summaryrefslogtreecommitdiffstats
path: root/Python/_warnings.c
blob: ef7f373bddcd92ba9cf4d432f4fa749b74effb00 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
#include "Python.h"
#include "frameobject.h"

#define MODULE_NAME "_warnings"

PyDoc_STRVAR(warnings__doc__,
MODULE_NAME " provides basic warning filtering support.\n"
"It is a helper module to speed up interpreter start-up.");

/* Both 'filters' and 'onceregistry' can be set in warnings.py;
   get_warnings_attr() will reset these variables accordingly. */
static PyObject *_filters;  /* List */
static PyObject *_once_registry;  /* Dict */
static PyObject *_default_action; /* String */


static int
check_matched(PyObject *obj, PyObject *arg)
{
    PyObject *result;
    int rc;

    if (obj == Py_None)
        return 1;
    result = PyObject_CallMethod(obj, "match", "O", arg);
    if (result == NULL)
        return -1;

    rc = PyObject_IsTrue(result);
    Py_DECREF(result);
    return rc;
}

/*
   Returns a new reference.
   A NULL return value can mean false or an error.
*/
static PyObject *
get_warnings_attr(const char *attr)
{
    static PyObject *warnings_str = NULL;
    PyObject *all_modules;
    PyObject *warnings_module;
    int result;

    if (warnings_str == NULL) {
        warnings_str = PyUnicode_InternFromString("warnings");
        if (warnings_str == NULL)
            return NULL;
    }

    all_modules = PyImport_GetModuleDict();
    result = PyDict_Contains(all_modules, warnings_str);
    if (result == -1 || result == 0)
        return NULL;

    warnings_module = PyDict_GetItem(all_modules, warnings_str);
    if (!PyObject_HasAttrString(warnings_module, attr))
            return NULL;
    return PyObject_GetAttrString(warnings_module, attr);
}


static PyObject *
get_once_registry(void)
{
    PyObject *registry;

    registry = get_warnings_attr("onceregistry");
    if (registry == NULL) {
        if (PyErr_Occurred())
            return NULL;
        return _once_registry;
    }
    Py_DECREF(_once_registry);
    _once_registry = registry;
    return registry;
}


static PyObject *
get_default_action(void)
{
    PyObject *default_action;

    default_action = get_warnings_attr("defaultaction");
    if (default_action == NULL) {
	if (PyErr_Occurred()) {
	    return NULL;
	}
	return _default_action;
    }

    Py_DECREF(_default_action);
    _default_action = default_action;
    return default_action;
}


/* The item is a borrowed reference. */
static const char *
get_filter(PyObject *category, PyObject *text, Py_ssize_t lineno,
           PyObject *module, PyObject **item)
{
    PyObject *action;
    Py_ssize_t i;
    PyObject *warnings_filters;

    warnings_filters = get_warnings_attr("filters");
    if (warnings_filters == NULL) {
        if (PyErr_Occurred())
            return NULL;
    }
    else {
        Py_DECREF(_filters);
        _filters = warnings_filters;
    }

    if (!PyList_Check(_filters)) {
        PyErr_SetString(PyExc_ValueError,
                        MODULE_NAME ".filters must be a list");
        return NULL;
    }

    /* _filters could change while we are iterating over it. */
    for (i = 0; i < PyList_GET_SIZE(_filters); i++) {
        PyObject *tmp_item, *action, *msg, *cat, *mod, *ln_obj;
        Py_ssize_t ln;
        int is_subclass, good_msg, good_mod;

        tmp_item = *item = PyList_GET_ITEM(_filters, i);
        if (PyTuple_Size(tmp_item) != 5) {
            PyErr_Format(PyExc_ValueError,
                         MODULE_NAME ".filters item %zd isn't a 5-tuple", i);
            return NULL;
        }

        /* Python code: action, msg, cat, mod, ln = item */
        action = PyTuple_GET_ITEM(tmp_item, 0);
        msg = PyTuple_GET_ITEM(tmp_item, 1);
        cat = PyTuple_GET_ITEM(tmp_item, 2);
        mod = PyTuple_GET_ITEM(tmp_item, 3);
        ln_obj = PyTuple_GET_ITEM(tmp_item, 4);

        good_msg = check_matched(msg, text);
        good_mod = check_matched(mod, module);
        is_subclass = PyObject_IsSubclass(category, cat);
        ln = PyLong_AsSsize_t(ln_obj);
        if (good_msg == -1 || good_mod == -1 || is_subclass == -1 ||
            (ln == -1 && PyErr_Occurred()))
            return NULL;

        if (good_msg && is_subclass && good_mod && (ln == 0 || lineno == ln))
            return _PyUnicode_AsString(action);
    }

    action = get_default_action();
    if (action != NULL) {
        return _PyUnicode_AsString(action);
    }

    PyErr_SetString(PyExc_ValueError,
                    MODULE_NAME ".defaultaction not found");
    return NULL;
}


static int
already_warned(PyObject *registry, PyObject *key, int should_set)
{
    PyObject *already_warned;

    if (key == NULL)
        return -1;

    already_warned = PyDict_GetItem(registry, key);
    if (already_warned != NULL) {
        int rc = PyObject_IsTrue(already_warned);
        if (rc != 0)
            return rc;
    }

    /* This warning wasn't found in the registry, set it. */
    if (should_set)
        return PyDict_SetItem(registry, key, Py_True);
    return 0;
}

/* New reference. */
static PyObject *
normalize_module(PyObject *filename)
{
    PyObject *module;
    const char *mod_str;
    Py_ssize_t len;

    int rc = PyObject_IsTrue(filename);
    if (rc == -1)
        return NULL;
    else if (rc == 0)
        return PyUnicode_FromString("<unknown>");

    mod_str = _PyUnicode_AsString(filename);
    if (mod_str == NULL)
	    return NULL;
    len = PyUnicode_GetSize(filename);
    if (len < 0)
        return NULL;
    if (len >= 3 &&
	strncmp(mod_str + (len - 3), ".py", 3) == 0) {
        module = PyUnicode_FromStringAndSize(mod_str, len-3);
    }
    else {
        module = filename;
        Py_INCREF(module);
    }
    return module;
}

static int
update_registry(PyObject *registry, PyObject *text, PyObject *category,
                int add_zero)
{
    PyObject *altkey, *zero = NULL;
    int rc;

    if (add_zero) {
        zero = PyLong_FromLong(0);
        if (zero == NULL)
            return -1;
        altkey = PyTuple_Pack(3, text, category, zero);
    }
    else
        altkey = PyTuple_Pack(2, text, category);

    rc = already_warned(registry, altkey, 1);
    Py_XDECREF(zero);
    Py_XDECREF(altkey);
    return rc;
}

static void
show_warning(PyObject *filename, int lineno, PyObject *text, PyObject
                *category, PyObject *sourceline)
{
    PyObject *f_stderr; 
    PyObject *name; 
    char lineno_str[128];

    PyOS_snprintf(lineno_str, sizeof(lineno_str), ":%d: ", lineno);

    name = PyObject_GetAttrString(category, "__name__");
    if (name == NULL)  /* XXX Can an object lack a '__name__' attribute? */
	    return;

    f_stderr = PySys_GetObject("stderr");
    if (f_stderr == NULL) {
        fprintf(stderr, "lost sys.stderr\n");
        Py_DECREF(name);
        return;
    }

    /* Print "filename:lineno: category: text\n" */
    PyFile_WriteObject(filename, f_stderr, Py_PRINT_RAW);
    PyFile_WriteString(lineno_str, f_stderr);
    PyFile_WriteObject(name, f_stderr, Py_PRINT_RAW);
    PyFile_WriteString(": ", f_stderr);
    PyFile_WriteObject(text, f_stderr, Py_PRINT_RAW);
    PyFile_WriteString("\n", f_stderr);
    Py_XDECREF(name);

    /* Print "  source_line\n" */
    if (sourceline) {
        char *source_line_str = _PyUnicode_AsString(sourceline);
	if (source_line_str == NULL)
		return;
        while (*source_line_str == ' ' || *source_line_str == '\t' ||
                *source_line_str == '\014')
            source_line_str++;

        PyFile_WriteString(source_line_str, f_stderr);
        PyFile_WriteString("\n", f_stderr);
    }
    else
        if (_Py_DisplaySourceLine(f_stderr, _PyUnicode_AsString(filename),
                              lineno, 2) < 0)
		return;
    PyErr_Clear();
}

static PyObject *
warn_explicit(PyObject *category, PyObject *message, 
              PyObject *filename, int lineno,
              PyObject *module, PyObject *registry, PyObject *sourceline)
{
    PyObject *key = NULL, *text = NULL, *result = NULL, *lineno_obj = NULL;
    PyObject *item = Py_None;
    const char *action;
    int rc;
    
    if (registry && !PyDict_Check(registry) && (registry != Py_None)) {
        PyErr_SetString(PyExc_TypeError, "'registry' must be a dict");
        return NULL;
    }

    /* Normalize module. */
    if (module == NULL) {
        module = normalize_module(filename);
        if (module == NULL)
            return NULL;
    }
    else
        Py_INCREF(module);

    /* Normalize message. */
    Py_INCREF(message);  /* DECREF'ed in cleanup. */
    rc = PyObject_IsInstance(message, PyExc_Warning);
    if (rc == -1) {
        goto cleanup;
    }
    if (rc == 1) {
        text = PyObject_Str(message);
        if (text == NULL)
            goto cleanup;
        category = (PyObject*)message->ob_type;
    }
    else {
        text = message;
        message = PyObject_CallFunction(category, "O", message);
        if (message == NULL)
            goto cleanup;
    }

    lineno_obj = PyLong_FromLong(lineno);
    if (lineno_obj == NULL)
        goto cleanup;

    /* Create key. */
    key = PyTuple_Pack(3, text, category, lineno_obj);
    if (key == NULL)
        goto cleanup;

    if ((registry != NULL) && (registry != Py_None)) {
        rc = already_warned(registry, key, 0);
        if (rc == -1)
            goto cleanup;
	else if (rc == 1)
            goto return_none;
        /* Else this warning hasn't been generated before. */
    }

    action = get_filter(category, text, lineno, module, &item);
    if (action == NULL)
        goto cleanup;

    if (strcmp(action, "error") == 0) {
        PyErr_SetObject(category, message);
        goto cleanup;
    }

    /* Store in the registry that we've been here, *except* when the action
       is "always". */
    rc = 0;
    if (strcmp(action, "always") != 0) {
        if (registry != NULL && registry != Py_None &&
                PyDict_SetItem(registry, key, Py_True) < 0)
            goto cleanup;
        else if (strcmp(action, "ignore") == 0)
            goto return_none;
        else if (strcmp(action, "once") == 0) {
            if (registry == NULL || registry == Py_None) {
                registry = get_once_registry();
                if (registry == NULL)
                    goto cleanup;
            }
            /* _once_registry[(text, category)] = 1 */
            rc = update_registry(registry, text, category, 0); 
        }
        else if (strcmp(action, "module") == 0) {
            /* registry[(text, category, 0)] = 1 */
            if (registry != NULL && registry != Py_None)
                rc = update_registry(registry, text, category, 0); 
        }
        else if (strcmp(action, "default") != 0) {
            PyObject *to_str = PyObject_Str(item);
            const char *err_str = "???";

            if (to_str != NULL) {
                err_str = _PyUnicode_AsString(to_str);
		if (err_str == NULL)
			goto cleanup;
	    }
            PyErr_Format(PyExc_RuntimeError,
                        "Unrecognized action (%s) in warnings.filters:\n %s",
                        action, err_str);
            Py_XDECREF(to_str);
            goto cleanup;
        }
    }

    if (rc == 1)  /* Already warned for this module. */
        goto return_none;
    if (rc == 0) {
        PyObject *show_fxn = get_warnings_attr("showwarning");
        if (show_fxn == NULL) {
            if (PyErr_Occurred())
                goto cleanup;
            show_warning(filename, lineno, text, category, sourceline);
        }
        else {
            PyObject *res;

            if (!PyMethod_Check(show_fxn) && !PyFunction_Check(show_fxn)) {
                PyErr_SetString(PyExc_TypeError,
                                "warnings.showwarning() must be set to a "
                                "function or method");
                Py_DECREF(show_fxn);
                goto cleanup;
            }

            res = PyObject_CallFunctionObjArgs(show_fxn, message, category,
                                                filename, lineno_obj,
                                                NULL);
            Py_DECREF(show_fxn);
            Py_XDECREF(res);
            if (res == NULL)
                goto cleanup;
        }
    }
    else /* if (rc == -1) */
        goto cleanup;

 return_none:
    result = Py_None;
    Py_INCREF(result);

 cleanup:
    Py_XDECREF(key);
    Py_XDECREF(text);
    Py_XDECREF(lineno_obj);
    Py_DECREF(module);
    Py_XDECREF(message);
    return result;  /* Py_None or NULL. */
}

/* filename, module, and registry are new refs, globals is borrowed */
/* Returns 0 on error (no new refs), 1 on success */
static int
setup_context(Py_ssize_t stack_level, PyObject **filename, int *lineno,
              PyObject **module, PyObject **registry)
{
    PyObject *globals;

    /* Setup globals and lineno. */
    PyFrameObject *f = PyThreadState_GET()->frame;
    while (--stack_level > 0 && f != NULL)
        f = f->f_back;

    if (f == NULL) {
        globals = PyThreadState_Get()->interp->sysdict;
        *lineno = 1;
    }
    else {
        globals = f->f_globals;
        *lineno = PyCode_Addr2Line(f->f_code, f->f_lasti);
    }

    *module = NULL;

    /* Setup registry. */
    assert(globals != NULL);
    assert(PyDict_Check(globals));
    *registry = PyDict_GetItemString(globals, "__warningregistry__");
    if (*registry == NULL) {
        int rc;

        *registry = PyDict_New();
        if (*registry == NULL)
            return 0;

         rc = PyDict_SetItemString(globals, "__warningregistry__", *registry);
         if (rc < 0)
            goto handle_error;
    }
    else
        Py_INCREF(*registry);

    /* Setup module. */
    *module = PyDict_GetItemString(globals, "__name__");
    if (*module == NULL) {
        *module = PyUnicode_FromString("<string>");
        if (*module == NULL)
            goto handle_error;
    }
    else
        Py_INCREF(*module);

    /* Setup filename. */
    *filename = PyDict_GetItemString(globals, "__file__");
    if (*filename != NULL) {
        Py_ssize_t len = PyUnicode_GetSize(*filename);
        const char *file_str = _PyUnicode_AsString(*filename);
	    if (file_str == NULL || (len < 0 && PyErr_Occurred()))
            goto handle_error;

        /* if filename.lower().endswith((".pyc", ".pyo")): */
        if (len >= 4 &&
            file_str[len-4] == '.' &&
            tolower(file_str[len-3]) == 'p' &&
            tolower(file_str[len-2]) == 'y' &&
            (tolower(file_str[len-1]) == 'c' ||
                tolower(file_str[len-1]) == 'o'))
        {
            *filename = PyUnicode_FromStringAndSize(file_str, len-1);
	        if (*filename == NULL)
		        goto handle_error;
	    }
	    else
            Py_INCREF(*filename);
    }
    else {
        const char *module_str = _PyUnicode_AsString(*module);
	if (module_str == NULL)
		goto handle_error;
        if (strcmp(module_str, "__main__") == 0) {
            PyObject *argv = PySys_GetObject("argv");
            if (argv != NULL && PyList_Size(argv) > 0) {
                int is_true;
                *filename = PyList_GetItem(argv, 0);
                Py_INCREF(*filename);
                /* If sys.argv[0] is false, then use '__main__'. */
                is_true = PyObject_IsTrue(*filename);
                if (is_true < 0) {
                    Py_DECREF(*filename);
                    goto handle_error;
                }
                else if (!is_true) {
                    Py_DECREF(*filename);
                    *filename = PyUnicode_FromString("__main__");
                    if (*filename == NULL)
                        goto handle_error;
                }
            }
            else {
                /* embedded interpreters don't have sys.argv, see bug #839151 */
                *filename = PyUnicode_FromString("__main__");
	            if (*filename == NULL)
	                goto handle_error;
            }
        }
        if (*filename == NULL) {
            *filename = *module;
            Py_INCREF(*filename);
        }
    }

    return 1;

 handle_error:
    /* filename not XDECREF'ed here as there is no way to jump here with a
       dangling reference. */
    Py_XDECREF(*registry);
    Py_XDECREF(*module);
    return 0;
}

static PyObject *
get_category(PyObject *message, PyObject *category)
{
    int rc;

    /* Get category. */
    rc = PyObject_IsInstance(message, PyExc_Warning);
    if (rc == -1)
        return NULL;

    if (rc == 1)
        category = (PyObject*)message->ob_type;
    else if (category == NULL)
        category = PyExc_UserWarning;

    /* Validate category. */
    rc = PyObject_IsSubclass(category, PyExc_Warning);
    if (rc == -1)
        return NULL;
    if (rc == 0) {
        PyErr_SetString(PyExc_ValueError,
                        "category is not a subclass of Warning");
        return NULL;
    }

    return category;
}

static PyObject *
do_warn(PyObject *message, PyObject *category, Py_ssize_t stack_level)
{
    PyObject *filename, *module, *registry, *res;
    int lineno;

    if (!setup_context(stack_level, &filename, &lineno, &module, &registry))
        return NULL;

    res = warn_explicit(category, message, filename, lineno, module, registry,
                        NULL);
    Py_DECREF(filename);
    Py_DECREF(registry);
    Py_DECREF(module);
    return res;
}

static PyObject *
warnings_warn(PyObject *self, PyObject *args, PyObject *kwds)
{
    static char *kw_list[] = { "message", "category", "stacklevel", 0 };
    PyObject *message, *category = NULL;
    Py_ssize_t stack_level = 1;

    if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|On:warn", kw_list, 
                                     &message, &category, &stack_level))
        return NULL;

    category = get_category(message, category);
    if (category == NULL)
        return NULL;
    return do_warn(message, category, stack_level);
}

static PyObject *
warnings_warn_explicit(PyObject *self, PyObject *args, PyObject *kwds)
{
    static char *kwd_list[] = {"message", "category", "filename", "lineno",
                                "module", "registry", "module_globals", 0};
    PyObject *message;
    PyObject *category;
    PyObject *filename;
    int lineno;
    PyObject *module = NULL;
    PyObject *registry = NULL;
    PyObject *module_globals = NULL;

    if (!PyArg_ParseTupleAndKeywords(args, kwds, "OOOi|OOO:warn_explicit",
                kwd_list, &message, &category, &filename, &lineno, &module,
                &registry, &module_globals))
        return NULL;

    if (module_globals) {
        static PyObject *get_source_name = NULL;
        static PyObject *splitlines_name = NULL;
        PyObject *loader;
        PyObject *module_name;
        PyObject *source;
        PyObject *source_list;
        PyObject *source_line;
        PyObject *returned;

        if (get_source_name == NULL) {
            get_source_name = PyUnicode_InternFromString("get_source");
            if (!get_source_name)
                return NULL;
        }
        if (splitlines_name == NULL) {
            splitlines_name = PyUnicode_InternFromString("splitlines");
            if (!splitlines_name)
                return NULL;
        }

        /* Check/get the requisite pieces needed for the loader. */
        loader = PyDict_GetItemString(module_globals, "__loader__");
        module_name = PyDict_GetItemString(module_globals, "__name__");

        if (loader == NULL || module_name == NULL)
            goto standard_call;

        /* Make sure the loader implements the optional get_source() method. */
        if (!PyObject_HasAttrString(loader, "get_source"))
                goto standard_call;
        /* Call get_source() to get the source code. */
        source = PyObject_CallMethodObjArgs(loader, get_source_name,
                                                module_name, NULL);
        if (!source)
            return NULL;
        else if (source == Py_None) {
            Py_DECREF(Py_None);
            goto standard_call;
        }

        /* Split the source into lines. */
        source_list = PyObject_CallMethodObjArgs(source, splitlines_name,
                                                    NULL);
        Py_DECREF(source);
        if (!source_list)
            return NULL;

        /* Get the source line. */
        source_line = PyList_GetItem(source_list, lineno-1);
        if (!source_line) {
            Py_DECREF(source_list);
            return NULL;
        }

        /* Handle the warning. */
        returned = warn_explicit(category, message, filename, lineno, module,
                            registry, source_line);
        Py_DECREF(source_list);
        return returned;
    }

 standard_call:
    return warn_explicit(category, message, filename, lineno, module,
                                registry, NULL);
}


/* Function to issue a warning message; may raise an exception. */
int
PyErr_WarnEx(PyObject *category, const char *text, Py_ssize_t stack_level)
{
    PyObject *res;
    PyObject *message = PyUnicode_FromString(text);
    if (message == NULL)
        return -1;

    if (category == NULL)
        category = PyExc_RuntimeWarning;

    res = do_warn(message, category, stack_level);
    Py_DECREF(message);
    if (res == NULL)
        return -1;
    Py_DECREF(res);

    return 0;
}

/* PyErr_Warn is only for backwards compatability and will be removed.
   Use PyErr_WarnEx instead. */

#undef PyErr_Warn

PyAPI_FUNC(int)
PyErr_Warn(PyObject *category, char *text)
{
    return PyErr_WarnEx(category, text, 1);
}

/* Warning with explicit origin */
int
PyErr_WarnExplicit(PyObject *category, const char *text,
                   const char *filename_str, int lineno,
                   const char *module_str, PyObject *registry)
{
    PyObject *res;
    PyObject *message = PyUnicode_FromString(text);
    PyObject *filename = PyUnicode_FromString(filename_str);
    PyObject *module = NULL;
    int ret = -1;

    if (message == NULL || filename == NULL)
        goto exit;
    if (module_str != NULL) {
        module = PyUnicode_FromString(module_str);
            if (module == NULL)
                goto exit;
    }

    if (category == NULL)
        category = PyExc_RuntimeWarning;
    res = warn_explicit(category, message, filename, lineno, module, registry,
                        NULL);
    if (res == NULL)
        goto exit;
    Py_DECREF(res);
    ret = 0;

 exit:
    Py_XDECREF(message);
    Py_XDECREF(module);
    Py_XDECREF(filename);
    return ret;
}


PyDoc_STRVAR(warn_doc,
"Issue a warning, or maybe ignore it or raise an exception.");

PyDoc_STRVAR(warn_explicit_doc,
"Low-level inferface to warnings functionality.");

static PyMethodDef warnings_functions[] = {
    {"warn", (PyCFunction)warnings_warn, METH_VARARGS | METH_KEYWORDS,
        warn_doc},
    {"warn_explicit", (PyCFunction)warnings_warn_explicit,
        METH_VARARGS | METH_KEYWORDS, warn_explicit_doc},
    /* XXX(brett.cannon): add showwarning? */
    /* XXX(brett.cannon): Reasonable to add formatwarning? */
    {NULL, NULL}	        /* sentinel */
};


static PyObject *
create_filter(PyObject *category, const char *action)
{
    static PyObject *ignore_str = NULL;
    static PyObject *error_str = NULL;
    static PyObject *default_str = NULL;
    PyObject *action_obj = NULL;
    PyObject *lineno, *result;

    if (!strcmp(action, "ignore")) {
        if (ignore_str == NULL) {
            ignore_str = PyUnicode_InternFromString("ignore");
            if (ignore_str == NULL)
                return NULL;
        }
        action_obj = ignore_str;
    }
    else if (!strcmp(action, "error")) {
        if (error_str == NULL) {
            error_str = PyUnicode_InternFromString("error");
            if (error_str == NULL)
                return NULL;
        }
        action_obj = error_str;
    }
    else if (!strcmp(action, "default")) {
        if (default_str == NULL) {
            default_str = PyUnicode_InternFromString("default");
            if (default_str == NULL)
                return NULL;
        }
        action_obj = default_str;
    }
    else {
        Py_FatalError("unknown action");
    }

    /* This assumes the line number is zero for now. */
    lineno = PyLong_FromLong(0);
    if (lineno == NULL)
        return NULL;
    result = PyTuple_Pack(5, action_obj, Py_None, category, Py_None, lineno);
    Py_DECREF(lineno);
    return result;
}

static PyObject *
init_filters(void)
{
    PyObject *filters = PyList_New(3);
    const char *bytes_action;
    if (filters == NULL)
        return NULL;

    PyList_SET_ITEM(filters, 0,
                    create_filter(PyExc_PendingDeprecationWarning, "ignore"));
    PyList_SET_ITEM(filters, 1, create_filter(PyExc_ImportWarning, "ignore"));
    if (Py_BytesWarningFlag > 1)
        bytes_action = "error";
    else if (Py_BytesWarningFlag)
        bytes_action = "default";
    else
        bytes_action = "ignore";
    PyList_SET_ITEM(filters, 2, create_filter(PyExc_BytesWarning,
                    bytes_action));

    if (PyList_GET_ITEM(filters, 0) == NULL ||
        PyList_GET_ITEM(filters, 1) == NULL ||
        PyList_GET_ITEM(filters, 2) == NULL) {
        Py_DECREF(filters);
        return NULL;
    }

    return filters;
}

static struct PyModuleDef warningsmodule = {
	PyModuleDef_HEAD_INIT,
	MODULE_NAME,
	warnings__doc__,
	0,
	warnings_functions,
	NULL,
	NULL,
	NULL,
	NULL
};


PyMODINIT_FUNC
_PyWarnings_Init(void)
{
    PyObject *m;

    m = PyModule_Create(&warningsmodule);
    if (m == NULL)
        return NULL;

    _filters = init_filters();
    if (_filters == NULL)
        return NULL;
    Py_INCREF(_filters);
    if (PyModule_AddObject(m, "filters", _filters) < 0)
        return NULL;

    _once_registry = PyDict_New();
    if (_once_registry == NULL)
        return NULL;
    Py_INCREF(_once_registry);
    if (PyModule_AddObject(m, "once_registry", _once_registry) < 0)
        return NULL;

    _default_action = PyUnicode_FromString("default");
    if (_default_action == NULL)
        return NULL;
    if (PyModule_AddObject(m, "default_action", _default_action) < 0)
        return NULL;
    return m;
}
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 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387
/*
 * tclEncoding.c --
 *
 *	Contains the implementation of the encoding conversion package.
 *
 * Copyright © 1996-1998 Sun Microsystems, Inc.
 *
 * See the file "license.terms" for information on usage and redistribution of
 * this file, and for a DISCLAIMER OF ALL WARRANTIES.
 */

#include "tclInt.h"
#include "tclIO.h"

typedef size_t (LengthProc)(const char *src);

/*
 * The following data structure represents an encoding, which describes how to
 * convert between various character sets and UTF-8.
 */

typedef struct {
    char *name;			/* Name of encoding. Malloced because (1) hash
				 * table entry that owns this encoding may be
				 * freed prior to this encoding being freed,
				 * (2) string passed in the Tcl_EncodingType
				 * structure may not be persistent. */
    Tcl_EncodingConvertProc *toUtfProc;
				/* Function to convert from external encoding
				 * into UTF-8. */
    Tcl_EncodingConvertProc *fromUtfProc;
				/* Function to convert from UTF-8 into
				 * external encoding. */
    Tcl_EncodingFreeProc *freeProc;
				/* If non-NULL, function to call when this
				 * encoding is deleted. */
    void *clientData;	/* Arbitrary value associated with encoding
				 * type. Passed to conversion functions. */
    int nullSize;		/* Number of 0x00 bytes that signify
				 * end-of-string in this encoding. This number
				 * is used to determine the source string
				 * length when the srcLen argument is
				 * negative. This number can be 1, 2, or 4. */
    LengthProc *lengthProc;	/* Function to compute length of
				 * null-terminated strings in this encoding.
				 * If nullSize is 1, this is strlen; if
				 * nullSize is 2, this is a function that
				 * returns the number of bytes in a 0x0000
				 * terminated string; if nullSize is 4, this
				 * is a function that returns the number of
				 * bytes in a 0x00000000 terminated string. */
    size_t refCount;		/* Number of uses of this structure. */
    Tcl_HashEntry *hPtr;	/* Hash table entry that owns this encoding. */
} Encoding;

/*
 * The following structure is the clientData for a dynamically-loaded,
 * table-driven encoding created by LoadTableEncoding(). It maps between
 * Unicode and a single-byte, double-byte, or multibyte (1 or 2 bytes only)
 * encoding.
 */

typedef struct {
    int fallback;		/* Character (in this encoding) to substitute
				 * when this encoding cannot represent a UTF-8
				 * character. */
    char prefixBytes[256];	/* If a byte in the input stream is a lead
				 * byte for a 2-byte sequence, the
				 * corresponding entry in this array is 1,
				 * otherwise it is 0. */
    unsigned short **toUnicode;	/* Two dimensional sparse matrix to map
				 * characters from the encoding to Unicode.
				 * Each element of the toUnicode array points
				 * to an array of 256 shorts. If there is no
				 * corresponding character in Unicode, the
				 * value in the matrix is 0x0000.
				 * malloc'd. */
    unsigned short **fromUnicode;
				/* Two dimensional sparse matrix to map
				 * characters from Unicode to the encoding.
				 * Each element of the fromUnicode array
				 * points to an array of 256 shorts. If there
				 * is no corresponding character the encoding,
				 * the value in the matrix is 0x0000.
				 * malloc'd. */
} TableEncodingData;

/*
 * Each of the following structures is the clientData for a dynamically-loaded
 * escape-driven encoding that is itself comprised of other simpler encodings.
 * An example is "iso-2022-jp", which uses escape sequences to switch between
 * ascii, jis0208, jis0212, gb2312, and ksc5601. Note that "escape-driven"
 * does not necessarily mean that the ESCAPE character is the character used
 * for switching character sets.
 */

typedef struct {
    unsigned sequenceLen;	/* Length of following string. */
    char sequence[16];		/* Escape code that marks this encoding. */
    char name[32];		/* Name for encoding. */
    Encoding *encodingPtr;	/* Encoding loaded using above name, or NULL
				 * if this sub-encoding has not been needed
				 * yet. */
} EscapeSubTable;

typedef struct {
    int fallback;		/* Character (in this encoding) to substitute
				 * when this encoding cannot represent a UTF-8
				 * character. */
    unsigned initLen;		/* Length of following string. */
    char init[16];		/* String to emit or expect before first char
				 * in conversion. */
    unsigned finalLen;		/* Length of following string. */
    char final[16];		/* String to emit or expect after last char in
				 * conversion. */
    char prefixBytes[256];	/* If a byte in the input stream is the first
				 * character of one of the escape sequences in
				 * the following array, the corresponding
				 * entry in this array is 1, otherwise it is
				 * 0. */
    int numSubTables;		/* Length of following array. */
    EscapeSubTable subTables[TCLFLEXARRAY];/* Information about each EscapeSubTable used
				 * by this encoding type. The actual size is
				 * as large as necessary to hold all
				 * EscapeSubTables. */
} EscapeEncodingData;

/*
 * Constants used when loading an encoding file to identify the type of the
 * file.
 */

#define ENCODING_SINGLEBYTE	0
#define ENCODING_DOUBLEBYTE	1
#define ENCODING_MULTIBYTE	2
#define ENCODING_ESCAPE		3

/*
 * A list of directories in which Tcl should look for *.enc files. This list
 * is shared by all threads. Access is governed by a mutex lock.
 */

static TclInitProcessGlobalValueProc InitializeEncodingSearchPath;
static ProcessGlobalValue encodingSearchPath = {
    0, 0, NULL, NULL, InitializeEncodingSearchPath, NULL, NULL
};

/*
 * A map from encoding names to the directories in which their data files have
 * been seen. The string value of the map is shared by all threads. Access to
 * the shared string is governed by a mutex lock.
 */

static ProcessGlobalValue encodingFileMap = {
    0, 0, NULL, NULL, NULL, NULL, NULL
};

/*
 * A list of directories making up the "library path". Historically this
 * search path has served many uses, but the only one remaining is a base for
 * the encodingSearchPath above. If the application does not explicitly set
 * the encodingSearchPath, then it is initialized by appending /encoding
 * to each directory in this "libraryPath".
 */

static ProcessGlobalValue libraryPath = {
    0, 0, NULL, NULL, TclpInitLibraryPath, NULL, NULL
};

static int encodingsInitialized = 0;

/*
 * Hash table that keeps track of all loaded Encodings. Keys are the string
 * names that represent the encoding, values are (Encoding *).
 */

static Tcl_HashTable encodingTable;
TCL_DECLARE_MUTEX(encodingMutex)

/*
 * The following are used to hold the default and current system encodings.
 * If NULL is passed to one of the conversion routines, the current setting of
 * the system encoding is used to perform the conversion.
 */

static Tcl_Encoding defaultEncoding = NULL;
static Tcl_Encoding systemEncoding = NULL;
Tcl_Encoding tclIdentityEncoding = NULL;

/*
 * Names of encoding profiles and corresponding integer values
 */
static struct TclEncodingProfiles {
    const char *name;
    int value;
} encodingProfiles[] = {
    {"tcl8", TCL_ENCODING_PROFILE_TCL8},
    {"strict", TCL_ENCODING_PROFILE_STRICT},
    {"replace", TCL_ENCODING_PROFILE_REPLACE},
};
#define PROFILE_STRICT(flags_)                                         \
    ((TCL_ENCODING_PROFILE_GET(flags_) == TCL_ENCODING_PROFILE_STRICT) \
     || (TCL_ENCODING_PROFILE_GET(flags_) == 0                         \
	 && TCL_ENCODING_PROFILE_DEFAULT == TCL_ENCODING_PROFILE_STRICT))

#define PROFILE_REPLACE(flags_)                                         \
    ((TCL_ENCODING_PROFILE_GET(flags_) == TCL_ENCODING_PROFILE_REPLACE) \
     || (TCL_ENCODING_PROFILE_GET(flags_) == 0                          \
	 && TCL_ENCODING_PROFILE_DEFAULT == TCL_ENCODING_PROFILE_REPLACE))

#define UNICODE_REPLACE_CHAR ((Tcl_UniChar)0xFFFD)
#define SURROGATE(c_)      (((c_) & ~0x7FF) == 0xD800)
#define HIGH_SURROGATE(c_) (((c_) & ~0x3FF) == 0xD800)
#define LOW_SURROGATE(c_)  (((c_) & ~0x3FF) == 0xDC00)

/*
 * The following variable is used in the sparse matrix code for a
 * TableEncoding to represent a page in the table that has no entries.
 */

static unsigned short emptyPage[256];

/*
 * Functions used only in this module.
 */

static Tcl_EncodingConvertProc	BinaryProc;
static Tcl_DupInternalRepProc	DupEncodingInternalRep;
static Tcl_EncodingFreeProc	EscapeFreeProc;
static Tcl_EncodingConvertProc	EscapeFromUtfProc;
static Tcl_EncodingConvertProc	EscapeToUtfProc;
static void			FillEncodingFileMap(void);
static void			FreeEncoding(Tcl_Encoding encoding);
static Tcl_FreeInternalRepProc	FreeEncodingInternalRep;
static Encoding *		GetTableEncoding(EscapeEncodingData *dataPtr,
				    int state);
static Tcl_Encoding		LoadEncodingFile(Tcl_Interp *interp,
				    const char *name);
static Tcl_Encoding		LoadTableEncoding(const char *name, int type,
				    Tcl_Channel chan);
static Tcl_Encoding		LoadEscapeEncoding(const char *name,
				    Tcl_Channel chan);
static Tcl_Channel		OpenEncodingFileChannel(Tcl_Interp *interp,
				    const char *name);
static Tcl_EncodingFreeProc	TableFreeProc;
static Tcl_EncodingConvertProc	TableFromUtfProc;
static Tcl_EncodingConvertProc	TableToUtfProc;
static size_t		unilen(const char *src);
static size_t		unilen4(const char *src);
static Tcl_EncodingConvertProc	Utf32ToUtfProc;
static Tcl_EncodingConvertProc	UtfToUtf32Proc;
static Tcl_EncodingConvertProc	Utf16ToUtfProc;
static Tcl_EncodingConvertProc	UtfToUtf16Proc;
static Tcl_EncodingConvertProc	UtfToUcs2Proc;
static Tcl_EncodingConvertProc	UtfToUtfProc;
static Tcl_EncodingConvertProc	Iso88591FromUtfProc;
static Tcl_EncodingConvertProc	Iso88591ToUtfProc;


/*
 * A Tcl_ObjType for holding a cached Tcl_Encoding in the twoPtrValue.ptr1 field
 * of the internalrep. This should help the lifetime of encodings be more useful.
 * See concerns raised in [Bug 1077262].
 */

static const Tcl_ObjType encodingType = {
    "encoding", FreeEncodingInternalRep, DupEncodingInternalRep, NULL, NULL
};
#define EncodingSetInternalRep(objPtr, encoding)				\
    do {								\
	Tcl_ObjInternalRep ir;						\
	ir.twoPtrValue.ptr1 = (encoding);				\
	ir.twoPtrValue.ptr2 = NULL;					\
	Tcl_StoreInternalRep((objPtr), &encodingType, &ir);			\
    } while (0)

#define EncodingGetInternalRep(objPtr, encoding)				\
    do {								\
	const Tcl_ObjInternalRep *irPtr;					\
	irPtr = TclFetchInternalRep ((objPtr), &encodingType);		\
	(encoding) = irPtr ? (Tcl_Encoding)irPtr->twoPtrValue.ptr1 : NULL;		\
    } while (0)


/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetEncodingFromObj --
 *
 *	Writes to (*encodingPtr) the Tcl_Encoding value of (*objPtr), if
 *	possible, and returns TCL_OK. If no such encoding exists, TCL_ERROR is
 *	returned, and if interp is non-NULL, an error message is written
 *	there.
 *
 * Results:
 *	Standard Tcl return code.
 *
 * Side effects:
 *	Caches the Tcl_Encoding value as the internal rep of (*objPtr).
 *
 *----------------------------------------------------------------------
 */

int
Tcl_GetEncodingFromObj(
    Tcl_Interp *interp,
    Tcl_Obj *objPtr,
    Tcl_Encoding *encodingPtr)
{
    Tcl_Encoding encoding;
    const char *name = TclGetString(objPtr);

    EncodingGetInternalRep(objPtr, encoding);
    if (encoding == NULL) {
	encoding = Tcl_GetEncoding(interp, name);
	if (encoding == NULL) {
	    return TCL_ERROR;
	}
	EncodingSetInternalRep(objPtr, encoding);
    }
    *encodingPtr = Tcl_GetEncoding(NULL, name);
    return TCL_OK;
}

/*
 *----------------------------------------------------------------------
 *
 * FreeEncodingInternalRep --
 *
 *	The Tcl_FreeInternalRepProc for the "encoding" Tcl_ObjType.
 *
 *----------------------------------------------------------------------
 */

static void
FreeEncodingInternalRep(
    Tcl_Obj *objPtr)
{
    Tcl_Encoding encoding;

    EncodingGetInternalRep(objPtr, encoding);
    Tcl_FreeEncoding(encoding);
}

/*
 *----------------------------------------------------------------------
 *
 * DupEncodingInternalRep --
 *
 *	The Tcl_DupInternalRepProc for the "encoding" Tcl_ObjType.
 *
 *----------------------------------------------------------------------
 */

static void
DupEncodingInternalRep(
    Tcl_Obj *srcPtr,
    Tcl_Obj *dupPtr)
{
    Tcl_Encoding encoding = Tcl_GetEncoding(NULL, TclGetString(srcPtr));
    EncodingSetInternalRep(dupPtr, encoding);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_GetEncodingSearchPath --
 *
 *	Keeps the per-thread copy of the encoding search path current with
 *	changes to the global copy.
 *
 * Results:
 *	Returns a "list" (Tcl_Obj *) that contains the encoding search path.
 *
 *----------------------------------------------------------------------
 */

Tcl_Obj *
Tcl_GetEncodingSearchPath(void)
{
    return TclGetProcessGlobalValue(&encodingSearchPath);
}

/*
 *----------------------------------------------------------------------
 *
 * Tcl_SetEncodingSearchPath --
 *
 *	Keeps the per-thread copy of the encoding search path current with
 *	changes to the global copy.
 *
 *----------------------------------------------------------------------
 */

int
Tcl_SetEncodingSearchPath(
    Tcl_Obj *searchPath)
{
    int dummy;

    if (TCL_ERROR == TclListObjLengthM(NULL, searchPath, &dummy)) {
	return TCL_ERROR;
    }
    TclSetProcessGlobalValue(&encodingSearchPath, searchPath, NULL);
    return TCL_OK;
}

/*
 *----------------------------------------------------------------------
 *
 * TclGetLibraryPath --
 *
 *	Keeps the per-thread copy of the library path current with changes to
 *	the global copy.
 *
 * Results:
 *	Returns a "list" (Tcl_Obj *) that contains the library path.
 *
 *----------------------------------------------------------------------
 */

Tcl_Obj *
TclGetLibraryPath(void)
{
    return TclGetProcessGlobalValue(&libraryPath);
}

/*
 *----------------------------------------------------------------------
 *
 * TclSetLibraryPath --
 *
 *	Keeps the per-thread copy of the library path current with changes to
 *	the global copy.
 *
 *	Since the result of this routine is void, if searchPath is not a valid
 *	list this routine silently does nothing.
 *
 *----------------------------------------------------------------------
 */

void
TclSetLibraryPath(
    Tcl_Obj *path)
{
    int dummy;

    if (TCL_ERROR == TclListObjLengthM(NULL, path, &dummy)) {
	return;
    }
    TclSetProcessGlobalValue(&libraryPath, path, NULL);
}

/*
 *---------------------------------------------------------------------------
 *
 * FillEncodingFileMap --
 *
 *	Called to update the encoding file map with the current value
 *	of the encoding search path.
 *
 *	Finds *.end files in the directories on the encoding search path and
 *	stores the found pathnames in a map associated with the encoding name.
 *
 *	If $dir is on the encoding search path and the file $dir/foo.enc is
 *	found, stores a "foo" -> $dir entry in the map.  if the "foo" encoding
 *	is needed later, the $dir/foo.enc name can be quickly constructed in
 *	order to read the encoding data.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Entries are added to the encoding file map.
 *
 *---------------------------------------------------------------------------
 */

static void
FillEncodingFileMap(void)
{
    int i, numDirs = 0;
    Tcl_Obj *map, *searchPath;

    searchPath = Tcl_GetEncodingSearchPath();
    Tcl_IncrRefCount(searchPath);
    TclListObjLengthM(NULL, searchPath, &numDirs);
    map = Tcl_NewDictObj();
    Tcl_IncrRefCount(map);

    for (i = numDirs-1; i >= 0; i--) {
	/*
	 * Iterate backwards through the search path so as we overwrite
	 * entries found, we favor files earlier on the search path.
	 */

	int j, numFiles;
	Tcl_Obj *directory, *matchFileList;
	Tcl_Obj **filev;
	Tcl_GlobTypeData readableFiles = {
	    TCL_GLOB_TYPE_FILE, TCL_GLOB_PERM_R, NULL, NULL
	};

	TclNewObj(matchFileList);
	Tcl_ListObjIndex(NULL, searchPath, i, &directory);
	Tcl_IncrRefCount(directory);
	Tcl_IncrRefCount(matchFileList);
	Tcl_FSMatchInDirectory(NULL, matchFileList, directory, "*.enc",
		&readableFiles);

	TclListObjGetElementsM(NULL, matchFileList, &numFiles, &filev);
	for (j=0; j<numFiles; j++) {
	    Tcl_Obj *encodingName, *fileObj;

	    fileObj = TclPathPart(NULL, filev[j], TCL_PATH_TAIL);
	    encodingName = TclPathPart(NULL, fileObj, TCL_PATH_ROOT);
	    Tcl_DictObjPut(NULL, map, encodingName, directory);
	    Tcl_DecrRefCount(fileObj);
	    Tcl_DecrRefCount(encodingName);
	}
	Tcl_DecrRefCount(matchFileList);
	Tcl_DecrRefCount(directory);
    }
    Tcl_DecrRefCount(searchPath);
    TclSetProcessGlobalValue(&encodingFileMap, map, NULL);
    Tcl_DecrRefCount(map);
}

/*
 *---------------------------------------------------------------------------
 *
 * TclInitEncodingSubsystem --
 *
 *	Initialize all resources used by this subsystem on a per-process
 *	basis.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Depends on the memory, object, and IO subsystems.
 *
 *---------------------------------------------------------------------------
 */

/* Since TCL_ENCODING_MODIFIED is only used for utf-8/cesu-8 and
 * TCL_ENCODING_LE is only used for  utf-16/utf-32/ucs-2. re-use the same value */
#define TCL_ENCODING_LE		TCL_ENCODING_MODIFIED	/* Little-endian encoding */
#define ENCODING_UTF	0x200	/* For UTF-8 encoding, allow 4-byte output sequences */
#define ENCODING_INPUT	0x400 /* For UTF-8/CESU-8 encoding, means external -> internal */

void
TclInitEncodingSubsystem(void)
{
    Tcl_EncodingType type;
    TableEncodingData *dataPtr;
    unsigned size;
    unsigned short i;
    union {
        char c;
        short s;
    } isLe;

    if (encodingsInitialized) {
	return;
    }

    isLe.s = TCL_ENCODING_LE;
    Tcl_MutexLock(&encodingMutex);
    Tcl_InitHashTable(&encodingTable, TCL_STRING_KEYS);
    Tcl_MutexUnlock(&encodingMutex);

    /*
     * Create a few initial encodings.  UTF-8 to UTF-8 translation is not a
     * no-op because it turns a stream of improperly formed UTF-8 into a
     * properly formed stream.
     */

    type.encodingName	= NULL;
    type.toUtfProc	= BinaryProc;
    type.fromUtfProc	= BinaryProc;
    type.freeProc	= NULL;
    type.nullSize	= 1;
    type.clientData	= NULL;
    tclIdentityEncoding = Tcl_CreateEncoding(&type);

    type.encodingName	= "utf-8";
    type.toUtfProc	= UtfToUtfProc;
    type.fromUtfProc	= UtfToUtfProc;
    type.freeProc	= NULL;
    type.nullSize	= 1;
    type.clientData	= INT2PTR(ENCODING_UTF);
    Tcl_CreateEncoding(&type);
    type.clientData	= INT2PTR(0);
    type.encodingName	= "cesu-8";
    Tcl_CreateEncoding(&type);

    type.toUtfProc	= Utf16ToUtfProc;
    type.fromUtfProc    = UtfToUcs2Proc;
    type.freeProc	= NULL;
    type.nullSize	= 2;
    type.encodingName   = "ucs-2le";
    type.clientData	= INT2PTR(TCL_ENCODING_LE);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "ucs-2be";
    type.clientData	= INT2PTR(0);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "ucs-2";
    type.clientData	= INT2PTR(isLe.c);
    Tcl_CreateEncoding(&type);

    type.toUtfProc	= Utf32ToUtfProc;
    type.fromUtfProc    = UtfToUtf32Proc;
    type.freeProc	= NULL;
    type.nullSize	= 4;
    type.encodingName   = "utf-32le";
    type.clientData	= INT2PTR(TCL_ENCODING_LE);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "utf-32be";
    type.clientData	= INT2PTR(0);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "utf-32";
    type.clientData	= INT2PTR(isLe.c);
    Tcl_CreateEncoding(&type);

    type.toUtfProc	= Utf16ToUtfProc;
    type.fromUtfProc    = UtfToUtf16Proc;
    type.freeProc	= NULL;
    type.nullSize	= 2;
    type.encodingName   = "utf-16le";
    type.clientData	= INT2PTR(TCL_ENCODING_LE);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "utf-16be";
    type.clientData	= INT2PTR(0);
    Tcl_CreateEncoding(&type);
    type.encodingName   = "utf-16";
    type.clientData	= INT2PTR(isLe.c);
    Tcl_CreateEncoding(&type);

#ifndef TCL_NO_DEPRECATED
    type.encodingName   = "unicode";
    Tcl_CreateEncoding(&type);
#endif

    /*
     * Need the iso8859-1 encoding in order to process binary data, so force
     * it to always be embedded. Note that this encoding *must* be a proper
     * table encoding or some of the escape encodings crash! Hence the ugly
     * code to duplicate the structure of a table encoding here.
     */

    dataPtr = (TableEncodingData *)ckalloc(sizeof(TableEncodingData));
    memset(dataPtr, 0, sizeof(TableEncodingData));
    dataPtr->fallback = '?';

    size = 256*(sizeof(unsigned short *) + sizeof(unsigned short));
    dataPtr->toUnicode = (unsigned short **)ckalloc(size);
    memset(dataPtr->toUnicode, 0, size);
    dataPtr->fromUnicode = (unsigned short **)ckalloc(size);
    memset(dataPtr->fromUnicode, 0, size);

    dataPtr->toUnicode[0] = (unsigned short *) (dataPtr->toUnicode + 256);
    dataPtr->fromUnicode[0] = (unsigned short *) (dataPtr->fromUnicode + 256);
    for (i=1 ; i<256 ; i++) {
	dataPtr->toUnicode[i] = emptyPage;
	dataPtr->fromUnicode[i] = emptyPage;
    }

    for (i=0 ; i<256 ; i++) {
	dataPtr->toUnicode[0][i] = i;
	dataPtr->fromUnicode[0][i] = i;
    }

    type.encodingName	= "iso8859-1";
    type.toUtfProc	= Iso88591ToUtfProc;
    type.fromUtfProc	= Iso88591FromUtfProc;
    type.freeProc	= TableFreeProc;
    type.nullSize	= 1;
    type.clientData	= dataPtr;
    defaultEncoding	= Tcl_CreateEncoding(&type);
    systemEncoding	= Tcl_GetEncoding(NULL, type.encodingName);

    encodingsInitialized = 1;
}

/*
 *----------------------------------------------------------------------
 *
 * TclFinalizeEncodingSubsystem --
 *
 *	Release the state associated with the encoding subsystem.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Frees all of the encodings.
 *
 *----------------------------------------------------------------------
 */

void
TclFinalizeEncodingSubsystem(void)
{
    Tcl_HashSearch search;
    Tcl_HashEntry *hPtr;

    Tcl_MutexLock(&encodingMutex);
    encodingsInitialized = 0;
    FreeEncoding(systemEncoding);
    systemEncoding = NULL;
    defaultEncoding = NULL;
    FreeEncoding(tclIdentityEncoding);
    tclIdentityEncoding = NULL;

    hPtr = Tcl_FirstHashEntry(&encodingTable, &search);
    while (hPtr != NULL) {
	/*
	 * Call FreeEncoding instead of doing it directly to handle refcounts
	 * like escape encodings use. [Bug 524674] Make sure to call
	 * Tcl_FirstHashEntry repeatedly so that all encodings are eventually
	 * cleaned up.
	 */

	FreeEncoding((Tcl_Encoding)Tcl_GetHashValue(hPtr));
	hPtr = Tcl_FirstHashEntry(&encodingTable, &search);
    }

    Tcl_DeleteHashTable(&encodingTable);
    Tcl_MutexUnlock(&encodingMutex);
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetDefaultEncodingDir --
 *
 *	Legacy public interface to retrieve first directory in the encoding
 *	searchPath.
 *
 * Results:
 *	The directory pathname, as a string, or NULL for an empty encoding
 *	search path.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

#if !defined(TCL_NO_DEPRECATED) && TCL_MAJOR_VERSION < 9
const char *
Tcl_GetDefaultEncodingDir(void)
{
    int numDirs;
    Tcl_Obj *first, *searchPath = Tcl_GetEncodingSearchPath();

    TclListObjLengthM(NULL, searchPath, &numDirs);
    if (numDirs == 0) {
	return NULL;
    }
    Tcl_ListObjIndex(NULL, searchPath, 0, &first);

    return TclGetString(first);
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_SetDefaultEncodingDir --
 *
 *	Legacy public interface to set the first directory in the encoding
 *	search path.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Modifies the encoding search path.
 *
 *-------------------------------------------------------------------------
 */

void
Tcl_SetDefaultEncodingDir(
    const char *path)
{
    Tcl_Obj *searchPath = Tcl_GetEncodingSearchPath();
    Tcl_Obj *directory = Tcl_NewStringObj(path, -1);

    searchPath = Tcl_DuplicateObj(searchPath);
    Tcl_ListObjReplace(NULL, searchPath, 0, 0, 1, &directory);
    Tcl_SetEncodingSearchPath(searchPath);
}
#endif

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetEncoding --
 *
 *	Given the name of a encoding, find the corresponding Tcl_Encoding
 *	token. If the encoding did not already exist, Tcl attempts to
 *	dynamically load an encoding by that name.
 *
 * Results:
 *	Returns a token that represents the encoding. If the name didn't refer
 *	to any known or loadable encoding, NULL is returned. If NULL was
 *	returned, an error message is left in interp's result object, unless
 *	interp was NULL.
 *
 * Side effects:
 *	LoadEncodingFile is called if necessary.
 *
 *-------------------------------------------------------------------------
 */

Tcl_Encoding
Tcl_GetEncoding(
    Tcl_Interp *interp,		/* Interp for error reporting, if not NULL. */
    const char *name)		/* The name of the desired encoding. */
{
    Tcl_HashEntry *hPtr;
    Encoding *encodingPtr;

    Tcl_MutexLock(&encodingMutex);
    if (name == NULL) {
	encodingPtr = (Encoding *) systemEncoding;
	encodingPtr->refCount++;
	Tcl_MutexUnlock(&encodingMutex);
	return systemEncoding;
    }

    hPtr = Tcl_FindHashEntry(&encodingTable, name);
    if (hPtr != NULL) {
	encodingPtr = (Encoding *)Tcl_GetHashValue(hPtr);
	encodingPtr->refCount++;
	Tcl_MutexUnlock(&encodingMutex);
	return (Tcl_Encoding) encodingPtr;
    }
    Tcl_MutexUnlock(&encodingMutex);

    return LoadEncodingFile(interp, name);
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_FreeEncoding --
 *
 *	Releases an encoding allocated by Tcl_CreateEncoding() or
 *	Tcl_GetEncoding().
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The reference count associated with the encoding is decremented and
 *	the encoding is deleted if nothing is using it anymore.
 *
 *---------------------------------------------------------------------------
 */

void
Tcl_FreeEncoding(
    Tcl_Encoding encoding)
{
    Tcl_MutexLock(&encodingMutex);
    FreeEncoding(encoding);
    Tcl_MutexUnlock(&encodingMutex);
}

/*
 *----------------------------------------------------------------------
 *
 * FreeEncoding --
 *
 *	Decrements the reference count of an encoding.  The caller must hold
 *	encodingMutes.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Releases the resource for an encoding if it is now unused.
 *	The reference count associated with the encoding is decremented and
 *	the encoding may be deleted if nothing is using it anymore.
 *
 *----------------------------------------------------------------------
 */

static void
FreeEncoding(
    Tcl_Encoding encoding)
{
    Encoding *encodingPtr = (Encoding *) encoding;

    if (encodingPtr == NULL) {
	return;
    }
    if (encodingPtr->refCount-- <= 1) {
	if (encodingPtr->freeProc != NULL) {
	    encodingPtr->freeProc(encodingPtr->clientData);
	}
	if (encodingPtr->hPtr != NULL) {
	    Tcl_DeleteHashEntry(encodingPtr->hPtr);
	}
	if (encodingPtr->name) {
	    ckfree(encodingPtr->name);
	}
	ckfree(encodingPtr);
    }
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetEncodingName --
 *
 *	Given an encoding, return the name that was used to construct the
 *	encoding.
 *
 * Results:
 *	The name of the encoding.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

const char *
Tcl_GetEncodingName(
    Tcl_Encoding encoding)	/* The encoding whose name to fetch. */
{
    if (encoding == NULL) {
	encoding = systemEncoding;
    }

    return ((Encoding *) encoding)->name;
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetEncodingNames --
 *
 *	Get the list of all known encodings, including the ones stored as
 *	files on disk in the encoding path.
 *
 * Results:
 *	Modifies interp's result object to hold a list of all the available
 *	encodings.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

void
Tcl_GetEncodingNames(
    Tcl_Interp *interp)		/* Interp to hold result. */
{
    Tcl_HashTable table;
    Tcl_HashSearch search;
    Tcl_HashEntry *hPtr;
    Tcl_Obj *map, *name, *result;
    Tcl_DictSearch mapSearch;
    int dummy, done = 0;

    TclNewObj(result);
    Tcl_InitObjHashTable(&table);

    /*
     * Copy encoding names from loaded encoding table to table.
     */

    Tcl_MutexLock(&encodingMutex);
    for (hPtr = Tcl_FirstHashEntry(&encodingTable, &search); hPtr != NULL;
	    hPtr = Tcl_NextHashEntry(&search)) {
	Encoding *encodingPtr = (Encoding *)Tcl_GetHashValue(hPtr);

	Tcl_CreateHashEntry(&table,
		Tcl_NewStringObj(encodingPtr->name, -1), &dummy);
    }
    Tcl_MutexUnlock(&encodingMutex);

    FillEncodingFileMap();
    map = TclGetProcessGlobalValue(&encodingFileMap);

    /*
     * Copy encoding names from encoding file map to table.
     */

    Tcl_DictObjFirst(NULL, map, &mapSearch, &name, NULL, &done);
    for (; !done; Tcl_DictObjNext(&mapSearch, &name, NULL, &done)) {
	Tcl_CreateHashEntry(&table, name, &dummy);
    }

    /*
     * Pull all encoding names from table into the result list.
     */

    for (hPtr = Tcl_FirstHashEntry(&table, &search); hPtr != NULL;
	    hPtr = Tcl_NextHashEntry(&search)) {
	Tcl_ListObjAppendElement(NULL, result,
		(Tcl_Obj *) Tcl_GetHashKey(&table, hPtr));
    }
    Tcl_SetObjResult(interp, result);
    Tcl_DeleteHashTable(&table);
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetEncodingNulLength --
 *
 *	Given an encoding, return the number of nul bytes used for the
 *      string termination.
 *
 * Results:
 *	The number of nul bytes used for the string termination.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */
int
Tcl_GetEncodingNulLength(
    Tcl_Encoding encoding)
{
    if (encoding == NULL) {
	encoding = systemEncoding;
    }

    return ((Encoding *) encoding)->nullSize;
}

/*
 *------------------------------------------------------------------------
 *
 * Tcl_SetSystemEncoding --
 *
 *	Sets the default encoding that should be used whenever the user passes
 *	a NULL value in to one of the conversion routines. If the supplied
 *	name is NULL, the system encoding is reset to the default system
 *	encoding.
 *
 * Results:
 *	The return value is TCL_OK if the system encoding was successfully set
 *	to the encoding specified by name, TCL_ERROR otherwise. If TCL_ERROR
 *	is returned, an error message is left in interp's result object,
 *	unless interp was NULL.
 *
 * Side effects:
 *	The reference count of the new system encoding is incremented. The
 *	reference count of the old system encoding is decremented and it may
 *	be freed. All VFS cached information is invalidated.
 *
 *------------------------------------------------------------------------
 */

int
Tcl_SetSystemEncoding(
    Tcl_Interp *interp,		/* Interp for error reporting, if not NULL. */
    const char *name)		/* The name of the desired encoding, or NULL/""
				 * to reset to default encoding. */
{
    Tcl_Encoding encoding;
    Encoding *encodingPtr;

    if (!name || !*name) {
	Tcl_MutexLock(&encodingMutex);
	encoding = defaultEncoding;
	encodingPtr = (Encoding *) encoding;
	encodingPtr->refCount++;
	Tcl_MutexUnlock(&encodingMutex);
    } else {
	encoding = Tcl_GetEncoding(interp, name);
	if (encoding == NULL) {
	    return TCL_ERROR;
	}
    }

    Tcl_MutexLock(&encodingMutex);
    FreeEncoding(systemEncoding);
    systemEncoding = encoding;
    Tcl_MutexUnlock(&encodingMutex);
    Tcl_FSMountsChanged(NULL);

    return TCL_OK;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_CreateEncoding --
 *
 *	Defines a new encoding, along with the functions that are used to
 *	convert to and from Unicode.
 *
 * Results:
 *	Returns a token that represents the encoding. If an encoding with the
 *	same name already existed, the old encoding token remains valid and
 *	continues to behave as it used to, and is eventually garbage collected
 *	when the last reference to it goes away. Any subsequent calls to
 *	Tcl_GetEncoding with the specified name retrieve the most recent
 *	encoding token.
 *
 * Side effects:
 *	A new record having the name of the encoding is entered into a table of
 *	encodings visible to all interpreters.  For each call to this function,
 *	there should eventually be a call to Tcl_FreeEncoding, which cleans
 *	deletes the record in the table when an encoding is no longer needed.
 *
 *---------------------------------------------------------------------------
 */

Tcl_Encoding
Tcl_CreateEncoding(
    const Tcl_EncodingType *typePtr)
				/* The encoding type. */
{
    Encoding *encodingPtr = (Encoding *)ckalloc(sizeof(Encoding));
    encodingPtr->name		= NULL;
    encodingPtr->toUtfProc	= typePtr->toUtfProc;
    encodingPtr->fromUtfProc	= typePtr->fromUtfProc;
    encodingPtr->freeProc	= typePtr->freeProc;
    encodingPtr->nullSize	= typePtr->nullSize;
    encodingPtr->clientData	= typePtr->clientData;
    if (typePtr->nullSize == 2) {
	encodingPtr->lengthProc = (LengthProc *) unilen;
    } else if (typePtr->nullSize == 4) {
	encodingPtr->lengthProc = (LengthProc *) unilen4;
    } else {
	encodingPtr->lengthProc = (LengthProc *) strlen;
    }
    encodingPtr->refCount	= 1;
    encodingPtr->hPtr		= NULL;

  if (typePtr->encodingName) {
    Tcl_HashEntry *hPtr;
    int isNew;
    char *name;

    Tcl_MutexLock(&encodingMutex);
    hPtr = Tcl_CreateHashEntry(&encodingTable, typePtr->encodingName, &isNew);
    if (isNew == 0) {
	/*
	 * Remove old encoding from hash table, but don't delete it until last
	 * reference goes away.
	 */

	Encoding *replaceMe = (Encoding *)Tcl_GetHashValue(hPtr);
	replaceMe->hPtr = NULL;
    }

    name = (char *)ckalloc(strlen(typePtr->encodingName) + 1);
    encodingPtr->name		= strcpy(name, typePtr->encodingName);
    encodingPtr->hPtr		= hPtr;
    Tcl_SetHashValue(hPtr, encodingPtr);

    Tcl_MutexUnlock(&encodingMutex);
  }
    return (Tcl_Encoding) encodingPtr;
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_ExternalToUtfDString --
 *
 *	Convert a source buffer from the specified encoding into UTF-8. If any
 *	of the bytes in the source buffer are invalid or cannot be represented
 *	in the target encoding, a default fallback character will be
 *	substituted.
 *
 * Results:
 *	The converted bytes are stored in the DString, which is then NULL
 *	terminated. The return value is a pointer to the value stored in the
 *	DString.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

char *
Tcl_ExternalToUtfDString(
    Tcl_Encoding encoding,	/* The encoding for the source string, or NULL
				 * for the default system encoding. */
    const char *src,		/* Source string in specified encoding. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * encoding-specific string length. */
    Tcl_DString *dstPtr)	/* Uninitialized or free DString in which the
				 * converted string is stored. */
{
    Tcl_ExternalToUtfDStringEx(encoding, src, srcLen, TCL_ENCODING_PROFILE_TCL8, dstPtr);
    return Tcl_DStringValue(dstPtr);
}


/*
 *-------------------------------------------------------------------------
 *
 * Tcl_ExternalToUtfDStringEx --
 *
 *	Convert a source buffer from the specified encoding into UTF-8.
*	The parameter flags controls the behavior, if any of the bytes in
 *	the source buffer are invalid or cannot be represented in utf-8.
 *	Possible flags values:
 *	target encoding. It should be composed by OR-ing the following:
 *	- *At most one* of TCL_ENCODING_PROFILE{DEFAULT,TCL8,STRICT}
 *	- TCL_ENCODING_STOPONERROR: Backward compatibility. Sets the profile
 *	  to TCL_ENCODING_PROFILE_STRICT overriding any specified profile flags
 *	- TCL_ENCODING_MODIFIED: enable Tcl internal conversion mapping \xC0\x80
 *        to 0x00. Only valid for "utf-8" and "cesu-8".
 *
 * Results:
 *	The converted bytes are stored in the DString, which is then NULL
 *	terminated in an encoding-specific manner. The return value is
 *	the error position in the source string or -1 if no conversion error
 *	is reported.
  *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

int
Tcl_ExternalToUtfDStringEx(
    Tcl_Encoding encoding,	/* The encoding for the source string, or NULL
				 * for the default system encoding. */
    const char *src,		/* Source string in specified encoding. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * encoding-specific string length. */
    int flags,			/* Conversion control flags. */
    Tcl_DString *dstPtr)	/* Uninitialized or free DString in which the
				 * converted string is stored. */
{
    char *dst;
    Tcl_EncodingState state;
    const Encoding *encodingPtr;
    int dstLen, result, soFar, srcRead, dstWrote, dstChars;
    const char *srcStart = src;

    Tcl_DStringInit(dstPtr);
    dst = Tcl_DStringValue(dstPtr);
    dstLen = dstPtr->spaceAvl - 1;

    if (encoding == NULL) {
	encoding = systemEncoding;
    }
    encodingPtr = (Encoding *) encoding;

    if (src == NULL) {
	srcLen = 0;
    } else if (srcLen < 0) {
	srcLen = encodingPtr->lengthProc(src);
    }

    flags = TclEncodingExternalFlagsToInternal(flags);
    flags |= TCL_ENCODING_START | TCL_ENCODING_END;
    if (encodingPtr->toUtfProc == UtfToUtfProc) {
	flags |= ENCODING_INPUT;
    }

    while (1) {
	result = encodingPtr->toUtfProc(encodingPtr->clientData, src, srcLen,
		flags, &state, dst, dstLen, &srcRead, &dstWrote, &dstChars);
	soFar = dst + dstWrote - Tcl_DStringValue(dstPtr);

	src += srcRead;
	if (result != TCL_CONVERT_NOSPACE) {
	    Tcl_DStringSetLength(dstPtr, soFar);
	    return (result == TCL_OK) ? TCL_INDEX_NONE : (int)(src - srcStart);
	}
	flags &= ~TCL_ENCODING_START;
	srcLen -= srcRead;
	if (Tcl_DStringLength(dstPtr) == 0) {
	    Tcl_DStringSetLength(dstPtr, dstLen);
	}
	Tcl_DStringSetLength(dstPtr, 2 * Tcl_DStringLength(dstPtr) + 1);
	dst = Tcl_DStringValue(dstPtr) + soFar;
	dstLen = Tcl_DStringLength(dstPtr) - soFar - 1;
    }
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_ExternalToUtf --
 *
 *	Convert a source buffer from the specified encoding into UTF-8.
 *
 * Results:
 *	The return value is one of TCL_OK, TCL_CONVERT_MULTIBYTE,
 *	TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, as
 *	documented in tcl.h.
 *
 * Side effects:
 *	The converted bytes are stored in the output buffer.
 *
 *-------------------------------------------------------------------------
 */

int
Tcl_ExternalToUtf(
    TCL_UNUSED(Tcl_Interp *),	/* TODO: Re-examine this. */
    Tcl_Encoding encoding,	/* The encoding for the source string, or NULL
				 * for the default system encoding. */
    const char *src,		/* Source string in specified encoding. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * encoding-specific string length. */
    int flags,			/* Conversion control flags. */
    Tcl_EncodingState *statePtr,/* Place for conversion routine to store state
				 * information used during a piecewise
				 * conversion. Contents of statePtr are
				 * initialized and/or reset by conversion
				 * routine under control of flags argument. */
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const Encoding *encodingPtr;
    int result, srcRead, dstWrote, dstChars = 0;
    int noTerminate = flags & TCL_ENCODING_NO_TERMINATE;
    int charLimited = (flags & TCL_ENCODING_CHAR_LIMIT) && dstCharsPtr;
    int maxChars = INT_MAX;
    Tcl_EncodingState state;

    if (encoding == NULL) {
	encoding = systemEncoding;
    }
    encodingPtr = (Encoding *) encoding;

    if (src == NULL) {
	srcLen = 0;
    } else if (srcLen < 0) {
	srcLen = encodingPtr->lengthProc(src);
    }
    if (statePtr == NULL) {
	flags |= TCL_ENCODING_START | TCL_ENCODING_END;
	statePtr = &state;
    }
    if (srcReadPtr == NULL) {
	srcReadPtr = &srcRead;
    }
    if (dstWrotePtr == NULL) {
	dstWrotePtr = &dstWrote;
    }
    if (dstCharsPtr == NULL) {
	dstCharsPtr = &dstChars;
	flags &= ~TCL_ENCODING_CHAR_LIMIT;
    } else if (charLimited) {
	maxChars = *dstCharsPtr;
    }

    if (!noTerminate) {
	/*
	 * If there are any null characters in the middle of the buffer,
	 * they will converted to the UTF-8 null character (\xC0\x80). To get
	 * the actual \0 at the end of the destination buffer, we need to
	 * append it manually.  First make room for it...
	 */

	dstLen--;
    }
    if (encodingPtr->toUtfProc == UtfToUtfProc) {
	flags |= ENCODING_INPUT;
    }
    do {
	Tcl_EncodingState savedState = *statePtr;

	result = encodingPtr->toUtfProc(encodingPtr->clientData, src, srcLen,
		flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr,
		dstCharsPtr);
	if (*dstCharsPtr <= maxChars) {
	    break;
	}
	dstLen = TclUtfAtIndex(dst, maxChars) - dst + (TCL_UTF_MAX - 1);
	*statePtr = savedState;
    } while (1);
    if (!noTerminate) {
	/* ...and then append it */

	dst[*dstWrotePtr] = '\0';
    }

    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_UtfToExternalDString --
 *
 *	Convert a source buffer from UTF-8 to the specified encoding. If any
 *	of the bytes in the source buffer are invalid or cannot be represented
 *	in the target encoding, a default fallback character is substituted.
 *
 * Results:
 *	The converted bytes are stored in the DString, which is then NULL
 *	terminated in an encoding-specific manner. The return value is a
 *	pointer to the value stored in the DString.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

char *
Tcl_UtfToExternalDString(
    Tcl_Encoding encoding,	/* The encoding for the converted string, or
				 * NULL for the default system encoding. */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * strlen(). */
    Tcl_DString *dstPtr)	/* Uninitialized or free DString in which the
				 * converted string is stored. */
{
    Tcl_UtfToExternalDStringEx(encoding, src, srcLen, TCL_ENCODING_PROFILE_DEFAULT, dstPtr);
    return Tcl_DStringValue(dstPtr);
}


/*
 *-------------------------------------------------------------------------
 *
 * Tcl_UtfToExternalDStringEx --
 *
 *	Convert a source buffer from UTF-8 to the specified encoding.
 *	The parameter flags controls the behavior, if any of the bytes in
 *	the source buffer are invalid or cannot be represented in the
 *	target encoding. It should be composed by OR-ing the following:
 *	- *At most one* of TCL_ENCODING_PROFILE{DEFAULT,TCL8,STRICT}
 *	- TCL_ENCODING_STOPONERROR: Backward compatibility. Sets the profile
 *	  to TCL_ENCODING_PROFILE_STRICT overriding any specified profile flags
 *	- TCL_ENCODING_MODIFIED: convert NULL bytes to \xC0\x80 instead
 *        of 0x00. Only valid for "utf-8" and "cesu-8".
 *
 * Results:
 *	The converted bytes are stored in the DString, which is then NULL
 *	terminated in an encoding-specific manner. The return value is
 *	the error position in the source string or -1 if no conversion error
 *	is reported.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

int
Tcl_UtfToExternalDStringEx(
    Tcl_Encoding encoding,	/* The encoding for the converted string, or
				 * NULL for the default system encoding. */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * strlen(). */
    int flags,			/* Conversion control flags. */
    Tcl_DString *dstPtr)	/* Uninitialized or free DString in which the
				 * converted string is stored. */
{
    char *dst;
    Tcl_EncodingState state;
    const Encoding *encodingPtr;
    int dstLen, result, soFar, srcRead, dstWrote, dstChars;
    const char *srcStart = src;

    Tcl_DStringInit(dstPtr);
    dst = Tcl_DStringValue(dstPtr);
    dstLen = dstPtr->spaceAvl - 1;

    if (encoding == NULL) {
	encoding = systemEncoding;
    }
    encodingPtr = (Encoding *) encoding;

    if (src == NULL) {
	srcLen = 0;
    } else if (srcLen < 0) {
	srcLen = strlen(src);
    }
    flags = TclEncodingExternalFlagsToInternal(flags);
    flags |= TCL_ENCODING_START | TCL_ENCODING_END;
    while (1) {
	result = encodingPtr->fromUtfProc(encodingPtr->clientData, src,
		srcLen, flags, &state, dst, dstLen,
		&srcRead, &dstWrote, &dstChars);
	soFar = dst + dstWrote - Tcl_DStringValue(dstPtr);

	src += srcRead;
	if (result != TCL_CONVERT_NOSPACE) {
	    int i = soFar + encodingPtr->nullSize - 1;
	    while (i >= soFar) {
		Tcl_DStringSetLength(dstPtr, i--);
	    }
	    return (result == TCL_OK) ? TCL_INDEX_NONE : (int)(src - srcStart);
	}

	flags &= ~TCL_ENCODING_START;
	srcLen -= srcRead;
	if (Tcl_DStringLength(dstPtr) == 0) {
	    Tcl_DStringSetLength(dstPtr, dstLen);
	}
	Tcl_DStringSetLength(dstPtr, 2 * Tcl_DStringLength(dstPtr) + 1);
	dst = Tcl_DStringValue(dstPtr) + soFar;
	dstLen = Tcl_DStringLength(dstPtr) - soFar - 1;
    }
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_UtfToExternal --
 *
 *	Convert a buffer from UTF-8 into the specified encoding.
 *
 * Results:
 *	The return value is one of TCL_OK, TCL_CONVERT_MULTIBYTE,
 *	TCL_CONVERT_SYNTAX, TCL_CONVERT_UNKNOWN, or TCL_CONVERT_NOSPACE, as
 *	documented in tcl.h.
 *
 * Side effects:
 *	The converted bytes are stored in the output buffer.
 *
 *-------------------------------------------------------------------------
 */

int
Tcl_UtfToExternal(
    TCL_UNUSED(Tcl_Interp *),	/* TODO: Re-examine this. */
    Tcl_Encoding encoding,	/* The encoding for the converted string, or
				 * NULL for the default system encoding. */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes, or < 0 for
				 * strlen(). */
    int flags,			/* Conversion control flags. */
    Tcl_EncodingState *statePtr,/* Place for conversion routine to store state
				 * information used during a piecewise
				 * conversion. Contents of statePtr are
				 * initialized and/or reset by conversion
				 * routine under control of flags argument. */
    char *dst,			/* Output buffer in which converted string
				 * is stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const Encoding *encodingPtr;
    int result, srcRead, dstWrote, dstChars;
    Tcl_EncodingState state;

    if (encoding == NULL) {
	encoding = systemEncoding;
    }
    encodingPtr = (Encoding *) encoding;

    if (src == NULL) {
	srcLen = 0;
    } else if (srcLen < 0) {
	srcLen = strlen(src);
    }
    if (statePtr == NULL) {
	flags |= TCL_ENCODING_START | TCL_ENCODING_END;
	statePtr = &state;
    }
    if (srcReadPtr == NULL) {
	srcReadPtr = &srcRead;
    }
    if (dstWrotePtr == NULL) {
	dstWrotePtr = &dstWrote;
    }
    if (dstCharsPtr == NULL) {
	dstCharsPtr = &dstChars;
    }

    dstLen -= encodingPtr->nullSize;
    result = encodingPtr->fromUtfProc(encodingPtr->clientData, src, srcLen,
	    flags, statePtr, dst, dstLen, srcReadPtr,
	    dstWrotePtr, dstCharsPtr);
    memset(&dst[*dstWrotePtr], '\0', encodingPtr->nullSize);

    return result;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_FindExecutable --
 *
 *	This function computes the absolute path name of the current
 *	application, given its argv[0] value.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The absolute pathname for the application is computed and stored to be
 *	returned later by [info nameofexecutable].
 *
 *---------------------------------------------------------------------------
 */
#undef Tcl_FindExecutable
const char *
Tcl_FindExecutable(
    const char *argv0)		/* The value of the application's argv[0]
				 * (native). */
{
    const char *version = Tcl_InitSubsystems();
    TclpSetInitialEncodings();
    TclpFindExecutable(argv0);
    return version;
}

/*
 *---------------------------------------------------------------------------
 *
 * OpenEncodingFileChannel --
 *
 *	Open the file believed to hold data for the encoding, "name".
 *
 * Results:
 *	Returns the readable Tcl_Channel from opening the file, or NULL if the
 *	file could not be successfully opened. If NULL was returned, an error
 *	message is left in interp's result object, unless interp was NULL.
 *
 * Side effects:
 *	Channel may be opened. Information about the filesystem may be cached
 *	to speed later calls.
 *
 *---------------------------------------------------------------------------
 */

static Tcl_Channel
OpenEncodingFileChannel(
    Tcl_Interp *interp,		/* Interp for error reporting, if not NULL. */
    const char *name)		/* The name of the encoding file on disk and
				 * also the name for new encoding. */
{
    Tcl_Obj *nameObj = Tcl_NewStringObj(name, -1);
    Tcl_Obj *fileNameObj = Tcl_DuplicateObj(nameObj);
    Tcl_Obj *searchPath = Tcl_DuplicateObj(Tcl_GetEncodingSearchPath());
    Tcl_Obj *map = TclGetProcessGlobalValue(&encodingFileMap);
    Tcl_Obj **dir, *path, *directory = NULL;
    Tcl_Channel chan = NULL;
    int i, numDirs;

    TclListObjGetElementsM(NULL, searchPath, &numDirs, &dir);
    Tcl_IncrRefCount(nameObj);
    Tcl_AppendToObj(fileNameObj, ".enc", -1);
    Tcl_IncrRefCount(fileNameObj);
    Tcl_DictObjGet(NULL, map, nameObj, &directory);

    /*
     * Check that any cached directory is still on the encoding search path.
     */

    if (NULL != directory) {
	int verified = 0;

	for (i=0; i<numDirs && !verified; i++) {
	    if (dir[i] == directory) {
		verified = 1;
	    }
	}
	if (!verified) {
	    const char *dirString = TclGetString(directory);

	    for (i=0; i<numDirs && !verified; i++) {
		if (strcmp(dirString, TclGetString(dir[i])) == 0) {
		    verified = 1;
		}
	    }
	}
	if (!verified) {
	    /*
	     * Directory no longer on the search path. Remove from cache.
	     */

	    map = Tcl_DuplicateObj(map);
	    Tcl_DictObjRemove(NULL, map, nameObj);
	    TclSetProcessGlobalValue(&encodingFileMap, map, NULL);
	    directory = NULL;
	}
    }

    if (NULL != directory) {
	/*
	 * Got a directory from the cache. Try to use it first.
	 */

	Tcl_IncrRefCount(directory);
	path = Tcl_FSJoinToPath(directory, 1, &fileNameObj);
	Tcl_IncrRefCount(path);
	Tcl_DecrRefCount(directory);
	chan = Tcl_FSOpenFileChannel(NULL, path, "r", 0);
	Tcl_DecrRefCount(path);
    }

    /*
     * Scan the search path until we find it.
     */

    for (i=0; i<numDirs && (chan == NULL); i++) {
	path = Tcl_FSJoinToPath(dir[i], 1, &fileNameObj);
	Tcl_IncrRefCount(path);
	chan = Tcl_FSOpenFileChannel(NULL, path, "r", 0);
	Tcl_DecrRefCount(path);
	if (chan != NULL) {
	    /*
	     * Save directory in the cache.
	     */

	    map = Tcl_DuplicateObj(TclGetProcessGlobalValue(&encodingFileMap));
	    Tcl_DictObjPut(NULL, map, nameObj, dir[i]);
	    TclSetProcessGlobalValue(&encodingFileMap, map, NULL);
	}
    }

    if ((NULL == chan) && (interp != NULL)) {
	Tcl_SetObjResult(interp, Tcl_ObjPrintf(
		"unknown encoding \"%s\"", name));
	Tcl_SetErrorCode(interp, "TCL", "LOOKUP", "ENCODING", name, NULL);
    }
    Tcl_DecrRefCount(fileNameObj);
    Tcl_DecrRefCount(nameObj);
    Tcl_DecrRefCount(searchPath);

    return chan;
}

/*
 *---------------------------------------------------------------------------
 *
 * LoadEncodingFile --
 *
 *	Read a file that describes an encoding and create a new Encoding from
 *	the data.
 *
 * Results:
 *	The return value is the newly loaded Tcl_Encoding or NULL if the file
 *	didn't exist or could not be processed. If NULL is returned and interp
 *	is not NULL, an error message is left in interp's result object.
 *
 * Side effects:
 *	A corresponding encoding file might be read from persistent storage, in
 *	which case LoadTableEncoding is called.
 *
 *---------------------------------------------------------------------------
 */

static Tcl_Encoding
LoadEncodingFile(
    Tcl_Interp *interp,		/* Interp for error reporting, if not NULL. */
    const char *name)		/* The name of both the encoding file
				 * and the new encoding. */
{
    Tcl_Channel chan = NULL;
    Tcl_Encoding encoding = NULL;
    int ch;

    chan = OpenEncodingFileChannel(interp, name);
    if (chan == NULL) {
	return NULL;
    }

    Tcl_SetChannelOption(NULL, chan, "-encoding", "utf-8");

    while (1) {
	Tcl_DString ds;

	Tcl_DStringInit(&ds);
	Tcl_Gets(chan, &ds);
	ch = Tcl_DStringValue(&ds)[0];
	Tcl_DStringFree(&ds);
	if (ch != '#') {
	    break;
	}
    }

    switch (ch) {
    case 'S':
	encoding = LoadTableEncoding(name, ENCODING_SINGLEBYTE, chan);
	break;
    case 'D':
	encoding = LoadTableEncoding(name, ENCODING_DOUBLEBYTE, chan);
	break;
    case 'M':
	encoding = LoadTableEncoding(name, ENCODING_MULTIBYTE, chan);
	break;
    case 'E':
	encoding = LoadEscapeEncoding(name, chan);
	break;
    }
    if ((encoding == NULL) && (interp != NULL)) {
	Tcl_SetObjResult(interp, Tcl_ObjPrintf(
		"invalid encoding file \"%s\"", name));
	Tcl_SetErrorCode(interp, "TCL", "LOOKUP", "ENCODING", name, NULL);
    }
    Tcl_Close(NULL, chan);

    return encoding;
}

/*
 *-------------------------------------------------------------------------
 *
 * LoadTableEncoding --
 *
 *	Helper function for LoadEncodingFile().  Creates a Tcl_EncodingType
 *	structure along with its corresponding TableEncodingData structure, and
 *	passes it to Tcl_Createncoding.
 *
 *	The file contains binary data but begins with a marker to indicate
 *	byte-ordering so a single binary file can be read on big or
 *	little-endian systems.
 *
 * Results:
 *	Returns the new Tcl_Encoding,  or NULL if it could could
 *	not be created because the file contained invalid data.
 *
 * Side effects:
 *	See Tcl_CreateEncoding().
 *
 *-------------------------------------------------------------------------
 */

static Tcl_Encoding
LoadTableEncoding(
    const char *name,		/* Name of the new encoding. */
    int type,			/* Type of encoding (ENCODING_?????). */
    Tcl_Channel chan)		/* File containing new encoding. */
{
    Tcl_DString lineString;
    Tcl_Obj *objPtr;
    char *line;
    int i, hi, lo, numPages, symbol, fallback, len;
    unsigned char used[256];
    unsigned size;
    TableEncodingData *dataPtr;
    unsigned short *pageMemPtr, *page;
    Tcl_EncodingType encType;

    /*
     * Speed over memory. Use a full 256 character table to decode hex
     * sequences in the encoding files.
     */

    static const char staticHex[] = {
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*   0 ...  15 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*  16 ...  31 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*  32 ...  47 */
      0,  1,  2,  3,  4,  5,  6, 7, 8, 9, 0, 0, 0, 0, 0, 0, /*  48 ...  63 */
      0, 10, 11, 12, 13, 14, 15, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*  64 ...  79 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*  80 ...  95 */
      0, 10, 11, 12, 13, 14, 15, 0, 0, 0, 0, 0, 0, 0, 0, 0, /*  96 ... 111 */
      0,  1,  2,  3,  4,  5,  6, 7, 8, 9, 0, 0, 0, 0, 0, 0, /* 112 ... 127 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 128 ... 143 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 144 ... 159 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 160 ... 175 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 176 ... 191 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 192 ... 207 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 208 ... 223 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 224 ... 239 */
      0,  0,  0,  0,  0,  0,  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, /* 240 ... 255 */
    };

    Tcl_DStringInit(&lineString);
    if (Tcl_Gets(chan, &lineString) == TCL_IO_FAILURE) {
	return NULL;
    }
    line = Tcl_DStringValue(&lineString);

    fallback = (int) strtol(line, &line, 16);
    symbol = (int) strtol(line, &line, 10);
    numPages = (int) strtol(line, &line, 10);
    Tcl_DStringFree(&lineString);

    if (numPages < 0) {
	numPages = 0;
    } else if (numPages > 256) {
	numPages = 256;
    }

    memset(used, 0, sizeof(used));

#undef PAGESIZE
#define PAGESIZE    (256 * sizeof(unsigned short))

    dataPtr = (TableEncodingData *)ckalloc(sizeof(TableEncodingData));
    memset(dataPtr, 0, sizeof(TableEncodingData));

    dataPtr->fallback = fallback;

    /*
     * Read the table that maps characters to Unicode. Performs a single
     * malloc to get the memory for the array and all the pages needed by the
     * array.
     */

    size = 256 * sizeof(unsigned short *) + numPages * PAGESIZE;
    dataPtr->toUnicode = (unsigned short **)ckalloc(size);
    memset(dataPtr->toUnicode, 0, size);
    pageMemPtr = (unsigned short *) (dataPtr->toUnicode + 256);

    TclNewObj(objPtr);
    Tcl_IncrRefCount(objPtr);
    for (i = 0; i < numPages; i++) {
	int ch;
	const char *p;
	int expected = 3 + 16 * (16 * 4 + 1);

	if (Tcl_ReadChars(chan, objPtr, expected, 0) != expected) {
	    return NULL;
	}
	p = TclGetString(objPtr);
	hi = (staticHex[UCHAR(p[0])] << 4) + staticHex[UCHAR(p[1])];
	dataPtr->toUnicode[hi] = pageMemPtr;
	p += 2;
	for (lo = 0; lo < 256; lo++) {
	    if ((lo & 0x0F) == 0) {
		p++;
	    }
	    ch = (staticHex[UCHAR(p[0])] << 12) + (staticHex[UCHAR(p[1])] << 8)
		    + (staticHex[UCHAR(p[2])] << 4) + staticHex[UCHAR(p[3])];
	    if (ch != 0) {
		used[ch >> 8] = 1;
	    }
	    *pageMemPtr = (unsigned short) ch;
	    pageMemPtr++;
	    p += 4;
	}
    }
    TclDecrRefCount(objPtr);

    if (type == ENCODING_DOUBLEBYTE) {
	memset(dataPtr->prefixBytes, 1, sizeof(dataPtr->prefixBytes));
    } else {
	for (hi = 1; hi < 256; hi++) {
	    if (dataPtr->toUnicode[hi] != NULL) {
		dataPtr->prefixBytes[hi] = 1;
	    }
	}
    }

    /*
     * Invert the toUnicode array to produce the fromUnicode array. Performs a
     * single malloc to get the memory for the array and all the pages needed
     * by the array. While reading in the toUnicode array remember what
     * pages are needed for the fromUnicode array.
     */

    if (symbol) {
	used[0] = 1;
    }
    numPages = 0;
    for (hi = 0; hi < 256; hi++) {
	if (used[hi]) {
	    numPages++;
	}
    }
    size = 256 * sizeof(unsigned short *) + numPages * PAGESIZE;
    dataPtr->fromUnicode = (unsigned short **)ckalloc(size);
    memset(dataPtr->fromUnicode, 0, size);
    pageMemPtr = (unsigned short *) (dataPtr->fromUnicode + 256);

    for (hi = 0; hi < 256; hi++) {
	if (dataPtr->toUnicode[hi] == NULL) {
	    dataPtr->toUnicode[hi] = emptyPage;
	    continue;
	}
	for (lo = 0; lo < 256; lo++) {
	    int ch = dataPtr->toUnicode[hi][lo];

	    if (ch != 0) {
		page = dataPtr->fromUnicode[ch >> 8];
		if (page == NULL) {
		    page = pageMemPtr;
		    pageMemPtr += 256;
		    dataPtr->fromUnicode[ch >> 8] = page;
		}
		page[ch & 0xFF] = (unsigned short) ((hi << 8) + lo);
	    }
	}
    }
    if (type == ENCODING_MULTIBYTE) {
	/*
	 * If multibyte encodings don't have a backslash character, define
	 * one. Otherwise, on Windows, native file names don't work because
	 * the backslash in the file name maps to the unknown character
	 * (question mark) when converting from UTF-8 to external encoding.
	 */

	if (dataPtr->fromUnicode[0] != NULL) {
	    if (dataPtr->fromUnicode[0][(int)'\\'] == '\0') {
		dataPtr->fromUnicode[0][(int)'\\'] = '\\';
	    }
	}
    }
    if (symbol) {
	/*
	 * Make a special symbol encoding that maps each symbol character from
	 * its Unicode code point down into page 0, and also ensure that each
	 * characters on page 0 maps to itself so that a symbol font can be
	 * used to display a simple string like "abcd" and have alpha, beta,
	 * chi, delta show up, rather than have "unknown" chars show up because
	 * strictly speaking the symbol font doesn't have glyphs for those low
	 * ASCII chars.
	 */

	page = dataPtr->fromUnicode[0];
	if (page == NULL) {
	    page = pageMemPtr;
	    dataPtr->fromUnicode[0] = page;
	}
	for (lo = 0; lo < 256; lo++) {
	    if (dataPtr->toUnicode[0][lo] != 0) {
		page[lo] = (unsigned short) lo;
	    }
	}
    }
    for (hi = 0; hi < 256; hi++) {
	if (dataPtr->fromUnicode[hi] == NULL) {
	    dataPtr->fromUnicode[hi] = emptyPage;
	}
    }

    /*
     * For trailing 'R'everse encoding, see [Patch 689341]
     */

    Tcl_DStringInit(&lineString);

    /*
     * Skip leading empty lines.
     */

    while ((len = Tcl_Gets(chan, &lineString)) == 0) {
	/* empty body */
    }
    if (len < 0) {
	goto doneParse;
    }

    /*
     * Require that it starts with an 'R'.
     */

    line = Tcl_DStringValue(&lineString);
    if (line[0] != 'R') {
	goto doneParse;
    }

    /*
     * Read lines until EOF.
     */

    for (TclDStringClear(&lineString);
	    (len = Tcl_Gets(chan, &lineString)) >= 0;
	    TclDStringClear(&lineString)) {
	const unsigned char *p;
	int to, from;

	/*
	 * Skip short lines.
	 */

	if (len < 5) {
	    continue;
	}

	/*
	 * Parse the line as a sequence of hex digits.
	 */

	p = (const unsigned char *) Tcl_DStringValue(&lineString);
	to = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8)
		+ (staticHex[p[2]] << 4) + staticHex[p[3]];
	if (to == 0) {
	    continue;
	}
	for (p += 5, len -= 5; len >= 0 && *p; p += 5, len -= 5) {
	    from = (staticHex[p[0]] << 12) + (staticHex[p[1]] << 8)
		    + (staticHex[p[2]] << 4) + staticHex[p[3]];
	    if (from == 0) {
		continue;
	    }
	    dataPtr->fromUnicode[from >> 8][from & 0xFF] = to;
	}
    }
  doneParse:
    Tcl_DStringFree(&lineString);

    /*
     * Package everything into an encoding structure.
     */

    encType.encodingName    = name;
    encType.toUtfProc	    = TableToUtfProc;
    encType.fromUtfProc	    = TableFromUtfProc;
    encType.freeProc	    = TableFreeProc;
    encType.nullSize	    = (type == ENCODING_DOUBLEBYTE) ? 2 : 1;
    encType.clientData	    = dataPtr;

    return Tcl_CreateEncoding(&encType);
}

/*
 *-------------------------------------------------------------------------
 *
 * LoadEscapeEncoding --
 *
 *	Helper function for LoadEncodingTable(). Loads a state machine that
 *	converts between Unicode and some other encoding.
 *
 *	File contains text data that describes the escape sequences that are
 *	used to choose an encoding and the associated names for the
 *	sub-encodings.
 *
 * Results:
 *	The return value is the new encoding, or NULL if the encoding could
 *	not be created (because the file contained invalid data).
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static Tcl_Encoding
LoadEscapeEncoding(
    const char *name,		/* Name of the new encoding. */
    Tcl_Channel chan)		/* File containing new encoding. */
{
    int i;
    unsigned size;
    Tcl_DString escapeData;
    char init[16], final[16];
    EscapeEncodingData *dataPtr;
    Tcl_EncodingType type;

    init[0] = '\0';
    final[0] = '\0';
    Tcl_DStringInit(&escapeData);

    while (1) {
	int argc;
	const char **argv;
	char *line;
	Tcl_DString lineString;

	Tcl_DStringInit(&lineString);
	if (Tcl_Gets(chan, &lineString) < 0) {
	    break;
	}
	line = Tcl_DStringValue(&lineString);
	if (Tcl_SplitList(NULL, line, &argc, &argv) != TCL_OK) {
	    Tcl_DStringFree(&lineString);
	    continue;
	}
	if (argc >= 2) {
	    if (strcmp(argv[0], "name") == 0) {
		/* do nothing */
	    } else if (strcmp(argv[0], "init") == 0) {
		strncpy(init, argv[1], sizeof(init));
		init[sizeof(init) - 1] = '\0';
	    } else if (strcmp(argv[0], "final") == 0) {
		strncpy(final, argv[1], sizeof(final));
		final[sizeof(final) - 1] = '\0';
	    } else {
		EscapeSubTable est;
		Encoding *e;

		strncpy(est.sequence, argv[1], sizeof(est.sequence));
		est.sequence[sizeof(est.sequence) - 1] = '\0';
		est.sequenceLen = strlen(est.sequence);

		strncpy(est.name, argv[0], sizeof(est.name));
		est.name[sizeof(est.name) - 1] = '\0';

		/*
		 * To avoid infinite recursion in [encoding system iso2022-*]
		 */

		e = (Encoding *) Tcl_GetEncoding(NULL, est.name);
		if ((e != NULL) && (e->toUtfProc != TableToUtfProc)
			&& (e->toUtfProc != Iso88591ToUtfProc)) {
		   Tcl_FreeEncoding((Tcl_Encoding) e);
		   e = NULL;
		}
		est.encodingPtr = e;
		Tcl_DStringAppend(&escapeData, (char *) &est, sizeof(est));
	    }
	}
	ckfree(argv);
	Tcl_DStringFree(&lineString);
    }

    size = offsetof(EscapeEncodingData, subTables)
	    + Tcl_DStringLength(&escapeData);
    dataPtr = (EscapeEncodingData *)ckalloc(size);
    dataPtr->initLen = strlen(init);
    memcpy(dataPtr->init, init, dataPtr->initLen + 1);
    dataPtr->finalLen = strlen(final);
    memcpy(dataPtr->final, final, dataPtr->finalLen + 1);
    dataPtr->numSubTables =
	    Tcl_DStringLength(&escapeData) / sizeof(EscapeSubTable);
    memcpy(dataPtr->subTables, Tcl_DStringValue(&escapeData),
	    Tcl_DStringLength(&escapeData));
    Tcl_DStringFree(&escapeData);

    memset(dataPtr->prefixBytes, 0, sizeof(dataPtr->prefixBytes));
    for (i = 0; i < dataPtr->numSubTables; i++) {
	dataPtr->prefixBytes[UCHAR(dataPtr->subTables[i].sequence[0])] = 1;
    }
    if (dataPtr->init[0] != '\0') {
	dataPtr->prefixBytes[UCHAR(dataPtr->init[0])] = 1;
    }
    if (dataPtr->final[0] != '\0') {
	dataPtr->prefixBytes[UCHAR(dataPtr->final[0])] = 1;
    }

    /*
     * Package everything into an encoding structure.
     */

    type.encodingName	= name;
    type.toUtfProc	= EscapeToUtfProc;
    type.fromUtfProc    = EscapeFromUtfProc;
    type.freeProc	= EscapeFreeProc;
    type.nullSize	= 1;
    type.clientData	= dataPtr;

    return Tcl_CreateEncoding(&type);
}

/*
 *-------------------------------------------------------------------------
 *
 * BinaryProc --
 *
 *	The default conversion when no other conversion is specified. No
 *	translation is done; source bytes are copied directly to destination
 *	bytes.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
BinaryProc(
    TCL_UNUSED(void *),
    const char *src,		/* Source string (unknown encoding). */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    int result;

    result = TCL_OK;
    dstLen -= TCL_UTF_MAX - 1;
    if (dstLen < 0) {
	dstLen = 0;
    }
    if ((flags & TCL_ENCODING_CHAR_LIMIT) && srcLen > *dstCharsPtr) {
	srcLen = *dstCharsPtr;
    }
    if (srcLen > dstLen) {
	srcLen = dstLen;
	result = TCL_CONVERT_NOSPACE;
    }

    *srcReadPtr = srcLen;
    *dstWrotePtr = srcLen;
    *dstCharsPtr = srcLen;
    memcpy(dst, src, srcLen);
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfToUtfProc --
 *
 *	Converts from UTF-8 to UTF-8. Note that the UTF-8 to UTF-8 translation
 *	is not a no-op, because it turns a stream of improperly formed
 *	UTF-8 into a properly-formed stream.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
UtfToUtfProc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_MODIFIED */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* TCL_ENCODING_* conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd, *srcClose;
    const char *dstStart, *dstEnd;
    int result, numChars, charLimit = INT_MAX;
    int ch;
    int profile;

    result = TCL_OK;

    srcStart = src;
    srcEnd = src + srcLen;
    srcClose = srcEnd;
    if ((flags & TCL_ENCODING_END) == 0) {
	srcClose -= 6;
    }
    if (flags & TCL_ENCODING_CHAR_LIMIT) {
	charLimit = *dstCharsPtr;
    }

    dstStart = dst;
    flags |= PTR2INT(clientData);
    dstEnd = dst + dstLen - ((flags & ENCODING_UTF) ? TCL_UTF_MAX : 6);

    profile = TCL_ENCODING_PROFILE_GET(flags);
    for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {

	if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
	    /*
	     * If there is more string to follow, this will ensure that the
	     * last UTF-8 character in the source buffer hasn't been cut off.
	     */

	    result = TCL_CONVERT_MULTIBYTE;
	    break;
	}
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	if (UCHAR(*src) < 0x80 && !((UCHAR(*src) == 0) && (flags & ENCODING_INPUT))) {
	    /*
	     * Copy 7bit characters, but skip null-bytes when target encoding
	     * is Tcl's "modified" UTF-8. These need to be converted to
	     * \xC0\x80 as is done in a later branch.
	     */

	    *dst++ = *src++;
	}
	else if ((UCHAR(*src) == 0xC0) && (src + 1 < srcEnd) &&
		 (UCHAR(src[1]) == 0x80) && !(flags & TCL_ENCODING_MODIFIED) &&
		 (!(flags & ENCODING_INPUT) || PROFILE_STRICT(profile) ||
		  PROFILE_REPLACE(profile))) {
	    /* Special sequence \xC0\x80 */
            if ((PROFILE_STRICT(profile) || PROFILE_REPLACE(profile)) && (flags & ENCODING_INPUT)) {
		if (PROFILE_REPLACE(profile)) {
		   dst += Tcl_UniCharToUtf(UNICODE_REPLACE_CHAR, dst);
		   src += 2;
		} else {
		   /* PROFILE_STRICT */
		   result = TCL_CONVERT_SYNTAX;
		   break;
		}
	    } else {
		/*
		 * Convert 0xC080 to real nulls when we are in output mode,
		 * irrespective of the profile.
		 */
		*dst++ = 0;
		src += 2;
	    }

	}
	else if (!Tcl_UtfCharComplete(src, srcEnd - src)) {
	    /*
	     * Incomplete byte sequence.
	     * Always check before using TclUtfToUCS4. Not doing can so
	     * cause it run beyond the end of the buffer! If we happen such an
	     * incomplete char its bytes are made to represent themselves
	     * unless the user has explicitly asked to be told.
	     */

	    if (flags & ENCODING_INPUT) {
		/* Incomplete bytes for modified UTF-8 target */
		if (PROFILE_STRICT(profile)) {
		    result = (flags & TCL_ENCODING_CHAR_LIMIT)
			       ? TCL_CONVERT_MULTIBYTE
			       : TCL_CONVERT_SYNTAX;
		    break;
		}
		if (PROFILE_REPLACE(profile)) {
		    ch = UNICODE_REPLACE_CHAR;
		} else {
		    /* TCL_ENCODING_PROFILE_TCL8 */
		    ch = UCHAR(*src);
		}
		++src;
	    }
	    else {
		/*
		 * Incomplete bytes for real UTF-8 target.
		 * TODO - no profile check here because did not have any
		 * checks in the pre-profile code. Why? Is it because on
		 * output a valid internal utf-8 stream is assumed?
		 */
		char chbuf[2];
		/*
		 * TODO - this code seems broken to me.
		 * - it does not check profiles
		 * - generates invalid output for real UTF-8 target
		 *   (consider \xC2)
		 * A possible explanation is this behavior matches the
		 * Tcl8 decoding behavior of mapping invalid bytes to the same
		 * code point value. Still, at least strictness checks should
		 * be made.
		 */
		chbuf[0] = UCHAR(*src++); chbuf[1] = 0;
		TclUtfToUCS4(chbuf, &ch);
	    }
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	else {
	    int low;
            int isInvalid = 0;
	    size_t len = TclUtfToUCS4(src, &ch);
	    if (flags & ENCODING_INPUT) {
		if ((len < 2) && (ch != 0)) {
                    isInvalid = 1;
		} else if ((ch > 0xFFFF) && !(flags & ENCODING_UTF)) {
                    isInvalid = 1;
		}
		if (isInvalid) {
		    if (PROFILE_STRICT(profile)) {
			result = TCL_CONVERT_SYNTAX;
			break;
		    }
		    else if (PROFILE_REPLACE(profile)) {
			ch = UNICODE_REPLACE_CHAR;
		    }
		}
	    }

	    const char *saveSrc = src;
	    src += len;
	    if (!(flags & ENCODING_UTF) && !(flags & ENCODING_INPUT) && (ch > 0x3FF)) {
		if (ch > 0xFFFF) {
		    /* CESU-8 6-byte sequence for chars > U+FFFF */
		    ch -= 0x10000;
		    *dst++ = 0xED;
		    *dst++ = (char) (((ch >> 16) & 0x0F) | 0xA0);
		    *dst++ = (char) (((ch >> 10) & 0x3F) | 0x80);
		    ch = (ch & 0x0CFF) | 0xDC00;
		}
		goto cesu8;
	    } else if ((ch | 0x7FF) == 0xDFFF) {
		/*
		 * A surrogate character is detected, handle especially.
		 */
		if (PROFILE_STRICT(profile) && (flags & ENCODING_UTF)) {
		    result = TCL_CONVERT_UNKNOWN;
		    src = saveSrc;
		    break;
		}
		if (0 && PROFILE_REPLACE(profile)) {
		    ch = UNICODE_REPLACE_CHAR;
		    src += len;
		    // dst += Tcl_UniCharToUtf(ch, dst);
		}
		else {
		    low = ch;
		    len = (src <= srcEnd - 3) ? TclUtfToUCS4(src, &low) : 0;

		    if ((!LOW_SURROGATE(low)) || (ch & 0x400)) {

			if (PROFILE_STRICT(profile)) {
			    result = TCL_CONVERT_UNKNOWN;
			    src = saveSrc;
			    break;
			}
cesu8:
			*dst++ = (char)(((ch >> 12) | 0xE0) & 0xEF);
			*dst++ = (char)(((ch >> 6) | 0x80) & 0xBF);
			*dst++ = (char)((ch | 0x80) & 0xBF);
			continue;
		    }
		    src += len;
		    dst += Tcl_UniCharToUtf(ch, dst);
		    ch = low;
		}
	    } else if (PROFILE_STRICT(profile) &&
		       (!(flags & ENCODING_INPUT)) &&
		       SURROGATE(ch)) {
		result = TCL_CONVERT_UNKNOWN;
		src = saveSrc;
		break;
	    } else if (PROFILE_STRICT(profile) &&
	               (flags & ENCODING_INPUT) &&
		       SURROGATE(ch)) {
		result = TCL_CONVERT_SYNTAX;
		src = saveSrc;
		break;
	    }
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
    }

    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Utf32ToUtfProc --
 *
 *	Convert from UTF-32 to UTF-8.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
Utf32ToUtfProc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_LE */
    const char *src,		/* Source string in Unicode. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd;
    const char *dstEnd, *dstStart;
    int result, numChars, charLimit = INT_MAX;
    int ch = 0;

    flags |= PTR2INT(clientData);
    if (flags & TCL_ENCODING_CHAR_LIMIT) {
	charLimit = *dstCharsPtr;
    }
    result = TCL_OK;

    /*
     * Check alignment with utf-32 (4 == sizeof(UTF-32))
     */

    if ((srcLen % 4) != 0) {
	result = TCL_CONVERT_MULTIBYTE;
	srcLen &= -4;
    }

    /*
     * If last code point is a high surrogate, we cannot handle that yet,
     * unless we are at the end.
     */

    if (!(flags & TCL_ENCODING_END) && (srcLen >= 4) &&
	    ((src[srcLen - ((flags & TCL_ENCODING_LE)?3:2)] & 0xFC) == 0xD8) &&
	    ((src[srcLen - ((flags & TCL_ENCODING_LE)?2:3)]) == 0) &&
	    ((src[srcLen - ((flags & TCL_ENCODING_LE)?1:4)]) == 0)) {
	result = TCL_CONVERT_MULTIBYTE;
	srcLen-= 4;
    }

    srcStart = src;
    srcEnd = src + srcLen;

    dstStart = dst;
    dstEnd = dst + dstLen - TCL_UTF_MAX;

    for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}

	int prev = ch;
	if (flags & TCL_ENCODING_LE) {
	    ch = (src[3] & 0xFF) << 24 | (src[2] & 0xFF) << 16 | (src[1] & 0xFF) << 8 | (src[0] & 0xFF);
	} else {
	    ch = (src[0] & 0xFF) << 24 | (src[1] & 0xFF) << 16 | (src[2] & 0xFF) << 8 | (src[3] & 0xFF);
	}
	if (((prev  & ~0x3FF) == 0xD800) && ((ch  & ~0x3FF) != 0xDC00)) {
	    /* Bug [10c2c17c32]. If Hi surrogate not followed by Lo surrogate, finish 3-byte UTF-8 */
	    dst += Tcl_UniCharToUtf(-1, dst);
	}
	
	if ((unsigned)ch > 0x10FFFF || SURROGATE(ch)) {
	    if (PROFILE_STRICT(flags)) {
		result = TCL_CONVERT_SYNTAX;
		ch = 0;
		break;
	    }
	    if (PROFILE_REPLACE(flags)) {
		ch = UNICODE_REPLACE_CHAR;
	    }
	}

	/*
	 * Special case for 1-byte utf chars for speed. Make sure we work with
	 * unsigned short-size data.
	 */

	if ((ch > 0) && (ch < 0x80)) {
	    *dst++ = (ch & 0xFF);
	} else {
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	src += sizeof(unsigned int);
    }

    if ((ch  & ~0x3FF) == 0xD800) {
	/* Bug [10c2c17c32]. If Hi surrogate, finish 3-byte UTF-8 */
	dst += Tcl_UniCharToUtf(-1, dst);
    }
    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfToUtf32Proc --
 *
 *	Convert from UTF-8 to UTF-32.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
UtfToUtf32Proc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_LE */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd, *srcClose, *dstStart, *dstEnd;
    int result, numChars;
    int ch, len;

    srcStart = src;
    srcEnd = src + srcLen;
    srcClose = srcEnd;
    if ((flags & TCL_ENCODING_END) == 0) {
	srcClose -= TCL_UTF_MAX;
    }

    dstStart = dst;
    dstEnd = dst + dstLen - sizeof(Tcl_UniChar);
    flags |= PTR2INT(clientData);

    result = TCL_OK;
    for (numChars = 0; src < srcEnd; numChars++) {
	if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
	    /*
	     * If there is more string to follow, this will ensure that the
	     * last UTF-8 character in the source buffer hasn't been cut off.
	     */

	    result = TCL_CONVERT_MULTIBYTE;
	    break;
	}
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	len = TclUtfToUCS4(src, &ch);
	if (SURROGATE(ch)) {
	    if (PROFILE_STRICT(flags)) {
		result = TCL_CONVERT_UNKNOWN;
		break;
	    }
	    if (PROFILE_REPLACE(flags)) {
		ch = UNICODE_REPLACE_CHAR;
	    }
	}
	src += len;
	if (flags & TCL_ENCODING_LE) {
	    *dst++ = (ch & 0xFF);
	    *dst++ = ((ch >> 8) & 0xFF);
	    *dst++ = ((ch >> 16) & 0xFF);
	    *dst++ = ((ch >> 24) & 0xFF);
	} else {
	    *dst++ = ((ch >> 24) & 0xFF);
	    *dst++ = ((ch >> 16) & 0xFF);
	    *dst++ = ((ch >> 8) & 0xFF);
	    *dst++ = (ch & 0xFF);
	}
    }

    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Utf16ToUtfProc --
 *
 *	Convert from UTF-16 to UTF-8.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
Utf16ToUtfProc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_LE */
    const char *src,		/* Source string in Unicode. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd;
    const char *dstEnd, *dstStart;
    int result, numChars, charLimit = INT_MAX;
    unsigned short ch = 0;

    flags |= PTR2INT(clientData);
    if (flags & TCL_ENCODING_CHAR_LIMIT) {
	charLimit = *dstCharsPtr;
    }
    result = TCL_OK;

    /*
     * Check alignment with utf-16 (2 == sizeof(UTF-16))
     */

    if ((srcLen % 2) != 0) {
	result = TCL_CONVERT_MULTIBYTE;
	srcLen--;
    }

    /*
     * If last code point is a high surrogate, we cannot handle that yet,
     * unless we are at the end.
     */

    if (!(flags & TCL_ENCODING_END) && (srcLen >= 2) &&
	    ((src[srcLen - ((flags & TCL_ENCODING_LE)?1:2)] & 0xFC) == 0xD8)) {
	result = TCL_CONVERT_MULTIBYTE;
	srcLen-= 2;
    }

    srcStart = src;
    srcEnd = src + srcLen;

    dstStart = dst;
    dstEnd = dst + dstLen - TCL_UTF_MAX;

    for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}

	unsigned short prev = ch;
	if (flags & TCL_ENCODING_LE) {
	    ch = (src[1] & 0xFF) << 8 | (src[0] & 0xFF);
	} else {
	    ch = (src[0] & 0xFF) << 8 | (src[1] & 0xFF);
	}
	if (((prev  & ~0x3FF) == 0xD800) && ((ch  & ~0x3FF) != 0xDC00)) {
	    /* Bug [10c2c17c32]. If Hi surrogate not followed by Lo surrogate, finish 3-byte UTF-8 */
	    dst += Tcl_UniCharToUtf(-1, dst);
	}

	/*
	 * Special case for 1-byte utf chars for speed. Make sure we work with
	 * unsigned short-size data.
	 */

	if (ch && ch < 0x80) {
	    *dst++ = (ch & 0xFF);
	} else {
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	src += sizeof(unsigned short);
    }

    if ((ch  & ~0x3FF) == 0xD800) {
	/* Bug [10c2c17c32]. If Hi surrogate, finish 3-byte UTF-8 */
	dst += Tcl_UniCharToUtf(-1, dst);
    }
    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfToUtf16Proc --
 *
 *	Convert from UTF-8 to UTF-16.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
UtfToUtf16Proc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_LE */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd, *srcClose, *dstStart, *dstEnd;
    int result, numChars;
    int ch, len;

    srcStart = src;
    srcEnd = src + srcLen;
    srcClose = srcEnd;
    if ((flags & TCL_ENCODING_END) == 0) {
	srcClose -= TCL_UTF_MAX;
    }

    dstStart = dst;
    dstEnd   = dst + dstLen - sizeof(Tcl_UniChar);
    flags |= PTR2INT(clientData);

    result = TCL_OK;
    for (numChars = 0; src < srcEnd; numChars++) {
	if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
	    /*
	     * If there is more string to follow, this will ensure that the
	     * last UTF-8 character in the source buffer hasn't been cut off.
	     */

	    result = TCL_CONVERT_MULTIBYTE;
	    break;
	}
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	len = TclUtfToUCS4(src, &ch);
	if (SURROGATE(ch)) {
	    if (PROFILE_STRICT(flags)) {
		result = TCL_CONVERT_UNKNOWN;
		break;
	    }
	    if (PROFILE_REPLACE(flags)) {
		ch = UNICODE_REPLACE_CHAR;
	    }
	}
	src += len;
	if (flags & TCL_ENCODING_LE) {
	    if (ch <= 0xFFFF) {
		*dst++ = (ch & 0xFF);
		*dst++ = (ch >> 8);
	    } else {
		*dst++ = (((ch - 0x10000) >> 10) & 0xFF);
		*dst++ = (((ch - 0x10000) >> 18) & 0x3) | 0xD8;
		*dst++ = (ch & 0xFF);
		*dst++ = ((ch >> 8) & 0x3) | 0xDC;
	    }
	} else {
	    if (ch <= 0xFFFF) {
		*dst++ = (ch >> 8);
		*dst++ = (ch & 0xFF);
	    } else {
		*dst++ = (((ch - 0x10000) >> 18) & 0x3) | 0xD8;
		*dst++ = (((ch - 0x10000) >> 10) & 0xFF);
		*dst++ = ((ch >> 8) & 0x3) | 0xDC;
		*dst++ = (ch & 0xFF);
	    }
	}
    }
    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfToUcs2Proc --
 *
 *	Convert from UTF-8 to UCS-2.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
UtfToUcs2Proc(
    void *clientData,	/* additional flags, e.g. TCL_ENCODING_LE */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd, *srcClose, *dstStart, *dstEnd;
    int result, numChars;
#if TCL_UTF_MAX < 4
    int len;
#endif
    Tcl_UniChar ch = 0;

    flags |= PTR2INT(clientData);
    srcStart = src;
    srcEnd = src + srcLen;
    srcClose = srcEnd;
    if ((flags & TCL_ENCODING_END) == 0) {
	srcClose -= TCL_UTF_MAX;
    }

    dstStart = dst;
    dstEnd   = dst + dstLen - sizeof(Tcl_UniChar);

    result = TCL_OK;
    for (numChars = 0; src < srcEnd; numChars++) {
	if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
	    /*
	     * If there is more string to follow, this will ensure that the
	     * last UTF-8 character in the source buffer hasn't been cut off.
	     */

	    result = TCL_CONVERT_MULTIBYTE;
	    break;
	}
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	/* TODO - there were no STRICT or NOCOMPLAIN checks here (why?)
	 * so no profile checks either for now. */

#if TCL_UTF_MAX < 4
	src += (len = TclUtfToUniChar(src, &ch));
	if ((ch >= 0xD800) && (len < 3)) {
	    src += TclUtfToUniChar(src, &ch);
	    ch = 0xFFFD;
	}
#else
	src += TclUtfToUniChar(src, &ch);
	if (ch > 0xFFFF) {
	    ch = 0xFFFD;
	}
#endif

	/*
	 * Need to handle this in a way that won't cause misalignment by
	 * casting dst to a Tcl_UniChar. [Bug 1122671]
	 */

	if (flags & TCL_ENCODING_LE) {
	    *dst++ = (ch & 0xFF);
	    *dst++ = (ch >> 8);
	} else {
	    *dst++ = (ch >> 8);
	    *dst++ = (ch & 0xFF);
	}
    }
    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * TableToUtfProc --
 *
 *	Convert from the encoding specified by the TableEncodingData into
 *	UTF-8.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
TableToUtfProc(
    void *clientData,	/* TableEncodingData that specifies
				 * encoding. */
    const char *src,		/* Source string in specified encoding. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd;
    const char *dstEnd, *dstStart, *prefixBytes;
    int result, byte, numChars, charLimit = INT_MAX;
    Tcl_UniChar ch = 0;
    const unsigned short *const *toUnicode;
    const unsigned short *pageZero;
    TableEncodingData *dataPtr = (TableEncodingData *)clientData;

    if (flags & TCL_ENCODING_CHAR_LIMIT) {
	charLimit = *dstCharsPtr;
    }
    srcStart = src;
    srcEnd = src + srcLen;

    dstStart = dst;
    dstEnd = dst + dstLen - TCL_UTF_MAX;

    toUnicode = (const unsigned short *const *) dataPtr->toUnicode;
    prefixBytes = dataPtr->prefixBytes;
    pageZero = toUnicode[0];

    result = TCL_OK;
    for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	byte = *((unsigned char *) src);
	if (prefixBytes[byte]) {
	    src++;
	    if (src >= srcEnd) {
		/*
		 * TODO - this is broken. For consistency with other
		 * decoders, an error should be raised only if strict.
		 * However, doing that check cause a whole bunch of test
		 * failures. Need to verify if those tests are in fact
		 * correct.
		 */
		src--;
		result = TCL_CONVERT_MULTIBYTE;
		break;
	    }
	    ch = toUnicode[byte][*((unsigned char *)src)];
	} else {
	    ch = pageZero[byte];
	}
	if ((ch == 0) && (byte != 0)) {
	    if (PROFILE_STRICT(flags)) {
		result = TCL_CONVERT_SYNTAX;
		break;
	    }
	    if (prefixBytes[byte]) {
		src--;
	    }
	    if (PROFILE_REPLACE(flags)) {
		ch = UNICODE_REPLACE_CHAR;
	    } else {
		ch = (Tcl_UniChar)byte;
	    }
	}

	/*
	 * Special case for 1-byte utf chars for speed.
	 */

	if (ch && ch < 0x80) {
	    *dst++ = (char) ch;
	} else {
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	src++;
    }

    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * TableFromUtfProc --
 *
 *	Convert from UTF-8 into the encoding specified by the
 *	TableEncodingData.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
TableFromUtfProc(
    void *clientData,	/* TableEncodingData that specifies
				 * encoding. */
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd, *srcClose;
    const char *dstStart, *dstEnd, *prefixBytes;
    Tcl_UniChar ch = 0;
    int result, len, word, numChars;
    TableEncodingData *dataPtr = (TableEncodingData *)clientData;
    const unsigned short *const *fromUnicode;

    result = TCL_OK;

    prefixBytes = dataPtr->prefixBytes;
    fromUnicode = (const unsigned short *const *) dataPtr->fromUnicode;

    srcStart = src;
    srcEnd = src + srcLen;
    srcClose = srcEnd;
    if ((flags & TCL_ENCODING_END) == 0) {
	srcClose -= TCL_UTF_MAX;
    }

    dstStart = dst;
    dstEnd = dst + dstLen - 1;

    for (numChars = 0; src < srcEnd; numChars++) {
	if ((src > srcClose) && (!Tcl_UtfCharComplete(src, srcEnd - src))) {
	    /*
	     * If there is more string to follow, this will ensure that the
	     * last UTF-8 character in the source buffer hasn't been cut off.
	     */

	    result = TCL_CONVERT_MULTIBYTE;
	    break;
	}
	len = TclUtfToUniChar(src, &ch);

#if TCL_UTF_MAX > 3
	/* Unicode chars > +U0FFFF cannot be represented in any table encoding */
	if (ch & 0xFFFF0000) {
	    word = 0;
	} else
#else
	if (!len) {
	    word = 0;
	} else
#endif
	    word = fromUnicode[(ch >> 8)][ch & 0xFF];

	if ((word == 0) && (ch != 0)) {
	    if (PROFILE_STRICT(flags)) {
		result = TCL_CONVERT_UNKNOWN;
		break;
	    }
	    word = dataPtr->fallback; /* Both profiles REPLACE and TCL8 */
	}
	if (prefixBytes[(word >> 8)] != 0) {
	    if (dst + 1 > dstEnd) {
		result = TCL_CONVERT_NOSPACE;
		break;
	    }
	    dst[0] = (char) (word >> 8);
	    dst[1] = (char) word;
	    dst += 2;
	} else {
	    if (dst > dstEnd) {
		result = TCL_CONVERT_NOSPACE;
		break;
	    }
	    dst[0] = (char) word;
	    dst++;
	}
	src += len;
    }

    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Iso88591ToUtfProc --
 *
 *	Convert from the "iso8859-1" encoding into UTF-8.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
Iso88591ToUtfProc(
    TCL_UNUSED(void *),
    const char *src,		/* Source string in specified encoding. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source
				 * characters. */
    int *dstWrotePtr,		/* Filled with the number of bytes that were
				 * stored in the output buffer as a result of
				 * the conversion. */
    int *dstCharsPtr)		/* Filled with the number of characters that
				 * correspond to the bytes stored in the
				 * output buffer. */
{
    const char *srcStart, *srcEnd;
    const char *dstEnd, *dstStart;
    int result, numChars, charLimit = INT_MAX;

    if (flags & TCL_ENCODING_CHAR_LIMIT) {
	charLimit = *dstCharsPtr;
    }
    srcStart = src;
    srcEnd = src + srcLen;

    dstStart = dst;
    dstEnd = dst + dstLen - TCL_UTF_MAX;

    result = TCL_OK;
    for (numChars = 0; src < srcEnd && numChars <= charLimit; numChars++) {
	Tcl_UniChar ch = 0;

	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	ch = (Tcl_UniChar) *((unsigned char *) src);

	/*
	 * Special case for 1-byte utf chars for speed.
	 */

	if (ch && ch < 0x80) {
	    *dst++ = (char) ch;
	} else {
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	src++;
    }

    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Iso88591FromUtfProc --
 *
 *	Convert from UTF-8 into the encoding "iso8859-1".
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int
Iso88591FromUtfProc(
    TCL_UNUSED(void *),
    const char *src,		/* Source string in UTF-8. */
    int srcLen,			/* Source string length in bytes. */
    int flags,			/* Conversion control flags. */
    TCL_UNUSED(Tcl_EncodingState *),
    char *dst,			/* Output buffer in which converted string is
				 * stored. */
    int dstLen,			/* The maximum length of output buffer in
				 * bytes. */
    int *srcReadPtr,		/* Filled with the number of bytes from the
				 * source string that were converted. This may
				 * be less than the original source length if
				 * there was a problem converting some source