summaryrefslogtreecommitdiffstats
path: root/Python/pystate.c
blob: 42bc3eceafa22224e6679ebbe283ff09a15476d4 (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

/* Thread and interpreter state structures and their interfaces */

#include "Python.h"

/* --------------------------------------------------------------------------
CAUTION

Always use malloc() and free() directly in this file.  A number of these
functions are advertised as safe to call when the GIL isn't held, and in
a debug build Python redirects (e.g.) PyMem_NEW (etc) to Python's debugging
obmalloc functions.  Those aren't thread-safe (they rely on the GIL to avoid
the expense of doing their own locking).
-------------------------------------------------------------------------- */

#ifdef HAVE_DLOPEN
#ifdef HAVE_DLFCN_H
#include <dlfcn.h>
#endif
#ifndef RTLD_LAZY
#define RTLD_LAZY 1
#endif
#endif

#ifdef __cplusplus
extern "C" {
#endif

#ifdef WITH_THREAD
#include "pythread.h"
static PyThread_type_lock head_mutex = NULL; /* Protects interp->tstate_head */
#define HEAD_INIT() (void)(head_mutex || (head_mutex = PyThread_allocate_lock()))
#define HEAD_LOCK() PyThread_acquire_lock(head_mutex, WAIT_LOCK)
#define HEAD_UNLOCK() PyThread_release_lock(head_mutex)

/* The single PyInterpreterState used by this process'
   GILState implementation
*/
static PyInterpreterState *autoInterpreterState = NULL;
static int autoTLSkey = 0;
#else
#define HEAD_INIT() /* Nothing */
#define HEAD_LOCK() /* Nothing */
#define HEAD_UNLOCK() /* Nothing */
#endif

static PyInterpreterState *interp_head = NULL;

/* Assuming the current thread holds the GIL, this is the
   PyThreadState for the current thread. */
_Py_atomic_address _PyThreadState_Current = {NULL};
PyThreadFrameGetter _PyThreadState_GetFrame = NULL;

#ifdef WITH_THREAD
static void _PyGILState_NoteThreadState(PyThreadState* tstate);
#endif


PyInterpreterState *
PyInterpreterState_New(void)
{
    PyInterpreterState *interp = (PyInterpreterState *)
                                 malloc(sizeof(PyInterpreterState));

    if (interp != NULL) {
        HEAD_INIT();
#ifdef WITH_THREAD
        if (head_mutex == NULL)
            Py_FatalError("Can't initialize threads for interpreter");
#endif
        interp->modules = NULL;
        interp->modules_reloading = NULL;
        interp->modules_by_index = NULL;
        interp->sysdict = NULL;
        interp->builtins = NULL;
        interp->tstate_head = NULL;
        interp->codec_search_path = NULL;
        interp->codec_search_cache = NULL;
        interp->codec_error_registry = NULL;
        interp->codecs_initialized = 0;
        interp->fscodec_initialized = 0;
#ifdef HAVE_DLOPEN
#ifdef RTLD_NOW
        interp->dlopenflags = RTLD_NOW;
#else
        interp->dlopenflags = RTLD_LAZY;
#endif
#endif
#ifdef WITH_TSC
        interp->tscdump = 0;
#endif

        HEAD_LOCK();
        interp->next = interp_head;
        interp_head = interp;
        HEAD_UNLOCK();
    }

    return interp;
}


void
PyInterpreterState_Clear(PyInterpreterState *interp)
{
    PyThreadState *p;
    HEAD_LOCK();
    for (p = interp->tstate_head; p != NULL; p = p->next)
        PyThreadState_Clear(p);
    HEAD_UNLOCK();
    Py_CLEAR(interp->codec_search_path);
    Py_CLEAR(interp->codec_search_cache);
    Py_CLEAR(interp->codec_error_registry);
    Py_CLEAR(interp->modules);
    Py_CLEAR(interp->modules_by_index);
    Py_CLEAR(interp->modules_reloading);
    Py_CLEAR(interp->sysdict);
    Py_CLEAR(interp->builtins);
}


static void
zapthreads(PyInterpreterState *interp)
{
    PyThreadState *p;
    /* No need to lock the mutex here because this should only happen
       when the threads are all really dead (XXX famous last words). */
    while ((p = interp->tstate_head) != NULL) {
        PyThreadState_Delete(p);
    }
}


void
PyInterpreterState_Delete(PyInterpreterState *interp)
{
    PyInterpreterState **p;
    zapthreads(interp);
    HEAD_LOCK();
    for (p = &interp_head; ; p = &(*p)->next) {
        if (*p == NULL)
            Py_FatalError(
                "PyInterpreterState_Delete: invalid interp");
        if (*p == interp)
            break;
    }
    if (interp->tstate_head != NULL)
        Py_FatalError("PyInterpreterState_Delete: remaining threads");
    *p = interp->next;
    HEAD_UNLOCK();
    free(interp);
#ifdef WITH_THREAD
    if (interp_head == NULL && head_mutex != NULL) {
        PyThread_free_lock(head_mutex);
        head_mutex = NULL;
    }
#endif
}


/* Default implementation for _PyThreadState_GetFrame */
static struct _frame *
threadstate_getframe(PyThreadState *self)
{
    return self->frame;
}

static PyThreadState *
new_threadstate(PyInterpreterState *interp, int init)
{
    PyThreadState *tstate = (PyThreadState *)malloc(sizeof(PyThreadState));

    if (_PyThreadState_GetFrame == NULL)
        _PyThreadState_GetFrame = threadstate_getframe;

    if (tstate != NULL) {
        tstate->interp = interp;

        tstate->frame = NULL;
        tstate->recursion_depth = 0;
        tstate->overflowed = 0;
        tstate->recursion_critical = 0;
        tstate->tracing = 0;
        tstate->use_tracing = 0;
        tstate->tick_counter = 0;
        tstate->gilstate_counter = 0;
        tstate->async_exc = NULL;
#ifdef WITH_THREAD
        tstate->thread_id = PyThread_get_thread_ident();
#else
        tstate->thread_id = 0;
#endif

        tstate->dict = NULL;

        tstate->curexc_type = NULL;
        tstate->curexc_value = NULL;
        tstate->curexc_traceback = NULL;

        tstate->exc_type = NULL;
        tstate->exc_value = NULL;
        tstate->exc_traceback = NULL;

        tstate->c_profilefunc = NULL;
        tstate->c_tracefunc = NULL;
        tstate->c_profileobj = NULL;
        tstate->c_traceobj = NULL;

        tstate->trash_delete_nesting = 0;
        tstate->trash_delete_later = NULL;

        if (init)
            _PyThreadState_Init(tstate);

        HEAD_LOCK();
        tstate->next = interp->tstate_head;
        interp->tstate_head = tstate;
        HEAD_UNLOCK();
    }

    return tstate;
}

PyThreadState *
PyThreadState_New(PyInterpreterState *interp)
{
    return new_threadstate(interp, 1);
}

PyThreadState *
_PyThreadState_Prealloc(PyInterpreterState *interp)
{
    return new_threadstate(interp, 0);
}

void
_PyThreadState_Init(PyThreadState *tstate)
{
#ifdef WITH_THREAD
    _PyGILState_NoteThreadState(tstate);
#endif
}

PyObject*
PyState_FindModule(struct PyModuleDef* m)
{
    Py_ssize_t index = m->m_base.m_index;
    PyInterpreterState *state = PyThreadState_GET()->interp;
    PyObject *res;
    if (index == 0)
        return NULL;
    if (state->modules_by_index == NULL)
        return NULL;
    if (index >= PyList_GET_SIZE(state->modules_by_index))
        return NULL;
    res = PyList_GET_ITEM(state->modules_by_index, index);
    return res==Py_None ? NULL : res;
}

int
_PyState_AddModule(PyObject* module, struct PyModuleDef* def)
{
    PyInterpreterState *state = PyThreadState_GET()->interp;
    if (!def)
        return -1;
    if (!state->modules_by_index) {
        state->modules_by_index = PyList_New(0);
        if (!state->modules_by_index)
            return -1;
    }
    while(PyList_GET_SIZE(state->modules_by_index) <= def->m_base.m_index)
        if (PyList_Append(state->modules_by_index, Py_None) < 0)
            return -1;
    Py_INCREF(module);
    return PyList_SetItem(state->modules_by_index,
                          def->m_base.m_index, module);
}

void
PyThreadState_Clear(PyThreadState *tstate)
{
    if (Py_VerboseFlag && tstate->frame != NULL)
        fprintf(stderr,
          "PyThreadState_Clear: warning: thread still has a frame\n");

    Py_CLEAR(tstate->frame);

    Py_CLEAR(tstate->dict);
    Py_CLEAR(tstate->async_exc);

    Py_CLEAR(tstate->curexc_type);
    Py_CLEAR(tstate->curexc_value);
    Py_CLEAR(tstate->curexc_traceback);

    Py_CLEAR(tstate->exc_type);
    Py_CLEAR(tstate->exc_value);
    Py_CLEAR(tstate->exc_traceback);

    tstate->c_profilefunc = NULL;
    tstate->c_tracefunc = NULL;
    Py_CLEAR(tstate->c_profileobj);
    Py_CLEAR(tstate->c_traceobj);
}


/* Common code for PyThreadState_Delete() and PyThreadState_DeleteCurrent() */
static void
tstate_delete_common(PyThreadState *tstate)
{
    PyInterpreterState *interp;
    PyThreadState **p;
    PyThreadState *prev_p = NULL;
    if (tstate == NULL)
        Py_FatalError("PyThreadState_Delete: NULL tstate");
    interp = tstate->interp;
    if (interp == NULL)
        Py_FatalError("PyThreadState_Delete: NULL interp");
    HEAD_LOCK();
    for (p = &interp->tstate_head; ; p = &(*p)->next) {
        if (*p == NULL)
            Py_FatalError(
                "PyThreadState_Delete: invalid tstate");
        if (*p == tstate)
            break;
        /* Sanity check.  These states should never happen but if
         * they do we must abort.  Otherwise we'll end up spinning in
         * in a tight loop with the lock held.  A similar check is done
         * in thread.c find_key().  */
        if (*p == prev_p)
            Py_FatalError(
                "PyThreadState_Delete: small circular list(!)"
                " and tstate not found.");
        prev_p = *p;
        if ((*p)->next == interp->tstate_head)
            Py_FatalError(
                "PyThreadState_Delete: circular list(!) and"
                " tstate not found.");
    }
    *p = tstate->next;
    HEAD_UNLOCK();
    free(tstate);
}


void
PyThreadState_Delete(PyThreadState *tstate)
{
    if (tstate == _Py_atomic_load_relaxed(&_PyThreadState_Current))
        Py_FatalError("PyThreadState_Delete: tstate is still current");
    tstate_delete_common(tstate);
#ifdef WITH_THREAD
    if (autoInterpreterState && PyThread_get_key_value(autoTLSkey) == tstate)
        PyThread_delete_key_value(autoTLSkey);
#endif /* WITH_THREAD */
}


#ifdef WITH_THREAD
void
PyThreadState_DeleteCurrent()
{
    PyThreadState *tstate = (PyThreadState*)_Py_atomic_load_relaxed(
        &_PyThreadState_Current);
    if (tstate == NULL)
        Py_FatalError(
            "PyThreadState_DeleteCurrent: no current tstate");
    _Py_atomic_store_relaxed(&_PyThreadState_Current, NULL);
    tstate_delete_common(tstate);
    if (autoInterpreterState && PyThread_get_key_value(autoTLSkey) == tstate)
        PyThread_delete_key_value(autoTLSkey);
    PyEval_ReleaseLock();
}
#endif /* WITH_THREAD */


PyThreadState *
PyThreadState_Get(void)
{
    PyThreadState *tstate = (PyThreadState*)_Py_atomic_load_relaxed(
        &_PyThreadState_Current);
    if (tstate == NULL)
        Py_FatalError("PyThreadState_Get: no current thread");

    return tstate;
}


PyThreadState *
PyThreadState_Swap(PyThreadState *newts)
{
    PyThreadState *oldts = (PyThreadState*)_Py_atomic_load_relaxed(
        &_PyThreadState_Current);

    _Py_atomic_store_relaxed(&_PyThreadState_Current, newts);
    /* It should not be possible for more than one thread state
       to be used for a thread.  Check this the best we can in debug
       builds.
    */
#if defined(Py_DEBUG) && defined(WITH_THREAD)
    if (newts) {
        /* This can be called from PyEval_RestoreThread(). Similar
           to it, we need to ensure errno doesn't change.
        */
        int err = errno;
        PyThreadState *check = PyGILState_GetThisThreadState();
        if (check && check->interp == newts->interp && check != newts)
            Py_FatalError("Invalid thread state for this thread");
        errno = err;
    }
#endif
    return oldts;
}

/* An extension mechanism to store arbitrary additional per-thread state.
   PyThreadState_GetDict() returns a dictionary that can be used to hold such
   state; the caller should pick a unique key and store its state there.  If
   PyThreadState_GetDict() returns NULL, an exception has *not* been raised
   and the caller should assume no per-thread state is available. */

PyObject *
PyThreadState_GetDict(void)
{
    PyThreadState *tstate = (PyThreadState*)_Py_atomic_load_relaxed(
        &_PyThreadState_Current);
    if (tstate == NULL)
        return NULL;

    if (tstate->dict == NULL) {
        PyObject *d;
        tstate->dict = d = PyDict_New();
        if (d == NULL)
            PyErr_Clear();
    }
    return tstate->dict;
}


/* Asynchronously raise an exception in a thread.
   Requested by Just van Rossum and Alex Martelli.
   To prevent naive misuse, you must write your own extension
   to call this, or use ctypes.  Must be called with the GIL held.
   Returns the number of tstates modified (normally 1, but 0 if `id` didn't
   match any known thread id).  Can be called with exc=NULL to clear an
   existing async exception.  This raises no exceptions. */

int
PyThreadState_SetAsyncExc(long id, PyObject *exc) {
    PyThreadState *tstate = PyThreadState_GET();
    PyInterpreterState *interp = tstate->interp;
    PyThreadState *p;

    /* Although the GIL is held, a few C API functions can be called
     * without the GIL held, and in particular some that create and
     * destroy thread and interpreter states.  Those can mutate the
     * list of thread states we're traversing, so to prevent that we lock
     * head_mutex for the duration.
     */
    HEAD_LOCK();
    for (p = interp->tstate_head; p != NULL; p = p->next) {
        if (p->thread_id == id) {
            /* Tricky:  we need to decref the current value
             * (if any) in p->async_exc, but that can in turn
             * allow arbitrary Python code to run, including
             * perhaps calls to this function.  To prevent
             * deadlock, we need to release head_mutex before
             * the decref.
             */
            PyObject *old_exc = p->async_exc;
            Py_XINCREF(exc);
            p->async_exc = exc;
            HEAD_UNLOCK();
            Py_XDECREF(old_exc);
            _PyEval_SignalAsyncExc();
            return 1;
        }
    }
    HEAD_UNLOCK();
    return 0;
}


/* Routines for advanced debuggers, requested by David Beazley.
   Don't use unless you know what you are doing! */

PyInterpreterState *
PyInterpreterState_Head(void)
{
    return interp_head;
}

PyInterpreterState *
PyInterpreterState_Next(PyInterpreterState *interp) {
    return interp->next;
}

PyThreadState *
PyInterpreterState_ThreadHead(PyInterpreterState *interp) {
    return interp->tstate_head;
}

PyThreadState *
PyThreadState_Next(PyThreadState *tstate) {
    return tstate->next;
}

/* The implementation of sys._current_frames().  This is intended to be
   called with the GIL held, as it will be when called via
   sys._current_frames().  It's possible it would work fine even without
   the GIL held, but haven't thought enough about that.
*/
PyObject *
_PyThread_CurrentFrames(void)
{
    PyObject *result;
    PyInterpreterState *i;

    result = PyDict_New();
    if (result == NULL)
        return NULL;

    /* for i in all interpreters:
     *     for t in all of i's thread states:
     *          if t's frame isn't NULL, map t's id to its frame
     * Because these lists can mutate even when the GIL is held, we
     * need to grab head_mutex for the duration.
     */
    HEAD_LOCK();
    for (i = interp_head; i != NULL; i = i->next) {
        PyThreadState *t;
        for (t = i->tstate_head; t != NULL; t = t->next) {
            PyObject *id;
            int stat;
            struct _frame *frame = t->frame;
            if (frame == NULL)
                continue;
            id = PyLong_FromLong(t->thread_id);
            if (id == NULL)
                goto Fail;
            stat = PyDict_SetItem(result, id, (PyObject *)frame);
            Py_DECREF(id);
            if (stat < 0)
                goto Fail;
        }
    }
    HEAD_UNLOCK();
    return result;

 Fail:
    HEAD_UNLOCK();
    Py_DECREF(result);
    return NULL;
}

/* Python "auto thread state" API. */
#ifdef WITH_THREAD

/* Keep this as a static, as it is not reliable!  It can only
   ever be compared to the state for the *current* thread.
   * If not equal, then it doesn't matter that the actual
     value may change immediately after comparison, as it can't
     possibly change to the current thread's state.
   * If equal, then the current thread holds the lock, so the value can't
     change until we yield the lock.
*/
static int
PyThreadState_IsCurrent(PyThreadState *tstate)
{
    /* Must be the tstate for this thread */
    assert(PyGILState_GetThisThreadState()==tstate);
    return tstate == _Py_atomic_load_relaxed(&_PyThreadState_Current);
}

/* Internal initialization/finalization functions called by
   Py_Initialize/Py_Finalize
*/
void
_PyGILState_Init(PyInterpreterState *i, PyThreadState *t)
{
    assert(i && t); /* must init with valid states */
    autoTLSkey = PyThread_create_key();
    if (autoTLSkey == -1)
        Py_FatalError("Could not allocate TLS entry");
    autoInterpreterState = i;
    assert(PyThread_get_key_value(autoTLSkey) == NULL);
    assert(t->gilstate_counter == 0);

    _PyGILState_NoteThreadState(t);
}

void
_PyGILState_Fini(void)
{
    PyThread_delete_key(autoTLSkey);
    autoInterpreterState = NULL;
}

/* Reset the TLS key - called by PyOS_AfterFork().
 * This should not be necessary, but some - buggy - pthread implementations
 * don't reset TLS upon fork(), see issue #10517.
 */
void
_PyGILState_Reinit(void)
{
    PyThreadState *tstate = PyGILState_GetThisThreadState();
    PyThread_delete_key(autoTLSkey);
    if ((autoTLSkey = PyThread_create_key()) == -1)
        Py_FatalError("Could not allocate TLS entry");

    /* If the thread had an associated auto thread state, reassociate it with
     * the new key. */
    if (tstate && PyThread_set_key_value(autoTLSkey, (void *)tstate) < 0)
        Py_FatalError("Couldn't create autoTLSkey mapping");
}

/* When a thread state is created for a thread by some mechanism other than
   PyGILState_Ensure, it's important that the GILState machinery knows about
   it so it doesn't try to create another thread state for the thread (this is
   a better fix for SF bug #1010677 than the first one attempted).
*/
static void
_PyGILState_NoteThreadState(PyThreadState* tstate)
{
    /* If autoTLSkey isn't initialized, this must be the very first
       threadstate created in Py_Initialize().  Don't do anything for now
       (we'll be back here when _PyGILState_Init is called). */
    if (!autoInterpreterState)
        return;

    /* Stick the thread state for this thread in thread local storage.

       The only situation where you can legitimately have more than one
       thread state for an OS level thread is when there are multiple
       interpreters, when:

           a) You shouldn't really be using the PyGILState_ APIs anyway,
          and:

           b) The slightly odd way PyThread_set_key_value works (see
          comments by its implementation) means that the first thread
          state created for that given OS level thread will "win",
          which seems reasonable behaviour.
    */
    if (PyThread_set_key_value(autoTLSkey, (void *)tstate) < 0)
        Py_FatalError("Couldn't create autoTLSkey mapping");

    /* PyGILState_Release must not try to delete this thread state. */
    tstate->gilstate_counter = 1;
}

/* The public functions */
PyThreadState *
PyGILState_GetThisThreadState(void)
{
    if (autoInterpreterState == NULL)
        return NULL;
    return (PyThreadState *)PyThread_get_key_value(autoTLSkey);
}

PyGILState_STATE
PyGILState_Ensure(void)
{
    int current;
    PyThreadState *tcur;
    /* Note that we do not auto-init Python here - apart from
       potential races with 2 threads auto-initializing, pep-311
       spells out other issues.  Embedders are expected to have
       called Py_Initialize() and usually PyEval_InitThreads().
    */
    assert(autoInterpreterState); /* Py_Initialize() hasn't been called! */
    tcur = (PyThreadState *)PyThread_get_key_value(autoTLSkey);
    if (tcur == NULL) {
        /* Create a new thread state for this thread */
        tcur = PyThreadState_New(autoInterpreterState);
        if (tcur == NULL)
            Py_FatalError("Couldn't create thread-state for new thread");
        /* This is our thread state!  We'll need to delete it in the
           matching call to PyGILState_Release(). */
        tcur->gilstate_counter = 0;
        current = 0; /* new thread state is never current */
    }
    else
        current = PyThreadState_IsCurrent(tcur);
    if (current == 0)
        PyEval_RestoreThread(tcur);
    /* Update our counter in the thread-state - no need for locks:
       - tcur will remain valid as we hold the GIL.
       - the counter is safe as we are the only thread "allowed"
         to modify this value
    */
    ++tcur->gilstate_counter;
    return current ? PyGILState_LOCKED : PyGILState_UNLOCKED;
}

void
PyGILState_Release(PyGILState_STATE oldstate)
{
    PyThreadState *tcur = (PyThreadState *)PyThread_get_key_value(
                                                            autoTLSkey);
    if (tcur == NULL)
        Py_FatalError("auto-releasing thread-state, "
                      "but no thread-state for this thread");
    /* We must hold the GIL and have our thread state current */
    /* XXX - remove the check - the assert should be fine,
       but while this is very new (April 2003), the extra check
       by release-only users can't hurt.
    */
    if (! PyThreadState_IsCurrent(tcur))
        Py_FatalError("This thread state must be current when releasing");
    assert(PyThreadState_IsCurrent(tcur));
    --tcur->gilstate_counter;
    assert(tcur->gilstate_counter >= 0); /* illegal counter value */

    /* If we're going to destroy this thread-state, we must
     * clear it while the GIL is held, as destructors may run.
     */
    if (tcur->gilstate_counter == 0) {
        /* can't have been locked when we created it */
        assert(oldstate == PyGILState_UNLOCKED);
        PyThreadState_Clear(tcur);
        /* Delete the thread-state.  Note this releases the GIL too!
         * It's vital that the GIL be held here, to avoid shutdown
         * races; see bugs 225673 and 1061968 (that nasty bug has a
         * habit of coming back).
         */
        PyThreadState_DeleteCurrent();
    }
    /* Release the lock if necessary */
    else if (oldstate == PyGILState_UNLOCKED)
        PyEval_SaveThread();
}

#endif /* WITH_THREAD */

#ifdef __cplusplus
}
#endif


> 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707 2708 2709 2710 2711 2712 2713 2714 2715 2716 2717 2718 2719 2720 2721 2722 2723 2724 2725 2726 2727 2728 2729 2730 2731 2732 2733 2734 2735 2736 2737 2738 2739 2740 2741 2742 2743 2744 2745 2746 2747 2748 2749 2750 2751 2752 2753 2754 2755 2756 2757 2758 2759 2760 2761 2762 2763 2764 2765 2766 2767 2768 2769 2770 2771 2772 2773 2774 2775 2776 2777 2778 2779 2780 2781 2782 2783 2784 2785 2786 2787 2788 2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805 2806 2807 2808 2809 2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823 2824 2825 2826 2827 2828 2829 2830 2831 2832 2833 2834 2835 2836 2837 2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850 2851 2852 2853 2854 2855 2856 2857 2858 2859 2860 2861 2862 2863 2864 2865 2866 2867 2868 2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880 2881 2882 2883 2884 2885 2886 2887 2888 2889 2890 2891 2892 2893 2894 2895 2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944 2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961 2962 2963 2964 2965 2966 2967 2968 2969 2970 2971 2972 2973 2974 2975 2976 2977 2978 2979 2980 2981 2982 2983 2984 2985 2986 2987 2988 2989 2990 2991 2992 2993 2994 2995 2996 2997 2998 2999 3000 3001 3002 3003 3004 3005 3006 3007 3008 3009 3010 3011 3012 3013 3014 3015 3016 3017 3018 3019 3020 3021 3022 3023 3024 3025 3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042 3043 3044 3045 3046 3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060 3061 3062 3063 3064 3065 3066 3067 3068 3069 3070 3071 3072 3073 3074 3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087 3088 3089 3090 3091 3092 3093 3094 3095 3096 3097 3098 3099 3100 3101 3102 3103 3104 3105 3106 3107 3108 3109 3110 3111 3112 3113 3114 3115 3116 3117 3118 3119 3120 3121 3122 3123 3124 3125 3126 3127 3128 3129 3130 3131 3132 3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164 3165 3166 3167 3168 3169 3170 3171 3172 3173
/*
 * tclEncoding.c --
 *
 *	Contains the implementation of the encoding conversion package.
 *
 * Copyright (c) 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.
 *
 * RCS: @(#) $Id: tclEncoding.c,v 1.16.2.14 2007/02/12 19:25:42 andreas_kupries Exp $
 */

#include "tclInt.h"
#include "tclPort.h"

typedef size_t (LengthProc)_ANSI_ARGS_((CONST char *src));

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

typedef struct Encoding {
    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;
				/* Procedure to convert from external
				 * encoding into UTF-8. */
    Tcl_EncodingConvertProc *fromUtfProc;
				/* Procedure to convert from UTF-8 into
				 * external encoding. */
    Tcl_EncodingFreeProc *freeProc;
				/* If non-NULL, procedure to call when this
				 * encoding is deleted. */
    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 or 2. */
    ClientData clientData;	/* Arbitrary value associated with encoding
				 * type.  Passed to conversion procedures. */
    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. */
    int 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 TableEncodingData {
    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;

/*
 * 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 EscapeSubTable {
    unsigned int 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 EscapeEncodingData {
    int fallback;		/* Character (in this encoding) to
				 * substitute when this encoding cannot
				 * represent a UTF-8 character. */
    unsigned int initLen;	/* Length of following string. */
    char init[16];		/* String to emit or expect before first char
				 * in conversion. */
    unsigned int 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[1];/* Information about each EscapeSubTable
				 * used by this encoding type.  The actual 
				 * size will be 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

/*
 * Initialize the default encoding directory.  If this variable contains
 * a non NULL value, it will be the first path used to locate the
 * system encoding files.
 */

char *tclDefaultEncodingDir = 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 will be used to perform the conversion.
 */

static Tcl_Encoding defaultEncoding;
static Tcl_Encoding systemEncoding;

/*
 * 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];

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

static int		BinaryProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static void		DupEncodingIntRep _ANSI_ARGS_((Tcl_Obj *srcPtr,
			    Tcl_Obj *dupPtr));
static void		EscapeFreeProc _ANSI_ARGS_((ClientData clientData));
static int		EscapeFromUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		EscapeToUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static void		FreeEncoding _ANSI_ARGS_((Tcl_Encoding encoding));
static void		FreeEncodingIntRep _ANSI_ARGS_((Tcl_Obj *objPtr));
static Encoding *	GetTableEncoding _ANSI_ARGS_((
			    EscapeEncodingData *dataPtr, int state));
static Tcl_Encoding	LoadEncodingFile _ANSI_ARGS_((Tcl_Interp *interp,
			    CONST char *name));
static Tcl_Encoding	LoadTableEncoding _ANSI_ARGS_((Tcl_Interp *interp,
			    CONST char *name, int type, Tcl_Channel chan));
static Tcl_Encoding	LoadEscapeEncoding _ANSI_ARGS_((CONST char *name, 
			    Tcl_Channel chan));
static Tcl_Channel	OpenEncodingFile _ANSI_ARGS_((CONST char *dir,
			    CONST char *name));
static void		TableFreeProc _ANSI_ARGS_((ClientData clientData));
static int		TableFromUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		TableToUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static size_t		unilen _ANSI_ARGS_((CONST char *src));
static int		UnicodeToUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		UtfToUnicodeProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		UtfToUtfProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr, int pureNullMode));
static int		UtfIntToUtfExtProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		UtfExtToUtfIntProc _ANSI_ARGS_((ClientData clientData,
			    CONST char *src, int srcLen, int flags,
			    Tcl_EncodingState *statePtr, char *dst, int dstLen,
			    int *srcReadPtr, int *dstWrotePtr,
			    int *dstCharsPtr));
static int		TclFindEncodings _ANSI_ARGS_((CONST char *argv0));

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

static Tcl_ObjType EncodingType = {
    "encoding", FreeEncodingIntRep, DupEncodingIntRep, NULL, NULL
};


/*
 *----------------------------------------------------------------------
 *
 * TclGetEncodingFromObj --
 *
 *      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 
TclGetEncodingFromObj(interp, objPtr, encodingPtr)
    Tcl_Interp *interp;
    Tcl_Obj *objPtr;
    Tcl_Encoding *encodingPtr;
{
    CONST char *name = Tcl_GetString(objPtr);
    if (objPtr->typePtr != &EncodingType) {
	Tcl_Encoding encoding = Tcl_GetEncoding(interp, name);

	if (encoding == NULL) {
	    return TCL_ERROR;
	}
	if (objPtr->typePtr && objPtr->typePtr->freeIntRepProc) {
	    objPtr->typePtr->freeIntRepProc(objPtr);
	}
	objPtr->internalRep.otherValuePtr = (VOID *) encoding;
	objPtr->typePtr = &EncodingType;
    }
    *encodingPtr = Tcl_GetEncoding(NULL, name);
    return TCL_OK;
}

/*
 *----------------------------------------------------------------------
 *
 * FreeEncodingIntRep --
 *
 *      The Tcl_FreeInternalRepProc for the "encoding" Tcl_ObjType.
 *
 *----------------------------------------------------------------------
 */
static void
FreeEncodingIntRep(objPtr)
    Tcl_Obj *objPtr;
{
    Tcl_FreeEncoding((Tcl_Encoding) objPtr->internalRep.otherValuePtr);
}

/*
 *----------------------------------------------------------------------
 *
 * DupEncodingIntRep --
 *
 *      The Tcl_DupInternalRepProc for the "encoding" Tcl_ObjType.
 *
 *----------------------------------------------------------------------
 */
static void
DupEncodingIntRep(srcPtr, dupPtr)
    Tcl_Obj *srcPtr;
    Tcl_Obj *dupPtr;
{
    dupPtr->internalRep.otherValuePtr = (VOID *)
	    Tcl_GetEncoding(NULL, srcPtr->bytes);
}

/*
 *---------------------------------------------------------------------------
 *
 * 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.
 *
 *---------------------------------------------------------------------------
 */

void
TclInitEncodingSubsystem()
{
    Tcl_EncodingType type;

    Tcl_MutexLock(&encodingMutex);
    Tcl_InitHashTable(&encodingTable, TCL_STRING_KEYS);
    Tcl_MutexUnlock(&encodingMutex);
    
    /*
     * Create a few initial encodings.  Note that the UTF-8 to UTF-8 
     * translation is not a no-op, because it will turn a stream of
     * improperly formed UTF-8 into a properly formed stream.
     */

    type.encodingName	= "identity";
    type.toUtfProc	= BinaryProc;
    type.fromUtfProc	= BinaryProc;
    type.freeProc	= NULL;
    type.nullSize	= 1;
    type.clientData	= NULL;

    defaultEncoding	= Tcl_CreateEncoding(&type);
    systemEncoding	= Tcl_GetEncoding(NULL, type.encodingName);

    type.encodingName	= "utf-8";
    type.toUtfProc	= UtfExtToUtfIntProc;
    type.fromUtfProc	= UtfIntToUtfExtProc;
    type.freeProc	= NULL;
    type.nullSize	= 1;
    type.clientData	= NULL;
    Tcl_CreateEncoding(&type);

    type.encodingName   = "unicode";
    type.toUtfProc	= UnicodeToUtfProc;
    type.fromUtfProc    = UtfToUnicodeProc;
    type.freeProc	= NULL;
    type.nullSize	= 2;
    type.clientData	= NULL;
    Tcl_CreateEncoding(&type);
}

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

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

    Tcl_MutexLock(&encodingMutex);
    encodingsInitialized  = 0;
    FreeEncoding(systemEncoding);
    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 --
 *
 *
 * Results:
 *
 * Side effects:
 *
 *-------------------------------------------------------------------------
 */

CONST char *
Tcl_GetDefaultEncodingDir()
{
    return tclDefaultEncodingDir;
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_SetDefaultEncodingDir --
 *
 *
 * Results:
 *
 * Side effects:
 *
 *-------------------------------------------------------------------------
 */

void
Tcl_SetDefaultEncodingDir(path)
    CONST char *path;
{
    tclDefaultEncodingDir = (char *)ckalloc((unsigned) strlen(path) + 1);
    strcpy(tclDefaultEncodingDir, path);
}

/*
 *-------------------------------------------------------------------------
 *
 * 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:
 *	The new encoding type is entered into a table visible to all
 *	interpreters, keyed off the encoding's name.  For each call to
 *	this procedure, there should eventually be a call to
 *	Tcl_FreeEncoding, so that the database can be cleaned up when
 *	encodings aren't needed anymore.
 *
 *-------------------------------------------------------------------------
 */

Tcl_Encoding
Tcl_GetEncoding(interp, name)
    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 --
 *
 *	This procedure is called to release 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 may be deleted if nothing is using it anymore.
 *
 *---------------------------------------------------------------------------
 */

void
Tcl_FreeEncoding(encoding)
    Tcl_Encoding encoding;
{
    Tcl_MutexLock(&encodingMutex);
    FreeEncoding(encoding);
    Tcl_MutexUnlock(&encodingMutex);
}

/*
 *----------------------------------------------------------------------
 *
 * FreeEncoding --
 *
 *	This procedure is called to release an encoding by procedures
 *	that already have the encodingMutex.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The reference count associated with the encoding is decremented
 *	and the encoding may be deleted if nothing is using it anymore.
 *
 *----------------------------------------------------------------------
 */

static void
FreeEncoding(encoding)
    Tcl_Encoding encoding;
{
    Encoding *encodingPtr;
    
    encodingPtr = (Encoding *) encoding;
    if (encodingPtr == NULL) {
	return;
    }
    encodingPtr->refCount--;
    if (encodingPtr->refCount == 0) {
	if (encodingPtr->freeProc != NULL) {
	    (*encodingPtr->freeProc)(encodingPtr->clientData);
	}
	if (encodingPtr->hPtr != NULL) {
	    Tcl_DeleteHashEntry(encodingPtr->hPtr);
	}
	ckfree((char *) encodingPtr->name);
	ckfree((char *) encodingPtr);
    }
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_GetEncodingName --
 *
 *	Given an encoding, return the name that was used to constuct
 *	the encoding.
 *
 * Results:
 *	The name of the encoding.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

CONST char *
Tcl_GetEncodingName(encoding)
    Tcl_Encoding encoding;	/* The encoding whose name to fetch. */
{
    Encoding *encodingPtr;

    if (encoding == NULL) {
	encoding = systemEncoding;
    }
    encodingPtr = (Encoding *) encoding;
    return encodingPtr->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(interp)
    Tcl_Interp *interp;		/* Interp to hold result. */
{
    Tcl_HashSearch search;
    Tcl_HashEntry *hPtr;
    Tcl_Obj *pathPtr, *resultPtr;
    int dummy;

    Tcl_HashTable table;

    Tcl_MutexLock(&encodingMutex);
    Tcl_InitHashTable(&table, TCL_STRING_KEYS);
    hPtr = Tcl_FirstHashEntry(&encodingTable, &search);
    while (hPtr != NULL) {
	Encoding *encodingPtr;
	
	encodingPtr = (Encoding *) Tcl_GetHashValue(hPtr);
	Tcl_CreateHashEntry(&table, encodingPtr->name, &dummy);
	hPtr = Tcl_NextHashEntry(&search);
    }
    Tcl_MutexUnlock(&encodingMutex);

    pathPtr = TclGetLibraryPath();
    if (pathPtr != NULL) {
	int i, objc;
	Tcl_Obj **objv;
	char globArgString[10];
	Tcl_Obj* encodingObj = Tcl_NewStringObj("encoding",-1);
	Tcl_IncrRefCount(encodingObj);
	
	objc = 0;
	Tcl_ListObjGetElements(NULL, pathPtr, &objc, &objv);

	for (i = 0; i < objc; i++) {
	    Tcl_Obj *searchIn;
	    
	    /* 
	     * Construct the path from the element of pathPtr,
	     * joined with 'encoding'.
	     */
	    searchIn = Tcl_FSJoinToPath(objv[i],1,&encodingObj);
	    Tcl_IncrRefCount(searchIn);
	    Tcl_ResetResult(interp);

	    /*
	     * TclGlob() changes the contents of globArgString, which causes
	     * a segfault if we pass in a pointer to non-writeable memory.
	     * TclGlob() puts its results directly into interp.
	     */

	    strcpy(globArgString, "*.enc");
	    /* 
	     * The GLOBMODE_TAILS flag returns just the tail of each file
	     * which is the encoding name with a .enc extension 
	     */
	    if ((TclGlob(interp, globArgString, searchIn, 
			 TCL_GLOBMODE_TAILS, NULL) == TCL_OK)) {
		int objc2 = 0;
		Tcl_Obj **objv2;
		int j;

		Tcl_ListObjGetElements(NULL, Tcl_GetObjResult(interp), &objc2,
			&objv2);

		for (j = 0; j < objc2; j++) {
		    int length;
		    char *string;
		    string = Tcl_GetStringFromObj(objv2[j], &length);
		    length -= 4;
		    if (length > 0) {
			string[length] = '\0';
			Tcl_CreateHashEntry(&table, string, &dummy);
			string[length] = '.';
		    }
		}
	    }
	    Tcl_DecrRefCount(searchIn);
	}
	Tcl_DecrRefCount(encodingObj);
    }

    /*
     * Clear any values placed in the result by globbing.
     */

    Tcl_ResetResult(interp);
    resultPtr = Tcl_GetObjResult(interp);

    hPtr = Tcl_FirstHashEntry(&table, &search);
    while (hPtr != NULL) {
	Tcl_Obj *strPtr;

	strPtr = Tcl_NewStringObj(Tcl_GetHashKey(&table, hPtr), -1);
	Tcl_ListObjAppendElement(NULL, resultPtr, strPtr);
	hPtr = Tcl_NextHashEntry(&search);
    }
    Tcl_DeleteHashTable(&table);
}

/*
 *------------------------------------------------------------------------
 *
 * 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.  
 *
 *------------------------------------------------------------------------
 */

int
Tcl_SetSystemEncoding(interp, name)
    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 == NULL) {
	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);

    return TCL_OK;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_CreateEncoding --
 *
 *	This procedure is called to define a new encoding and the procedures
 *	that are used to convert between the specified encoding and 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 will eventually
 *	be garbage collected when the last reference to it goes away.  Any
 *	subsequent calls to Tcl_GetEncoding with the specified name will
 *	retrieve the most recent encoding token.
 *
 * Side effects:
 *	The new encoding type is entered into a table visible to all
 *	interpreters, keyed off the encoding's name.  For each call to
 *	this procedure, there should eventually be a call to
 *	Tcl_FreeEncoding, so that the database can be cleaned up when
 *	encodings aren't needed anymore.
 *
 *---------------------------------------------------------------------------
 */ 

Tcl_Encoding
Tcl_CreateEncoding(typePtr)
    Tcl_EncodingType *typePtr;	/* The encoding type. */
{
    Tcl_HashEntry *hPtr;
    int new;
    Encoding *encodingPtr;
    char *name;

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

    name = ckalloc((unsigned) strlen(typePtr->encodingName) + 1);
    
    encodingPtr = (Encoding *) ckalloc(sizeof(Encoding));
    encodingPtr->name		= strcpy(name, typePtr->encodingName);
    encodingPtr->toUtfProc	= typePtr->toUtfProc;
    encodingPtr->fromUtfProc	= typePtr->fromUtfProc;
    encodingPtr->freeProc	= typePtr->freeProc;
    encodingPtr->nullSize	= typePtr->nullSize;
    encodingPtr->clientData	= typePtr->clientData;
    if (typePtr->nullSize == 1) {
	encodingPtr->lengthProc = (LengthProc *) strlen;
    } else {
	encodingPtr->lengthProc = (LengthProc *) unilen;
    }
    encodingPtr->refCount	= 1;
    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(encoding, src, srcLen, dstPtr)
    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. */
{
    char *dst;
    Tcl_EncodingState state;
    Encoding *encodingPtr;
    int flags, dstLen, result, soFar, srcRead, dstWrote, dstChars;

    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 = TCL_ENCODING_START | TCL_ENCODING_END;
    while (1) {
	result = (*encodingPtr->toUtfProc)(encodingPtr->clientData, src,
		srcLen, flags, &state, dst, dstLen, &srcRead, &dstWrote,
		&dstChars);
	soFar = dst + dstWrote - Tcl_DStringValue(dstPtr);
	if (result != TCL_CONVERT_NOSPACE) {
	    Tcl_DStringSetLength(dstPtr, soFar);
	    return Tcl_DStringValue(dstPtr);
	}
	flags &= ~TCL_ENCODING_START;
	src += srcRead;
	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(interp, encoding, src, srcLen, flags, statePtr, dst,
	dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr)
    Tcl_Interp *interp;		/* Interp for error return, if not NULL. */
    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. */
{
    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 = (*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;
    }

    /*
     * If there are any null characters in the middle of the buffer, they will
     * converted to the UTF-8 null character (\xC080).  To get the actual 
     * \0 at the end of the destination buffer, we need to append it manually.
     */

    dstLen--;
    result = (*encodingPtr->toUtfProc)(encodingPtr->clientData, src, srcLen,
	    flags, statePtr, dst, dstLen, srcReadPtr, dstWrotePtr,
	    dstCharsPtr);
    dst[*dstWrotePtr] = '\0';
    return result;
}

/*
 *-------------------------------------------------------------------------
 *
 * Tcl_UtfToExternalDString --
 *
 *	Convert a source buffer from UTF-8 into 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 will be 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(encoding, src, srcLen, dstPtr)
    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. */
{
    char *dst;
    Tcl_EncodingState state;
    Encoding *encodingPtr;
    int flags, dstLen, result, soFar, srcRead, dstWrote, dstChars;
    
    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 = 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);
	if (result != TCL_CONVERT_NOSPACE) {
	    if (encodingPtr->nullSize == 2) {
	        Tcl_DStringSetLength(dstPtr, soFar + 1);
	    }
	    Tcl_DStringSetLength(dstPtr, soFar);
	    return Tcl_DStringValue(dstPtr);
	}
	flags &= ~TCL_ENCODING_START;
	src += srcRead;
	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(interp, encoding, src, srcLen, flags, statePtr, dst,
	dstLen, srcReadPtr, dstWrotePtr, dstCharsPtr)
    Tcl_Interp *interp;		/* Interp for error return, if not NULL. */
    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. */
{
    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);
    if (encodingPtr->nullSize == 2) {
	dst[*dstWrotePtr + 1] = '\0';
    }
    dst[*dstWrotePtr] = '\0';
    
    return result;
}

/*
 *---------------------------------------------------------------------------
 *
 * Tcl_FindExecutable --
 *
 *	This procedure computes the absolute path name of the current
 *	application, given its argv[0] value.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	The variable tclExecutableName gets filled in with the file
 *	name for the application, if we figured it out.  If we couldn't
 *	figure it out, tclExecutableName is set to NULL.
 *
 *---------------------------------------------------------------------------
 */

void
Tcl_FindExecutable(argv0)
    CONST char *argv0;		/* The value of the application's argv[0]
				 * (native). */
{
    int mustCleanUtf;
    CONST char *name;
    Tcl_DString buffer, nameString;

    TclInitSubsystems(argv0);

    if (argv0 == NULL) {
	goto done;
    }
    if (tclExecutableName != NULL) {
	ckfree(tclExecutableName);
	tclExecutableName = NULL;
    }
    if ((name = TclpFindExecutable(argv0)) == NULL) {
	goto done;
    }

    /*
     * The value returned from TclpNameOfExecutable is a UTF string that
     * is possibly dirty depending on when it was initialized.
     * TclFindEncodings will indicate whether we must "clean" the UTF (as
     * reported by the underlying system).  To assure that the UTF string
     * is a properly encoded native string for this system, convert the
     * UTF string to the default native encoding before the default
     * encoding is initialized.  Then, convert it back to UTF after the
     * system encoding is loaded.
     */
    
    Tcl_UtfToExternalDString(NULL, name, -1, &buffer);
    mustCleanUtf = TclFindEncodings(argv0);

    /*
     * Now it is OK to convert the native string back to UTF and set
     * the value of the tclExecutableName.
     */
    
    if (mustCleanUtf) {
	Tcl_ExternalToUtfDString(NULL, Tcl_DStringValue(&buffer), -1,
		&nameString);
	tclExecutableName = (char *)
	    ckalloc((unsigned) (Tcl_DStringLength(&nameString) + 1));
	strcpy(tclExecutableName, Tcl_DStringValue(&nameString));

	Tcl_DStringFree(&nameString);
    } else {
	tclExecutableName = (char *) ckalloc((unsigned) (strlen(name) + 1));
	strcpy(tclExecutableName, name);
    }
    Tcl_DStringFree(&buffer);
    return;
	
    done:
    (void) TclFindEncodings(argv0);
}

/*
 *---------------------------------------------------------------------------
 *
 * LoadEncodingFile --
 *
 *	Read a file that describes an encoding and create a new Encoding
 *	from the data.  
 *
 * Results:
 *	The return value is the newly loaded Encoding, or NULL if
 *	the file didn't exist of was in the incorrect format.  If NULL was
 *	returned, an error message is left in interp's result object,
 *	unless interp was NULL.
 *
 * Side effects:
 *	File read from disk.  
 *
 *---------------------------------------------------------------------------
 */

static Tcl_Encoding
LoadEncodingFile(interp, name)
    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. */
{
    int objc, i, ch;
    Tcl_Obj **objv;
    Tcl_Obj *pathPtr;
    Tcl_Channel chan;
    Tcl_Encoding encoding;

    pathPtr = TclGetLibraryPath();
    if (pathPtr == NULL) {
	goto unknown;
    }
    objc = 0;
    Tcl_ListObjGetElements(NULL, pathPtr, &objc, &objv);

    chan = NULL;
    for (i = 0; i < objc; i++) {
	chan = OpenEncodingFile(Tcl_GetString(objv[i]), name);
	if (chan != NULL) {
	    break;
	}
    }

    if (chan == NULL) {
	goto unknown;
    }

    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;
	}
    }

    encoding = NULL;
    switch (ch) {
	case 'S': {
	    encoding = LoadTableEncoding(interp, name, ENCODING_SINGLEBYTE,
		    chan);
	    break;
	}
	case 'D': {
	    encoding = LoadTableEncoding(interp, name, ENCODING_DOUBLEBYTE,
		    chan);
	    break;
	}
	case 'M': {
	    encoding = LoadTableEncoding(interp, name, ENCODING_MULTIBYTE,
		    chan);
	    break;
	}
	case 'E': {
	    encoding = LoadEscapeEncoding(name, chan);
	    break;
	}
    }
    if ((encoding == NULL) && (interp != NULL)) {
	Tcl_AppendResult(interp, "invalid encoding file \"", name, "\"", NULL);
	if (ch == 'E') {
	    Tcl_AppendResult(interp, " or missing sub-encoding", NULL);
	}
    }
    Tcl_Close(NULL, chan);
    return encoding;

    unknown:
    if (interp != NULL) {
	Tcl_AppendResult(interp, "unknown encoding \"", name, "\"", NULL);
    }
    return NULL;
}

/*
 *----------------------------------------------------------------------
 *
 * OpenEncodingFile --
 *
 *	Look for the file encoding/<name>.enc in the specified
 *	directory.
 *
 * Results:
 *	Returns an open file channel if the file exists.
 *
 * Side effects:
 *	None.
 *
 *----------------------------------------------------------------------
 */

static Tcl_Channel
OpenEncodingFile(dir, name)
    CONST char *dir;
    CONST char *name;

{
    CONST char *argv[3];
    Tcl_DString pathString;
    CONST char *path;
    Tcl_Channel chan;
    Tcl_Obj *pathPtr;
    
    argv[0] = dir;
    argv[1] = "encoding";
    argv[2] = name;

    Tcl_DStringInit(&pathString);
    Tcl_JoinPath(3, argv, &pathString);
    path = Tcl_DStringAppend(&pathString, ".enc", -1);
    pathPtr = Tcl_NewStringObj(path,-1);

    Tcl_IncrRefCount(pathPtr);
    chan = Tcl_FSOpenFileChannel(NULL, pathPtr, "r", 0);
    Tcl_DecrRefCount(pathPtr);

    Tcl_DStringFree(&pathString);

    return chan;
}

/*
 *-------------------------------------------------------------------------
 *
 * LoadTableEncoding --
 *
 *	Helper function for LoadEncodingTable().  Loads a table to that 
 *	converts between Unicode and some other encoding and creates an 
 *	encoding (using a TableEncoding structure) from that information.
 *
 *	File contains binary data, but begins with a marker to indicate 
 *	byte-ordering, so that same binary file can be read on either
 *	endian platforms.
 *
 * 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
LoadTableEncoding(interp, name, type, chan)
    Tcl_Interp *interp;		/* Interp for temporary obj while reading. */
    CONST char *name;		/* Name for 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;
    unsigned char used[256];
    unsigned int size;
    TableEncodingData *dataPtr;
    unsigned short *pageMemPtr;
    Tcl_EncodingType encType;

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

    static 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);
    Tcl_Gets(chan, &lineString);
    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);

    if (interp == NULL) {
	objPtr = Tcl_NewObj();
    } else {
	objPtr = Tcl_GetObjResult(interp);
    }
    for (i = 0; i < numPages; i++) {
	int ch;
	char *p;

	Tcl_ReadChars(chan, objPtr, 3 + 16 * (16 * 4 + 1), 0);
	p = Tcl_GetString(objPtr);
	hi = (staticHex[(unsigned int)p[0]] << 4) + staticHex[(unsigned int)p[1]];
	dataPtr->toUnicode[hi] = pageMemPtr;
	p += 2;
	for (lo = 0; lo < 256; lo++) {
	    if ((lo & 0x0f) == 0) {
		p++;
	    }
	    ch = (staticHex[(unsigned int)p[0]] << 12) + (staticHex[(unsigned int)p[1]] << 8)
		+ (staticHex[(unsigned int)p[2]] << 4) + staticHex[(unsigned int)p[3]];
	    if (ch != 0) {
		used[ch >> 8] = 1;
	    }
	    *pageMemPtr = (unsigned short) ch;
	    pageMemPtr++;
	    p += 4;
	}
    }
    if (interp == NULL) {
	Tcl_DecrRefCount(objPtr);
    } else {
	Tcl_ResetResult(interp);
    }
	
    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 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, we
     * remembered what pages that would be 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;
	} else {
	    for (lo = 0; lo < 256; lo++) {
		int ch;

		ch = dataPtr->toUnicode[hi][lo];
		if (ch != 0) {
		    unsigned short *page;
		    
		    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 won't work because
	 * the backslash in the file name will map to the unknown character
	 * (question mark) when converting from UTF-8 to external encoding.
	 */

	if (dataPtr->fromUnicode[0] != NULL) {
	    if (dataPtr->fromUnicode[0]['\\'] == '\0') {
		dataPtr->fromUnicode[0]['\\'] = '\\';
	    }
	}
    }
    if (symbol) {
	unsigned short *page;
	
	/*
	 * Make a special symbol encoding that not only maps the symbol
	 * characters from their Unicode code points down into page 0, but
	 * also ensure that the characters on page 0 map to themselves.
	 * This is 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);
    do {
	int len;
	/* skip leading empty lines */
	while ((len = Tcl_Gets(chan, &lineString)) == 0)
	    ;
	if (len < 0) {
	    break;
	}
	line = Tcl_DStringValue(&lineString);
	if (line[0] != 'R') {
	    break;
	}
	for (Tcl_DStringSetLength(&lineString, 0);
	     (len = Tcl_Gets(chan, &lineString)) >= 0;
	     Tcl_DStringSetLength(&lineString, 0)) {
	    unsigned char* p;
	    int to, from;
	    if (len < 5) {
		continue;
	    }
	    p = (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;
	    }
	}
    } while (0);
    Tcl_DStringFree(&lineString);

    encType.encodingName    = name;
    encType.toUtfProc	    = TableToUtfProc;
    encType.fromUtfProc	    = TableFromUtfProc;
    encType.freeProc	    = TableFreeProc;
    encType.nullSize	    = (type == ENCODING_DOUBLEBYTE) ? 2 : 1;
    encType.clientData	    = (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(name, chan)
    CONST char *name;		/* Name for new encoding. */
    Tcl_Channel chan;		/* File containing new encoding. */
{
    int i, missingSubEncoding = 0;
    unsigned int 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) {
	    continue;
	}
	if (argc >= 2) {
	    if (strcmp(argv[0], "name") == 0) {
		;
	    } 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;

		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';

		/*
		 * Load the subencodings first so we're never stuck
		 * trying to use a half-loaded system encoding to
		 * open/read a *.enc file.
		 */

		est.encodingPtr = (Encoding *) Tcl_GetEncoding(NULL, est.name);
		if ((est.encodingPtr == NULL) 
			|| (est.encodingPtr->toUtfProc != TableToUtfProc)) {
		    missingSubEncoding = 1;
		}
		Tcl_DStringAppend(&escapeData, (char *) &est, sizeof(est));
	    }
	}
	ckfree((char *) argv);
	Tcl_DStringFree(&lineString);
    }
    if (missingSubEncoding) {
	Tcl_DStringFree(&escapeData);
	return NULL;
    }

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

    type.encodingName	= name;
    type.toUtfProc	= EscapeToUtfProc;
    type.fromUtfProc    = EscapeFromUtfProc;
    type.freeProc	= EscapeFreeProc;
    type.nullSize	= 1;
    type.clientData	= (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(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* Not used. */
    CONST char *src;		/* Source string (unknown encoding). */
    int srcLen;			/* Source string length in bytes. */
    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. */
    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 (srcLen > dstLen) {
	srcLen = dstLen;
	result = TCL_CONVERT_NOSPACE;
    }

    *srcReadPtr = srcLen;
    *dstWrotePtr = srcLen;
    *dstCharsPtr = srcLen;
    memcpy((void *) dst, (void *) src, (size_t) srcLen);
    return result;
}


/*
 *-------------------------------------------------------------------------
 *
 * UtfExtToUtfIntProc --
 *
 *	Convert from UTF-8 to UTF-8. While converting null-bytes from
 *	the Tcl's internal representation (0xc0, 0x80) to the official
 *	representation (0x00). See UtfToUtfProc for details.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */
static int 
UtfIntToUtfExtProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	     srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* Not used. */
    CONST char *src;		/* Source string in UTF-8. */
    int srcLen;			/* Source string length in bytes. */
    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. */
{
    return UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
			srcReadPtr, dstWrotePtr, dstCharsPtr, 1);
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfExtToUtfIntProc --
 *
 *	Convert from UTF-8 to UTF-8 while converting null-bytes from
 *	the official representation (0x00) to Tcl's internal
 *	representation (0xc0, 0x80). See UtfToUtfProc for details.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */
static int 
UtfExtToUtfIntProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	     srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* Not used. */
    CONST char *src;		/* Source string in UTF-8. */
    int srcLen;			/* Source string length in bytes. */
    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. */
{
    return UtfToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
			srcReadPtr, dstWrotePtr, dstCharsPtr, 0);
}

/*
 *-------------------------------------------------------------------------
 *
 * UtfToUtfProc --
 *
 *	Convert from UTF-8 to UTF-8.  Note that the UTF-8 to UTF-8 
 *	translation is not a no-op, because it will turn 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(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	     srcReadPtr, dstWrotePtr, dstCharsPtr, pureNullMode)
    ClientData clientData;	/* Not used. */
    CONST char *src;		/* Source string in UTF-8. */
    int srcLen;			/* Source string length in bytes. */
    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. */
    int pureNullMode;		/* Convert embedded nulls from
				 * internal representation to real
				 * null-bytes or vice versa */

{
    CONST char *srcStart, *srcEnd, *srcClose;
    char *dstStart, *dstEnd;
    int result, numChars;
    Tcl_UniChar ch;

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

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

    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;
	}
	if (UCHAR(*src) < 0x80 &&
	    !(UCHAR(*src) == 0 && pureNullMode == 0)) {
	    /*
	     * Copy 7bit chatacters, but skip null-bytes when we are
	     * in input mode, so that they get converted to 0xc080.
	     */
	    *dst++ = *src++;
	} else if (pureNullMode == 1 &&
		   UCHAR(*src) == 0xc0 &&
		   UCHAR(*(src+1)) == 0x80) {
	    /* 
	     * Convert 0xc080 to real nulls when we are in output mode.
	     */
	    *dst++ = 0;
	    src += 2;
	} else if (!Tcl_UtfCharComplete(src, srcEnd - src)) {
	    /* Always check before using Tcl_UtfToUniChar. Not doing
	     * can so cause it run beyond the endof the buffer!  If we
	     * * happen such an incomplete char its byts are made to *
	     * represent themselves.
	     */

	    ch = (Tcl_UniChar) *src;
	    src += 1;
	    dst += Tcl_UniCharToUtf(ch, dst);
	} else {
	    src += Tcl_UtfToUniChar(src, &ch);
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
    }

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

static int 
UnicodeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* Not used. */
    CONST char *src;		/* Source string in Unicode. */
    int srcLen;			/* Source string length in bytes. */
    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 char *srcStart, *srcEnd;
    char *dstEnd, *dstStart;
    int result, numChars;
    Tcl_UniChar ch;

    result = TCL_OK;
    if ((srcLen % sizeof(Tcl_UniChar)) != 0) {
	result = TCL_CONVERT_MULTIBYTE;
	srcLen /= sizeof(Tcl_UniChar);
	srcLen *= sizeof(Tcl_UniChar);
    }

    srcStart = src;
    srcEnd = src + srcLen;

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

    for (numChars = 0; src < srcEnd; numChars++) {
	if (dst > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	    break;
	}
	/*
	 * Special case for 1-byte utf chars for speed.  Make sure we
	 * work with Tcl_UniChar-size data.
	 */
	ch = *(Tcl_UniChar *)src;
	if (ch && ch < 0x80) {
	    *dst++ = (ch & 0xFF);
	} else {
	    dst += Tcl_UniCharToUtf(ch, dst);
	}
	src += sizeof(Tcl_UniChar);
    }

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

static int 
UtfToUnicodeProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData 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_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 char *srcStart, *srcEnd, *srcClose, *dstStart, *dstEnd;
    int result, numChars;
    Tcl_UniChar ch;

    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;
        }
	src += TclUtfToUniChar(src, &ch);
	/*
	 * Need to handle this in a way that won't cause misalignment
	 * by casting dst to a Tcl_UniChar. [Bug 1122671]
	 * XXX: This hard-codes the assumed size of Tcl_UniChar as 2.
	 */
#ifdef WORDS_BIGENDIAN
	*dst++ = (ch >> 8);
	*dst++ = (ch & 0xFF);
#else
	*dst++ = (ch & 0xFF);
	*dst++ = (ch >> 8);
#endif
    }
    *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(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData 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_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 char *srcStart, *srcEnd;
    char *dstEnd, *dstStart, *prefixBytes;
    int result, byte, numChars;
    Tcl_UniChar ch;
    unsigned short **toUnicode;
    unsigned short *pageZero;
    TableEncodingData *dataPtr;
    
    srcStart = src;
    srcEnd = src + srcLen;

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

    dataPtr = (TableEncodingData *) clientData;
    toUnicode = dataPtr->toUnicode;
    prefixBytes = dataPtr->prefixBytes;
    pageZero = toUnicode[0];

    result = TCL_OK;
    for (numChars = 0; src < srcEnd; numChars++) {
        if (dst > dstEnd) {
            result = TCL_CONVERT_NOSPACE;
            break;
        }
	byte = *((unsigned char *) src);
	if (prefixBytes[byte]) {
	    src++;
	    if (src >= srcEnd) {
		src--;
		result = TCL_CONVERT_MULTIBYTE;
		break;
	    }
	    ch = toUnicode[byte][*((unsigned char *) src)];
	} else {
	    ch = pageZero[byte];
	}
	if ((ch == 0) && (byte != 0)) {
	    if (flags & TCL_ENCODING_STOPONERROR) {
		result = TCL_CONVERT_SYNTAX;
		break;
	    }
	    if (prefixBytes[byte]) {
		src--;
	    }
	    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(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData 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_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 char *srcStart, *srcEnd, *srcClose;
    char *dstStart, *dstEnd, *prefixBytes;
    Tcl_UniChar ch;
    int result, len, word, numChars;
    TableEncodingData *dataPtr;
    unsigned short **fromUnicode;
    
    result = TCL_OK;    

    dataPtr = (TableEncodingData *) clientData;
    prefixBytes = dataPtr->prefixBytes;
    fromUnicode = 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
	/*
	 * This prevents a crash condition.  More evaluation is required
	 * for full support of int Tcl_UniChar. [Bug 1004065]
	 */
	if (ch & 0xffff0000) {
	    word = 0;
	} else
#endif
	    word = fromUnicode[(ch >> 8)][ch & 0xff];

	if ((word == 0) && (ch != 0)) {
	    if (flags & TCL_ENCODING_STOPONERROR) {
		result = TCL_CONVERT_UNKNOWN;
		break;
	    }
	    word = dataPtr->fallback; 
	}
	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;
}

/*
 *---------------------------------------------------------------------------
 *
 * TableFreeProc --
 *
 *	This procedure is invoked when an encoding is deleted.  It deletes
 *	the memory used by the TableEncodingData.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Memory freed.
 *
 *---------------------------------------------------------------------------
 */

static void
TableFreeProc(clientData)
    ClientData clientData;	/* TableEncodingData that specifies
				 * encoding. */
{
    TableEncodingData *dataPtr;

    /*
     * Make sure we aren't freeing twice on shutdown.  [Bug #219314]
     */

    dataPtr = (TableEncodingData *) clientData;
    ckfree((char *) dataPtr->toUnicode);
    ckfree((char *) dataPtr->fromUnicode);
    ckfree((char *) dataPtr);
}

/*
 *-------------------------------------------------------------------------
 *
 * EscapeToUtfProc --
 *
 *	Convert from the encoding specified by the EscapeEncodingData into
 *	UTF-8.
 *
 * Results:
 *	Returns TCL_OK if conversion was successful.
 *
 * Side effects:
 *	None.
 *
 *-------------------------------------------------------------------------
 */

static int 
EscapeToUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* EscapeEncodingData that specifies
				 * encoding. */
    CONST char *src;		/* Source string in specified encoding. */
    int srcLen;			/* Source string length in bytes. */
    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. */
{
    EscapeEncodingData *dataPtr;
    char *prefixBytes, *tablePrefixBytes;
    unsigned short **tableToUnicode;
    Encoding *encodingPtr;
    int state, result, numChars;
    CONST char *srcStart, *srcEnd;
    char *dstStart, *dstEnd;

    result = TCL_OK;

    tablePrefixBytes = NULL;	/* lint. */
    tableToUnicode = NULL;	/* lint. */

    dataPtr = (EscapeEncodingData *) clientData;
    prefixBytes = dataPtr->prefixBytes;
    encodingPtr = NULL;

    srcStart = src;
    srcEnd = src + srcLen;

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

    state = (int) *statePtr;
    if (flags & TCL_ENCODING_START) {
	state = 0;
    }

    for (numChars = 0; src < srcEnd; ) {
	int byte, hi, lo, ch;

        if (dst > dstEnd) {
            result = TCL_CONVERT_NOSPACE;
            break;
        }
	byte = *((unsigned char *) src);
	if (prefixBytes[byte]) {
	    unsigned int left, len, longest;
	    int checked, i;
	    EscapeSubTable *subTablePtr;
	    
	    /*
	     * Saw the beginning of an escape sequence. 
	     */
	     
	    left = srcEnd - src;
	    len = dataPtr->initLen;
	    longest = len;
	    checked = 0;
	    if (len <= left) {
		checked++;
		if ((len > 0) && 
			(memcmp(src, dataPtr->init, len) == 0)) {
		    /*
		     * If we see initialization string, skip it, even if we're
		     * not at the beginning of the buffer. 
		     */
		     
		    src += len;
		    continue;
		}
	    }
	    len = dataPtr->finalLen;
	    if (len > longest) {
		longest = len;
	    }
	    if (len <= left) {
		checked++;
		if ((len > 0) && 
			(memcmp(src, dataPtr->final, len) == 0)) {
		    /*
		     * If we see finalization string, skip it, even if we're
		     * not at the end of the buffer. 
		     */
		     
		    src += len;
		    continue;
		}
	    }
	    subTablePtr = dataPtr->subTables;
	    for (i = 0; i < dataPtr->numSubTables; i++) {
		len = subTablePtr->sequenceLen;
		if (len > longest) {
		    longest = len;
		}
		if (len <= left) {
		    checked++;
		    if ((len > 0) && 
			    (memcmp(src, subTablePtr->sequence, len) == 0)) {
			state = i;
			encodingPtr = NULL;
			subTablePtr = NULL;
			src += len;
			break;
		    }
		}
		subTablePtr++;
	    }
	    if (subTablePtr == NULL) {
		/*
		 * A match was found, the escape sequence was consumed, and
		 * the state was updated.
		 */

		continue;
	    }

	    /*
	     * We have a split-up or unrecognized escape sequence.  If we
	     * checked all the sequences, then it's a syntax error,
	     * otherwise we need more bytes to determine a match.
	     */

	    if ((checked == dataPtr->numSubTables + 2)
		    || (flags & TCL_ENCODING_END)) {
		if ((flags & TCL_ENCODING_STOPONERROR) == 0) {
		    /*
		     * Skip the unknown escape sequence.
		     */

		    src += longest;
		    continue;
		}
		result = TCL_CONVERT_SYNTAX;
	    } else {
		result = TCL_CONVERT_MULTIBYTE;
	    }
	    break;
	}

	if (encodingPtr == NULL) {
	    TableEncodingData *tableDataPtr;

	    encodingPtr = GetTableEncoding(dataPtr, state);
	    tableDataPtr = (TableEncodingData *) encodingPtr->clientData;
	    tablePrefixBytes = tableDataPtr->prefixBytes;
	    tableToUnicode = tableDataPtr->toUnicode;
	}
	if (tablePrefixBytes[byte]) {
	    src++;
	    if (src >= srcEnd) {
		src--;
		result = TCL_CONVERT_MULTIBYTE;
		break;
	    }
	    hi = byte;
	    lo = *((unsigned char *) src);
	} else {
	    hi = 0;
	    lo = byte;
	}
	ch = tableToUnicode[hi][lo];
	dst += Tcl_UniCharToUtf(ch, dst);
	src++;
	numChars++;
    }

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

static int 
EscapeFromUtfProc(clientData, src, srcLen, flags, statePtr, dst, dstLen,
	srcReadPtr, dstWrotePtr, dstCharsPtr)
    ClientData clientData;	/* EscapeEncodingData that specifies
				 * encoding. */
    CONST char *src;		/* Source string in UTF-8. */
    int srcLen;			/* Source string length in bytes. */
    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. */
{
    EscapeEncodingData *dataPtr;
    Encoding *encodingPtr;
    CONST char *srcStart, *srcEnd, *srcClose;
    char *dstStart, *dstEnd;
    int state, result, numChars;
    TableEncodingData *tableDataPtr;
    char *tablePrefixBytes;
    unsigned short **tableFromUnicode;
    
    result = TCL_OK;    

    dataPtr = (EscapeEncodingData *) clientData;

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

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

    /*
     * RFC1468 states that the text starts in ASCII, and switches to Japanese
     * characters, and that the text must end in ASCII. [Patch #474358]
     */

    if (flags & TCL_ENCODING_START) {
	state = 0;
	if ((dst + dataPtr->initLen) > dstEnd) {
	    *srcReadPtr = 0;
	    *dstWrotePtr = 0;
	    return TCL_CONVERT_NOSPACE;
	}
	memcpy((VOID *) dst, (VOID *) dataPtr->init,
		(size_t) dataPtr->initLen);
	dst += dataPtr->initLen;
    } else {
        state = (int) *statePtr;
    }

    encodingPtr = GetTableEncoding(dataPtr, state);
    tableDataPtr = (TableEncodingData *) encodingPtr->clientData;
    tablePrefixBytes = tableDataPtr->prefixBytes;
    tableFromUnicode = tableDataPtr->fromUnicode;

    for (numChars = 0; src < srcEnd; numChars++) {
	unsigned int len;
	int word;
	Tcl_UniChar ch;
	
	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);
	word = tableFromUnicode[(ch >> 8)][ch & 0xff];

	if ((word == 0) && (ch != 0)) {
	    int oldState;
	    EscapeSubTable *subTablePtr;
	    
	    oldState = state;
	    for (state = 0; state < dataPtr->numSubTables; state++) {
		encodingPtr = GetTableEncoding(dataPtr, state);
		tableDataPtr = (TableEncodingData *) encodingPtr->clientData;
	    	word = tableDataPtr->fromUnicode[(ch >> 8)][ch & 0xff];
		if (word != 0) {
		    break;
		}
	    }

	    if (word == 0) {
		state = oldState;
		if (flags & TCL_ENCODING_STOPONERROR) {
		    result = TCL_CONVERT_UNKNOWN;
		    break;
		}
		encodingPtr = GetTableEncoding(dataPtr, state);
		tableDataPtr = (TableEncodingData *) encodingPtr->clientData;
		word = tableDataPtr->fallback;
	    } 
	    
	    tablePrefixBytes = tableDataPtr->prefixBytes;
	    tableFromUnicode = tableDataPtr->fromUnicode;

	    /*
	     * The state variable has the value of oldState when word is 0.
	     * In this case, the escape sequense should not be copied to dst 
	     * because the current character set is not changed.
	     */
	    if (state != oldState) {
		subTablePtr = &dataPtr->subTables[state];
		if ((dst + subTablePtr->sequenceLen) > dstEnd) {
		    /*
		     * If there is no space to write the escape sequence, the
		     * state variable must be changed to the value of oldState
		     * variable because this escape sequence must be written
		     * in the next conversion.
		     */
		    state = oldState;
		    result = TCL_CONVERT_NOSPACE;
		    break;
		}
		memcpy((VOID *) dst, (VOID *) subTablePtr->sequence,
			(size_t) subTablePtr->sequenceLen);
		dst += subTablePtr->sequenceLen;
	    }
	}

	if (tablePrefixBytes[(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;
    }

    if ((result == TCL_OK) && (flags & TCL_ENCODING_END)) {
	unsigned int len = dataPtr->subTables[0].sequenceLen;
	/*
	 * [Bug 1516109].
	 * Certain encodings like iso2022-jp need to write
	 * an escape sequence after all characters have
	 * been converted. This logic checks that enough
	 * room is available in the buffer for the escape bytes.
	 * The TCL_ENCODING_END flag is cleared after a final
	 * escape sequence has been added to the buffer so
	 * that another call to this method does not attempt
	 * to append escape bytes a second time.
	 */
	if ((dst + dataPtr->finalLen + (state?len:0)) > dstEnd) {
	    result = TCL_CONVERT_NOSPACE;
	} else {
	    if (state) {
		memcpy((VOID *) dst, (VOID *) dataPtr->subTables[0].sequence,
			(size_t) len);
		dst += len;
	    }
	    memcpy((VOID *) dst, (VOID *) dataPtr->final,
		    (size_t) dataPtr->finalLen);
	    dst += dataPtr->finalLen;
	    state &= ~TCL_ENCODING_END;
	}
    }

    *statePtr = (Tcl_EncodingState) state;
    *srcReadPtr = src - srcStart;
    *dstWrotePtr = dst - dstStart;
    *dstCharsPtr = numChars;
    return result;
}

/*
 *---------------------------------------------------------------------------
 *
 * EscapeFreeProc --
 *
 *	This procedure is invoked when an EscapeEncodingData encoding is 
 *	deleted.  It deletes the memory used by the encoding.
 *
 * Results:
 *	None.
 *
 * Side effects:
 *	Memory freed.
 *
 *---------------------------------------------------------------------------
 */

static void
EscapeFreeProc(clientData)
    ClientData clientData;	/* EscapeEncodingData that specifies encoding. */
{
    EscapeEncodingData *dataPtr;
    EscapeSubTable *subTablePtr;
    int i;

    dataPtr = (EscapeEncodingData *) clientData;
    if (dataPtr == NULL) {
	return;
    }
    subTablePtr = dataPtr->subTables;
    for (i = 0; i < dataPtr->numSubTables; i++) {
	FreeEncoding((Tcl_Encoding) subTablePtr->encodingPtr);
	subTablePtr++;
    }
    ckfree((char *) dataPtr);
}

/*
 *---------------------------------------------------------------------------
 *
 * GetTableEncoding --
 *
 *	Helper function for the EscapeEncodingData conversions.  Gets the
 *	encoding (of type TextEncodingData) that represents the specified
 *	state.
 *
 * Results:
 *	The return value is the encoding.
 *
 * Side effects:
 *	If the encoding that represents the specified state has not
 *	already been used by this EscapeEncoding, it will be loaded
 *	and cached in the dataPtr.
 *
 *---------------------------------------------------------------------------
 */

static Encoding *
GetTableEncoding(dataPtr, state)
    EscapeEncodingData *dataPtr;/* Contains names of encodings. */
    int state;			/* Index in dataPtr of desired Encoding. */
{
    EscapeSubTable *subTablePtr;
    Encoding *encodingPtr;
    
    subTablePtr = &dataPtr->subTables[state];
    encodingPtr = subTablePtr->encodingPtr;
    if (encodingPtr == NULL) {
	/*
	 * Now that escape encodings load their sub-encodings first, and
	 * fail to load if any sub-encodings are missing, this branch should
	 * never happen.  
	 */
	encodingPtr = (Encoding *) Tcl_GetEncoding(NULL, subTablePtr->name);
	if ((encodingPtr == NULL) 
		|| (encodingPtr->toUtfProc != TableToUtfProc)) {
	    panic("EscapeToUtfProc: invalid sub table");
	}
	subTablePtr->encodingPtr = encodingPtr;
    }
    return encodingPtr;
}

/*
 *---------------------------------------------------------------------------
 *
 * unilen --
 *
 *	A helper function for the Tcl_ExternalToUtf functions.  This
 *	function is similar to strlen for double-byte characters: it
 *	returns the number of bytes in a 0x0000 terminated string.
 *
 * Results:
 *	As above.
 *
 * Side effects:
 *	None.
 *
 *---------------------------------------------------------------------------
 */

static size_t
unilen(src)
    CONST char *src;
{
    unsigned short *p;

    p = (unsigned short *) src;
    while (*p != 0x0000) {
	p++;
    }
    return (char *) p - src;
}

/*
 *-------------------------------------------------------------------------
 *
 * TclFindEncodings --
 *
 *	Find and load the encoding file for this operating system.
 *	Before this is called, Tcl makes assumptions about the
 *	native string representation, but the true encoding is not
 *	assured.
 *
 * Results:
 *	Return result of TclpInitLibraryPath, which reports whether the
 *	path is clean (0) or dirty (1) UTF.
 *
 * Side effects:
 *	Varied, see the respective initialization routines.
 *
 *-------------------------------------------------------------------------
 */

static int
TclFindEncodings(argv0)
    CONST char *argv0;		/* Name of executable from argv[0] to main()
				 * in native multi-byte encoding. */
{
    int mustCleanUtf = 0;

    if (encodingsInitialized == 0) {
	/* 
	 * Double check inside the mutex.  There may be calls
	 * back into this routine from some of the procedures below.
	 */

	TclpInitLock();
	if (encodingsInitialized == 0) {
	    char *native;
	    Tcl_Obj *pathPtr;
	    Tcl_DString libPath, buffer;

	    /*
	     * Have to set this bit here to avoid deadlock with the
	     * routines below us that call into TclInitSubsystems.
	     */

	    encodingsInitialized = 1;

	    native = TclpFindExecutable(argv0);
	    mustCleanUtf = TclpInitLibraryPath(native);

	    /*
	     * The library path was set in the TclpInitLibraryPath routine.
	     * The string set is a dirty UTF string.  To preserve the value
	     * convert the UTF string back to native before setting the new
	     * default encoding.
	     */

	    pathPtr = TclGetLibraryPath();
	    if ((pathPtr != NULL) && mustCleanUtf) {
		Tcl_UtfToExternalDString(NULL, Tcl_GetString(pathPtr), -1,
			&libPath);
	    }

	    TclpSetInitialEncodings();

	    /*
	     * Now convert the native string back to UTF.
	     */

	    if ((pathPtr != NULL) && mustCleanUtf) {
		Tcl_ExternalToUtfDString(NULL, Tcl_DStringValue(&libPath), -1,
			&buffer);
		pathPtr = Tcl_NewStringObj(Tcl_DStringValue(&buffer), -1);
		TclSetLibraryPath(pathPtr);

		Tcl_DStringFree(&libPath);
		Tcl_DStringFree(&buffer);
	    }
	}
	TclpInitUnlock();
    }

    return mustCleanUtf;
}