
        $this_tag = ''
            unless defined( $this_tag = ( $this = shift @pile )->{'_tag'} )
        ;    # dance around the strange case of having an undef tagname.
        push @{ $map{$this_tag} ||= [] }, $this;    # add to map
        unshift @pile, grep ref($_),
            @{ $this->{'_content'} || next };       # traverse
    }
    return \%map;
}

=head2 $h->extract_links() or $h->extract_links(@wantedTypes)

Returns links found by traversing the element and all of its children
and looking for attributes (like "href" in an "a" element, or "src" in
an "img" element) whose values represent links.  The return value is a
I<reference> to an array.  Each element of the array is reference to
an array with I<four> items: the link-value, the element that has the
attribute with that link-value, and the name of that attribute, and
the tagname of that element.
(Example: C<['http://www.suck.com/',> I<$elem_obj> C<, 'href', 'a']>.)
You may or may not end up using the
element itself -- for some purposes, you may use only the link value.

You might specify that you want to extract links from just some kinds
of elements (instead of the default, which is to extract links from
I<all> the kinds of elements known to have attributes whose values
represent links).  For instance, if you want to extract links from
only "a" and "img" elements, you could code it like this:

  for (@{  $e->extract_links('a', 'img')  }) {
      my($link, $element, $attr, $tag) = @$_;
      print
        "Hey, there's a $tag that links to ",
        $link, ", in its $attr attribute, at ",
        $element->address(), ".\n";
  }

=cut

sub extract_links {
    my $start = shift;

    my %wantType;
    @wantType{ $start->_fold_case(@_) } = (1) x @_;    # if there were any
    my $wantType = scalar(@_);

    my @links;

    # TODO: add xml:link?

    my ( $link_attrs, $tag, $self, $val );    # scratch for each iteration
    $start->traverse(
        [   sub {                             # pre-order call only
                $self = $_[0];

                $tag = $self->{'_tag'};
                return 1
                    if $wantType && !$wantType{$tag};    # if we're selective

                if (defined(
                        $link_attrs = $HTML::Element::linkElements{$tag}
                    )
                    )
                {

                    # If this is a tag that has any link attributes,
                    #  look over possibly present link attributes,
                    #  saving the value, if found.
                    for ( ref($link_attrs) ? @$link_attrs : $link_attrs ) {
                        if ( defined( $val = $self->attr($_) ) ) {
                            push( @links, [ $val, $self, $_, $tag ] );
                        }
                    }
                }
                1;    # return true, so we keep recursing
            },
            undef
        ],
        1,            # ignore text nodes
    );
    \@links;
}

=head2 $h->simplify_pres

In text bits under PRE elements that are at/under $h, this routine
nativizes all newlines, and expands all tabs.

That is, if you read a file with lines delimited by C<\cm\cj>'s, the
text under PRE areas will have C<\cm\cj>'s instead of C<\n>'s. Calling
$h->nativize_pre_newlines on such a tree will turn C<\cm\cj>'s into
C<\n>'s.

Tabs are expanded to however many spaces it takes to get
to the next 8th column -- the usual way of expanding them.

=cut

sub simplify_pres {
    my $pre = 0;

    my $sub;
    my $line;
    $sub = sub {
        ++$pre if $_[0]->{'_tag'} eq 'pre';
        foreach my $it ( @{ $_[0]->{'_content'} || return } ) {
            if ( ref $it ) {
                $sub->($it);    # recurse!
            }
            elsif ($pre) {

                #$it =~ s/(?:(?:\cm\cj*)|(?:\cj))/\n/g;

                $it = join "\n", map {
                    ;
                    $line = $_;
                    while (
                        $line
                        =~ s/^([^\t]*)(\t+)/$1.(" " x ((length($2)<<3)-(length($1)&7)))/e

              # Sort of adapted from Text::Tabs -- yes, it's hardwired-in that
              # tabs are at every EIGHTH column.
                        )
                    {
                    }
                    $line;
                    }
                    split /(?:(?:\cm\cj*)|(?:\cj))/, $it, -1;
            }
        }
        --$pre if $_[0]->{'_tag'} eq 'pre';
        return;
    };
    $sub->( $_[0] );

    undef $sub;
    return;

}

=head2 $h->same_as($i)

Returns true if $h and $i are both elements representing the same tree
of elements, each with the same tag name, with the same explicit
attributes (i.e., not counting attributes whose names start with "_"),
and with the same content (textual, comments, etc.).

Sameness of descendant elements is tested, recursively, with
C<$child1-E<gt>same_as($child_2)>, and sameness of text segments is tested
with C<$segment1 eq $segment2>.

=cut

sub same_as {
    die 'same_as() takes only one argument: $h->same_as($i)' unless @_ == 2;
    my ( $h, $i ) = @_[ 0, 1 ];
    die "same_as() can be called only as an object method" unless ref $h;

    return 0 unless defined $i and ref $i;

    # An element can't be same_as anything but another element!
    # They needn't be of the same class, tho.

    return 1 if $h eq $i;

    # special (if rare) case: anything is the same as... itself!

    # assumes that no content lists in/under $h or $i contain subsequent
    #  text segments, like: ['foo', ' bar']

    # compare attributes now.
    #print "Comparing tags of $h and $i...\n";

    return 0 unless $h->{'_tag'} eq $i->{'_tag'};

    # only significant attribute whose name starts with "_"

    #print "Comparing attributes of $h and $i...\n";
    # Compare attributes, but only the real ones.
    {

        # Bear in mind that the average element has very few attributes,
        #  and that element names are rather short.
        # (Values are a different story.)

    # XXX I would think that /^[^_]/ would be faster, at least easier to read.
        my @keys_h
            = sort grep { length $_ and substr( $_, 0, 1 ) ne '_' } keys %$h;
        my @keys_i
            = sort grep { length $_ and substr( $_, 0, 1 ) ne '_' } keys %$i;

        return 0 unless @keys_h == @keys_i;

        # different number of real attributes?  they're different.
        for ( my $x = 0; $x < @keys_h; ++$x ) {
            return 0
                unless $keys_h[$x] eq $keys_i[$x] and    # same key name
                    $h->{ $keys_h[$x] } eq $i->{ $keys_h[$x] };   # same value
             # Should this test for definedness on values?
             # People shouldn't be putting undef in attribute values, I think.
        }
    }

    #print "Comparing children of $h and $i...\n";
    my $hcl = $h->{'_content'} || [];
    my $icl = $i->{'_content'} || [];

    return 0 unless @$hcl == @$icl;

    # different numbers of children?  they're different.

    if (@$hcl) {

        # compare each of the children:
        for ( my $x = 0; $x < @$hcl; ++$x ) {
            if ( ref $hcl->[$x] ) {
                return 0 unless ref( $icl->[$x] );

                # an element can't be the same as a text segment
                # Both elements:
                return 0 unless $hcl->[$x]->same_as( $icl->[$x] );  # RECURSE!
            }
            else {
                return 0 if ref( $icl->[$x] );

                # a text segment can't be the same as an element
                # Both text segments:
                return 0 unless $hcl->[$x] eq $icl->[$x];
            }
        }
    }

    return 1;    # passed all the tests!
}

=head2 $h = HTML::Element->new_from_lol(ARRAYREF)

Resursively constructs a tree of nodes, based on the (non-cyclic)
data structure represented by ARRAYREF, where that is a reference
to an array of arrays (of arrays (of arrays (etc.))).

In each arrayref in that structure, different kinds of values are
treated as follows:

=over

=item * Arrayrefs

Arrayrefs are considered to
designate a sub-tree representing children for the node constructed
from the current arrayref.

=item * Hashrefs

Hashrefs are considered to contain
attribute-value pairs to add to the element to be constructed from
the current arrayref

=item * Text segments

Text segments at the start of any arrayref
will be considered to specify the name of the element to be
constructed from the current araryref; all other text segments will
be considered to specify text segments as children for the current
arrayref.

=item * Elements

Existing element objects are either inserted into the treelet
constructed, or clones of them are.  That is, when the lol-tree is
being traversed and elements constructed based what's in it, if
an existing element object is found, if it has no parent, then it is
added directly to the treelet constructed; but if it has a parent,
then C<$that_node-E<gt>clone> is added to the treelet at the
appropriate place.

=back

An example will hopefully make this more obvious:

  my $h = HTML::Element->new_from_lol(
    ['html',
      ['head',
        [ 'title', 'I like stuff!' ],
      ],
      ['body',
        {'lang', 'en-JP', _implicit => 1},
        'stuff',
        ['p', 'um, p < 4!', {'class' => 'par123'}],
        ['div', {foo => 'bar'}, '123'],
      ]
    ]
  );
  $h->dump;

Will print this:

  <html> @0
    <head> @0.0
      <title> @0.0.0
        "I like stuff!"
    <body lang="en-JP"> @0.1 (IMPLICIT)
      "stuff"
      <p class="par123"> @0.1.1
        "um, p < 4!"
      <div foo="bar"> @0.1.2
        "123"

And printing $h->as_HTML will give something like:

  <html><head><title>I like stuff!</title></head>
  <body lang="en-JP">stuff<p class="par123">um, p &lt; 4!
  <div foo="bar">123</div></body></html>

You can even do fancy things with C<map>:

  $body->push_content(
    # push_content implicitly calls new_from_lol on arrayrefs...
    ['br'],
    ['blockquote',
      ['h2', 'Pictures!'],
      map ['p', $_],
      $body2->look_down("_tag", "img"),
        # images, to be copied from that other tree.
    ],
    # and more stuff:
    ['ul',
      map ['li', ['a', {'href'=>"$_.png"}, $_ ] ],
      qw(Peaches Apples Pears Mangos)
    ],
  );

=head2 @elements = HTML::Element->new_from_lol(ARRAYREFS)

Constructs I<several> elements, by calling
new_from_lol for every arrayref in the ARRAYREFS list.

  @elements = HTML::Element->new_from_lol(
    ['hr'],
    ['p', 'And there, on the door, was a hook!'],
  );
   # constructs two elements.

=cut

sub new_from_lol {
    my $class = shift;
    $class = ref($class) || $class;

  # calling as an object method is just the same as ref($h)->new_from_lol(...)
    my $lol = $_[1];

    my @ancestor_lols;

    # So we can make sure there's no cyclicities in this lol.
    # That would be perverse, but one never knows.
    my ( $sub, $k, $v, $node );    # last three are scratch values
    $sub = sub {

        #print "Building for $_[0]\n";
        my $lol = $_[0];
        return unless @$lol;
        my ( @attributes, @children );
        Carp::croak "Cyclicity detected in source LOL tree, around $lol?!?"
            if grep( $_ eq $lol, @ancestor_lols );
        push @ancestor_lols, $lol;

        my $tag_name = 'null';

        # Recursion in in here:
        for ( my $i = 0; $i < @$lol; ++$i ) {    # Iterate over children
            if ( ref( $lol->[$i] ) eq 'ARRAY' )
            {    # subtree: most common thing in loltree
                push @children, $sub->( $lol->[$i] );
            }
            elsif ( !ref( $lol->[$i] ) ) {
                if ( $i == 0 ) {    # name
                    $tag_name = $lol->[$i];
                    Carp::croak "\"$tag_name\" isn't a good tag name!"
                        if $tag_name =~ m/[<>\/\x00-\x20]/
                    ;               # minimal sanity, certainly!
                }
                else {              # text segment child
                    push @children, $lol->[$i];
                }
            }
            elsif ( ref( $lol->[$i] ) eq 'HASH' ) {    # attribute hashref
                keys %{ $lol->[$i] };   # reset the each-counter, just in case
                while ( ( $k, $v ) = each %{ $lol->[$i] } ) {
                    push @attributes, $class->_fold_case($k), $v
                        if defined $v
                            and $k ne '_name'
                            and $k ne '_content'
                            and $k ne '_parent';

                    # enforce /some/ sanity!
                }
            }
            elsif ( UNIVERSAL::isa( $lol->[$i], __PACKAGE__ ) ) {
                if ( $lol->[$i]->{'_parent'} ) {    # if claimed
                        #print "About to clone ", $lol->[$i], "\n";
                    push @children, $lol->[$i]->clone();
                }
                else {
                    push @children, $lol->[$i];    # if unclaimed...
                         #print "Claiming ", $lol->[$i], "\n";
                    $lol->[$i]->{'_parent'} = 1;    # claim it NOW
                      # This WILL be replaced by the correct value once we actually
                      #  construct the parent, just after the end of this loop...
                }
            }
            else {
                Carp::croak "new_from_lol doesn't handle references of type "
                    . ref( $lol->[$i] );
            }
        }

        pop @ancestor_lols;
        $node = $class->new($tag_name);

        #print "Children: @children\n";

        if ( $class eq __PACKAGE__ ) {    # Special-case it, for speed:
            %$node = ( %$node, @attributes ) if @attributes;

            #print join(' ', $node, ' ' , map("<$_>", %$node), "\n");
            if (@children) {
                $node->{'_content'} = \@children;
                foreach my $c (@children) {
                    $c->{'_parent'} = $node
                        if ref $c;
                }
            }
        }
        else {                            # Do it the clean way...
                                          #print "Done neatly\n";
            while (@attributes) { $node->attr( splice @attributes, 0, 2 ) }
            $node->push_content( map { $_->{'_parent'} = $node if ref $_; $_ }
                    @children )
                if @children;
        }

        return $node;
    };

    # End of sub definition.

    if (wantarray) {
        my (@nodes) = map { ; ( ref($_) eq 'ARRAY' ) ? $sub->($_) : $_ } @_;

        # Let text bits pass thru, I guess.  This makes this act more like
        #  unshift_content et al.  Undocumented.
        undef $sub;

        # so it won't be in its own frame, so its refcount can hit 0
        return @nodes;
    }
    else {
        Carp::croak "new_from_lol in scalar context needs exactly one lol"
            unless @_ == 1;
        return $_[0] unless ref( $_[0] ) eq 'ARRAY';

        # used to be a fatal error.  still undocumented tho.
        $node = $sub->( $_[0] );
        undef $sub;

        # so it won't be in its own frame, so its refcount can hit 0
        return $node;
    }
}

=head2 $h->objectify_text()

This turns any text nodes under $h from mere text segments (strings)
into real objects, pseudo-elements with a tag-name of "~text", and the
actual text content in an attribute called "text".  (For a discussion
of pseudo-elements, see the "tag" method, far above.)  This method is
provided because, for some purposes, it is convenient or necessary to
be able, for a given text node, to ask what element is its parent; and
clearly this is not possible if a node is just a text string.

Note that these "~text" objects are not recognized as text nodes by
methods like as_text.  Presumably you will want to call
$h->objectify_text, perform whatever task that you needed that for,
and then call $h->deobjectify_text before calling anything like
$h->as_text.

=head2 $h->deobjectify_text()

This undoes the effect of $h->objectify_text.  That is, it takes any
"~text" pseudo-elements in the tree at/under $h, and deletes each one,
replacing each with the content of its "text" attribute. 

Note that if $h itself is a "~text" pseudo-element, it will be
destroyed -- a condition you may need to treat specially in your
calling code (since it means you can't very well do anything with $h
after that).  So that you can detect that condition, if $h is itself a
"~text" pseudo-element, then this method returns the value of the
"text" attribute, which should be a defined value; in all other cases,
it returns undef.

(This method assumes that no "~text" pseudo-element has any children.)

=cut

sub objectify_text {
    my (@stack) = ( $_[0] );

    my ($this);
    while (@stack) {
        foreach my $c ( @{ ( $this = shift @stack )->{'_content'} } ) {
            if ( ref($c) ) {
                unshift @stack, $c;    # visit it later.
            }
            else {
                $c = $this->element_class->new(
                    '~text',
                    'text'    => $c,
                    '_parent' => $this
                );
            }
        }
    }
    return;
}

sub deobjectify_text {
    my (@stack) = ( $_[0] );
    my ($old_node);

    if ( $_[0]{'_tag'} eq '~text' ) {    # special case
            # Puts the $old_node variable to a different purpose
        if ( $_[0]{'_parent'} ) {
            $_[0]->replace_with( $old_node = delete $_[0]{'text'} )->delete;
        }
        else {    # well, that's that, then!
            $old_node = delete $_[0]{'text'};
        }

        if ( ref( $_[0] ) eq __PACKAGE__ ) {    # common case
            %{ $_[0] } = ();                    # poof!
        }
        else {

            # play nice:
            delete $_[0]{'_parent'};
            $_[0]->delete;
        }
        return '' unless defined $old_node;     # sanity!
        return $old_node;
    }

    while (@stack) {
        foreach my $c ( @{ ( shift @stack )->{'_content'} } ) {
            if ( ref($c) ) {
                if ( $c->{'_tag'} eq '~text' ) {
                    $c = ( $old_node = $c )->{'text'};
                    if ( ref($old_node) eq __PACKAGE__ ) {    # common case
                        %$old_node = ();                      # poof!
                    }
                    else {

                        # play nice:
                        delete $old_node->{'_parent'};
                        $old_node->delete;
                    }
                }
                else {
                    unshift @stack, $c;    # visit it later.
                }
            }
        }
    }

    return;
}

=head2 $h->number_lists()

For every UL, OL, DIR, and MENU element at/under $h, this sets a
"_bullet" attribute for every child LI element.  For LI children of an
OL, the "_bullet" attribute's value will be something like "4.", "d.",
"D.", "IV.", or "iv.", depending on the OL element's "type" attribute.
LI children of a UL, DIR, or MENU get their "_bullet" attribute set
to "*".
There should be no other LIs (i.e., except as children of OL, UL, DIR,
or MENU elements), and if there are, they are unaffected.

=cut

{

    # The next three subs are basically copied from Number::Latin,
    # based on a one-liner by Abigail.  Yes, I could simply require that
    # module, and a Roman numeral module too, but really, HTML-Tree already
    # has enough dependecies as it is; and anyhow, I don't need the functions
    # that do latin2int or roman2int.
    no integer;

    sub _int2latin {
        return unless defined $_[0];
        return '0' if $_[0] < 1 and $_[0] > -1;
        return '-' . _i2l( abs int $_[0] )
            if $_[0] <= -1;    # tolerate negatives
        return _i2l( int $_[0] );
    }

    sub _int2LATIN {

        # just the above plus uc
        return unless defined $_[0];
        return '0' if $_[0] < 1 and $_[0] > -1;
        return '-' . uc( _i2l( abs int $_[0] ) )
            if $_[0] <= -1;    # tolerate negs
        return uc( _i2l( int $_[0] ) );
    }

    my @alpha = ( 'a' .. 'z' );

    sub _i2l {                 # the real work
        my $int = $_[0] || return "";
        _i2l( int( ( $int - 1 ) / 26 ) )
            . $alpha[ $int % 26 - 1 ];    # yes, recursive
            # Yes, 26 => is (26 % 26 - 1), which is -1 => Z!
    }
}

{

    # And now, some much less impressive Roman numerals code:

    my (@i) = ( '', qw(I II III IV V VI VII VIII IX) );
    my (@x) = ( '', qw(X XX XXX XL L LX LXX LXXX XC) );
    my (@c) = ( '', qw(C CC CCC CD D DC DCC DCCC CM) );
    my (@m) = ( '', qw(M MM MMM) );

    sub _int2ROMAN {
        my ( $i, $pref );
        return '0'
            if 0 == ( $i = int( $_[0] || 0 ) );    # zero is a special case
        return $i + 0 if $i <= -4000 or $i >= 4000;

       # Because over 3999 would require non-ASCII chars, like D-with-)-inside
        if ( $i < 0 ) {    # grumble grumble tolerate negatives grumble
            $pref = '-';
            $i    = abs($i);
        }
        else {
            $pref = '';    # normal case
        }

        my ( $x, $c, $m ) = ( 0, 0, 0 );
        if ( $i >= 10 ) {
            $x = $i / 10;
            $i %= 10;
            if ( $x >= 10 ) {
                $c = $x / 10;
                $x %= 10;
                if ( $c >= 10 ) { $m = $c / 10; $c %= 10; }
            }
        }

        #print "m$m c$c x$x i$i\n";

        return join( '', $pref, $m[$m], $c[$c], $x[$x], $i[$i] );
    }

    sub _int2roman { lc( _int2ROMAN( $_[0] ) ) }
}

sub _int2int { $_[0] }    # dummy

%list_type_to_sub = (
    'I' => \&_int2ROMAN,
    'i' => \&_int2roman,
    'A' => \&_int2LATIN,
    'a' => \&_int2latin,
    '1' => \&_int2int,
);

sub number_lists {
    my (@stack) = ( $_[0] );
    my ( $this, $tag, $counter, $numberer );    # scratch
    while (@stack) {    # yup, pre-order-traverser idiom
        if ( ( $tag = ( $this = shift @stack )->{'_tag'} ) eq 'ol' ) {

            # Prep some things:
            $counter
                = ( ( $this->{'start'} || '' ) =~ m<^\s*(\d{1,7})\s*$>s )
                ? $1
                : 1;
            $numberer = $list_type_to_sub{ $this->{'type'} || '' }
                || $list_type_to_sub{'1'};

            # Immeditately iterate over all children
            foreach my $c ( @{ $this->{'_content'} || next } ) {
                next unless ref $c;
                unshift @stack, $c;
                if ( $c->{'_tag'} eq 'li' ) {
                    $counter = $1
                        if (
                        ( $c->{'value'} || '' ) =~ m<^\s*(\d{1,7})\s*$>s );
                    $c->{'_bullet'} = $numberer->($counter) . '.';
                    ++$counter;
                }
            }

        }
        elsif ( $tag eq 'ul' or $tag eq 'dir' or $tag eq 'menu' ) {

            # Immeditately iterate over all children
            foreach my $c ( @{ $this->{'_content'} || next } ) {
                next unless ref $c;
                unshift @stack, $c;
                $c->{'_bullet'} = '*' if $c->{'_tag'} eq 'li';
            }

        }
        else {
            foreach my $c ( @{ $this->{'_content'} || next } ) {
                unshift @stack, $c if ref $c;
            }
        }
    }
    return;
}

=head2 $h->has_insane_linkage

This method is for testing whether this element or the elements
under it have linkage attributes (_parent and _content) whose values
are deeply aberrant: if there are undefs in a content list; if an
element appears in the content lists of more than one element;
if the _parent attribute of an element doesn't match its actual
parent; or if an element appears as its own descendant (i.e.,
if there is a cyclicity in the tree).

This returns empty list (or false, in scalar context) if the subtree's
linkage methods are sane; otherwise it returns two items (or true, in
scalar context): the element where the error occurred, and a string
describing the error.

This method is provided is mainly for debugging and troubleshooting --
it should be I<quite impossible> for any document constructed via
HTML::TreeBuilder to parse into a non-sane tree (since it's not
the content of the tree per se that's in question, but whether
the tree in memory was properly constructed); and it I<should> be
impossible for you to produce an insane tree just thru reasonable
use of normal documented structure-modifying methods.  But if you're
constructing your own trees, and your program is going into infinite
loops as during calls to traverse() or any of the secondary
structural methods, as part of debugging, consider calling is_insane
on the tree.

=cut

sub has_insane_linkage {
    my @pile = ( $_[0] );
    my ( $c, $i, $p, $this );    # scratch

    # Another iterative traverser; this time much simpler because
    #  only in pre-order:
    my %parent_of = ( $_[0], 'TOP-OF-SCAN' );
    while (@pile) {
        $this = shift @pile;
        $c = $this->{'_content'} || next;
        return ( $this, "_content attribute is true but nonref." )
            unless ref($c) eq 'ARRAY';
        next unless @$c;
        for ( $i = 0; $i < @$c; ++$i ) {
            return ( $this, "Child $i is undef" )
                unless defined $c->[$i];
            if ( ref( $c->[$i] ) ) {
                return ( $c->[$i], "appears in its own content list" )
                    if $c->[$i] eq $this;
                return ( $c->[$i],
                    "appears twice in the tree: once under $this, once under $parent_of{$c->[$i]}"
                ) if exists $parent_of{ $c->[$i] };
                $parent_of{ $c->[$i] } = '' . $this;

                # might as well just use the stringification of it.

                return ( $c->[$i],
                    "_parent attribute is wrong (not defined)" )
                    unless defined( $p = $c->[$i]{'_parent'} );
                return ( $c->[$i], "_parent attribute is wrong (nonref)" )
                    unless ref($p);
                return ( $c->[$i],
                    "_parent attribute is wrong (is $p; should be $this)" )
                    unless $p eq $this;
            }
        }
        unshift @pile, grep ref($_), @$c;

        # queue up more things on the pile stack
    }
    return;    #okay
}

sub _asserts_fail {    # to be run on trusted documents only
    my (@pile) = ( $_[0] );
    my ( @errors, $this, $id, $assert, $parent, $rv );
    while (@pile) {
        $this = shift @pile;
        if ( defined( $assert = $this->{'assert'} ) ) {
            $id = ( $this->{'id'} ||= $this->address )
                ;      # don't use '0' as an ID, okay?
            unless ( ref($assert) ) {

                package main;
## no critic
                $assert = $this->{'assert'} = (
                    $assert =~ m/\bsub\b/
                    ? eval($assert)
                    : eval("sub {  $assert\n}")
                );
## use critic
                if ($@) {
                    push @errors,
                        [ $this, "assertion at $id broke in eval: $@" ];
                    $assert = $this->{'assert'} = sub { };
                }
            }
            $parent = $this->{'_parent'};
            $rv     = undef;
            eval {
                $rv = $assert->(
                    $this, $this->{'_tag'}, $this->{'_id'},    # 0,1,2
                    $parent
                    ? ( $parent, $parent->{'_tag'}, $parent->{'id'} )
                    : ()                                       # 3,4,5
                );
            };
            if ($@) {
                push @errors, [ $this, "assertion at $id died: $@" ];
            }
            elsif ( !$rv ) {
                push @errors, [ $this, "assertion at $id failed" ];
            }

            # else OK
        }
        push @pile, grep ref($_), @{ $this->{'_content'} || next };
    }
    return @errors;
}

## _valid_name
#  validate XML style attribute names
#  http://www.w3.org/TR/2006/REC-xml11-20060816/#NT-Name

sub _valid_name {
    my $self = shift;
    my $attr = shift
        or Carp::croak("sub valid_name requires an attribute name");

    return (0) unless ( $attr =~ /^$START_CHAR$NAME_CHAR+$/ );

    return (1);
}

=head2 $h->element_class

This method returns the class which will be used for new elements.  It
defaults to HTML::Element, but can be overridden by subclassing or esoteric
means best left to those will will read the source and then not complain when
those esoteric means change.  (Just subclass.)

=cut

sub element_class {
    $_[0]->{_element_class} || __PACKAGE__;
}

1;

=head1 BUGS

* If you want to free the memory associated with a tree built of
HTML::Element nodes, then you will have to delete it explicitly.
See the $h->delete method, above.

* There's almost nothing to stop you from making a "tree" with
cyclicities (loops) in it, which could, for example, make the
traverse method go into an infinite loop.  So don't make
cyclicities!  (If all you're doing is parsing HTML files,
and looking at the resulting trees, this will never be a problem
for you.)

* There's no way to represent comments or processing directives
in a tree with HTML::Elements.  Not yet, at least.

* There's (currently) nothing to stop you from using an undefined
value as a text segment.  If you're running under C<perl -w>, however,
this may make HTML::Element's code produce a slew of warnings.

=head1 NOTES ON SUBCLASSING

You are welcome to derive subclasses from HTML::Element, but you
should be aware that the code in HTML::Element makes certain
assumptions about elements (and I'm using "element" to mean ONLY an
object of class HTML::Element, or of a subclass of HTML::Element):

* The value of an element's _parent attribute must either be undef or
otherwise false, or must be an element.

* The value of an element's _content attribute must either be undef or
otherwise false, or a reference to an (unblessed) array.  The array
may be empty; but if it has items, they must ALL be either mere
strings (text segments), or elements.

* The value of an element's _tag attribute should, at least, be a 
string of printable characters.

Moreover, bear these rules in mind:

* Do not break encapsulation on objects.  That is, access their
contents only thru $obj->attr or more specific methods.

* You should think twice before completely overriding any of the
methods that HTML::Element provides.  (Overriding with a method that
calls the superclass method is not so bad, though.)

=head1 SEE ALSO

L<HTML::Tree>; L<HTML::TreeBuilder>; L<HTML::AsSubs>; L<HTML::Tagset>; 
and, for the morbidly curious, L<HTML::Element::traverse>.

=head1 COPYRIGHT

Copyright 1995-1998 Gisle Aas, 1999-2004 Sean M. Burke, 2005 Andy Lester,
2006 Pete Krawczyk, 2010 Jeff Fearn.

This library is free software; you can redistribute it and/or
modify it under the same terms as Perl itself.

This program is distributed in the hope that it will be useful, but
without any warranty; without even the implied warranty of
merchantability or fitness for a particular purpose.

=head1 AUTHOR

Current Author:
	Jeff Fearn C<< <jfearn@cpan.org> >>.

Original HTML-Tree author:
	Gisle Aas.

Former Authors:
	Sean M. Burke.
	Andy Lester.
	Pete Krawczyk C<< <petek@cpan.org> >>.

Thanks to Mark-Jason Dominus for a POD suggestion.

=cut

1;