summaryrefslogtreecommitdiffstats
path: root/Lib/test/mapping_tests.py
blob: bc12c7756f2d2c77280078be78f2dc8ba9a66112 (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
# tests common to dict and UserDict
import unittest
import collections


class BasicTestMappingProtocol(unittest.TestCase):
    # This base class can be used to check that an object conforms to the
    # mapping protocol

    # Functions that can be useful to override to adapt to dictionary
    # semantics
    type2test = None # which class is being tested (overwrite in subclasses)

    def _reference(self):
        """Return a dictionary of values which are invariant by storage
        in the object under test."""
        return {"1": "2", "key1":"value1", "key2":(1,2,3)}
    def _empty_mapping(self):
        """Return an empty mapping object"""
        return self.type2test()
    def _full_mapping(self, data):
        """Return a mapping object with the value contained in data
        dictionary"""
        x = self._empty_mapping()
        for key, value in data.items():
            x[key] = value
        return x

    def __init__(self, *args, **kw):
        unittest.TestCase.__init__(self, *args, **kw)
        self.reference = self._reference().copy()

        # A (key, value) pair not in the mapping
        key, value = self.reference.popitem()
        self.other = {key:value}

        # A (key, value) pair in the mapping
        key, value = self.reference.popitem()
        self.inmapping = {key:value}
        self.reference[key] = value

    def test_read(self):
        # Test for read only operations on mapping
        p = self._empty_mapping()
        p1 = dict(p) #workaround for singleton objects
        d = self._full_mapping(self.reference)
        if d is p:
            p = p1
        #Indexing
        for key, value in self.reference.items():
            self.assertEqual(d[key], value)
        knownkey = list(self.other.keys())[0]
        self.assertRaises(KeyError, lambda:d[knownkey])
        #len
        self.assertEqual(len(p), 0)
        self.assertEqual(len(d), len(self.reference))
        #__contains__
        for k in self.reference:
            self.assertIn(k, d)
        for k in self.other:
            self.assertNotIn(k, d)
        #cmp
        self.assertEqual(p, p)
        self.assertEqual(d, d)
        self.assertNotEqual(p, d)
        self.assertNotEqual(d, p)
        #__non__zero__
        if p: self.fail("Empty mapping must compare to False")
        if not d: self.fail("Full mapping must compare to True")
        # keys(), items(), iterkeys() ...
        def check_iterandlist(iter, lst, ref):
            self.assertTrue(hasattr(iter, '__next__'))
            self.assertTrue(hasattr(iter, '__iter__'))
            x = list(iter)
            self.assertTrue(set(x)==set(lst)==set(ref))
        check_iterandlist(iter(d.keys()), list(d.keys()),
                          self.reference.keys())
        check_iterandlist(iter(d), list(d.keys()), self.reference.keys())
        check_iterandlist(iter(d.values()), list(d.values()),
                          self.reference.values())
        check_iterandlist(iter(d.items()), list(d.items()),
                          self.reference.items())
        #get
        key, value = next(iter(d.items()))
        knownkey, knownvalue = next(iter(self.other.items()))
        self.assertEqual(d.get(key, knownvalue), value)
        self.assertEqual(d.get(knownkey, knownvalue), knownvalue)
        self.assertNotIn(knownkey, d)

    def test_write(self):
        # Test for write operations on mapping
        p = self._empty_mapping()
        #Indexing
        for key, value in self.reference.items():
            p[key] = value
            self.assertEqual(p[key], value)
        for key in self.reference.keys():
            del p[key]
            self.assertRaises(KeyError, lambda:p[key])
        p = self._empty_mapping()
        #update
        p.update(self.reference)
        self.assertEqual(dict(p), self.reference)
        items = list(p.items())
        p = self._empty_mapping()
        p.update(items)
        self.assertEqual(dict(p), self.reference)
        d = self._full_mapping(self.reference)
        #setdefault
        key, value = next(iter(d.items()))
        knownkey, knownvalue = next(iter(self.other.items()))
        self.assertEqual(d.setdefault(key, knownvalue), value)
        self.assertEqual(d[key], value)
        self.assertEqual(d.setdefault(knownkey, knownvalue), knownvalue)
        self.assertEqual(d[knownkey], knownvalue)
        #pop
        self.assertEqual(d.pop(knownkey), knownvalue)
        self.assertNotIn(knownkey, d)
        self.assertRaises(KeyError, d.pop, knownkey)
        default = 909
        d[knownkey] = knownvalue
        self.assertEqual(d.pop(knownkey, default), knownvalue)
        self.assertNotIn(knownkey, d)
        self.assertEqual(d.pop(knownkey, default), default)
        #popitem
        key, value = d.popitem()
        self.assertNotIn(key, d)
        self.assertEqual(value, self.reference[key])
        p=self._empty_mapping()
        self.assertRaises(KeyError, p.popitem)

    def test_constructor(self):
        self.assertEqual(self._empty_mapping(), self._empty_mapping())

    def test_bool(self):
        self.assertTrue(not self._empty_mapping())
        self.assertTrue(self.reference)
        self.assertTrue(bool(self._empty_mapping()) is False)
        self.assertTrue(bool(self.reference) is True)

    def test_keys(self):
        d = self._empty_mapping()
        self.assertEqual(list(d.keys()), [])
        d = self.reference
        self.assertIn(list(self.inmapping.keys())[0], d.keys())
        self.assertNotIn(list(self.other.keys())[0], d.keys())
        self.assertRaises(TypeError, d.keys, None)

    def test_values(self):
        d = self._empty_mapping()
        self.assertEqual(list(d.values()), [])

        self.assertRaises(TypeError, d.values, None)

    def test_items(self):
        d = self._empty_mapping()
        self.assertEqual(list(d.items()), [])

        self.assertRaises(TypeError, d.items, None)

    def test_len(self):
        d = self._empty_mapping()
        self.assertEqual(len(d), 0)

    def test_getitem(self):
        d = self.reference
        self.assertEqual(d[list(self.inmapping.keys())[0]],
                         list(self.inmapping.values())[0])

        self.assertRaises(TypeError, d.__getitem__)

    def test_update(self):
        # mapping argument
        d = self._empty_mapping()
        d.update(self.other)
        self.assertEqual(list(d.items()), list(self.other.items()))

        # No argument
        d = self._empty_mapping()
        d.update()
        self.assertEqual(d, self._empty_mapping())

        # item sequence
        d = self._empty_mapping()
        d.update(self.other.items())
        self.assertEqual(list(d.items()), list(self.other.items()))

        # Iterator
        d = self._empty_mapping()
        d.update(self.other.items())
        self.assertEqual(list(d.items()), list(self.other.items()))

        # FIXME: Doesn't work with UserDict
        # self.assertRaises((TypeError, AttributeError), d.update, None)
        self.assertRaises((TypeError, AttributeError), d.update, 42)

        outerself = self
        class SimpleUserDict:
            def __init__(self):
                self.d = outerself.reference
            def keys(self):
                return self.d.keys()
            def __getitem__(self, i):
                return self.d[i]
        d.clear()
        d.update(SimpleUserDict())
        i1 = sorted(d.items())
        i2 = sorted(self.reference.items())
        self.assertEqual(i1, i2)

        class Exc(Exception): pass

        d = self._empty_mapping()
        class FailingUserDict:
            def keys(self):
                raise Exc
        self.assertRaises(Exc, d.update, FailingUserDict())

        d.clear()

        class FailingUserDict:
            def keys(self):
                class BogonIter:
                    def __init__(self):
                        self.i = 1
                    def __iter__(self):
                        return self
                    def __next__(self):
                        if self.i:
                            self.i = 0
                            return 'a'
                        raise Exc
                return BogonIter()
            def __getitem__(self, key):
                return key
        self.assertRaises(Exc, d.update, FailingUserDict())

        class FailingUserDict:
            def keys(self):
                class BogonIter:
                    def __init__(self):
                        self.i = ord('a')
                    def __iter__(self):
                        return self
                    def __next__(self):
                        if self.i <= ord('z'):
                            rtn = chr(self.i)
                            self.i += 1
                            return rtn
                        raise StopIteration
                return BogonIter()
            def __getitem__(self, key):
                raise Exc
        self.assertRaises(Exc, d.update, FailingUserDict())

        d = self._empty_mapping()
        class badseq(object):
            def __iter__(self):
                return self
            def __next__(self):
                raise Exc()

        self.assertRaises(Exc, d.update, badseq())

        self.assertRaises(ValueError, d.update, [(1, 2, 3)])

    # no test_fromkeys or test_copy as both os.environ and selves don't support it

    def test_get(self):
        d = self._empty_mapping()
        self.assertTrue(d.get(list(self.other.keys())[0]) is None)
        self.assertEqual(d.get(list(self.other.keys())[0], 3), 3)
        d = self.reference
        self.assertTrue(d.get(list(self.other.keys())[0]) is None)
        self.assertEqual(d.get(list(self.other.keys())[0], 3), 3)
        self.assertEqual(d.get(list(self.inmapping.keys())[0]),
                         list(self.inmapping.values())[0])
        self.assertEqual(d.get(list(self.inmapping.keys())[0], 3),
                         list(self.inmapping.values())[0])
        self.assertRaises(TypeError, d.get)
        self.assertRaises(TypeError, d.get, None, None, None)

    def test_setdefault(self):
        d = self._empty_mapping()
        self.assertRaises(TypeError, d.setdefault)

    def test_popitem(self):
        d = self._empty_mapping()
        self.assertRaises(KeyError, d.popitem)
        self.assertRaises(TypeError, d.popitem, 42)

    def test_pop(self):
        d = self._empty_mapping()
        k, v = list(self.inmapping.items())[0]
        d[k] = v
        self.assertRaises(KeyError, d.pop, list(self.other.keys())[0])

        self.assertEqual(d.pop(k), v)
        self.assertEqual(len(d), 0)

        self.assertRaises(KeyError, d.pop, k)


class TestMappingProtocol(BasicTestMappingProtocol):
    def test_constructor(self):
        BasicTestMappingProtocol.test_constructor(self)
        self.assertTrue(self._empty_mapping() is not self._empty_mapping())
        self.assertEqual(self.type2test(x=1, y=2), {"x": 1, "y": 2})

    def test_bool(self):
        BasicTestMappingProtocol.test_bool(self)
        self.assertTrue(not self._empty_mapping())
        self.assertTrue(self._full_mapping({"x": "y"}))
        self.assertTrue(bool(self._empty_mapping()) is False)
        self.assertTrue(bool(self._full_mapping({"x": "y"})) is True)

    def test_keys(self):
        BasicTestMappingProtocol.test_keys(self)
        d = self._empty_mapping()
        self.assertEqual(list(d.keys()), [])
        d = self._full_mapping({'a': 1, 'b': 2})
        k = d.keys()
        self.assertIn('a', k)
        self.assertIn('b', k)
        self.assertNotIn('c', k)

    def test_values(self):
        BasicTestMappingProtocol.test_values(self)
        d = self._full_mapping({1:2})
        self.assertEqual(list(d.values()), [2])

    def test_items(self):
        BasicTestMappingProtocol.test_items(self)

        d = self._full_mapping({1:2})
        self.assertEqual(list(d.items()), [(1, 2)])

    def test_contains(self):
        d = self._empty_mapping()
        self.assertNotIn('a', d)
        self.assertTrue(not ('a' in d))
        self.assertTrue('a' not in d)
        d = self._full_mapping({'a': 1, 'b': 2})
        self.assertIn('a', d)
        self.assertIn('b', d)
        self.assertNotIn('c', d)

        self.assertRaises(TypeError, d.__contains__)

    def test_len(self):
        BasicTestMappingProtocol.test_len(self)
        d = self._full_mapping({'a': 1, 'b': 2})
        self.assertEqual(len(d), 2)

    def test_getitem(self):
        BasicTestMappingProtocol.test_getitem(self)
        d = self._full_mapping({'a': 1, 'b': 2})
        self.assertEqual(d['a'], 1)
        self.assertEqual(d['b'], 2)
        d['c'] = 3
        d['a'] = 4
        self.assertEqual(d['c'], 3)
        self.assertEqual(d['a'], 4)
        del d['b']
        self.assertEqual(d, {'a': 4, 'c': 3})

        self.assertRaises(TypeError, d.__getitem__)

    def test_clear(self):
        d = self._full_mapping({1:1, 2:2, 3:3})
        d.clear()
        self.assertEqual(d, {})

        self.assertRaises(TypeError, d.clear, None)

    def test_update(self):
        BasicTestMappingProtocol.test_update(self)
        # mapping argument
        d = self._empty_mapping()
        d.update({1:100})
        d.update({2:20})
        d.update({1:1, 2:2, 3:3})
        self.assertEqual(d, {1:1, 2:2, 3:3})

        # no argument
        d.update()
        self.assertEqual(d, {1:1, 2:2, 3:3})

        # keyword arguments
        d = self._empty_mapping()
        d.update(x=100)
        d.update(y=20)
        d.update(x=1, y=2, z=3)
        self.assertEqual(d, {"x":1, "y":2, "z":3})

        # item sequence
        d = self._empty_mapping()
        d.update([("x", 100), ("y", 20)])
        self.assertEqual(d, {"x":100, "y":20})

        # Both item sequence and keyword arguments
        d = self._empty_mapping()
        d.update([("x", 100), ("y", 20)], x=1, y=2)
        self.assertEqual(d, {"x":1, "y":2})

        # iterator
        d = self._full_mapping({1:3, 2:4})
        d.update(self._full_mapping({1:2, 3:4, 5:6}).items())
        self.assertEqual(d, {1:2, 2:4, 3:4, 5:6})

        class SimpleUserDict:
            def __init__(self):
                self.d = {1:1, 2:2, 3:3}
            def keys(self):
                return self.d.keys()
            def __getitem__(self, i):
                return self.d[i]
        d.clear()
        d.update(SimpleUserDict())
        self.assertEqual(d, {1:1, 2:2, 3:3})

    def test_fromkeys(self):
        self.assertEqual(self.type2test.fromkeys('abc'), {'a':None, 'b':None, 'c':None})
        d = self._empty_mapping()
        self.assertTrue(not(d.fromkeys('abc') is d))
        self.assertEqual(d.fromkeys('abc'), {'a':None, 'b':None, 'c':None})
        self.assertEqual(d.fromkeys((4,5),0), {4:0, 5:0})
        self.assertEqual(d.fromkeys([]), {})
        def g():
            yield 1
        self.assertEqual(d.fromkeys(g()), {1:None})
        self.assertRaises(TypeError, {}.fromkeys, 3)
        class dictlike(self.type2test): pass
        self.assertEqual(dictlike.fromkeys('a'), {'a':None})
        self.assertEqual(dictlike().fromkeys('a'), {'a':None})
        self.assertTrue(dictlike.fromkeys('a').__class__ is dictlike)
        self.assertTrue(dictlike().fromkeys('a').__class__ is dictlike)
        self.assertTrue(type(dictlike.fromkeys('a')) is dictlike)
        class mydict(self.type2test):
            def __new__(cls):
                return collections.UserDict()
        ud = mydict.fromkeys('ab')
        self.assertEqual(ud, {'a':None, 'b':None})
        self.assertIsInstance(ud, collections.UserDict)
        self.assertRaises(TypeError, dict.fromkeys)

        class Exc(Exception): pass

        class baddict1(self.type2test):
            def __init__(self):
                raise Exc()

        self.assertRaises(Exc, baddict1.fromkeys, [1])

        class BadSeq(object):
            def __iter__(self):
                return self
            def __next__(self):
                raise Exc()

        self.assertRaises(Exc, self.type2test.fromkeys, BadSeq())

        class baddict2(self.type2test):
            def __setitem__(self, key, value):
                raise Exc()

        self.assertRaises(Exc, baddict2.fromkeys, [1])

    def test_copy(self):
        d = self._full_mapping({1:1, 2:2, 3:3})
        self.assertEqual(d.copy(), {1:1, 2:2, 3:3})
        d = self._empty_mapping()
        self.assertEqual(d.copy(), d)
        self.assertIsInstance(d.copy(), d.__class__)
        self.assertRaises(TypeError, d.copy, None)

    def test_get(self):
        BasicTestMappingProtocol.test_get(self)
        d = self._empty_mapping()
        self.assertTrue(d.get('c') is None)
        self.assertEqual(d.get('c', 3), 3)
        d = self._full_mapping({'a' : 1, 'b' : 2})
        self.assertTrue(d.get('c') is None)
        self.assertEqual(d.get('c', 3), 3)
        self.assertEqual(d.get('a'), 1)
        self.assertEqual(d.get('a', 3), 1)

    def test_setdefault(self):
        BasicTestMappingProtocol.test_setdefault(self)
        d = self._empty_mapping()
        self.assertTrue(d.setdefault('key0') is None)
        d.setdefault('key0', [])
        self.assertTrue(d.setdefault('key0') is None)
        d.setdefault('key', []).append(3)
        self.assertEqual(d['key'][0], 3)
        d.setdefault('key', []).append(4)
        self.assertEqual(len(d['key']), 2)

    def test_popitem(self):
        BasicTestMappingProtocol.test_popitem(self)
        for copymode in -1, +1:
            # -1: b has same structure as a
            # +1: b is a.copy()
            for log2size in range(12):
                size = 2**log2size
                a = self._empty_mapping()
                b = self._empty_mapping()
                for i in range(size):
                    a[repr(i)] = i
                    if copymode < 0:
                        b[repr(i)] = i
                if copymode > 0:
                    b = a.copy()
                for i in range(size):
                    ka, va = ta = a.popitem()
                    self.assertEqual(va, int(ka))
                    kb, vb = tb = b.popitem()
                    self.assertEqual(vb, int(kb))
                    self.assertTrue(not(copymode < 0 and ta != tb))
                self.assertTrue(not a)
                self.assertTrue(not b)

    def test_pop(self):
        BasicTestMappingProtocol.test_pop(self)

        # Tests for pop with specified key
        d = self._empty_mapping()
        k, v = 'abc', 'def'

        self.assertEqual(d.pop(k, v), v)
        d[k] = v
        self.assertEqual(d.pop(k, 1), v)


class TestHashMappingProtocol(TestMappingProtocol):

    def test_getitem(self):
        TestMappingProtocol.test_getitem(self)
        class Exc(Exception): pass

        class BadEq(object):
            def __eq__(self, other):
                raise Exc()
            def __hash__(self):
                return 24

        d = self._empty_mapping()
        d[BadEq()] = 42
        self.assertRaises(KeyError, d.__getitem__, 23)

        class BadHash(object):
            fail = False
            def __hash__(self):
                if self.fail:
                    raise Exc()
                else:
                    return 42

        d = self._empty_mapping()
        x = BadHash()
        d[x] = 42
        x.fail = True
        self.assertRaises(Exc, d.__getitem__, x)

    def test_fromkeys(self):
        TestMappingProtocol.test_fromkeys(self)
        class mydict(self.type2test):
            def __new__(cls):
                return collections.UserDict()
        ud = mydict.fromkeys('ab')
        self.assertEqual(ud, {'a':None, 'b':None})
        self.assertIsInstance(ud, collections.UserDict)

    def test_pop(self):
        TestMappingProtocol.test_pop(self)

        class Exc(Exception): pass

        class BadHash(object):
            fail = False
            def __hash__(self):
                if self.fail:
                    raise Exc()
                else:
                    return 42

        d = self._empty_mapping()
        x = BadHash()
        d[x] = 42
        x.fail = True
        self.assertRaises(Exc, d.pop, x)

    def test_mutatingiteration(self):
        d = self._empty_mapping()
        d[1] = 1
        try:
            for i in d:
                d[i+1] = 1
        except RuntimeError:
            pass
        else:
            self.fail("changing dict size during iteration doesn't raise Error")

    def test_repr(self):
        d = self._empty_mapping()
        self.assertEqual(repr(d), '{}')
        d[1] = 2
        self.assertEqual(repr(d), '{1: 2}')
        d = self._empty_mapping()
        d[1] = d
        self.assertEqual(repr(d), '{1: {...}}')

        class Exc(Exception): pass

        class BadRepr(object):
            def __repr__(self):
                raise Exc()

        d = self._full_mapping({1: BadRepr()})
        self.assertRaises(Exc, repr, d)

    def test_eq(self):
        self.assertEqual(self._empty_mapping(), self._empty_mapping())
        self.assertEqual(self._full_mapping({1: 2}),
                         self._full_mapping({1: 2}))

        class Exc(Exception): pass

        class BadCmp(object):
            def __eq__(self, other):
                raise Exc()
            def __hash__(self):
                return 1

        d1 = self._full_mapping({BadCmp(): 1})
        d2 = self._full_mapping({1: 1})
        self.assertRaises(Exc, lambda: BadCmp()==1)
        self.assertRaises(Exc, lambda: d1==d2)

    def test_setdefault(self):
        TestMappingProtocol.test_setdefault(self)

        class Exc(Exception): pass

        class BadHash(object):
            fail = False
            def __hash__(self):
                if self.fail:
                    raise Exc()
                else:
                    return 42

        d = self._empty_mapping()
        x = BadHash()
        d[x] = 42
        x.fail = True
        self.assertRaises(Exc, d.setdefault, x, [])
> 3743 3744 3745 3746 3747 3748 3749 3750 3751 3752 3753 3754 3755 3756 3757 3758 3759 3760 3761 3762 3763 3764 3765 3766 3767 3768 3769 3770 3771 3772 3773 3774 3775 3776 3777 3778 3779 3780 3781 3782 3783 3784 3785 3786 3787 3788 3789 3790 3791 3792 3793 3794 3795 3796 3797 3798 3799 3800 3801 3802 3803 3804 3805 3806 3807 3808 3809 3810 3811 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3822 3823 3824 3825 3826 3827 3828 3829 3830 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3867 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3885 3886 3887 3888 3889 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3912 3913 3914 3915 3916 3917 3918 3919 3920 3921 3922 3923 3924 3925 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3947 3948 3949 3950 3951 3952 3953 3954 3955 3956 3957 3958 3959 3960 3961 3962 3963 3964 3965 3966 3967 3968 3969 3970 3971 3972 3973 3974 3975 3976 3977 3978 3979 3980 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3991 3992 3993 3994 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4014 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4030 4031 4032 4033 4034 4035 4036 4037 4038 4039 4040 4041 4042 4043 4044 4045 4046 4047 4048 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4076 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4087 4088 4089 4090 4091 4092 4093 4094 4095 4096 4097 4098 4099 4100 4101 4102 4103 4104 4105 4106 4107 4108 4109 4110 4111 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4132 4133 4134 4135 4136 4137 4138 4139 4140 4141 4142 4143 4144 4145 4146 4147 4148 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4159 4160 4161 4162 4163 4164 4165 4166 4167 4168 4169 4170 4171 4172 4173 4174 4175 4176 4177 4178 4179 4180 4181 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4206 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4228 4229 4230 4231 4232 4233 4234 4235 4236 4237 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4263 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4278 4279 4280 4281 4282 4283 4284 4285 4286 4287 4288 4289 4290 4291 4292 4293 4294 4295 4296 4297 4298 4299 4300 4301 4302 4303 4304 4305 4306 4307 4308 4309 4310 4311 4312 4313 4314 4315 4316 4317 4318 4319 4320 4321 4322 4323 4324 4325 4326 4327 4328 4329 4330 4331 4332 4333 4334 4335 4336 4337 4338 4339 4340 4341 4342 4343 4344 4345 4346 4347 4348 4349 4350 4351 4352 4353 4354 4355 4356 4357 4358 4359 4360 4361 4362 4363 4364 4365 4366 4367 4368 4369 4370 4371 4372 4373 4374 4375 4376 4377 4378 4379 4380 4381 4382 4383 4384 4385 4386 4387 4388 4389 4390 4391 4392 4393 4394 4395 4396 4397 4398 4399 4400 4401 4402 4403 4404 4405 4406 4407 4408 4409 4410 4411 4412 4413 4414 4415 4416 4417 4418 4419 4420 4421 4422 4423 4424 4425 4426 4427 4428 4429 4430 4431 4432 4433 4434 4435 4436 4437 4438 4439 4440 4441 4442 4443 4444 4445 4446 4447 4448 4449 4450 4451 4452 4453 4454 4455 4456 4457 4458 4459 4460 4461 4462 4463 4464 4465 4466 4467 4468 4469 4470 4471 4472 4473 4474 4475 4476 4477 4478 4479 4480 4481 4482 4483 4484 4485 4486 4487 4488 4489 4490 4491 4492 4493 4494 4495 4496 4497 4498 4499 4500 4501 4502 4503 4504 4505 4506 4507 4508 4509 4510 4511 4512 4513 4514 4515 4516 4517 4518 4519 4520 4521 4522 4523 4524 4525 4526 4527 4528 4529 4530 4531 4532 4533 4534 4535 4536 4537 4538 4539 4540 4541 4542 4543 4544 4545 4546 4547 4548 4549 4550 4551 4552 4553 4554 4555 4556 4557 4558 4559 4560 4561 4562 4563 4564 4565 4566 4567 4568 4569 4570 4571 4572 4573 4574 4575 4576 4577 4578 4579 4580 4581 4582 4583 4584 4585 4586 4587 4588 4589 4590 4591 4592 4593 4594 4595 4596 4597 4598 4599 4600 4601 4602 4603 4604 4605 4606 4607 4608 4609 4610 4611 4612 4613 4614 4615 4616 4617 4618 4619 4620 4621 4622 4623 4624 4625 4626 4627 4628 4629 4630 4631 4632 4633 4634 4635 4636 4637 4638 4639 4640 4641 4642 4643 4644 4645 4646 4647 4648 4649 4650 4651 4652 4653 4654 4655 4656 4657 4658 4659 4660 4661 4662 4663 4664 4665 4666 4667 4668 4669 4670 4671 4672 4673 4674 4675 4676 4677 4678 4679 4680 4681 4682 4683 4684 4685 4686 4687 4688 4689 4690 4691 4692 4693 4694 4695 4696 4697 4698 4699 4700 4701 4702 4703 4704 4705 4706 4707 4708 4709 4710 4711 4712 4713 4714 4715 4716 4717 4718 4719 4720 4721 4722 4723 4724 4725 4726 4727 4728 4729 4730 4731 4732 4733 4734 4735 4736 4737 4738 4739 4740 4741 4742 4743 4744 4745 4746 4747 4748 4749 4750 4751 4752 4753 4754 4755 4756 4757 4758 4759 4760 4761 4762 4763 4764 4765 4766 4767 4768 4769 4770 4771 4772 4773 4774 4775 4776 4777 4778 4779 4780 4781 4782 4783 4784 4785 4786 4787 4788 4789 4790 4791 4792 4793 4794 4795 4796 4797 4798 4799 4800 4801 4802 4803 4804 4805 4806 4807 4808 4809 4810 4811 4812 4813 4814 4815 4816 4817 4818 4819 4820 4821 4822 4823 4824 4825 4826 4827 4828 4829 4830 4831 4832 4833 4834 4835 4836 4837 4838 4839 4840 4841 4842 4843 4844 4845 4846 4847 4848 4849 4850 4851 4852 4853 4854 4855 4856 4857 4858 4859 4860 4861 4862 4863 4864 4865 4866 4867 4868 4869 4870 4871 4872 4873 4874 4875 4876 4877 4878 4879 4880 4881 4882 4883 4884 4885 4886 4887 4888 4889 4890 4891 4892 4893 4894 4895 4896 4897 4898 4899 4900 4901 4902 4903 4904 4905 4906 4907 4908 4909 4910 4911 4912 4913 4914 4915 4916 4917 4918 4919 4920 4921 4922 4923 4924 4925 4926 4927 4928 4929 4930 4931 4932 4933 4934 4935 4936 4937 4938 4939 4940 4941 4942 4943 4944 4945 4946 4947 4948 4949 4950 4951 4952 4953 4954 4955 4956 4957 4958 4959 4960 4961 4962 4963 4964 4965 4966 4967 4968 4969 4970 4971 4972 4973 4974 4975 4976 4977 4978 4979 4980 4981 4982 4983 4984 4985 4986 4987 4988 4989 4990 4991 4992 4993 4994 4995 4996 4997 4998 4999 5000 5001 5002 5003 5004 5005 5006 5007 5008 5009 5010 5011 5012 5013 5014 5015 5016 5017 5018 5019 5020 5021 5022 5023 5024 5025 5026 5027 5028 5029 5030 5031 5032 5033 5034 5035 5036 5037 5038 5039 5040 5041 5042 5043 5044 5045 5046 5047 5048 5049 5050 5051 5052 5053 5054 5055 5056 5057 5058 5059 5060 5061 5062 5063 5064 5065 5066 5067 5068 5069 5070 5071 5072 5073 5074 5075 5076 5077 5078 5079 5080 5081 5082 5083 5084 5085 5086 5087 5088 5089 5090 5091 5092 5093 5094 5095 5096 5097 5098 5099 5100 5101 5102 5103 5104 5105 5106 5107 5108 5109 5110 5111 5112 5113 5114 5115 5116 5117 5118 5119 5120 5121 5122 5123 5124 5125 5126 5127 5128 5129 5130 5131 5132 5133 5134 5135 5136 5137 5138 5139 5140 5141 5142 5143 5144 5145 5146 5147 5148 5149 5150 5151 5152 5153 5154 5155 5156 5157 5158 5159 5160 5161 5162 5163 5164 5165 5166 5167 5168 5169 5170 5171 5172 5173 5174 5175 5176 5177 5178 5179 5180 5181 5182 5183 5184 5185 5186 5187 5188 5189 5190 5191 5192 5193 5194 5195 5196 5197 5198 5199 5200 5201 5202 5203 5204 5205 5206 5207 5208 5209 5210 5211 5212 5213 5214 5215 5216 5217 5218 5219 5220 5221 5222 5223 5224 5225 5226 5227 5228 5229 5230 5231 5232 5233 5234 5235 5236 5237 5238 5239 5240 5241 5242 5243 5244 5245 5246 5247 5248 5249 5250 5251 5252 5253 5254 5255 5256 5257 5258 5259 5260 5261 5262 5263 5264 5265 5266 5267 5268 5269 5270 5271 5272 5273 5274 5275 5276 5277 5278 5279 5280 5281 5282 5283 5284 5285 5286 5287 5288 5289 5290 5291 5292 5293 5294 5295 5296 5297 5298 5299 5300 5301 5302 5303 5304 5305 5306 5307 5308 5309 5310 5311 5312 5313 5314 5315 5316 5317 5318 5319 5320 5321 5322 5323 5324 5325 5326 5327 5328 5329 5330 5331 5332 5333 5334 5335 5336 5337 5338 5339 5340 5341 5342 5343 5344 5345 5346 5347 5348 5349 5350 5351 5352 5353 5354 5355 5356 5357 5358 5359 5360 5361 5362 5363 5364 5365 5366 5367 5368 5369 5370 5371 5372 5373 5374 5375 5376 5377 5378 5379 5380 5381 5382 5383 5384 5385 5386 5387 5388 5389 5390 5391 5392 5393 5394 5395 5396 5397 5398 5399 5400 5401 5402 5403 5404 5405 5406 5407 5408 5409 5410 5411 5412 5413 5414 5415 5416 5417 5418 5419 5420 5421 5422 5423 5424 5425 5426 5427 5428 5429 5430 5431 5432 5433 5434 5435 5436 5437 5438 5439 5440 5441 5442 5443 5444 5445 5446 5447 5448 5449 5450 5451 5452 5453 5454 5455 5456 5457 5458 5459 5460 5461 5462 5463 5464 5465 5466 5467 5468 5469 5470 5471 5472 5473 5474 5475 5476 5477 5478 5479 5480 5481 5482 5483 5484 5485 5486 5487 5488 5489 5490 5491 5492 5493 5494 5495 5496 5497 5498 5499 5500 5501 5502 5503 5504 5505 5506 5507 5508 5509 5510 5511 5512 5513 5514 5515 5516 5517 5518 5519 5520 5521 5522 5523 5524 5525 5526 5527 5528 5529 5530 5531 5532 5533 5534 5535 5536 5537 5538 5539 5540 5541 5542 5543 5544 5545 5546 5547 5548 5549 5550 5551 5552 5553 5554 5555 5556 5557 5558 5559 5560 5561 5562 5563 5564 5565 5566 5567 5568 5569 5570 5571 5572 5573 5574 5575 5576 5577 5578 5579 5580 5581 5582 5583 5584 5585 5586 5587 5588 5589 5590 5591 5592 5593 5594 5595 5596 5597 5598 5599 5600 5601 5602 5603 5604 5605 5606 5607 5608 5609 5610 5611 5612 5613 5614 5615 5616 5617 5618 5619 5620 5621 5622 5623 5624 5625 5626 5627 5628 5629 5630 5631 5632 5633 5634 5635 5636 5637 5638 5639 5640 5641 5642 5643 5644 5645 5646 5647 5648 5649 5650 5651 5652 5653 5654 5655 5656 5657 5658 5659 5660 5661 5662 5663 5664 5665 5666 5667 5668 5669 5670 5671 5672 5673 5674 5675 5676 5677 5678 5679 5680 5681 5682 5683 5684 5685 5686 5687 5688 5689 5690 5691 5692 5693 5694 5695 5696 5697 5698 5699 5700 5701 5702 5703 5704 5705 5706 5707 5708 5709 5710 5711 5712 5713 5714 5715 5716 5717 5718 5719 5720 5721 5722 5723 5724 5725 5726 5727 5728 5729 5730 5731 5732 5733 5734 5735 5736 5737 5738 5739 5740 5741 5742 5743 5744 5745 5746 5747 5748 5749 5750 5751 5752 5753 5754 5755 5756 5757 5758 5759 5760 5761 5762 5763 5764 5765 5766 5767 5768 5769 5770 5771 5772 5773 5774 5775 5776 5777 5778 5779 5780 5781 5782 5783 5784 5785 5786 5787 5788 5789 5790 5791 5792 5793 5794 5795 5796 5797 5798 5799 5800 5801 5802 5803 5804 5805 5806 5807 5808 5809 5810 5811 5812 5813 5814 5815 5816 5817 5818 5819 5820 5821 5822 5823 5824 5825 5826 5827 5828 5829 5830 5831 5832 5833 5834 5835 5836 5837 5838 5839 5840 5841 5842 5843 5844 5845 5846 5847 5848 5849 5850 5851 5852 5853 5854 5855 5856 5857 5858 5859 5860 5861 5862 5863 5864 5865 5866 5867 5868 5869 5870 5871 5872 5873 5874 5875 5876 5877 5878 5879 5880 5881 5882 5883 5884 5885 5886 5887 5888 5889 5890 5891 5892 5893 5894 5895 5896 5897 5898 5899 5900 5901 5902 5903 5904 5905 5906 5907 5908 5909 5910 5911 5912 5913 5914 5915 5916 5917 5918 5919 5920 5921 5922 5923 5924 5925 5926 5927 5928 5929 5930 5931 5932 5933 5934 5935 5936 5937 5938 5939 5940 5941 5942 5943 5944 5945 5946 5947 5948 5949 5950 5951 5952 5953 5954 5955 5956 5957 5958 5959 5960 5961 5962 5963 5964 5965 5966 5967 5968 5969 5970 5971 5972 5973 5974 5975 5976 5977 5978 5979 5980 5981 5982 5983 5984 5985 5986 5987 5988 5989 5990 5991 5992 5993 5994 5995 5996 5997 5998 5999 6000 6001 6002 6003 6004 6005 6006 6007 6008 6009 6010 6011 6012 6013 6014 6015 6016 6017 6018 6019 6020 6021 6022 6023 6024 6025 6026 6027 6028 6029 6030 6031 6032 6033 6034 6035 6036 6037 6038 6039 6040 6041 6042 6043 6044 6045 6046 6047 6048 6049 6050 6051 6052 6053 6054 6055 6056 6057 6058 6059 6060 6061 6062 6063 6064 6065 6066 6067 6068 6069 6070 6071 6072 6073 6074 6075 6076 6077 6078 6079 6080 6081 6082 6083 6084 6085 6086 6087 6088 6089 6090 6091 6092 6093 6094 6095 6096 6097 6098 6099 6100 6101 6102 6103 6104 6105 6106 6107 6108 6109 6110 6111 6112 6113 6114 6115 6116 6117 6118 6119 6120 6121 6122 6123 6124 6125 6126 6127 6128 6129 6130 6131 6132 6133 6134 6135 6136 6137 6138 6139 6140 6141 6142 6143 6144 6145 6146 6147 6148 6149 6150 6151 6152 6153 6154 6155 6156 6157 6158 6159 6160 6161 6162 6163 6164 6165 6166 6167 6168 6169 6170 6171 6172 6173 6174 6175 6176 6177 6178 6179 6180 6181 6182 6183 6184 6185 6186 6187 6188 6189 6190 6191 6192 6193 6194 6195 6196 6197 6198 6199 6200 6201 6202 6203 6204 6205 6206 6207 6208 6209 6210 6211 6212 6213 6214 6215 6216 6217 6218 6219 6220 6221 6222 6223 6224 6225 6226 6227 6228 6229 6230 6231 6232 6233 6234 6235 6236 6237 6238 6239 6240 6241 6242 6243 6244 6245 6246 6247 6248 6249 6250 6251 6252 6253 6254 6255 6256 6257 6258 6259 6260 6261 6262 6263 6264 6265 6266 6267 6268 6269 6270 6271 6272 6273 6274 6275 6276 6277 6278 6279 6280 6281 6282 6283 6284 6285 6286 6287 6288 6289 6290 6291 6292 6293 6294 6295 6296 6297 6298 6299 6300 6301 6302 6303 6304 6305 6306 6307 6308 6309 6310 6311 6312 6313 6314 6315 6316 6317 6318 6319 6320 6321 6322 6323 6324 6325 6326 6327 6328 6329 6330 6331 6332 6333 6334 6335 6336 6337 6338 6339 6340 6341 6342 6343 6344 6345 6346 6347 6348 6349 6350 6351 6352 6353 6354 6355 6356 6357 6358 6359 6360 6361 6362 6363 6364 6365 6366 6367 6368 6369 6370 6371 6372 6373 6374 6375 6376 6377 6378 6379 6380 6381 6382 6383 6384 6385 6386 6387 6388 6389 6390 6391 6392 6393 6394 6395 6396 6397 6398 6399 6400 6401 6402 6403 6404 6405 6406 6407 6408 6409 6410 6411 6412 6413 6414 6415 6416 6417 6418 6419 6420 6421 6422 6423 6424 6425 6426 6427 6428 6429 6430 6431 6432 6433 6434 6435 6436 6437 6438 6439 6440 6441 6442 6443 6444 6445 6446 6447 6448 6449 6450 6451 6452 6453 6454 6455 6456 6457 6458 6459 6460 6461 6462 6463 6464 6465 6466 6467 6468 6469 6470 6471 6472 6473 6474 6475 6476 6477 6478 6479 6480 6481 6482 6483 6484 6485 6486 6487 6488 6489 6490 6491 6492 6493 6494 6495 6496 6497 6498 6499 6500 6501 6502 6503 6504 6505 6506 6507 6508 6509 6510 6511 6512 6513 6514 6515 6516 6517 6518 6519 6520 6521 6522 6523 6524 6525 6526 6527 6528 6529 6530 6531 6532 6533 6534 6535 6536 6537 6538 6539 6540 6541 6542 6543 6544 6545 6546 6547 6548 6549 6550 6551 6552 6553 6554 6555 6556 6557 6558 6559 6560 6561 6562 6563 6564 6565 6566 6567 6568 6569 6570 6571 6572 6573 6574 6575 6576 6577 6578 6579 6580 6581 6582 6583 6584 6585 6586 6587 6588 6589 6590 6591 6592 6593 6594 6595 6596 6597 6598 6599 6600 6601 6602 6603 6604 6605 6606 6607 6608 6609 6610 6611 6612 6613 6614 6615 6616 6617 6618 6619 6620 6621 6622 6623 6624 6625 6626 6627 6628 6629 6630 6631 6632 6633 6634 6635 6636 6637 6638 6639 6640 6641 6642 6643 6644 6645 6646 6647 6648 6649 6650 6651 6652 6653 6654 6655 6656 6657 6658 6659 6660 6661 6662 6663 6664 6665 6666 6667 6668 6669 6670 6671 6672 6673 6674 6675 6676 6677 6678 6679 6680 6681 6682 6683 6684 6685 6686 6687 6688 6689 6690 6691 6692 6693 6694 6695 6696 6697 6698 6699 6700 6701 6702 6703 6704 6705 6706 6707 6708 6709 6710 6711 6712 6713 6714 6715 6716 6717 6718 6719 6720 6721 6722 6723 6724 6725 6726 6727 6728 6729 6730 6731 6732 6733 6734 6735 6736 6737 6738 6739 6740 6741 6742 6743 6744 6745 6746 6747 6748 6749 6750 6751 6752 6753 6754 6755 6756 6757 6758 6759 6760 6761 6762 6763 6764 6765 6766 6767 6768 6769 6770 6771 6772 6773 6774 6775 6776 6777 6778 6779 6780 6781 6782 6783 6784 6785 6786 6787 6788 6789 6790 6791 6792 6793 6794 6795 6796 6797 6798 6799 6800 6801 6802 6803 6804 6805 6806 6807 6808 6809 6810 6811 6812 6813 6814 6815 6816 6817 6818 6819 6820 6821 6822 6823 6824 6825 6826 6827 6828 6829 6830 6831 6832 6833 6834 6835 6836 6837 6838 6839 6840 6841 6842 6843 6844 6845 6846 6847 6848 6849 6850 6851 6852 6853 6854 6855 6856 6857 6858 6859 6860 6861 6862 6863 6864 6865 6866 6867 6868 6869 6870 6871 6872 6873 6874 6875 6876 6877 6878 6879 6880 6881 6882 6883 6884 6885 6886 6887 6888 6889 6890 6891 6892 6893 6894 6895 6896 6897 6898 6899 6900 6901 6902 6903 6904 6905 6906 6907 6908 6909 6910 6911 6912 6913 6914 6915 6916 6917 6918 6919 6920 6921 6922 6923 6924 6925 6926 6927 6928 6929 6930 6931 6932 6933 6934 6935 6936 6937 6938 6939 6940 6941 6942 6943 6944 6945 6946 6947 6948 6949 6950 6951 6952 6953 6954 6955 6956 6957 6958 6959 6960 6961 6962 6963 6964 6965 6966 6967 6968 6969 6970 6971 6972 6973 6974 6975 6976 6977 6978 6979 6980 6981 6982 6983 6984 6985 6986 6987 6988 6989 6990 6991 6992 6993 6994 6995 6996 6997 6998 6999 7000 7001 7002 7003 7004 7005 7006 7007 7008 7009 7010 7011 7012 7013 7014 7015 7016 7017 7018 7019 7020 7021 7022 7023 7024 7025 7026 7027 7028 7029 7030 7031 7032 7033 7034 7035 7036 7037 7038 7039 7040 7041 7042 7043 7044 7045 7046 7047 7048 7049 7050 7051 7052 7053 7054 7055 7056 7057 7058 7059 7060 7061 7062 7063 7064 7065 7066 7067 7068 7069 7070 7071 7072 7073 7074 7075 7076 7077 7078 7079 7080 7081 7082 7083 7084 7085 7086 7087 7088 7089 7090 7091 7092 7093 7094 7095 7096 7097 7098 7099 7100 7101 7102 7103 7104 7105 7106 7107 7108 7109 7110 7111 7112 7113 7114 7115 7116 7117 7118 7119 7120 7121 7122 7123 7124 7125 7126 7127 7128 7129 7130 7131 7132 7133 7134 7135 7136 7137 7138 7139 7140 7141 7142 7143 7144 7145 7146 7147 7148 7149 7150 7151 7152 7153 7154 7155 7156 7157 7158 7159 7160 7161 7162 7163 7164 7165 7166 7167 7168 7169 7170 7171 7172 7173 7174 7175 7176 7177 7178 7179 7180 7181 7182 7183 7184 7185 7186 7187 7188 7189 7190 7191 7192 7193 7194 7195 7196 7197 7198 7199 7200 7201 7202 7203 7204 7205 7206 7207 7208 7209 7210 7211 7212 7213 7214 7215 7216 7217 7218 7219 7220 7221 7222 7223 7224 7225 7226 7227 7228 7229 7230 7231 7232 7233 7234 7235 7236 7237 7238 7239 7240 7241 7242 7243 7244 7245 7246 7247 7248 7249 7250 7251 7252 7253 7254 7255 7256 7257 7258 7259 7260 7261 7262 7263 7264 7265 7266 7267 7268 7269 7270 7271 7272 7273 7274 7275 7276 7277 7278 7279 7280 7281 7282 7283 7284 7285 7286 7287 7288 7289 7290 7291 7292 7293 7294 7295 7296 7297 7298 7299 7300 7301 7302 7303 7304 7305 7306 7307 7308 7309 7310 7311 7312 7313 7314 7315 7316 7317 7318 7319 7320 7321 7322 7323 7324 7325 7326 7327 7328 7329 7330 7331 7332 7333 7334 7335 7336 7337 7338 7339 7340 7341 7342 7343 7344 7345 7346 7347 7348 7349 7350 7351 7352 7353 7354 7355 7356 7357 7358 7359 7360 7361 7362 7363 7364 7365 7366 7367 7368 7369 7370 7371 7372 7373 7374 7375 7376 7377 7378 7379 7380 7381 7382 7383 7384 7385 7386 7387 7388 7389 7390 7391 7392 7393 7394 7395 7396 7397 7398 7399 7400 7401 7402 7403 7404 7405 7406 7407 7408 7409 7410 7411 7412 7413 7414 7415 7416 7417 7418 7419 7420 7421 7422 7423 7424 7425 7426 7427 7428 7429 7430 7431 7432 7433 7434 7435 7436 7437 7438 7439 7440 7441 7442 7443 7444 7445 7446 7447 7448 7449 7450 7451 7452 7453 7454 7455 7456 7457 7458 7459 7460 7461 7462 7463 7464 7465 7466 7467 7468 7469 7470 7471 7472 7473 7474 7475 7476 7477 7478 7479 7480 7481 7482 7483 7484 7485 7486 7487 7488 7489 7490 7491 7492 7493 7494 7495 7496 7497 7498 7499 7500 7501 7502 7503 7504 7505 7506 7507 7508 7509 7510 7511 7512 7513 7514 7515 7516 7517 7518 7519 7520 7521 7522 7523 7524 7525 7526 7527 7528 7529 7530 7531 7532 7533 7534 7535 7536 7537 7538 7539 7540 7541 7542 7543 7544 7545 7546 7547 7548 7549 7550 7551 7552 7553 7554 7555 7556 7557 7558 7559 7560 7561 7562 7563 7564 7565 7566 7567 7568 7569 7570 7571 7572 7573 7574 7575 7576 7577 7578 7579 7580 7581 7582 7583 7584 7585 7586 7587 7588 7589 7590 7591 7592 7593 7594 7595 7596 7597 7598 7599 7600 7601 7602 7603 7604 7605 7606 7607 7608 7609 7610 7611 7612 7613 7614 7615 7616 7617 7618 7619 7620 7621 7622 7623 7624 7625 7626 7627 7628 7629 7630 7631 7632 7633 7634 7635 7636 7637 7638 7639 7640 7641 7642 7643 7644 7645 7646 7647 7648 7649 7650 7651 7652 7653 7654 7655 7656 7657 7658 7659 7660 7661 7662 7663 7664 7665 7666 7667 7668 7669 7670 7671 7672 7673 7674 7675 7676 7677 7678 7679 7680 7681 7682 7683 7684 7685 7686 7687 7688 7689 7690 7691 7692 7693 7694 7695 7696 7697 7698 7699 7700 7701 7702 7703 7704 7705 7706 7707 7708 7709 7710 7711 7712 7713 7714 7715 7716 7717 7718 7719 7720 7721 7722 7723 7724 7725 7726 7727 7728 7729 7730 7731 7732 7733 7734 7735 7736 7737 7738 7739 7740 7741 7742 7743 7744 7745 7746 7747 7748 7749 7750 7751 7752 7753 7754 7755 7756 7757 7758 7759 7760 7761 7762 7763 7764 7765 7766 7767 7768 7769 7770 7771 7772 7773 7774 7775 7776 7777 7778 7779 7780 7781 7782 7783 7784 7785 7786 7787 7788 7789 7790 7791 7792 7793 7794 7795 7796 7797 7798 7799 7800 7801 7802 7803 7804 7805 7806 7807 7808 7809 7810 7811 7812 7813 7814 7815 7816 7817 7818 7819 7820 7821 7822 7823 7824 7825 7826 7827 7828 7829 7830 7831 7832 7833 7834 7835 7836 7837 7838 7839 7840 7841 7842 7843 7844 7845 7846 7847 7848 7849 7850 7851 7852 7853 7854 7855 7856 7857 7858 7859 7860 7861 7862 7863 7864 7865 7866 7867 7868 7869 7870 7871 7872 7873 7874 7875 7876 7877 7878 7879 7880 7881 7882 7883 7884 7885 7886 7887 7888 7889 7890 7891 7892 7893 7894 7895 7896 7897 7898 7899 7900 7901 7902 7903 7904 7905 7906 7907 7908 7909 7910 7911 7912 7913 7914 7915 7916 7917 7918 7919 7920 7921 7922 7923 7924 7925 7926 7927 7928 7929 7930 7931 7932 7933 7934 7935 7936 7937 7938 7939 7940 7941 7942 7943 7944 7945 7946 7947 7948 7949 7950 7951 7952 7953 7954 7955 7956 7957 7958 7959 7960 7961 7962 7963 7964 7965 7966 7967 7968 7969 7970 7971 7972 7973 7974 7975 7976 7977 7978 7979 7980 7981 7982 7983 7984 7985 7986 7987 7988 7989 7990 7991 7992 7993 7994 7995 7996 7997 7998 7999 8000 8001 8002 8003 8004 8005 8006 8007 8008 8009 8010 8011 8012 8013 8014 8015 8016 8017 8018 8019 8020 8021 8022 8023 8024 8025 8026 8027 8028 8029 8030 8031 8032 8033 8034 8035 8036 8037 8038 8039 8040 8041 8042 8043 8044 8045 8046 8047 8048 8049 8050 8051 8052 8053 8054 8055 8056 8057 8058 8059 8060 8061 8062 8063 8064 8065 8066 8067 8068 8069 8070 8071 8072 8073 8074 8075 8076 8077 8078 8079 8080 8081 8082 8083 8084 8085 8086 8087 8088 8089 8090 8091 8092 8093 8094 8095 8096 8097 8098 8099 8100 8101 8102 8103 8104 8105 8106 8107 8108 8109 8110 8111 8112 8113 8114 8115 8116 8117 8118 8119 8120 8121 8122 8123 8124 8125 8126 8127 8128 8129 8130 8131 8132 8133 8134 8135 8136 8137 8138 8139 8140 8141 8142 8143 8144 8145 8146 8147 8148 8149 8150 8151 8152 8153 8154 8155 8156 8157 8158 8159 8160 8161 8162 8163 8164 8165 8166 8167 8168 8169 8170 8171 8172 8173 8174 8175 8176 8177 8178 8179 8180 8181 8182 8183 8184 8185 8186 8187 8188 8189 8190 8191 8192 8193 8194 8195 8196 8197 8198 8199 8200 8201 8202 8203 8204 8205 8206 8207 8208 8209 8210 8211 8212 8213 8214 8215 8216 8217 8218 8219 8220 8221 8222 8223 8224 8225 8226 8227 8228 8229 8230 8231 8232 8233 8234 8235 8236 8237 8238 8239 8240 8241 8242 8243 8244 8245 8246 8247 8248 8249 8250 8251 8252 8253 8254 8255 8256 8257 8258 8259 8260 8261 8262 8263 8264 8265 8266 8267 8268 8269 8270 8271 8272 8273 8274 8275 8276 8277 8278 8279 8280 8281 8282 8283 8284 8285 8286 8287 8288 8289 8290 8291 8292 8293 8294 8295 8296 8297 8298 8299 8300 8301 8302 8303 8304 8305 8306 8307 8308 8309 8310 8311 8312 8313 8314 8315 8316 8317 8318 8319 8320 8321 8322 8323 8324 8325 8326 8327 8328 8329 8330 8331 8332 8333 8334 8335 8336 8337 8338 8339 8340 8341 8342 8343 8344 8345 8346 8347 8348 8349 8350 8351 8352 8353 8354 8355 8356 8357 8358 8359 8360 8361 8362 8363 8364 8365 8366 8367 8368 8369 8370 8371 8372 8373 8374 8375 8376 8377 8378 8379 8380 8381 8382 8383 8384 8385 8386 8387 8388 8389 8390 8391 8392 8393 8394 8395 8396 8397 8398 8399 8400 8401 8402 8403 8404 8405 8406 8407 8408 8409 8410 8411 8412 8413 8414 8415 8416 8417 8418 8419 8420 8421 8422 8423 8424 8425 8426 8427 8428 8429 8430 8431 8432 8433 8434 8435 8436 8437 8438 8439 8440 8441 8442 8443 8444 8445 8446 8447 8448 8449 8450 8451 8452 8453 8454 8455 8456 8457 8458 8459 8460 8461 8462 8463 8464 8465 8466 8467 8468 8469 8470 8471 8472 8473 8474 8475 8476 8477 8478 8479 8480 8481 8482 8483 8484 8485 8486 8487 8488 8489 8490 8491 8492 8493 8494 8495 8496 8497 8498 8499 8500 8501 8502 8503 8504 8505 8506 8507 8508 8509 8510 8511 8512 8513 8514 8515 8516 8517 8518 8519 8520 8521 8522 8523 8524 8525 8526 8527 8528 8529 8530 8531 8532 8533 8534 8535 8536 8537 8538 8539 8540 8541 8542 8543 8544 8545 8546 8547 8548 8549 8550 8551 8552 8553 8554 8555 8556 8557 8558 8559 8560 8561 8562 8563 8564 8565 8566 8567 8568 8569 8570 8571 8572 8573 8574 8575 8576 8577 8578 8579 8580 8581 8582 8583 8584 8585 8586 8587 8588 8589 8590 8591 8592 8593 8594 8595 8596 8597 8598 8599 8600 8601 8602 8603 8604 8605 8606 8607 8608 8609 8610 8611 8612 8613 8614 8615 8616 8617 8618 8619 8620 8621 8622 8623 8624 8625 8626 8627 8628 8629 8630 8631 8632 8633 8634 8635 8636 8637 8638 8639 8640 8641 8642 8643 8644 8645 8646 8647 8648 8649 8650 8651 8652 8653 8654 8655 8656 8657 8658 8659 8660 8661 8662 8663 8664 8665 8666 8667 8668 8669 8670 8671 8672 8673 8674 8675 8676 8677 8678 8679 8680 8681 8682 8683 8684 8685 8686 8687 8688 8689 8690 8691 8692 8693 8694 8695 8696 8697 8698 8699 8700 8701 8702 8703 8704 8705 8706 8707 8708 8709 8710 8711 8712 8713 8714 8715 8716 8717 8718 8719 8720 8721 8722 8723 8724 8725 8726 8727 8728 8729 8730 8731 8732 8733 8734 8735 8736 8737 8738 8739 8740 8741 8742 8743 8744 8745 8746 8747 8748 8749 8750 8751 8752 8753 8754 8755 8756 8757 8758 8759 8760 8761 8762 8763 8764 8765 8766 8767 8768 8769 8770 8771 8772 8773 8774 8775 8776 8777 8778 8779 8780 8781 8782 8783 8784 8785 8786 8787 8788 8789 8790 8791 8792 8793 8794 8795 8796 8797 8798 8799 8800 8801 8802 8803 8804 8805 8806 8807 8808 8809 8810 8811 8812 8813 8814 8815 8816 8817 8818 8819 8820 8821 8822 8823 8824 8825 8826 8827 8828 8829 8830 8831 8832 8833 8834 8835 8836 8837 8838 8839 8840 8841 8842 8843 8844 8845 8846 8847 8848 8849 8850 8851 8852 8853 8854 8855 8856 8857 8858 8859 8860 8861 8862 8863 8864 8865 8866 8867 8868 8869 8870 8871 8872 8873 8874 8875 8876 8877 8878 8879 8880 8881 8882 8883 8884 8885 8886 8887 8888 8889 8890 8891 8892 8893 8894 8895 8896 8897 8898 8899 8900 8901 8902 8903 8904 8905 8906 8907 8908 8909 8910 8911 8912 8913 8914 8915 8916 8917 8918 8919 8920 8921 8922 8923 8924 8925 8926 8927 8928 8929 8930 8931 8932 8933 8934 8935 8936 8937 8938 8939 8940 8941 8942 8943 8944 8945 8946 8947 8948 8949 8950 8951 8952 8953 8954 8955 8956 8957 8958 8959 8960 8961 8962 8963 8964 8965 8966 8967 8968 8969 8970 8971 8972 8973 8974 8975 8976 8977 8978 8979 8980 8981 8982 8983 8984 8985 8986 8987 8988
/*

Unicode implementation based on original code by Fredrik Lundh,
modified by Marc-Andre Lemburg <mal@lemburg.com> according to the
Unicode Integration Proposal (see file Misc/unicode.txt).

Major speed upgrades to the method implementations at the Reykjavik
NeedForSpeed sprint, by Fredrik Lundh and Andrew Dalke.

Copyright (c) Corporation for National Research Initiatives.

--------------------------------------------------------------------
The original string type implementation is:

    Copyright (c) 1999 by Secret Labs AB
    Copyright (c) 1999 by Fredrik Lundh

By obtaining, using, and/or copying this software and/or its
associated documentation, you agree that you have read, understood,
and will comply with the following terms and conditions:

Permission to use, copy, modify, and distribute this software and its
associated documentation for any purpose and without fee is hereby
granted, provided that the above copyright notice appears in all
copies, and that both that copyright notice and this permission notice
appear in supporting documentation, and that the name of Secret Labs
AB or the author not be used in advertising or publicity pertaining to
distribution of the software without specific, written prior
permission.

SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO
THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
FITNESS.  IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR
ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT
OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
--------------------------------------------------------------------

*/

#define PY_SSIZE_T_CLEAN
#include "Python.h"

#include "formatter_unicode.h"

#include "unicodeobject.h"
#include "ucnhash.h"

#ifdef MS_WINDOWS
#include <windows.h>
#endif

/* Limit for the Unicode object free list */

#define PyUnicode_MAXFREELIST       1024

/* Limit for the Unicode object free list stay alive optimization.

   The implementation will keep allocated Unicode memory intact for
   all objects on the free list having a size less than this
   limit. This reduces malloc() overhead for small Unicode objects.

   At worst this will result in PyUnicode_MAXFREELIST *
   (sizeof(PyUnicodeObject) + KEEPALIVE_SIZE_LIMIT +
   malloc()-overhead) bytes of unused garbage.

   Setting the limit to 0 effectively turns the feature off.

   Note: This is an experimental feature ! If you get core dumps when
   using Unicode objects, turn this feature off.

*/

#define KEEPALIVE_SIZE_LIMIT       9

/* Endianness switches; defaults to little endian */

#ifdef WORDS_BIGENDIAN
# define BYTEORDER_IS_BIG_ENDIAN
#else
# define BYTEORDER_IS_LITTLE_ENDIAN
#endif

/* --- Globals ------------------------------------------------------------

   The globals are initialized by the _PyUnicode_Init() API and should
   not be used before calling that API.

*/


#ifdef __cplusplus
extern "C" {
#endif

/* Free list for Unicode objects */
static PyUnicodeObject *free_list;
static int numfree;

/* The empty Unicode object is shared to improve performance. */
static PyUnicodeObject *unicode_empty;

/* Single character Unicode strings in the Latin-1 range are being
   shared as well. */
static PyUnicodeObject *unicode_latin1[256];

/* Default encoding to use and assume when NULL is passed as encoding
   parameter; it is initialized by _PyUnicode_Init().

   Always use the PyUnicode_SetDefaultEncoding() and
   PyUnicode_GetDefaultEncoding() APIs to access this global.

*/
static char unicode_default_encoding[100];

/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
	0, 0, 0, 0, 0, 0, 0, 0,
//     case 0x0009: /* HORIZONTAL TABULATION */
//     case 0x000A: /* LINE FEED */
//     case 0x000B: /* VERTICAL TABULATION */
//     case 0x000C: /* FORM FEED */
//     case 0x000D: /* CARRIAGE RETURN */
	0, 1, 1, 1, 1, 1, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
//     case 0x001C: /* FILE SEPARATOR */
//     case 0x001D: /* GROUP SEPARATOR */
//     case 0x001E: /* RECORD SEPARATOR */
//     case 0x001F: /* UNIT SEPARATOR */
	0, 0, 0, 0, 1, 1, 1, 1,
//     case 0x0020: /* SPACE */
	1, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,

	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0
};

/* Same for linebreaks */
static unsigned char ascii_linebreak[] = {
	0, 0, 0, 0, 0, 0, 0, 0,
//         0x000A, /* LINE FEED */
//         0x000D, /* CARRIAGE RETURN */
	0, 0, 1, 0, 0, 1, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
//         0x001C, /* FILE SEPARATOR */
//         0x001D, /* GROUP SEPARATOR */
//         0x001E, /* RECORD SEPARATOR */
	0, 0, 0, 0, 1, 1, 1, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,

	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0,
	0, 0, 0, 0, 0, 0, 0, 0
};


Py_UNICODE
PyUnicode_GetMax(void)
{
#ifdef Py_UNICODE_WIDE
	return 0x10FFFF;
#else
	/* This is actually an illegal character, so it should
	   not be passed to unichr. */
	return 0xFFFF;
#endif
}

/* --- Bloom Filters ----------------------------------------------------- */

/* stuff to implement simple "bloom filters" for Unicode characters.
   to keep things simple, we use a single bitmask, using the least 5
   bits from each unicode characters as the bit index. */

/* the linebreak mask is set up by Unicode_Init below */

#define BLOOM_MASK unsigned long

static BLOOM_MASK bloom_linebreak;

#define BLOOM(mask, ch) ((mask & (1 << ((ch) & 0x1F))))

#define BLOOM_LINEBREAK(ch) \
    ((ch) < 128U ? ascii_linebreak[(ch)] : \
    (BLOOM(bloom_linebreak, (ch)) && Py_UNICODE_ISLINEBREAK(ch)))

Py_LOCAL_INLINE(BLOOM_MASK) make_bloom_mask(Py_UNICODE* ptr, Py_ssize_t len)
{
    /* calculate simple bloom-style bitmask for a given unicode string */

    long mask;
    Py_ssize_t i;

    mask = 0;
    for (i = 0; i < len; i++)
        mask |= (1 << (ptr[i] & 0x1F));

    return mask;
}

Py_LOCAL_INLINE(int) unicode_member(Py_UNICODE chr, Py_UNICODE* set, Py_ssize_t setlen)
{
    Py_ssize_t i;

    for (i = 0; i < setlen; i++)
        if (set[i] == chr)
            return 1;

    return 0;
}

#define BLOOM_MEMBER(mask, chr, set, setlen)\
    BLOOM(mask, chr) && unicode_member(chr, set, setlen)

/* --- Unicode Object ----------------------------------------------------- */

static
int unicode_resize(register PyUnicodeObject *unicode,
                      Py_ssize_t length)
{
    void *oldstr;

    /* Shortcut if there's nothing much to do. */
    if (unicode->length == length)
	goto reset;

    /* Resizing shared object (unicode_empty or single character
       objects) in-place is not allowed. Use PyUnicode_Resize()
       instead ! */

    if (unicode == unicode_empty || 
	(unicode->length == 1 && 
	 unicode->str[0] < 256U &&
	 unicode_latin1[unicode->str[0]] == unicode)) {
        PyErr_SetString(PyExc_SystemError,
                        "can't resize shared unicode objects");
        return -1;
    }

    /* We allocate one more byte to make sure the string is Ux0000 terminated.
       The overallocation is also used by fastsearch, which assumes that it's
       safe to look at str[length] (without making any assumptions about what
       it contains). */

    oldstr = unicode->str;
    unicode->str = PyObject_REALLOC(unicode->str,
				    sizeof(Py_UNICODE) * (length + 1));
    if (!unicode->str) {
	unicode->str = (Py_UNICODE *)oldstr;
        PyErr_NoMemory();
        return -1;
    }
    unicode->str[length] = 0;
    unicode->length = length;

 reset:
    /* Reset the object caches */
    if (unicode->defenc) {
        Py_DECREF(unicode->defenc);
        unicode->defenc = NULL;
    }
    unicode->hash = -1;

    return 0;
}

/* We allocate one more byte to make sure the string is
   Ux0000 terminated -- XXX is this needed ?

   XXX This allocator could further be enhanced by assuring that the
       free list never reduces its size below 1.

*/

static
PyUnicodeObject *_PyUnicode_New(Py_ssize_t length)
{
    register PyUnicodeObject *unicode;

    /* Optimization for empty strings */
    if (length == 0 && unicode_empty != NULL) {
        Py_INCREF(unicode_empty);
        return unicode_empty;
    }

    /* Unicode freelist & memory allocation */
    if (free_list) {
        unicode = free_list;
        free_list = *(PyUnicodeObject **)unicode;
        numfree--;
	if (unicode->str) {
	    /* Keep-Alive optimization: we only upsize the buffer,
	       never downsize it. */
	    if ((unicode->length < length) &&
                unicode_resize(unicode, length) < 0) {
		PyObject_DEL(unicode->str);
		goto onError;
	    }
	}
        else {
	    size_t new_size = sizeof(Py_UNICODE) * ((size_t)length + 1);
	    unicode->str = (Py_UNICODE*) PyObject_MALLOC(new_size);
        }
        PyObject_INIT(unicode, &PyUnicode_Type);
    }
    else {
	size_t new_size;
        unicode = PyObject_New(PyUnicodeObject, &PyUnicode_Type);
        if (unicode == NULL)
            return NULL;
	new_size = sizeof(Py_UNICODE) * ((size_t)length + 1);
	unicode->str = (Py_UNICODE*) PyObject_MALLOC(new_size);
    }

    if (!unicode->str) {
	PyErr_NoMemory();
	goto onError;
    }
    /* Initialize the first element to guard against cases where
     * the caller fails before initializing str -- unicode_resize()
     * reads str[0], and the Keep-Alive optimization can keep memory
     * allocated for str alive across a call to unicode_dealloc(unicode).
     * We don't want unicode_resize to read uninitialized memory in
     * that case.
     */
    unicode->str[0] = 0;
    unicode->str[length] = 0;
    unicode->length = length;
    unicode->hash = -1;
    unicode->defenc = NULL;
    return unicode;

 onError:
    _Py_ForgetReference((PyObject *)unicode);
    PyObject_Del(unicode);
    return NULL;
}

static
void unicode_dealloc(register PyUnicodeObject *unicode)
{
    if (PyUnicode_CheckExact(unicode) &&
	numfree < PyUnicode_MAXFREELIST) {
        /* Keep-Alive optimization */
	if (unicode->length >= KEEPALIVE_SIZE_LIMIT) {
	    PyObject_DEL(unicode->str);
	    unicode->str = NULL;
	    unicode->length = 0;
	}
	if (unicode->defenc) {
	    Py_DECREF(unicode->defenc);
	    unicode->defenc = NULL;
	}
	/* Add to free list */
        *(PyUnicodeObject **)unicode = free_list;
        free_list = unicode;
        numfree++;
    }
    else {
	PyObject_DEL(unicode->str);
	Py_XDECREF(unicode->defenc);
	Py_TYPE(unicode)->tp_free((PyObject *)unicode);
    }
}

int PyUnicode_Resize(PyObject **unicode, Py_ssize_t length)
{
    register PyUnicodeObject *v;

    /* Argument checks */
    if (unicode == NULL) {
	PyErr_BadInternalCall();
	return -1;
    }
    v = (PyUnicodeObject *)*unicode;
    if (v == NULL || !PyUnicode_Check(v) || Py_REFCNT(v) != 1 || length < 0) {
	PyErr_BadInternalCall();
	return -1;
    }

    /* Resizing unicode_empty and single character objects is not
       possible since these are being shared. We simply return a fresh
       copy with the same Unicode content. */
    if (v->length != length &&
	(v == unicode_empty || v->length == 1)) {
	PyUnicodeObject *w = _PyUnicode_New(length);
	if (w == NULL)
	    return -1;
	Py_UNICODE_COPY(w->str, v->str,
			length < v->length ? length : v->length);
	Py_DECREF(*unicode);
	*unicode = (PyObject *)w;
	return 0;
    }

    /* Note that we don't have to modify *unicode for unshared Unicode
       objects, since we can modify them in-place. */
    return unicode_resize(v, length);
}

/* Internal API for use in unicodeobject.c only ! */
#define _PyUnicode_Resize(unicodevar, length) \
        PyUnicode_Resize(((PyObject **)(unicodevar)), length)

PyObject *PyUnicode_FromUnicode(const Py_UNICODE *u,
				Py_ssize_t size)
{
    PyUnicodeObject *unicode;

    /* If the Unicode data is known at construction time, we can apply
       some optimizations which share commonly used objects. */
    if (u != NULL) {

	/* Optimization for empty strings */
	if (size == 0 && unicode_empty != NULL) {
	    Py_INCREF(unicode_empty);
	    return (PyObject *)unicode_empty;
	}

	/* Single character Unicode objects in the Latin-1 range are
	   shared when using this constructor */
	if (size == 1 && *u < 256) {
	    unicode = unicode_latin1[*u];
	    if (!unicode) {
		unicode = _PyUnicode_New(1);
		if (!unicode)
		    return NULL;
		unicode->str[0] = *u;
		unicode_latin1[*u] = unicode;
	    }
	    Py_INCREF(unicode);
	    return (PyObject *)unicode;
	}
    }

    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;

    /* Copy the Unicode data into the new object */
    if (u != NULL)
	Py_UNICODE_COPY(unicode->str, u, size);

    return (PyObject *)unicode;
}

PyObject *PyUnicode_FromStringAndSize(const char *u, Py_ssize_t size)
{
    PyUnicodeObject *unicode;
    /* If the Unicode data is known at construction time, we can apply
       some optimizations which share commonly used objects.
       Also, this means the input must be UTF-8, so fall back to the
       UTF-8 decoder at the end. */
    if (u != NULL) {

	/* Optimization for empty strings */
	if (size == 0 && unicode_empty != NULL) {
	    Py_INCREF(unicode_empty);
	    return (PyObject *)unicode_empty;
	}

	/* Single characters are shared when using this constructor.
           Restrict to ASCII, since the input must be UTF-8. */
	if (size == 1 && Py_CHARMASK(*u) < 128) {
	    unicode = unicode_latin1[Py_CHARMASK(*u)];
	    if (!unicode) {
		unicode = _PyUnicode_New(1);
		if (!unicode)
		    return NULL;
		unicode->str[0] = Py_CHARMASK(*u);
		unicode_latin1[Py_CHARMASK(*u)] = unicode;
	    }
	    Py_INCREF(unicode);
	    return (PyObject *)unicode;
	}

        return PyUnicode_DecodeUTF8(u, size, NULL);
    }

    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;

    return (PyObject *)unicode;
}

PyObject *PyUnicode_FromString(const char *u)
{
    size_t size = strlen(u);
    if (size > PY_SSIZE_T_MAX) {
        PyErr_SetString(PyExc_OverflowError, "input too long");
        return NULL;
    }

    return PyUnicode_FromStringAndSize(u, size);
}

#ifdef HAVE_WCHAR_H

PyObject *PyUnicode_FromWideChar(register const wchar_t *w,
				 Py_ssize_t size)
{
    PyUnicodeObject *unicode;

    if (w == NULL) {
	PyErr_BadInternalCall();
	return NULL;
    }

    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;

    /* Copy the wchar_t data into the new object */
#ifdef HAVE_USABLE_WCHAR_T
    memcpy(unicode->str, w, size * sizeof(wchar_t));
#else
    {
	register Py_UNICODE *u;
	register Py_ssize_t i;
	u = PyUnicode_AS_UNICODE(unicode);
	for (i = size; i > 0; i--)
	    *u++ = *w++;
    }
#endif

    return (PyObject *)unicode;
}

static void
makefmt(char *fmt, int longflag, int size_tflag, int zeropad, int width, int precision, char c)
{
	*fmt++ = '%';
	if (width) {
		if (zeropad)
			*fmt++ = '0';
		fmt += sprintf(fmt, "%d", width);
	}
	if (precision)
		fmt += sprintf(fmt, ".%d", precision);
	if (longflag)
		*fmt++ = 'l';
	else if (size_tflag) {
		char *f = PY_FORMAT_SIZE_T;
		while (*f)
			*fmt++ = *f++;
	}
	*fmt++ = c;
	*fmt = '\0';
}

#define appendstring(string) {for (copy = string;*copy;) *s++ = *copy++;}

PyObject *
PyUnicode_FromFormatV(const char *format, va_list vargs)
{
	va_list count;
	Py_ssize_t callcount = 0;
	PyObject **callresults = NULL;
	PyObject **callresult = NULL;
	Py_ssize_t n = 0;
	int width = 0;
	int precision = 0;
	int zeropad;
	const char* f;
	Py_UNICODE *s;
	PyObject *string;
	/* used by sprintf */
	char buffer[21];
	/* use abuffer instead of buffer, if we need more space
	 * (which can happen if there's a format specifier with width). */
	char *abuffer = NULL;
	char *realbuffer;
	Py_ssize_t abuffersize = 0;
	char fmt[60]; /* should be enough for %0width.precisionld */
	const char *copy;

#ifdef VA_LIST_IS_ARRAY
	Py_MEMCPY(count, vargs, sizeof(va_list));
#else
#ifdef  __va_copy
	__va_copy(count, vargs);
#else
	count = vargs;
#endif
#endif
	/* step 1: count the number of %S/%R format specifications
	 * (we call PyObject_Str()/PyObject_Repr() for these objects
	 * once during step 3 and put the result in an array) */
	for (f = format; *f; f++) {
		if (*f == '%' && (*(f+1)=='S' || *(f+1)=='R'))
			++callcount;
	}
	/* step 2: allocate memory for the results of
	 * PyObject_Str()/PyObject_Repr() calls */
	if (callcount) {
		callresults = PyObject_Malloc(sizeof(PyObject *)*callcount);
		if (!callresults) {
			PyErr_NoMemory();
			return NULL;
		}
		callresult = callresults;
	}
	/* step 3: figure out how large a buffer we need */
	for (f = format; *f; f++) {
		if (*f == '%') {
			const char* p = f;
			width = 0;
			while (isdigit((unsigned)*f))
				width = (width*10) + *f++ - '0';
			while (*++f && *f != '%' && !isalpha((unsigned)*f))
				;

			/* skip the 'l' or 'z' in {%ld, %zd, %lu, %zu} since
			 * they don't affect the amount of space we reserve.
			 */
			if ((*f == 'l' || *f == 'z') &&
					(f[1] == 'd' || f[1] == 'u'))
                                ++f;

			switch (*f) {
			case 'c':
				(void)va_arg(count, int);
				/* fall through... */
			case '%':
				n++;
				break;
			case 'd': case 'u': case 'i': case 'x':
				(void) va_arg(count, int);
				/* 20 bytes is enough to hold a 64-bit
				   integer.  Decimal takes the most space.
				   This isn't enough for octal.
				   If a width is specified we need more
				   (which we allocate later). */
				if (width < 20)
					width = 20;
				n += width;
				if (abuffersize < width)
					abuffersize = width;
				break;
			case 's':
			{
				/* UTF-8 */
				unsigned char*s;
				s = va_arg(count, unsigned char*);
				while (*s) {
					if (*s < 128) {
						n++; s++;
					} else if (*s < 0xc0) {
						/* invalid UTF-8 */
						n++; s++;
					} else if (*s < 0xc0) {
						n++;
						s++; if(!*s)break;
						s++;
					} else if (*s < 0xe0) {
						n++;
						s++; if(!*s)break;
						s++; if(!*s)break;
						s++;
					} else {
						#ifdef Py_UNICODE_WIDE
						n++;
						#else
						n+=2;
						#endif
						s++; if(!*s)break;
						s++; if(!*s)break;
						s++; if(!*s)break;
						s++;
					}
				}
				break;
			}
			case 'U':
			{
				PyObject *obj = va_arg(count, PyObject *);
				assert(obj && PyUnicode_Check(obj));
				n += PyUnicode_GET_SIZE(obj);
				break;
			}
			case 'V':
			{
				PyObject *obj = va_arg(count, PyObject *);
				const char *str = va_arg(count, const char *);
				assert(obj || str);
				assert(!obj || PyUnicode_Check(obj));
				if (obj)
					n += PyUnicode_GET_SIZE(obj);
				else
					n += strlen(str);
				break;
			}
			case 'S':
			{
				PyObject *obj = va_arg(count, PyObject *);
				PyObject *str;
				assert(obj);
				str = PyObject_Str(obj);
				if (!str)
					goto fail;
				n += PyUnicode_GET_SIZE(str);
				/* Remember the str and switch to the next slot */
				*callresult++ = str;
				break;
			}
			case 'R':
			{
				PyObject *obj = va_arg(count, PyObject *);
				PyObject *repr;
				assert(obj);
				repr = PyObject_Repr(obj);
				if (!repr)
					goto fail;
				n += PyUnicode_GET_SIZE(repr);
				/* Remember the repr and switch to the next slot */
				*callresult++ = repr;
				break;
			}
			case 'p':
				(void) va_arg(count, int);
				/* maximum 64-bit pointer representation:
				 * 0xffffffffffffffff
				 * so 19 characters is enough.
				 * XXX I count 18 -- what's the extra for?
				 */
				n += 19;
				break;
			default:
				/* if we stumble upon an unknown
				   formatting code, copy the rest of
				   the format string to the output
				   string. (we cannot just skip the
				   code, since there's no way to know
				   what's in the argument list) */
				n += strlen(p);
				goto expand;
			}
		} else
			n++;
	}
 expand:
	if (abuffersize > 20) {
		abuffer = PyObject_Malloc(abuffersize);
		if (!abuffer) {
			PyErr_NoMemory();
			goto fail;
		}
		realbuffer = abuffer;
	}
	else
		realbuffer = buffer;
	/* step 4: fill the buffer */
	/* Since we've analyzed how much space we need for the worst case,
	   we don't have to resize the string.
	   There can be no errors beyond this point. */
	string = PyUnicode_FromUnicode(NULL, n);
	if (!string)
		goto fail;

	s = PyUnicode_AS_UNICODE(string);
	callresult = callresults;

	for (f = format; *f; f++) {
		if (*f == '%') {
			const char* p = f++;
			int longflag = 0;
			int size_tflag = 0;
			zeropad = (*f == '0');
			/* parse the width.precision part */
			width = 0;
			while (isdigit((unsigned)*f))
				width = (width*10) + *f++ - '0';
			precision = 0;
			if (*f == '.') {
				f++;
				while (isdigit((unsigned)*f))
					precision = (precision*10) + *f++ - '0';
			}
			/* handle the long flag, but only for %ld and %lu.
			   others can be added when necessary. */
			if (*f == 'l' && (f[1] == 'd' || f[1] == 'u')) {
				longflag = 1;
				++f;
			}
			/* handle the size_t flag. */
			if (*f == 'z' && (f[1] == 'd' || f[1] == 'u')) {
				size_tflag = 1;
				++f;
			}

			switch (*f) {
			case 'c':
				*s++ = va_arg(vargs, int);
				break;
			case 'd':
				makefmt(fmt, longflag, size_tflag, zeropad, width, precision, 'd');
				if (longflag)
					sprintf(realbuffer, fmt, va_arg(vargs, long));
				else if (size_tflag)
					sprintf(realbuffer, fmt, va_arg(vargs, Py_ssize_t));
				else
					sprintf(realbuffer, fmt, va_arg(vargs, int));
				appendstring(realbuffer);
				break;
			case 'u':
				makefmt(fmt, longflag, size_tflag, zeropad, width, precision, 'u');
				if (longflag)
					sprintf(realbuffer, fmt, va_arg(vargs, unsigned long));
				else if (size_tflag)
					sprintf(realbuffer, fmt, va_arg(vargs, size_t));
				else
					sprintf(realbuffer, fmt, va_arg(vargs, unsigned int));
				appendstring(realbuffer);
				break;
			case 'i':
				makefmt(fmt, 0, 0, zeropad, width, precision, 'i');
				sprintf(realbuffer, fmt, va_arg(vargs, int));
				appendstring(realbuffer);
				break;
			case 'x':
				makefmt(fmt, 0, 0, zeropad, width, precision, 'x');
				sprintf(realbuffer, fmt, va_arg(vargs, int));
				appendstring(realbuffer);
				break;
			case 's':
			{
				/* Parameter must be UTF-8 encoded.
				   In case of encoding errors, use
				   the replacement character. */
				PyObject *u;
				p = va_arg(vargs, char*);
				u = PyUnicode_DecodeUTF8(p, strlen(p), 
							 "replace");
				if (!u)
					goto fail;
				Py_UNICODE_COPY(s, PyUnicode_AS_UNICODE(u),
						PyUnicode_GET_SIZE(u));
				s += PyUnicode_GET_SIZE(u);
				Py_DECREF(u);
				break;
			}
			case 'U':
			{
				PyObject *obj = va_arg(vargs, PyObject *);
				Py_ssize_t size = PyUnicode_GET_SIZE(obj);
				Py_UNICODE_COPY(s, PyUnicode_AS_UNICODE(obj), size);
				s += size;
				break;
			}
			case 'V':
			{
				PyObject *obj = va_arg(vargs, PyObject *);
				const char *str = va_arg(vargs, const char *);
				if (obj) {
					Py_ssize_t size = PyUnicode_GET_SIZE(obj);
					Py_UNICODE_COPY(s, PyUnicode_AS_UNICODE(obj), size);
					s += size;
				} else {
					appendstring(str);
				}
				break;
			}
			case 'S':
			case 'R':
			{
				Py_UNICODE *ucopy;
				Py_ssize_t usize;
				Py_ssize_t upos;
				/* unused, since we already have the result */
				(void) va_arg(vargs, PyObject *);
				ucopy = PyUnicode_AS_UNICODE(*callresult);
				usize = PyUnicode_GET_SIZE(*callresult);
				for (upos = 0; upos<usize;)
					*s++ = ucopy[upos++];
				/* We're done with the unicode()/repr() => forget it */
				Py_DECREF(*callresult);
				/* switch to next unicode()/repr() result */
				++callresult;
				break;
			}
			case 'p':
				sprintf(buffer, "%p", va_arg(vargs, void*));
				/* %p is ill-defined:  ensure leading 0x. */
				if (buffer[1] == 'X')
					buffer[1] = 'x';
				else if (buffer[1] != 'x') {
					memmove(buffer+2, buffer, strlen(buffer)+1);
					buffer[0] = '0';
					buffer[1] = 'x';
				}
				appendstring(buffer);
				break;
			case '%':
				*s++ = '%';
				break;
			default:
				appendstring(p);
				goto end;
			}
		} else
			*s++ = *f;
	}

 end:
	if (callresults)
		PyObject_Free(callresults);
	if (abuffer)
		PyObject_Free(abuffer);
	_PyUnicode_Resize(&string, s - PyUnicode_AS_UNICODE(string));
	return string;
 fail:
	if (callresults) {
		PyObject **callresult2 = callresults;
		while (callresult2 < callresult) {
			Py_DECREF(*callresult2);
			++callresult2;
		}
		PyObject_Free(callresults);
	}
	if (abuffer)
		PyObject_Free(abuffer);
	return NULL;
}

#undef appendstring

PyObject *
PyUnicode_FromFormat(const char *format, ...)
{
	PyObject* ret;
	va_list vargs;

#ifdef HAVE_STDARG_PROTOTYPES
	va_start(vargs, format);
#else
	va_start(vargs);
#endif
	ret = PyUnicode_FromFormatV(format, vargs);
	va_end(vargs);
	return ret;
}

Py_ssize_t PyUnicode_AsWideChar(PyUnicodeObject *unicode,
				wchar_t *w,
				Py_ssize_t size)
{
    if (unicode == NULL) {
	PyErr_BadInternalCall();
	return -1;
    }

    /* If possible, try to copy the 0-termination as well */
    if (size > PyUnicode_GET_SIZE(unicode))
	size = PyUnicode_GET_SIZE(unicode) + 1;

#ifdef HAVE_USABLE_WCHAR_T
    memcpy(w, unicode->str, size * sizeof(wchar_t));
#else
    {
	register Py_UNICODE *u;
	register Py_ssize_t i;
	u = PyUnicode_AS_UNICODE(unicode);
	for (i = size; i > 0; i--)
	    *w++ = *u++;
    }
#endif

    if (size > PyUnicode_GET_SIZE(unicode))
        return PyUnicode_GET_SIZE(unicode);
    else
    return size;
}

#endif

PyObject *PyUnicode_FromOrdinal(int ordinal)
{
    Py_UNICODE s[1];

#ifdef Py_UNICODE_WIDE
    if (ordinal < 0 || ordinal > 0x10ffff) {
	PyErr_SetString(PyExc_ValueError,
			"unichr() arg not in range(0x110000) "
			"(wide Python build)");
	return NULL;
    }
#else
    if (ordinal < 0 || ordinal > 0xffff) {
	PyErr_SetString(PyExc_ValueError,
			"unichr() arg not in range(0x10000) "
			"(narrow Python build)");
	return NULL;
    }
#endif

    s[0] = (Py_UNICODE)ordinal;
    return PyUnicode_FromUnicode(s, 1);
}

PyObject *PyUnicode_FromObject(register PyObject *obj)
{
    /* XXX Perhaps we should make this API an alias of
           PyObject_Unicode() instead ?! */
    if (PyUnicode_CheckExact(obj)) {
	Py_INCREF(obj);
	return obj;
    }
    if (PyUnicode_Check(obj)) {
	/* For a Unicode subtype that's not a Unicode object,
	   return a true Unicode object with the same data. */
	return PyUnicode_FromUnicode(PyUnicode_AS_UNICODE(obj),
				     PyUnicode_GET_SIZE(obj));
    }
    return PyUnicode_FromEncodedObject(obj, NULL, "strict");
}

PyObject *PyUnicode_FromEncodedObject(register PyObject *obj,
				      const char *encoding,
				      const char *errors)
{
    const char *s = NULL;
    Py_ssize_t len;
    PyObject *v;

    if (obj == NULL) {
	PyErr_BadInternalCall();
	return NULL;
    }

#if 0
    /* For b/w compatibility we also accept Unicode objects provided
       that no encodings is given and then redirect to
       PyObject_Unicode() which then applies the additional logic for
       Unicode subclasses.

       NOTE: This API should really only be used for object which
             represent *encoded* Unicode !

    */
	if (PyUnicode_Check(obj)) {
	    if (encoding) {
		PyErr_SetString(PyExc_TypeError,
				"decoding Unicode is not supported");
	    return NULL;
	    }
	return PyObject_Unicode(obj);
	    }
#else
    if (PyUnicode_Check(obj)) {
	PyErr_SetString(PyExc_TypeError,
			"decoding Unicode is not supported");
	return NULL;
	}
#endif

    /* Coerce object */
    if (PyString_Check(obj)) {
	    s = PyString_AS_STRING(obj);
	    len = PyString_GET_SIZE(obj);
	    }
    else if (PyObject_AsCharBuffer(obj, &s, &len)) {
	/* Overwrite the error message with something more useful in
	   case of a TypeError. */
	if (PyErr_ExceptionMatches(PyExc_TypeError))
	PyErr_Format(PyExc_TypeError,
			 "coercing to Unicode: need string or buffer, "
			 "%.80s found",
		     Py_TYPE(obj)->tp_name);
	goto onError;
    }

    /* Convert to Unicode */
    if (len == 0) {
	Py_INCREF(unicode_empty);
	v = (PyObject *)unicode_empty;
    }
    else
	v = PyUnicode_Decode(s, len, encoding, errors);

    return v;

 onError:
    return NULL;
}

PyObject *PyUnicode_Decode(const char *s,
			   Py_ssize_t size,
			   const char *encoding,
			   const char *errors)
{
    PyObject *buffer = NULL, *unicode;

    if (encoding == NULL)
	encoding = PyUnicode_GetDefaultEncoding();

    /* Shortcuts for common default encodings */
    if (strcmp(encoding, "utf-8") == 0)
        return PyUnicode_DecodeUTF8(s, size, errors);
    else if (strcmp(encoding, "latin-1") == 0)
        return PyUnicode_DecodeLatin1(s, size, errors);
#if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T)
    else if (strcmp(encoding, "mbcs") == 0)
        return PyUnicode_DecodeMBCS(s, size, errors);
#endif
    else if (strcmp(encoding, "ascii") == 0)
        return PyUnicode_DecodeASCII(s, size, errors);

    /* Decode via the codec registry */
    buffer = PyBuffer_FromMemory((void *)s, size);
    if (buffer == NULL)
        goto onError;
    unicode = PyCodec_Decode(buffer, encoding, errors);
    if (unicode == NULL)
        goto onError;
    if (!PyUnicode_Check(unicode)) {
        PyErr_Format(PyExc_TypeError,
                     "decoder did not return an unicode object (type=%.400s)",
                     Py_TYPE(unicode)->tp_name);
        Py_DECREF(unicode);
        goto onError;
    }
    Py_DECREF(buffer);
    return unicode;

 onError:
    Py_XDECREF(buffer);
    return NULL;
}

PyObject *PyUnicode_AsDecodedObject(PyObject *unicode,
                                    const char *encoding,
                                    const char *errors)
{
    PyObject *v;

    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        goto onError;
    }

    if (encoding == NULL)
	encoding = PyUnicode_GetDefaultEncoding();

    /* Decode via the codec registry */
    v = PyCodec_Decode(unicode, encoding, errors);
    if (v == NULL)
        goto onError;
    return v;

 onError:
    return NULL;
}

PyObject *PyUnicode_Encode(const Py_UNICODE *s,
			   Py_ssize_t size,
			   const char *encoding,
			   const char *errors)
{
    PyObject *v, *unicode;

    unicode = PyUnicode_FromUnicode(s, size);
    if (unicode == NULL)
	return NULL;
    v = PyUnicode_AsEncodedString(unicode, encoding, errors);
    Py_DECREF(unicode);
    return v;
}

PyObject *PyUnicode_AsEncodedObject(PyObject *unicode,
                                    const char *encoding,
                                    const char *errors)
{
    PyObject *v;

    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        goto onError;
    }

    if (encoding == NULL)
	encoding = PyUnicode_GetDefaultEncoding();

    /* Encode via the codec registry */
    v = PyCodec_Encode(unicode, encoding, errors);
    if (v == NULL)
        goto onError;
    return v;

 onError:
    return NULL;
}

PyObject *PyUnicode_AsEncodedString(PyObject *unicode,
                                    const char *encoding,
                                    const char *errors)
{
    PyObject *v;

    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        goto onError;
    }

    if (encoding == NULL)
	encoding = PyUnicode_GetDefaultEncoding();

    /* Shortcuts for common default encodings */
    if (errors == NULL) {
	if (strcmp(encoding, "utf-8") == 0)
	    return PyUnicode_AsUTF8String(unicode);
	else if (strcmp(encoding, "latin-1") == 0)
	    return PyUnicode_AsLatin1String(unicode);
#if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T)
	else if (strcmp(encoding, "mbcs") == 0)
	    return PyUnicode_AsMBCSString(unicode);
#endif
	else if (strcmp(encoding, "ascii") == 0)
	    return PyUnicode_AsASCIIString(unicode);
    }

    /* Encode via the codec registry */
    v = PyCodec_Encode(unicode, encoding, errors);
    if (v == NULL)
        goto onError;
    if (!PyString_Check(v)) {
        PyErr_Format(PyExc_TypeError,
                     "encoder did not return a string object (type=%.400s)",
                     Py_TYPE(v)->tp_name);
        Py_DECREF(v);
        goto onError;
    }
    return v;

 onError:
    return NULL;
}

PyObject *_PyUnicode_AsDefaultEncodedString(PyObject *unicode,
					    const char *errors)
{
    PyObject *v = ((PyUnicodeObject *)unicode)->defenc;

    if (v)
        return v;
    v = PyUnicode_AsEncodedString(unicode, NULL, errors);
    if (v && errors == NULL)
        ((PyUnicodeObject *)unicode)->defenc = v;
    return v;
}

Py_UNICODE *PyUnicode_AsUnicode(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        goto onError;
    }
    return PyUnicode_AS_UNICODE(unicode);

 onError:
    return NULL;
}

Py_ssize_t PyUnicode_GetSize(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        goto onError;
    }
    return PyUnicode_GET_SIZE(unicode);

 onError:
    return -1;
}

const char *PyUnicode_GetDefaultEncoding(void)
{
    return unicode_default_encoding;
}

int PyUnicode_SetDefaultEncoding(const char *encoding)
{
    PyObject *v;

    /* Make sure the encoding is valid. As side effect, this also
       loads the encoding into the codec registry cache. */
    v = _PyCodec_Lookup(encoding);
    if (v == NULL)
	goto onError;
    Py_DECREF(v);
    strncpy(unicode_default_encoding,
	    encoding,
	    sizeof(unicode_default_encoding));
    return 0;

 onError:
    return -1;
}

/* error handling callback helper:
   build arguments, call the callback and check the arguments,
   if no exception occurred, copy the replacement to the output
   and adjust various state variables.
   return 0 on success, -1 on error
*/

static
int unicode_decode_call_errorhandler(const char *errors, PyObject **errorHandler,
                 const char *encoding, const char *reason,
                 const char *input, Py_ssize_t insize, Py_ssize_t *startinpos,
                 Py_ssize_t *endinpos, PyObject **exceptionObject, const char **inptr,
                 PyObject **output, Py_ssize_t *outpos, Py_UNICODE **outptr)
{
    static char *argparse = "O!n;decoding error handler must return (unicode, int) tuple";

    PyObject *restuple = NULL;
    PyObject *repunicode = NULL;
    Py_ssize_t outsize = PyUnicode_GET_SIZE(*output);
    Py_ssize_t requiredsize;
    Py_ssize_t newpos;
    Py_UNICODE *repptr;
    Py_ssize_t repsize;
    int res = -1;

    if (*errorHandler == NULL) {
	*errorHandler = PyCodec_LookupError(errors);
	if (*errorHandler == NULL)
	   goto onError;
    }

    if (*exceptionObject == NULL) {
    	*exceptionObject = PyUnicodeDecodeError_Create(
	    encoding, input, insize, *startinpos, *endinpos, reason);
	if (*exceptionObject == NULL)
	   goto onError;
    }
    else {
	if (PyUnicodeDecodeError_SetStart(*exceptionObject, *startinpos))
	    goto onError;
	if (PyUnicodeDecodeError_SetEnd(*exceptionObject, *endinpos))
	    goto onError;
	if (PyUnicodeDecodeError_SetReason(*exceptionObject, reason))
	    goto onError;
    }

    restuple = PyObject_CallFunctionObjArgs(*errorHandler, *exceptionObject, NULL);
    if (restuple == NULL)
	goto onError;
    if (!PyTuple_Check(restuple)) {
	PyErr_Format(PyExc_TypeError, &argparse[4]);
	goto onError;
    }
    if (!PyArg_ParseTuple(restuple, argparse, &PyUnicode_Type, &repunicode, &newpos))
	goto onError;
    if (newpos<0)
	newpos = insize+newpos;
    if (newpos<0 || newpos>insize) {
	PyErr_Format(PyExc_IndexError, "position %zd from error handler out of bounds", newpos);
	goto onError;
    }

    /* need more space? (at least enough for what we
       have+the replacement+the rest of the string (starting
       at the new input position), so we won't have to check space
       when there are no errors in the rest of the string) */
    repptr = PyUnicode_AS_UNICODE(repunicode);
    repsize = PyUnicode_GET_SIZE(repunicode);
    requiredsize = *outpos + repsize + insize-newpos;
    if (requiredsize > outsize) {
	if (requiredsize<2*outsize)
	    requiredsize = 2*outsize;
	if (PyUnicode_Resize(output, requiredsize) < 0)
	    goto onError;
	*outptr = PyUnicode_AS_UNICODE(*output) + *outpos;
    }
    *endinpos = newpos;
    *inptr = input + newpos;
    Py_UNICODE_COPY(*outptr, repptr, repsize);
    *outptr += repsize;
    *outpos += repsize;
    /* we made it! */
    res = 0;

    onError:
    Py_XDECREF(restuple);
    return res;
}

/* --- UTF-7 Codec -------------------------------------------------------- */

/* see RFC2152 for details */

static
char utf7_special[128] = {
    /* indicate whether a UTF-7 character is special i.e. cannot be directly
       encoded:
	   0 - not special
	   1 - special
	   2 - whitespace (optional)
	   3 - RFC2152 Set O (optional) */
    1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 1, 1, 2, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    2, 3, 3, 3, 3, 3, 3, 0, 0, 0, 3, 1, 0, 0, 0, 1,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 3, 0,
    3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 1, 3, 3, 3,
    3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 3, 3, 1, 1,

};

/* Note: The comparison (c) <= 0 is a trick to work-around gcc
   warnings about the comparison always being false; since
   utf7_special[0] is 1, we can safely make that one comparison
   true  */

#define SPECIAL(c, encodeO, encodeWS) \
    ((c) > 127 || (c) <= 0 || utf7_special[(c)] == 1 || \
     (encodeWS && (utf7_special[(c)] == 2)) || \
     (encodeO && (utf7_special[(c)] == 3)))

#define B64(n)  \
    ("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"[(n) & 0x3f])
#define B64CHAR(c) \
    (isalnum(c) || (c) == '+' || (c) == '/')
#define UB64(c) \
    ((c) == '+' ? 62 : (c) == '/' ? 63 : (c) >= 'a' ?                   \
     (c) - 71 : (c) >= 'A' ? (c) - 65 : (c) + 4 )

#define ENCODE(out, ch, bits)                   \
    while (bits >= 6) {                         \
        *out++ = B64(ch >> (bits-6));           \
        bits -= 6;                              \
    }

#define DECODE(out, ch, bits, surrogate)                                \
    while (bits >= 16) {                                                \
        Py_UNICODE outCh = (Py_UNICODE) ((ch >> (bits-16)) & 0xffff);   \
        bits -= 16;                                                     \
        if (surrogate) {                                                \
            /* We have already generated an error for the high surrogate \
               so let's not bother seeing if the low surrogate is correct or not */ \
            surrogate = 0;                                              \
        } else if (0xDC00 <= outCh && outCh <= 0xDFFF) {                \
            /* This is a surrogate pair. Unfortunately we can't represent \
               it in a 16-bit character */                              \
            surrogate = 1;                                              \
            errmsg = "code pairs are not supported";                    \
            goto utf7Error;                                             \
        } else {                                                        \
            *out++ = outCh;                                             \
        }                                                               \
    }

PyObject *PyUnicode_DecodeUTF7(const char *s,
			       Py_ssize_t size,
			       const char *errors)
{
    return PyUnicode_DecodeUTF7Stateful(s, size, errors, NULL);
}

PyObject *PyUnicode_DecodeUTF7Stateful(const char *s,
			       Py_ssize_t size,
			       const char *errors,
			       Py_ssize_t *consumed)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    const char *e;
    PyUnicodeObject *unicode;
    Py_UNICODE *p;
    const char *errmsg = "";
    int inShift = 0;
    unsigned int bitsleft = 0;
    unsigned long charsleft = 0;
    int surrogate = 0;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;
    if (size == 0) {
        if (consumed)
            *consumed = 0;
        return (PyObject *)unicode;
    }

    p = unicode->str;
    e = s + size;

    while (s < e) {
        Py_UNICODE ch;
        restart:
        ch = *s;

        if (inShift) {
            if ((ch == '-') || !B64CHAR(ch)) {
                inShift = 0;
                s++;

                /* p, charsleft, bitsleft, surrogate = */ DECODE(p, charsleft, bitsleft, surrogate);
                if (bitsleft >= 6) {
                    /* The shift sequence has a partial character in it. If
                       bitsleft < 6 then we could just classify it as padding
                       but that is not the case here */

                    errmsg = "partial character in shift sequence";
                    goto utf7Error;
                }
                /* According to RFC2152 the remaining bits should be zero. We
                   choose to signal an error/insert a replacement character
                   here so indicate the potential of a misencoded character. */

                /* On x86, a << b == a << (b%32) so make sure that bitsleft != 0 */
                if (bitsleft && charsleft << (sizeof(charsleft) * 8 - bitsleft)) {
                    errmsg = "non-zero padding bits in shift sequence";
                    goto utf7Error;
                }

                if (ch == '-') {
                    if ((s < e) && (*(s) == '-')) {
                        *p++ = '-';
                        inShift = 1;
                    }
                } else if (SPECIAL(ch,0,0)) {
                    errmsg = "unexpected special character";
	                goto utf7Error;
                } else  {
                    *p++ = ch;
                }
            } else {
                charsleft = (charsleft << 6) | UB64(ch);
                bitsleft += 6;
                s++;
                /* p, charsleft, bitsleft, surrogate = */ DECODE(p, charsleft, bitsleft, surrogate);
            }
        }
        else if ( ch == '+' ) {
            startinpos = s-starts;
            s++;
            if (s < e && *s == '-') {
                s++;
                *p++ = '+';
            } else
            {
                inShift = 1;
                bitsleft = 0;
            }
        }
        else if (SPECIAL(ch,0,0)) {
            startinpos = s-starts;
            errmsg = "unexpected special character";
            s++;
	        goto utf7Error;
        }
        else {
            *p++ = ch;
            s++;
        }
        continue;
    utf7Error:
        outpos = p-PyUnicode_AS_UNICODE(unicode);
        endinpos = s-starts;
        if (unicode_decode_call_errorhandler(
             errors, &errorHandler,
             "utf7", errmsg,
             starts, size, &startinpos, &endinpos, &exc, &s,
             (PyObject **)&unicode, &outpos, &p))
        goto onError;
    }

    if (inShift && !consumed) {
        outpos = p-PyUnicode_AS_UNICODE(unicode);
        endinpos = size;
        if (unicode_decode_call_errorhandler(
             errors, &errorHandler,
             "utf7", "unterminated shift sequence",
             starts, size, &startinpos, &endinpos, &exc, &s,
             (PyObject **)&unicode, &outpos, &p))
            goto onError;
        if (s < e)
           goto restart;
    }
    if (consumed) {
        if(inShift)
            *consumed = startinpos;
        else
            *consumed = s-starts;
    }

    if (_PyUnicode_Resize(&unicode, p - PyUnicode_AS_UNICODE(unicode)) < 0)
        goto onError;

    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)unicode;

onError:
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    Py_DECREF(unicode);
    return NULL;
}


PyObject *PyUnicode_EncodeUTF7(const Py_UNICODE *s,
                   Py_ssize_t size,
                   int encodeSetO,
                   int encodeWhiteSpace,
                   const char *errors)
{
    PyObject *v;
    /* It might be possible to tighten this worst case */
    Py_ssize_t cbAllocated = 5 * size;
    int inShift = 0;
    Py_ssize_t i = 0;
    unsigned int bitsleft = 0;
    unsigned long charsleft = 0;
    char * out;
    char * start;

    if (size == 0)
		return PyString_FromStringAndSize(NULL, 0);

    v = PyString_FromStringAndSize(NULL, cbAllocated);
    if (v == NULL)
        return NULL;

    start = out = PyString_AS_STRING(v);
    for (;i < size; ++i) {
        Py_UNICODE ch = s[i];

        if (!inShift) {
            if (ch == '+') {
                *out++ = '+';
                *out++ = '-';
            } else if (SPECIAL(ch, encodeSetO, encodeWhiteSpace)) {
                charsleft = ch;
                bitsleft = 16;
                *out++ = '+';
                /* out, charsleft, bitsleft = */ ENCODE(out, charsleft, bitsleft);
                inShift = bitsleft > 0;
            } else {
                *out++ = (char) ch;
            }
        } else {
            if (!SPECIAL(ch, encodeSetO, encodeWhiteSpace)) {
                *out++ = B64(charsleft << (6-bitsleft));
                charsleft = 0;
                bitsleft = 0;
                /* Characters not in the BASE64 set implicitly unshift the sequence
                   so no '-' is required, except if the character is itself a '-' */
                if (B64CHAR(ch) || ch == '-') {
                    *out++ = '-';
                }
                inShift = 0;
                *out++ = (char) ch;
            } else {
                bitsleft += 16;
                charsleft = (charsleft << 16) | ch;
                /* out, charsleft, bitsleft = */ ENCODE(out, charsleft, bitsleft);

                /* If the next character is special then we dont' need to terminate
                   the shift sequence. If the next character is not a BASE64 character
                   or '-' then the shift sequence will be terminated implicitly and we
                   don't have to insert a '-'. */

                if (bitsleft == 0) {
                    if (i + 1 < size) {
                        Py_UNICODE ch2 = s[i+1];

                        if (SPECIAL(ch2, encodeSetO, encodeWhiteSpace)) {

                        } else if (B64CHAR(ch2) || ch2 == '-') {
                            *out++ = '-';
                            inShift = 0;
                        } else {
                            inShift = 0;
                        }

                    }
                    else {
                        *out++ = '-';
                        inShift = 0;
                    }
                }
            }
        }
    }
    if (bitsleft) {
        *out++= B64(charsleft << (6-bitsleft) );
        *out++ = '-';
    }

    _PyString_Resize(&v, out - start);
    return v;
}

#undef SPECIAL
#undef B64
#undef B64CHAR
#undef UB64
#undef ENCODE
#undef DECODE

/* --- UTF-8 Codec -------------------------------------------------------- */

static
char utf8_code_length[256] = {
    /* Map UTF-8 encoded prefix byte to sequence length.  zero means
       illegal prefix.  see RFC 2279 for details */
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
    2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
    3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
    4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
};

PyObject *PyUnicode_DecodeUTF8(const char *s,
			       Py_ssize_t size,
			       const char *errors)
{
    return PyUnicode_DecodeUTF8Stateful(s, size, errors, NULL);
}

PyObject *PyUnicode_DecodeUTF8Stateful(const char *s,
			                Py_ssize_t size,
			                const char *errors,
			                Py_ssize_t *consumed)
{
    const char *starts = s;
    int n;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    const char *e;
    PyUnicodeObject *unicode;
    Py_UNICODE *p;
    const char *errmsg = "";
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    /* Note: size will always be longer than the resulting Unicode
       character count */
    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;
    if (size == 0) {
        if (consumed)
            *consumed = 0;
        return (PyObject *)unicode;
    }

    /* Unpack UTF-8 encoded data */
    p = unicode->str;
    e = s + size;

    while (s < e) {
        Py_UCS4 ch = (unsigned char)*s;

        if (ch < 0x80) {
            *p++ = (Py_UNICODE)ch;
            s++;
            continue;
        }

        n = utf8_code_length[ch];

        if (s + n > e) {
	    if (consumed)
		break;
	    else {
		errmsg = "unexpected end of data";
		startinpos = s-starts;
		endinpos = size;
		goto utf8Error;
	    }
	}

        switch (n) {

        case 0:
            errmsg = "unexpected code byte";
	    startinpos = s-starts;
	    endinpos = startinpos+1;
	    goto utf8Error;

        case 1:
            errmsg = "internal error";
	    startinpos = s-starts;
	    endinpos = startinpos+1;
	    goto utf8Error;

        case 2:
            if ((s[1] & 0xc0) != 0x80) {
                errmsg = "invalid data";
		startinpos = s-starts;
		endinpos = startinpos+2;
		goto utf8Error;
	    }
            ch = ((s[0] & 0x1f) << 6) + (s[1] & 0x3f);
            if (ch < 0x80) {
		startinpos = s-starts;
		endinpos = startinpos+2;
                errmsg = "illegal encoding";
		goto utf8Error;
	    }
	    else
		*p++ = (Py_UNICODE)ch;
            break;

        case 3:
            if ((s[1] & 0xc0) != 0x80 ||
                (s[2] & 0xc0) != 0x80) {
                errmsg = "invalid data";
		startinpos = s-starts;
		endinpos = startinpos+3;
		goto utf8Error;
	    }
            ch = ((s[0] & 0x0f) << 12) + ((s[1] & 0x3f) << 6) + (s[2] & 0x3f);
            if (ch < 0x0800) {
		/* Note: UTF-8 encodings of surrogates are considered
		   legal UTF-8 sequences;

		   XXX For wide builds (UCS-4) we should probably try
		       to recombine the surrogates into a single code
		       unit.
		*/
                errmsg = "illegal encoding";
		startinpos = s-starts;
		endinpos = startinpos+3;
		goto utf8Error;
	    }
	    else
		*p++ = (Py_UNICODE)ch;
            break;

        case 4:
            if ((s[1] & 0xc0) != 0x80 ||
                (s[2] & 0xc0) != 0x80 ||
                (s[3] & 0xc0) != 0x80) {
                errmsg = "invalid data";
		startinpos = s-starts;
		endinpos = startinpos+4;
		goto utf8Error;
	    }
            ch = ((s[0] & 0x7) << 18) + ((s[1] & 0x3f) << 12) +
                 ((s[2] & 0x3f) << 6) + (s[3] & 0x3f);
            /* validate and convert to UTF-16 */
            if ((ch < 0x10000)        /* minimum value allowed for 4
					 byte encoding */
                || (ch > 0x10ffff))   /* maximum value allowed for
					 UTF-16 */
	    {
                errmsg = "illegal encoding";
		startinpos = s-starts;
		endinpos = startinpos+4;
		goto utf8Error;
	    }
#ifdef Py_UNICODE_WIDE
	    *p++ = (Py_UNICODE)ch;
#else
            /*  compute and append the two surrogates: */

            /*  translate from 10000..10FFFF to 0..FFFF */
            ch -= 0x10000;

            /*  high surrogate = top 10 bits added to D800 */
            *p++ = (Py_UNICODE)(0xD800 + (ch >> 10));

            /*  low surrogate = bottom 10 bits added to DC00 */
            *p++ = (Py_UNICODE)(0xDC00 + (ch & 0x03FF));
#endif
            break;

        default:
            /* Other sizes are only needed for UCS-4 */
            errmsg = "unsupported Unicode code range";
	    startinpos = s-starts;
	    endinpos = startinpos+n;
	    goto utf8Error;
        }
        s += n;
	continue;

    utf8Error:
    outpos = p-PyUnicode_AS_UNICODE(unicode);
    if (unicode_decode_call_errorhandler(
	     errors, &errorHandler,
	     "utf8", errmsg,
	     starts, size, &startinpos, &endinpos, &exc, &s,
	     (PyObject **)&unicode, &outpos, &p))
	goto onError;
    }
    if (consumed)
	*consumed = s-starts;

    /* Adjust length */
    if (_PyUnicode_Resize(&unicode, p - unicode->str) < 0)
        goto onError;

    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)unicode;

onError:
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    Py_DECREF(unicode);
    return NULL;
}

/* Allocation strategy:  if the string is short, convert into a stack buffer
   and allocate exactly as much space needed at the end.  Else allocate the
   maximum possible needed (4 result bytes per Unicode character), and return
   the excess memory at the end.
*/
PyObject *
PyUnicode_EncodeUTF8(const Py_UNICODE *s,
		     Py_ssize_t size,
		     const char *errors)
{
#define MAX_SHORT_UNICHARS 300  /* largest size we'll do on the stack */

    Py_ssize_t i;           /* index into s of next input byte */
    PyObject *v;        /* result string object */
    char *p;            /* next free byte in output buffer */
    Py_ssize_t nallocated;  /* number of result bytes allocated */
    Py_ssize_t nneeded;        /* number of result bytes needed */
    char stackbuf[MAX_SHORT_UNICHARS * 4];

    assert(s != NULL);
    assert(size >= 0);

    if (size <= MAX_SHORT_UNICHARS) {
        /* Write into the stack buffer; nallocated can't overflow.
         * At the end, we'll allocate exactly as much heap space as it
         * turns out we need.
         */
        nallocated = Py_SAFE_DOWNCAST(sizeof(stackbuf), size_t, int);
        v = NULL;   /* will allocate after we're done */
        p = stackbuf;
    }
    else {
        /* Overallocate on the heap, and give the excess back at the end. */
        nallocated = size * 4;
        if (nallocated / 4 != size)  /* overflow! */
            return PyErr_NoMemory();
        v = PyString_FromStringAndSize(NULL, nallocated);
        if (v == NULL)
            return NULL;
        p = PyString_AS_STRING(v);
    }

    for (i = 0; i < size;) {
        Py_UCS4 ch = s[i++];

        if (ch < 0x80)
            /* Encode ASCII */
            *p++ = (char) ch;

        else if (ch < 0x0800) {
            /* Encode Latin-1 */
            *p++ = (char)(0xc0 | (ch >> 6));
            *p++ = (char)(0x80 | (ch & 0x3f));
        }
        else {
            /* Encode UCS2 Unicode ordinals */
            if (ch < 0x10000) {
                /* Special case: check for high surrogate */
                if (0xD800 <= ch && ch <= 0xDBFF && i != size) {
                    Py_UCS4 ch2 = s[i];
                    /* Check for low surrogate and combine the two to
                       form a UCS4 value */
                    if (0xDC00 <= ch2 && ch2 <= 0xDFFF) {
                        ch = ((ch - 0xD800) << 10 | (ch2 - 0xDC00)) + 0x10000;
                        i++;
                        goto encodeUCS4;
                    }
                    /* Fall through: handles isolated high surrogates */
                }
                *p++ = (char)(0xe0 | (ch >> 12));
                *p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
                *p++ = (char)(0x80 | (ch & 0x3f));
                continue;
    	    }
encodeUCS4:
            /* Encode UCS4 Unicode ordinals */
            *p++ = (char)(0xf0 | (ch >> 18));
            *p++ = (char)(0x80 | ((ch >> 12) & 0x3f));
            *p++ = (char)(0x80 | ((ch >> 6) & 0x3f));
            *p++ = (char)(0x80 | (ch & 0x3f));
        }
    }

    if (v == NULL) {
        /* This was stack allocated. */
        nneeded = p - stackbuf;
        assert(nneeded <= nallocated);
        v = PyString_FromStringAndSize(stackbuf, nneeded);
    }
    else {
    	/* Cut back to size actually needed. */
        nneeded = p - PyString_AS_STRING(v);
        assert(nneeded <= nallocated);
        _PyString_Resize(&v, nneeded);
    }
    return v;

#undef MAX_SHORT_UNICHARS
}

PyObject *PyUnicode_AsUTF8String(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        return NULL;
    }
    return PyUnicode_EncodeUTF8(PyUnicode_AS_UNICODE(unicode),
				PyUnicode_GET_SIZE(unicode),
				NULL);
}

/* --- UTF-32 Codec ------------------------------------------------------- */

PyObject *
PyUnicode_DecodeUTF32(const char *s,
		      Py_ssize_t size,
		      const char *errors,
		      int *byteorder)
{
    return PyUnicode_DecodeUTF32Stateful(s, size, errors, byteorder, NULL);
}

PyObject *
PyUnicode_DecodeUTF32Stateful(const char *s,
			      Py_ssize_t size,
			      const char *errors,
			      int *byteorder,
			      Py_ssize_t *consumed)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    PyUnicodeObject *unicode;
    Py_UNICODE *p;
#ifndef Py_UNICODE_WIDE
    int i, pairs;
#else
    const int pairs = 0;
#endif
    const unsigned char *q, *e;
    int bo = 0;       /* assume native ordering by default */
    const char *errmsg = "";
    /* Offsets from q for retrieving bytes in the right order. */
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
    int iorder[] = {0, 1, 2, 3};
#else
    int iorder[] = {3, 2, 1, 0};
#endif
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    /* On narrow builds we split characters outside the BMP into two
       codepoints => count how much extra space we need. */
#ifndef Py_UNICODE_WIDE
    for (i = pairs = 0; i < size/4; i++)
	if (((Py_UCS4 *)s)[i] >= 0x10000)
	    pairs++;
#endif

    /* This might be one to much, because of a BOM */
    unicode = _PyUnicode_New((size+3)/4+pairs);
    if (!unicode)
        return NULL;
    if (size == 0)
        return (PyObject *)unicode;

    /* Unpack UTF-32 encoded data */
    p = unicode->str;
    q = (unsigned char *)s;
    e = q + size;

    if (byteorder)
        bo = *byteorder;

    /* Check for BOM marks (U+FEFF) in the input and adjust current
       byte order setting accordingly. In native mode, the leading BOM
       mark is skipped, in all other modes, it is copied to the output
       stream as-is (giving a ZWNBSP character). */
    if (bo == 0) {
        if (size >= 4) {
            const Py_UCS4 bom = (q[iorder[3]] << 24) | (q[iorder[2]] << 16) |
                                (q[iorder[1]] << 8) | q[iorder[0]];
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
	    if (bom == 0x0000FEFF) {
		q += 4;
		bo = -1;
	    }
	    else if (bom == 0xFFFE0000) {
		q += 4;
		bo = 1;
	    }
#else
	    if (bom == 0x0000FEFF) {
		q += 4;
		bo = 1;
	    }
	    else if (bom == 0xFFFE0000) {
		q += 4;
		bo = -1;
	    }
#endif
	}
    }

    if (bo == -1) {
        /* force LE */
        iorder[0] = 0;
        iorder[1] = 1;
        iorder[2] = 2;
        iorder[3] = 3;
    }
    else if (bo == 1) {
        /* force BE */
        iorder[0] = 3;
        iorder[1] = 2;
        iorder[2] = 1;
        iorder[3] = 0;
    }

    while (q < e) {
	Py_UCS4 ch;
	/* remaining bytes at the end? (size should be divisible by 4) */
	if (e-q<4) {
	    if (consumed)
		break;
	    errmsg = "truncated data";
	    startinpos = ((const char *)q)-starts;
	    endinpos = ((const char *)e)-starts;
	    goto utf32Error;
	    /* The remaining input chars are ignored if the callback
	       chooses to skip the input */
	}
	ch = (q[iorder[3]] << 24) | (q[iorder[2]] << 16) |
	     (q[iorder[1]] << 8) | q[iorder[0]];

	if (ch >= 0x110000)
	{
	    errmsg = "codepoint not in range(0x110000)";
	    startinpos = ((const char *)q)-starts;
	    endinpos = startinpos+4;
	    goto utf32Error;
	}
#ifndef Py_UNICODE_WIDE
	if (ch >= 0x10000)
	{
	    *p++ = 0xD800 | ((ch-0x10000) >> 10);
	    *p++ = 0xDC00 | ((ch-0x10000) & 0x3FF);
	}
	else
#endif
	    *p++ = ch;
	q += 4;
	continue;
    utf32Error:
	outpos = p-PyUnicode_AS_UNICODE(unicode);
    if (unicode_decode_call_errorhandler(
         errors, &errorHandler,
         "utf32", errmsg,
         starts, size, &startinpos, &endinpos, &exc, &s,
         (PyObject **)&unicode, &outpos, &p))
	    goto onError;
    }

    if (byteorder)
        *byteorder = bo;

    if (consumed)
	*consumed = (const char *)q-starts;

    /* Adjust length */
    if (_PyUnicode_Resize(&unicode, p - unicode->str) < 0)
        goto onError;

    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)unicode;

onError:
    Py_DECREF(unicode);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

PyObject *
PyUnicode_EncodeUTF32(const Py_UNICODE *s,
		      Py_ssize_t size,
		      const char *errors,
		      int byteorder)
{
    PyObject *v;
    unsigned char *p;
#ifndef Py_UNICODE_WIDE
    int i, pairs;
#else
    const int pairs = 0;
#endif
    /* Offsets from p for storing byte pairs in the right order. */
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
    int iorder[] = {0, 1, 2, 3};
#else
    int iorder[] = {3, 2, 1, 0};
#endif

#define STORECHAR(CH)                       \
    do {                                    \
        p[iorder[3]] = ((CH) >> 24) & 0xff; \
        p[iorder[2]] = ((CH) >> 16) & 0xff; \
        p[iorder[1]] = ((CH) >> 8) & 0xff;  \
        p[iorder[0]] = (CH) & 0xff;         \
        p += 4;                             \
    } while(0)

    /* In narrow builds we can output surrogate pairs as one codepoint,
       so we need less space. */
#ifndef Py_UNICODE_WIDE
    for (i = pairs = 0; i < size-1; i++)
	if (0xD800 <= s[i] && s[i] <= 0xDBFF &&
	    0xDC00 <= s[i+1] && s[i+1] <= 0xDFFF)
	    pairs++;
#endif
    v = PyString_FromStringAndSize(NULL,
		  4 * (size - pairs + (byteorder == 0)));
    if (v == NULL)
        return NULL;

    p = (unsigned char *)PyString_AS_STRING(v);
    if (byteorder == 0)
	STORECHAR(0xFEFF);
    if (size == 0)
        return v;

    if (byteorder == -1) {
        /* force LE */
        iorder[0] = 0;
        iorder[1] = 1;
        iorder[2] = 2;
        iorder[3] = 3;
    }
    else if (byteorder == 1) {
        /* force BE */
        iorder[0] = 3;
        iorder[1] = 2;
        iorder[2] = 1;
        iorder[3] = 0;
    }

    while (size-- > 0) {
	Py_UCS4 ch = *s++;
#ifndef Py_UNICODE_WIDE
	if (0xD800 <= ch && ch <= 0xDBFF && size > 0) {
	    Py_UCS4 ch2 = *s;
	    if (0xDC00 <= ch2 && ch2 <= 0xDFFF) {
		ch = (((ch & 0x3FF)<<10) | (ch2 & 0x3FF)) + 0x10000;
		s++;
		size--;
	    }
	}
#endif
        STORECHAR(ch);
    }
    return v;
#undef STORECHAR
}

PyObject *PyUnicode_AsUTF32String(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        return NULL;
    }
    return PyUnicode_EncodeUTF32(PyUnicode_AS_UNICODE(unicode),
				 PyUnicode_GET_SIZE(unicode),
				 NULL,
				 0);
}

/* --- UTF-16 Codec ------------------------------------------------------- */

PyObject *
PyUnicode_DecodeUTF16(const char *s,
		      Py_ssize_t size,
		      const char *errors,
		      int *byteorder)
{
    return PyUnicode_DecodeUTF16Stateful(s, size, errors, byteorder, NULL);
}

PyObject *
PyUnicode_DecodeUTF16Stateful(const char *s,
			      Py_ssize_t size,
			      const char *errors,
			      int *byteorder,
			      Py_ssize_t *consumed)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    PyUnicodeObject *unicode;
    Py_UNICODE *p;
    const unsigned char *q, *e;
    int bo = 0;       /* assume native ordering by default */
    const char *errmsg = "";
    /* Offsets from q for retrieving byte pairs in the right order. */
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
    int ihi = 1, ilo = 0;
#else
    int ihi = 0, ilo = 1;
#endif
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    /* Note: size will always be longer than the resulting Unicode
       character count */
    unicode = _PyUnicode_New(size);
    if (!unicode)
        return NULL;
    if (size == 0)
        return (PyObject *)unicode;

    /* Unpack UTF-16 encoded data */
    p = unicode->str;
    q = (unsigned char *)s;
    e = q + size;

    if (byteorder)
        bo = *byteorder;

    /* Check for BOM marks (U+FEFF) in the input and adjust current
       byte order setting accordingly. In native mode, the leading BOM
       mark is skipped, in all other modes, it is copied to the output
       stream as-is (giving a ZWNBSP character). */
    if (bo == 0) {
        if (size >= 2) {
            const Py_UNICODE bom = (q[ihi] << 8) | q[ilo];
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
	    if (bom == 0xFEFF) {
		q += 2;
		bo = -1;
	    }
	    else if (bom == 0xFFFE) {
		q += 2;
		bo = 1;
	    }
#else
	    if (bom == 0xFEFF) {
		q += 2;
		bo = 1;
	    }
	    else if (bom == 0xFFFE) {
		q += 2;
		bo = -1;
	    }
#endif
	}
    }

    if (bo == -1) {
        /* force LE */
        ihi = 1;
        ilo = 0;
    }
    else if (bo == 1) {
        /* force BE */
        ihi = 0;
        ilo = 1;
    }

    while (q < e) {
	Py_UNICODE ch;
	/* remaining bytes at the end? (size should be even) */
	if (e-q<2) {
	    if (consumed)
		break;
	    errmsg = "truncated data";
	    startinpos = ((const char *)q)-starts;
	    endinpos = ((const char *)e)-starts;
	    goto utf16Error;
	    /* The remaining input chars are ignored if the callback
	       chooses to skip the input */
	}
	ch = (q[ihi] << 8) | q[ilo];

	q += 2;

	if (ch < 0xD800 || ch > 0xDFFF) {
	    *p++ = ch;
	    continue;
	}

	/* UTF-16 code pair: */
	if (q >= e) {
	    errmsg = "unexpected end of data";
	    startinpos = (((const char *)q)-2)-starts;
	    endinpos = ((const char *)e)-starts;
	    goto utf16Error;
	}
	if (0xD800 <= ch && ch <= 0xDBFF) {
	    Py_UNICODE ch2 = (q[ihi] << 8) | q[ilo];
	    q += 2;
	    if (0xDC00 <= ch2 && ch2 <= 0xDFFF) {
#ifndef Py_UNICODE_WIDE
		*p++ = ch;
		*p++ = ch2;
#else
		*p++ = (((ch & 0x3FF)<<10) | (ch2 & 0x3FF)) + 0x10000;
#endif
		continue;
	    }
	    else {
                errmsg = "illegal UTF-16 surrogate";
		startinpos = (((const char *)q)-4)-starts;
		endinpos = startinpos+2;
		goto utf16Error;
	    }

	}
	errmsg = "illegal encoding";
	startinpos = (((const char *)q)-2)-starts;
	endinpos = startinpos+2;
	/* Fall through to report the error */

    utf16Error:
	outpos = p-PyUnicode_AS_UNICODE(unicode);
	if (unicode_decode_call_errorhandler(
	         errors, &errorHandler,
	         "utf16", errmsg,
	         starts, size, &startinpos, &endinpos, &exc, (const char **)&q,
	         (PyObject **)&unicode, &outpos, &p))
	    goto onError;
    }

    if (byteorder)
        *byteorder = bo;

    if (consumed)
	*consumed = (const char *)q-starts;

    /* Adjust length */
    if (_PyUnicode_Resize(&unicode, p - unicode->str) < 0)
        goto onError;

    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)unicode;

onError:
    Py_DECREF(unicode);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

PyObject *
PyUnicode_EncodeUTF16(const Py_UNICODE *s,
		      Py_ssize_t size,
		      const char *errors,
		      int byteorder)
{
    PyObject *v;
    unsigned char *p;
#ifdef Py_UNICODE_WIDE
    int i, pairs;
#else
    const int pairs = 0;
#endif
    /* Offsets from p for storing byte pairs in the right order. */
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
    int ihi = 1, ilo = 0;
#else
    int ihi = 0, ilo = 1;
#endif

#define STORECHAR(CH)                   \
    do {                                \
        p[ihi] = ((CH) >> 8) & 0xff;    \
        p[ilo] = (CH) & 0xff;           \
        p += 2;                         \
    } while(0)

#ifdef Py_UNICODE_WIDE
    for (i = pairs = 0; i < size; i++)
	if (s[i] >= 0x10000)
	    pairs++;
#endif
    v = PyString_FromStringAndSize(NULL,
		  2 * (size + pairs + (byteorder == 0)));
    if (v == NULL)
        return NULL;

    p = (unsigned char *)PyString_AS_STRING(v);
    if (byteorder == 0)
	STORECHAR(0xFEFF);
    if (size == 0)
        return v;

    if (byteorder == -1) {
        /* force LE */
        ihi = 1;
        ilo = 0;
    }
    else if (byteorder == 1) {
        /* force BE */
        ihi = 0;
        ilo = 1;
    }

    while (size-- > 0) {
	Py_UNICODE ch = *s++;
	Py_UNICODE ch2 = 0;
#ifdef Py_UNICODE_WIDE
	if (ch >= 0x10000) {
	    ch2 = 0xDC00 | ((ch-0x10000) & 0x3FF);
	    ch  = 0xD800 | ((ch-0x10000) >> 10);
	}
#endif
        STORECHAR(ch);
        if (ch2)
            STORECHAR(ch2);
    }
    return v;
#undef STORECHAR
}

PyObject *PyUnicode_AsUTF16String(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        return NULL;
    }
    return PyUnicode_EncodeUTF16(PyUnicode_AS_UNICODE(unicode),
				 PyUnicode_GET_SIZE(unicode),
				 NULL,
				 0);
}

/* --- Unicode Escape Codec ----------------------------------------------- */

static _PyUnicode_Name_CAPI *ucnhash_CAPI = NULL;

PyObject *PyUnicode_DecodeUnicodeEscape(const char *s,
					Py_ssize_t size,
					const char *errors)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    int i;
    PyUnicodeObject *v;
    Py_UNICODE *p;
    const char *end;
    char* message;
    Py_UCS4 chr = 0xffffffff; /* in case 'getcode' messes up */
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    /* Escaped strings will always be longer than the resulting
       Unicode string, so we start with size here and then reduce the
       length after conversion to the true value.
       (but if the error callback returns a long replacement string
       we'll have to allocate more space) */
    v = _PyUnicode_New(size);
    if (v == NULL)
        goto onError;
    if (size == 0)
        return (PyObject *)v;

    p = PyUnicode_AS_UNICODE(v);
    end = s + size;

    while (s < end) {
        unsigned char c;
        Py_UNICODE x;
        int digits;

        /* Non-escape characters are interpreted as Unicode ordinals */
        if (*s != '\\') {
            *p++ = (unsigned char) *s++;
            continue;
        }

        startinpos = s-starts;
        /* \ - Escapes */
        s++;
        c = *s++;
        if (s > end)
            c = '\0'; /* Invalid after \ */
        switch (c) {

        /* \x escapes */
        case '\n': break;
        case '\\': *p++ = '\\'; break;
        case '\'': *p++ = '\''; break;
        case '\"': *p++ = '\"'; break;
        case 'b': *p++ = '\b'; break;
        case 'f': *p++ = '\014'; break; /* FF */
        case 't': *p++ = '\t'; break;
        case 'n': *p++ = '\n'; break;
        case 'r': *p++ = '\r'; break;
        case 'v': *p++ = '\013'; break; /* VT */
        case 'a': *p++ = '\007'; break; /* BEL, not classic C */

        /* \OOO (octal) escapes */
        case '0': case '1': case '2': case '3':
        case '4': case '5': case '6': case '7':
            x = s[-1] - '0';
            if (s < end && '0' <= *s && *s <= '7') {
                x = (x<<3) + *s++ - '0';
                if (s < end && '0' <= *s && *s <= '7')
                    x = (x<<3) + *s++ - '0';
            }
            *p++ = x;
            break;

        /* hex escapes */
        /* \xXX */
        case 'x':
            digits = 2;
            message = "truncated \\xXX escape";
            goto hexescape;

        /* \uXXXX */
        case 'u':
            digits = 4;
            message = "truncated \\uXXXX escape";
            goto hexescape;

        /* \UXXXXXXXX */
        case 'U':
            digits = 8;
            message = "truncated \\UXXXXXXXX escape";
        hexescape:
            chr = 0;
            outpos = p-PyUnicode_AS_UNICODE(v);
            if (s+digits>end) {
                endinpos = size;
                if (unicode_decode_call_errorhandler(
                    errors, &errorHandler,
                    "unicodeescape", "end of string in escape sequence",
                    starts, size, &startinpos, &endinpos, &exc, &s,
                    (PyObject **)&v, &outpos, &p))
                    goto onError;
                goto nextByte;
            }
            for (i = 0; i < digits; ++i) {
                c = (unsigned char) s[i];
                if (!isxdigit(c)) {
                    endinpos = (s+i+1)-starts;
                    if (unicode_decode_call_errorhandler(
                        errors, &errorHandler,
                        "unicodeescape", message,
                        starts, size, &startinpos, &endinpos, &exc, &s,
                        (PyObject **)&v, &outpos, &p))
                        goto onError;
                    goto nextByte;
                }
                chr = (chr<<4) & ~0xF;
                if (c >= '0' && c <= '9')
                    chr += c - '0';
                else if (c >= 'a' && c <= 'f')
                    chr += 10 + c - 'a';
                else
                    chr += 10 + c - 'A';
            }
            s += i;
            if (chr == 0xffffffff && PyErr_Occurred())
                /* _decoding_error will have already written into the
                   target buffer. */
                break;
        store:
            /* when we get here, chr is a 32-bit unicode character */
            if (chr <= 0xffff)
                /* UCS-2 character */
                *p++ = (Py_UNICODE) chr;
            else if (chr <= 0x10ffff) {
                /* UCS-4 character. Either store directly, or as
                   surrogate pair. */
#ifdef Py_UNICODE_WIDE
                *p++ = chr;
#else
                chr -= 0x10000L;
                *p++ = 0xD800 + (Py_UNICODE) (chr >> 10);
                *p++ = 0xDC00 + (Py_UNICODE) (chr & 0x03FF);
#endif
            } else {
                endinpos = s-starts;
                outpos = p-PyUnicode_AS_UNICODE(v);
                if (unicode_decode_call_errorhandler(
                    errors, &errorHandler,
                    "unicodeescape", "illegal Unicode character",
                    starts, size, &startinpos, &endinpos, &exc, &s,
                    (PyObject **)&v, &outpos, &p))
                    goto onError;
            }
            break;

        /* \N{name} */
        case 'N':
            message = "malformed \\N character escape";
            if (ucnhash_CAPI == NULL) {
                /* load the unicode data module */
                PyObject *m, *api;
                m = PyImport_ImportModuleNoBlock("unicodedata");
                if (m == NULL)
                    goto ucnhashError;
                api = PyObject_GetAttrString(m, "ucnhash_CAPI");
                Py_DECREF(m);
                if (api == NULL)
                    goto ucnhashError;
                ucnhash_CAPI = (_PyUnicode_Name_CAPI *)PyCObject_AsVoidPtr(api);
                Py_DECREF(api);
                if (ucnhash_CAPI == NULL)
                    goto ucnhashError;
            }
            if (*s == '{') {
                const char *start = s+1;
                /* look for the closing brace */
                while (*s != '}' && s < end)
                    s++;
                if (s > start && s < end && *s == '}') {
                    /* found a name.  look it up in the unicode database */
                    message = "unknown Unicode character name";
                    s++;
                    if (ucnhash_CAPI->getcode(NULL, start, (int)(s-start-1), &chr))
                        goto store;
                }
            }
            endinpos = s-starts;
            outpos = p-PyUnicode_AS_UNICODE(v);
            if (unicode_decode_call_errorhandler(
                errors, &errorHandler,
                "unicodeescape", message,
                starts, size, &startinpos, &endinpos, &exc, &s,
                (PyObject **)&v, &outpos, &p))
                goto onError;
            break;

        default:
            if (s > end) {
                message = "\\ at end of string";
                s--;
                endinpos = s-starts;
                outpos = p-PyUnicode_AS_UNICODE(v);
                if (unicode_decode_call_errorhandler(
                    errors, &errorHandler,
                    "unicodeescape", message,
                    starts, size, &startinpos, &endinpos, &exc, &s,
                    (PyObject **)&v, &outpos, &p))
                    goto onError;
            }
            else {
                *p++ = '\\';
                *p++ = (unsigned char)s[-1];
            }
            break;
        }
        nextByte:
        ;
    }
    if (_PyUnicode_Resize(&v, p - PyUnicode_AS_UNICODE(v)) < 0)
        goto onError;
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)v;

ucnhashError:
    PyErr_SetString(
        PyExc_UnicodeError,
        "\\N escapes not supported (can't load unicodedata module)"
        );
    Py_XDECREF(v);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;

onError:
    Py_XDECREF(v);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

/* Return a Unicode-Escape string version of the Unicode object.

   If quotes is true, the string is enclosed in u"" or u'' quotes as
   appropriate.

*/

Py_LOCAL_INLINE(const Py_UNICODE *) findchar(const Py_UNICODE *s,
                                      Py_ssize_t size,
                                      Py_UNICODE ch)
{
    /* like wcschr, but doesn't stop at NULL characters */

    while (size-- > 0) {
        if (*s == ch)
            return s;
        s++;
    }

    return NULL;
}

static
PyObject *unicodeescape_string(const Py_UNICODE *s,
                               Py_ssize_t size,
                               int quotes)
{
    PyObject *repr;
    char *p;

    static const char *hexdigit = "0123456789abcdef";

    /* XXX(nnorwitz): rather than over-allocating, it would be
       better to choose a different scheme.  Perhaps scan the
       first N-chars of the string and allocate based on that size.
    */
    /* Initial allocation is based on the longest-possible unichr
       escape.

       In wide (UTF-32) builds '\U00xxxxxx' is 10 chars per source
       unichr, so in this case it's the longest unichr escape. In
       narrow (UTF-16) builds this is five chars per source unichr
       since there are two unichrs in the surrogate pair, so in narrow
       (UTF-16) builds it's not the longest unichr escape.

       In wide or narrow builds '\uxxxx' is 6 chars per source unichr,
       so in the narrow (UTF-16) build case it's the longest unichr
       escape.
    */

    repr = PyString_FromStringAndSize(NULL,
        2
#ifdef Py_UNICODE_WIDE
        + 10*size
#else
        + 6*size
#endif
        + 1);
    if (repr == NULL)
        return NULL;

    p = PyString_AS_STRING(repr);

    if (quotes) {
        *p++ = 'u';
        *p++ = (findchar(s, size, '\'') &&
                !findchar(s, size, '"')) ? '"' : '\'';
    }
    while (size-- > 0) {
        Py_UNICODE ch = *s++;

        /* Escape quotes and backslashes */
        if ((quotes &&
	     ch == (Py_UNICODE) PyString_AS_STRING(repr)[1]) || ch == '\\') {
            *p++ = '\\';
            *p++ = (char) ch;
	    continue;
        }

#ifdef Py_UNICODE_WIDE
        /* Map 21-bit characters to '\U00xxxxxx' */
        else if (ch >= 0x10000) {
            *p++ = '\\';
            *p++ = 'U';
            *p++ = hexdigit[(ch >> 28) & 0x0000000F];
            *p++ = hexdigit[(ch >> 24) & 0x0000000F];
            *p++ = hexdigit[(ch >> 20) & 0x0000000F];
            *p++ = hexdigit[(ch >> 16) & 0x0000000F];
            *p++ = hexdigit[(ch >> 12) & 0x0000000F];
            *p++ = hexdigit[(ch >> 8) & 0x0000000F];
            *p++ = hexdigit[(ch >> 4) & 0x0000000F];
            *p++ = hexdigit[ch & 0x0000000F];
	    continue;
        }
#else
	/* Map UTF-16 surrogate pairs to '\U00xxxxxx' */
	else if (ch >= 0xD800 && ch < 0xDC00) {
	    Py_UNICODE ch2;
	    Py_UCS4 ucs;

	    ch2 = *s++;
	    size--;
	    if (ch2 >= 0xDC00 && ch2 <= 0xDFFF) {
		ucs = (((ch & 0x03FF) << 10) | (ch2 & 0x03FF)) + 0x00010000;
		*p++ = '\\';
		*p++ = 'U';
		*p++ = hexdigit[(ucs >> 28) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 24) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 20) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 16) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 12) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 8) & 0x0000000F];
		*p++ = hexdigit[(ucs >> 4) & 0x0000000F];
		*p++ = hexdigit[ucs & 0x0000000F];
		continue;
	    }
	    /* Fall through: isolated surrogates are copied as-is */
	    s--;
	    size++;
	}
#endif

        /* Map 16-bit characters to '\uxxxx' */
        if (ch >= 256) {
            *p++ = '\\';
            *p++ = 'u';
            *p++ = hexdigit[(ch >> 12) & 0x000F];
            *p++ = hexdigit[(ch >> 8) & 0x000F];
            *p++ = hexdigit[(ch >> 4) & 0x000F];
            *p++ = hexdigit[ch & 0x000F];
        }

        /* Map special whitespace to '\t', \n', '\r' */
        else if (ch == '\t') {
            *p++ = '\\';
            *p++ = 't';
        }
        else if (ch == '\n') {
            *p++ = '\\';
            *p++ = 'n';
        }
        else if (ch == '\r') {
            *p++ = '\\';
            *p++ = 'r';
        }

        /* Map non-printable US ASCII to '\xhh' */
        else if (ch < ' ' || ch >= 0x7F) {
            *p++ = '\\';
            *p++ = 'x';
            *p++ = hexdigit[(ch >> 4) & 0x000F];
            *p++ = hexdigit[ch & 0x000F];
        }

        /* Copy everything else as-is */
        else
            *p++ = (char) ch;
    }
    if (quotes)
        *p++ = PyString_AS_STRING(repr)[1];

    *p = '\0';
    _PyString_Resize(&repr, p - PyString_AS_STRING(repr));
    return repr;
}

PyObject *PyUnicode_EncodeUnicodeEscape(const Py_UNICODE *s,
					Py_ssize_t size)
{
    return unicodeescape_string(s, size, 0);
}

PyObject *PyUnicode_AsUnicodeEscapeString(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        return NULL;
    }
    return PyUnicode_EncodeUnicodeEscape(PyUnicode_AS_UNICODE(unicode),
					 PyUnicode_GET_SIZE(unicode));
}

/* --- Raw Unicode Escape Codec ------------------------------------------- */

PyObject *PyUnicode_DecodeRawUnicodeEscape(const char *s,
					   Py_ssize_t size,
					   const char *errors)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    PyUnicodeObject *v;
    Py_UNICODE *p;
    const char *end;
    const char *bs;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    /* Escaped strings will always be longer than the resulting
       Unicode string, so we start with size here and then reduce the
       length after conversion to the true value. (But decoding error
       handler might have to resize the string) */
    v = _PyUnicode_New(size);
    if (v == NULL)
	goto onError;
    if (size == 0)
	return (PyObject *)v;
    p = PyUnicode_AS_UNICODE(v);
    end = s + size;
    while (s < end) {
	unsigned char c;
	Py_UCS4 x;
	int i;
        int count;

	/* Non-escape characters are interpreted as Unicode ordinals */
	if (*s != '\\') {
	    *p++ = (unsigned char)*s++;
	    continue;
	}
	startinpos = s-starts;

	/* \u-escapes are only interpreted iff the number of leading
	   backslashes if odd */
	bs = s;
	for (;s < end;) {
	    if (*s != '\\')
		break;
	    *p++ = (unsigned char)*s++;
	}
	if (((s - bs) & 1) == 0 ||
	    s >= end ||
	    (*s != 'u' && *s != 'U')) {
	    continue;
	}
	p--;
        count = *s=='u' ? 4 : 8;
	s++;

	/* \uXXXX with 4 hex digits, \Uxxxxxxxx with 8 */
	outpos = p-PyUnicode_AS_UNICODE(v);
	for (x = 0, i = 0; i < count; ++i, ++s) {
	    c = (unsigned char)*s;
	    if (!isxdigit(c)) {
		endinpos = s-starts;
		if (unicode_decode_call_errorhandler(
		    errors, &errorHandler,
		    "rawunicodeescape", "truncated \\uXXXX",
		    starts, size, &startinpos, &endinpos, &exc, &s,
		    (PyObject **)&v, &outpos, &p))
		    goto onError;
		goto nextByte;
	    }
	    x = (x<<4) & ~0xF;
	    if (c >= '0' && c <= '9')
		x += c - '0';
	    else if (c >= 'a' && c <= 'f')
		x += 10 + c - 'a';
	    else
		x += 10 + c - 'A';
	}
        if (x <= 0xffff)
                /* UCS-2 character */
                *p++ = (Py_UNICODE) x;
        else if (x <= 0x10ffff) {
                /* UCS-4 character. Either store directly, or as
                   surrogate pair. */
#ifdef Py_UNICODE_WIDE
                *p++ = (Py_UNIC0DE) x;
#else
                x -= 0x10000L;
                *p++ = 0xD800 + (Py_UNICODE) (x >> 10);
                *p++ = 0xDC00 + (Py_UNICODE) (x & 0x03FF);
#endif
        } else {
            endinpos = s-starts;
            outpos = p-PyUnicode_AS_UNICODE(v);
            if (unicode_decode_call_errorhandler(
                    errors, &errorHandler,
                    "rawunicodeescape", "\\Uxxxxxxxx out of range",
		    starts, size, &startinpos, &endinpos, &exc, &s,
		    (PyObject **)&v, &outpos, &p))
		    goto onError;
        }
	nextByte:
	;
    }
    if (_PyUnicode_Resize(&v, p - PyUnicode_AS_UNICODE(v)) < 0)
	goto onError;
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)v;

 onError:
    Py_XDECREF(v);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

PyObject *PyUnicode_EncodeRawUnicodeEscape(const Py_UNICODE *s,
					   Py_ssize_t size)
{
    PyObject *repr;
    char *p;
    char *q;

    static const char *hexdigit = "0123456789abcdef";

#ifdef Py_UNICODE_WIDE
    repr = PyString_FromStringAndSize(NULL, 10 * size);
#else
    repr = PyString_FromStringAndSize(NULL, 6 * size);
#endif
    if (repr == NULL)
        return NULL;
    if (size == 0)
	return repr;

    p = q = PyString_AS_STRING(repr);
    while (size-- > 0) {
        Py_UNICODE ch = *s++;
#ifdef Py_UNICODE_WIDE
	/* Map 32-bit characters to '\Uxxxxxxxx' */
	if (ch >= 0x10000) {
            *p++ = '\\';
            *p++ = 'U';
            *p++ = hexdigit[(ch >> 28) & 0xf];
            *p++ = hexdigit[(ch >> 24) & 0xf];
            *p++ = hexdigit[(ch >> 20) & 0xf];
            *p++ = hexdigit[(ch >> 16) & 0xf];
            *p++ = hexdigit[(ch >> 12) & 0xf];
            *p++ = hexdigit[(ch >> 8) & 0xf];
            *p++ = hexdigit[(ch >> 4) & 0xf];
            *p++ = hexdigit[ch & 15];
        }
        else
#else
	/* Map UTF-16 surrogate pairs to '\U00xxxxxx' */
	if (ch >= 0xD800 && ch < 0xDC00) {
	    Py_UNICODE ch2;
	    Py_UCS4 ucs;

	    ch2 = *s++;
	    size--;
	    if (ch2 >= 0xDC00 && ch2 <= 0xDFFF) {
		ucs = (((ch & 0x03FF) << 10) | (ch2 & 0x03FF)) + 0x00010000;
		*p++ = '\\';
		*p++ = 'U';
		*p++ = hexdigit[(ucs >> 28) & 0xf];
		*p++ = hexdigit[(ucs >> 24) & 0xf];
		*p++ = hexdigit[(ucs >> 20) & 0xf];
		*p++ = hexdigit[(ucs >> 16) & 0xf];
		*p++ = hexdigit[(ucs >> 12) & 0xf];
		*p++ = hexdigit[(ucs >> 8) & 0xf];
		*p++ = hexdigit[(ucs >> 4) & 0xf];
		*p++ = hexdigit[ucs & 0xf];
		continue;
	    }
	    /* Fall through: isolated surrogates are copied as-is */
	    s--;
	    size++;
	}
#endif
	/* Map 16-bit characters to '\uxxxx' */
	if (ch >= 256) {
            *p++ = '\\';
            *p++ = 'u';
            *p++ = hexdigit[(ch >> 12) & 0xf];
            *p++ = hexdigit[(ch >> 8) & 0xf];
            *p++ = hexdigit[(ch >> 4) & 0xf];
            *p++ = hexdigit[ch & 15];
        }
	/* Copy everything else as-is */
	else
            *p++ = (char) ch;
    }
    *p = '\0';
    _PyString_Resize(&repr, p - q);
    return repr;
}

PyObject *PyUnicode_AsRawUnicodeEscapeString(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
	PyErr_BadArgument();
	return NULL;
    }
    return PyUnicode_EncodeRawUnicodeEscape(PyUnicode_AS_UNICODE(unicode),
					    PyUnicode_GET_SIZE(unicode));
}

/* --- Unicode Internal Codec ------------------------------------------- */

PyObject *_PyUnicode_DecodeUnicodeInternal(const char *s,
					   Py_ssize_t size,
					   const char *errors)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    PyUnicodeObject *v;
    Py_UNICODE *p;
    const char *end;
    const char *reason;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

#ifdef Py_UNICODE_WIDE
    Py_UNICODE unimax = PyUnicode_GetMax();
#endif

    /* XXX overflow detection missing */
    v = _PyUnicode_New((size+Py_UNICODE_SIZE-1)/ Py_UNICODE_SIZE);
    if (v == NULL)
	goto onError;
    if (PyUnicode_GetSize((PyObject *)v) == 0)
	return (PyObject *)v;
    p = PyUnicode_AS_UNICODE(v);
    end = s + size;

    while (s < end) {
        memcpy(p, s, sizeof(Py_UNICODE));
        /* We have to sanity check the raw data, otherwise doom looms for
           some malformed UCS-4 data. */
        if (
            #ifdef Py_UNICODE_WIDE
            *p > unimax || *p < 0 ||
            #endif
            end-s < Py_UNICODE_SIZE
            )
            {
            startinpos = s - starts;
            if (end-s < Py_UNICODE_SIZE) {
                endinpos = end-starts;
                reason = "truncated input";
            }
            else {
                endinpos = s - starts + Py_UNICODE_SIZE;
                reason = "illegal code point (> 0x10FFFF)";
            }
            outpos = p - PyUnicode_AS_UNICODE(v);
            if (unicode_decode_call_errorhandler(
                    errors, &errorHandler,
                    "unicode_internal", reason,
                    starts, size, &startinpos, &endinpos, &exc, &s,
                    (PyObject **)&v, &outpos, &p)) {
                goto onError;
            }
        }
        else {
            p++;
            s += Py_UNICODE_SIZE;
        }
    }

    if (_PyUnicode_Resize(&v, p - PyUnicode_AS_UNICODE(v)) < 0)
        goto onError;
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)v;

 onError:
    Py_XDECREF(v);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

/* --- Latin-1 Codec ------------------------------------------------------ */

PyObject *PyUnicode_DecodeLatin1(const char *s,
				 Py_ssize_t size,
				 const char *errors)
{
    PyUnicodeObject *v;
    Py_UNICODE *p;

    /* Latin-1 is equivalent to the first 256 ordinals in Unicode. */
    if (size == 1) {
	Py_UNICODE r = *(unsigned char*)s;
	return PyUnicode_FromUnicode(&r, 1);
    }

    v = _PyUnicode_New(size);
    if (v == NULL)
	goto onError;
    if (size == 0)
	return (PyObject *)v;
    p = PyUnicode_AS_UNICODE(v);
    while (size-- > 0)
	*p++ = (unsigned char)*s++;
    return (PyObject *)v;

 onError:
    Py_XDECREF(v);
    return NULL;
}

/* create or adjust a UnicodeEncodeError */
static void make_encode_exception(PyObject **exceptionObject,
    const char *encoding,
    const Py_UNICODE *unicode, Py_ssize_t size,
    Py_ssize_t startpos, Py_ssize_t endpos,
    const char *reason)
{
    if (*exceptionObject == NULL) {
	*exceptionObject = PyUnicodeEncodeError_Create(
	    encoding, unicode, size, startpos, endpos, reason);
    }
    else {
	if (PyUnicodeEncodeError_SetStart(*exceptionObject, startpos))
	    goto onError;
	if (PyUnicodeEncodeError_SetEnd(*exceptionObject, endpos))
	    goto onError;
	if (PyUnicodeEncodeError_SetReason(*exceptionObject, reason))
	    goto onError;
	return;
	onError:
	Py_DECREF(*exceptionObject);
	*exceptionObject = NULL;
    }
}

/* raises a UnicodeEncodeError */
static void raise_encode_exception(PyObject **exceptionObject,
    const char *encoding,
    const Py_UNICODE *unicode, Py_ssize_t size,
    Py_ssize_t startpos, Py_ssize_t endpos,
    const char *reason)
{
    make_encode_exception(exceptionObject,
	encoding, unicode, size, startpos, endpos, reason);
    if (*exceptionObject != NULL)
	PyCodec_StrictErrors(*exceptionObject);
}

/* error handling callback helper:
   build arguments, call the callback and check the arguments,
   put the result into newpos and return the replacement string, which
   has to be freed by the caller */
static PyObject *unicode_encode_call_errorhandler(const char *errors,
    PyObject **errorHandler,
    const char *encoding, const char *reason,
    const Py_UNICODE *unicode, Py_ssize_t size, PyObject **exceptionObject,
    Py_ssize_t startpos, Py_ssize_t endpos,
    Py_ssize_t *newpos)
{
    static char *argparse = "O!n;encoding error handler must return (unicode, int) tuple";

    PyObject *restuple;
    PyObject *resunicode;

    if (*errorHandler == NULL) {
	*errorHandler = PyCodec_LookupError(errors);
        if (*errorHandler == NULL)
	    return NULL;
    }

    make_encode_exception(exceptionObject,
	encoding, unicode, size, startpos, endpos, reason);
    if (*exceptionObject == NULL)
	return NULL;

    restuple = PyObject_CallFunctionObjArgs(
	*errorHandler, *exceptionObject, NULL);
    if (restuple == NULL)
	return NULL;
    if (!PyTuple_Check(restuple)) {
	PyErr_Format(PyExc_TypeError, &argparse[4]);
	Py_DECREF(restuple);
	return NULL;
    }
    if (!PyArg_ParseTuple(restuple, argparse, &PyUnicode_Type,
	&resunicode, newpos)) {
	Py_DECREF(restuple);
	return NULL;
    }
    if (*newpos<0)
	*newpos = size+*newpos;
    if (*newpos<0 || *newpos>size) {
	PyErr_Format(PyExc_IndexError, "position %zd from error handler out of bounds", *newpos);
	Py_DECREF(restuple);
	return NULL;
    }
    Py_INCREF(resunicode);
    Py_DECREF(restuple);
    return resunicode;
}

static PyObject *unicode_encode_ucs1(const Py_UNICODE *p,
				 Py_ssize_t size,
				 const char *errors,
				 int limit)
{
    /* output object */
    PyObject *res;
    /* pointers to the beginning and end+1 of input */
    const Py_UNICODE *startp = p;
    const Py_UNICODE *endp = p + size;
    /* pointer to the beginning of the unencodable characters */
    /* const Py_UNICODE *badp = NULL; */
    /* pointer into the output */
    char *str;
    /* current output position */
    Py_ssize_t respos = 0;
    Py_ssize_t ressize;
    const char *encoding = (limit == 256) ? "latin-1" : "ascii";
    const char *reason = (limit == 256) ? "ordinal not in range(256)" : "ordinal not in range(128)";
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    /* the following variable is used for caching string comparisons
     * -1=not initialized, 0=unknown, 1=strict, 2=replace, 3=ignore, 4=xmlcharrefreplace */
    int known_errorHandler = -1;

    /* allocate enough for a simple encoding without
       replacements, if we need more, we'll resize */
    res = PyString_FromStringAndSize(NULL, size);
    if (res == NULL)
        goto onError;
    if (size == 0)
	return res;
    str = PyString_AS_STRING(res);
    ressize = size;

    while (p<endp) {
	Py_UNICODE c = *p;

	/* can we encode this? */
	if (c<limit) {
	    /* no overflow check, because we know that the space is enough */
	    *str++ = (char)c;
	    ++p;
	}
	else {
	    Py_ssize_t unicodepos = p-startp;
	    Py_ssize_t requiredsize;
	    PyObject *repunicode;
	    Py_ssize_t repsize;
	    Py_ssize_t newpos;
	    Py_ssize_t respos;
	    Py_UNICODE *uni2;
	    /* startpos for collecting unencodable chars */
	    const Py_UNICODE *collstart = p;
	    const Py_UNICODE *collend = p;
	    /* find all unecodable characters */
	    while ((collend < endp) && ((*collend)>=limit))
		++collend;
	    /* cache callback name lookup (if not done yet, i.e. it's the first error) */
	    if (known_errorHandler==-1) {
		if ((errors==NULL) || (!strcmp(errors, "strict")))
		    known_errorHandler = 1;
		else if (!strcmp(errors, "replace"))
		    known_errorHandler = 2;
		else if (!strcmp(errors, "ignore"))
		    known_errorHandler = 3;
		else if (!strcmp(errors, "xmlcharrefreplace"))
		    known_errorHandler = 4;
		else
		    known_errorHandler = 0;
	    }
	    switch (known_errorHandler) {
		case 1: /* strict */
		    raise_encode_exception(&exc, encoding, startp, size, collstart-startp, collend-startp, reason);
		    goto onError;
		case 2: /* replace */
		    while (collstart++<collend)
			*str++ = '?'; /* fall through */
		case 3: /* ignore */
		    p = collend;
		    break;
		case 4: /* xmlcharrefreplace */
		    respos = str-PyString_AS_STRING(res);
		    /* determine replacement size (temporarily (mis)uses p) */
		    for (p = collstart, repsize = 0; p < collend; ++p) {
			if (*p<10)
			    repsize += 2+1+1;
			else if (*p<100)
			    repsize += 2+2+1;
			else if (*p<1000)
			    repsize += 2+3+1;
			else if (*p<10000)
			    repsize += 2+4+1;
#ifndef Py_UNICODE_WIDE
			else
			    repsize += 2+5+1;
#else
			else if (*p<100000)
			    repsize += 2+5+1;
			else if (*p<1000000)
			    repsize += 2+6+1;
			else
			    repsize += 2+7+1;
#endif
		    }
		    requiredsize = respos+repsize+(endp-collend);
		    if (requiredsize > ressize) {
			if (requiredsize<2*ressize)
			    requiredsize = 2*ressize;
			if (_PyString_Resize(&res, requiredsize))
			    goto onError;
			str = PyString_AS_STRING(res) + respos;
			ressize = requiredsize;
		    }
		    /* generate replacement (temporarily (mis)uses p) */
		    for (p = collstart; p < collend; ++p) {
			str += sprintf(str, "&#%d;", (int)*p);
		    }
		    p = collend;
		    break;
		default:
		    repunicode = unicode_encode_call_errorhandler(errors, &errorHandler,
			encoding, reason, startp, size, &exc,
			collstart-startp, collend-startp, &newpos);
		    if (repunicode == NULL)
			goto onError;
		    /* need more space? (at least enough for what we
		       have+the replacement+the rest of the string, so
		       we won't have to check space for encodable characters) */
		    respos = str-PyString_AS_STRING(res);
		    repsize = PyUnicode_GET_SIZE(repunicode);
		    requiredsize = respos+repsize+(endp-collend);
		    if (requiredsize > ressize) {
			if (requiredsize<2*ressize)
			    requiredsize = 2*ressize;
			if (_PyString_Resize(&res, requiredsize)) {
			    Py_DECREF(repunicode);
			    goto onError;
			}
			str = PyString_AS_STRING(res) + respos;
			ressize = requiredsize;
		    }
		    /* check if there is anything unencodable in the replacement
		       and copy it to the output */
		    for (uni2 = PyUnicode_AS_UNICODE(repunicode);repsize-->0; ++uni2, ++str) {
			c = *uni2;
			if (c >= limit) {
			    raise_encode_exception(&exc, encoding, startp, size,
				unicodepos, unicodepos+1, reason);
			    Py_DECREF(repunicode);
			    goto onError;
			}
			*str = (char)c;
		    }
		    p = startp + newpos;
		    Py_DECREF(repunicode);
	    }
	}
    }
    /* Resize if we allocated to much */
    respos = str-PyString_AS_STRING(res);
    if (respos<ressize)
       /* If this falls res will be NULL */
	_PyString_Resize(&res, respos);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return res;

    onError:
    Py_XDECREF(res);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

PyObject *PyUnicode_EncodeLatin1(const Py_UNICODE *p,
				 Py_ssize_t size,
				 const char *errors)
{
    return unicode_encode_ucs1(p, size, errors, 256);
}

PyObject *PyUnicode_AsLatin1String(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
	PyErr_BadArgument();
	return NULL;
    }
    return PyUnicode_EncodeLatin1(PyUnicode_AS_UNICODE(unicode),
				  PyUnicode_GET_SIZE(unicode),
				  NULL);
}

/* --- 7-bit ASCII Codec -------------------------------------------------- */

PyObject *PyUnicode_DecodeASCII(const char *s,
				Py_ssize_t size,
				const char *errors)
{
    const char *starts = s;
    PyUnicodeObject *v;
    Py_UNICODE *p;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    const char *e;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;

    /* ASCII is equivalent to the first 128 ordinals in Unicode. */
    if (size == 1 && *(unsigned char*)s < 128) {
	Py_UNICODE r = *(unsigned char*)s;
	return PyUnicode_FromUnicode(&r, 1);
    }

    v = _PyUnicode_New(size);
    if (v == NULL)
	goto onError;
    if (size == 0)
	return (PyObject *)v;
    p = PyUnicode_AS_UNICODE(v);
    e = s + size;
    while (s < e) {
	register unsigned char c = (unsigned char)*s;
	if (c < 128) {
	    *p++ = c;
	    ++s;
	}
	else {
	    startinpos = s-starts;
	    endinpos = startinpos + 1;
	    outpos = p - (Py_UNICODE *)PyUnicode_AS_UNICODE(v);
	    if (unicode_decode_call_errorhandler(
		 errors, &errorHandler,
		 "ascii", "ordinal not in range(128)",
		 starts, size, &startinpos, &endinpos, &exc, &s,
		 (PyObject **)&v, &outpos, &p))
		goto onError;
	}
    }
    if (p - PyUnicode_AS_UNICODE(v) < PyString_GET_SIZE(v))
	if (_PyUnicode_Resize(&v, p - PyUnicode_AS_UNICODE(v)) < 0)
	    goto onError;
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)v;

 onError:
    Py_XDECREF(v);
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return NULL;
}

PyObject *PyUnicode_EncodeASCII(const Py_UNICODE *p,
				Py_ssize_t size,
				const char *errors)
{
    return unicode_encode_ucs1(p, size, errors, 128);
}

PyObject *PyUnicode_AsASCIIString(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
	PyErr_BadArgument();
	return NULL;
    }
    return PyUnicode_EncodeASCII(PyUnicode_AS_UNICODE(unicode),
				 PyUnicode_GET_SIZE(unicode),
				 NULL);
}

#if defined(MS_WINDOWS) && defined(HAVE_USABLE_WCHAR_T)

/* --- MBCS codecs for Windows -------------------------------------------- */

#if SIZEOF_INT < SIZEOF_SSIZE_T
#define NEED_RETRY
#endif

/* XXX This code is limited to "true" double-byte encodings, as
   a) it assumes an incomplete character consists of a single byte, and
   b) IsDBCSLeadByte (probably) does not work for non-DBCS multi-byte
      encodings, see IsDBCSLeadByteEx documentation. */

static int is_dbcs_lead_byte(const char *s, int offset)
{
    const char *curr = s + offset;

    if (IsDBCSLeadByte(*curr)) {
	const char *prev = CharPrev(s, curr);
	return (prev == curr) || !IsDBCSLeadByte(*prev) || (curr - prev == 2);
    }
    return 0;
}

/*
 * Decode MBCS string into unicode object. If 'final' is set, converts
 * trailing lead-byte too. Returns consumed size if succeed, -1 otherwise.
 */
static int decode_mbcs(PyUnicodeObject **v,
			const char *s, /* MBCS string */
			int size, /* sizeof MBCS string */
			int final)
{
    Py_UNICODE *p;
    Py_ssize_t n = 0;
    int usize = 0;

    assert(size >= 0);

    /* Skip trailing lead-byte unless 'final' is set */
    if (!final && size >= 1 && is_dbcs_lead_byte(s, size - 1))
	--size;

    /* First get the size of the result */
    if (size > 0) {
	usize = MultiByteToWideChar(CP_ACP, 0, s, size, NULL, 0);
	if (usize == 0) {
	    PyErr_SetFromWindowsErrWithFilename(0, NULL);
	    return -1;
	}
    }

    if (*v == NULL) {
	/* Create unicode object */
	*v = _PyUnicode_New(usize);
	if (*v == NULL)
	    return -1;
    }
    else {
	/* Extend unicode object */
	n = PyUnicode_GET_SIZE(*v);
	if (_PyUnicode_Resize(v, n + usize) < 0)
	    return -1;
    }

    /* Do the conversion */
    if (size > 0) {
	p = PyUnicode_AS_UNICODE(*v) + n;
	if (0 == MultiByteToWideChar(CP_ACP, 0, s, size, p, usize)) {
	    PyErr_SetFromWindowsErrWithFilename(0, NULL);
	    return -1;
	}
    }

    return size;
}

PyObject *PyUnicode_DecodeMBCSStateful(const char *s,
					Py_ssize_t size,
					const char *errors,
					Py_ssize_t *consumed)
{
    PyUnicodeObject *v = NULL;
    int done;

    if (consumed)
	*consumed = 0;

#ifdef NEED_RETRY
  retry:
    if (size > INT_MAX)
	done = decode_mbcs(&v, s, INT_MAX, 0);
    else
#endif
	done = decode_mbcs(&v, s, (int)size, !consumed);

    if (done < 0) {
        Py_XDECREF(v);
	return NULL;
    }

    if (consumed)
	*consumed += done;

#ifdef NEED_RETRY
    if (size > INT_MAX) {
	s += done;
	size -= done;
	goto retry;
    }
#endif

    return (PyObject *)v;
}

PyObject *PyUnicode_DecodeMBCS(const char *s,
				Py_ssize_t size,
				const char *errors)
{
    return PyUnicode_DecodeMBCSStateful(s, size, errors, NULL);
}

/*
 * Convert unicode into string object (MBCS).
 * Returns 0 if succeed, -1 otherwise.
 */
static int encode_mbcs(PyObject **repr,
			const Py_UNICODE *p, /* unicode */
			int size) /* size of unicode */
{
    int mbcssize = 0;
    Py_ssize_t n = 0;

    assert(size >= 0);

    /* First get the size of the result */
    if (size > 0) {
	mbcssize = WideCharToMultiByte(CP_ACP, 0, p, size, NULL, 0, NULL, NULL);
	if (mbcssize == 0) {
	    PyErr_SetFromWindowsErrWithFilename(0, NULL);
	    return -1;
	}
    }

    if (*repr == NULL) {
	/* Create string object */
	*repr = PyString_FromStringAndSize(NULL, mbcssize);
	if (*repr == NULL)
	    return -1;
    }
    else {
	/* Extend string object */
	n = PyString_Size(*repr);
	if (_PyString_Resize(repr, n + mbcssize) < 0)
	    return -1;
    }

    /* Do the conversion */
    if (size > 0) {
	char *s = PyString_AS_STRING(*repr) + n;
	if (0 == WideCharToMultiByte(CP_ACP, 0, p, size, s, mbcssize, NULL, NULL)) {
	    PyErr_SetFromWindowsErrWithFilename(0, NULL);
	    return -1;
	}
    }

    return 0;
}

PyObject *PyUnicode_EncodeMBCS(const Py_UNICODE *p,
				Py_ssize_t size,
				const char *errors)
{
    PyObject *repr = NULL;
    int ret;

#ifdef NEED_RETRY
 retry:
    if (size > INT_MAX)
	ret = encode_mbcs(&repr, p, INT_MAX);
    else
#endif
	ret = encode_mbcs(&repr, p, (int)size);

    if (ret < 0) {
	Py_XDECREF(repr);
	return NULL;
    }

#ifdef NEED_RETRY
    if (size > INT_MAX) {
	p += INT_MAX;
	size -= INT_MAX;
	goto retry;
    }
#endif

    return repr;
}

PyObject *PyUnicode_AsMBCSString(PyObject *unicode)
{
    if (!PyUnicode_Check(unicode)) {
        PyErr_BadArgument();
        return NULL;
    }
    return PyUnicode_EncodeMBCS(PyUnicode_AS_UNICODE(unicode),
				PyUnicode_GET_SIZE(unicode),
				NULL);
}

#undef NEED_RETRY

#endif /* MS_WINDOWS */

/* --- Character Mapping Codec -------------------------------------------- */

PyObject *PyUnicode_DecodeCharmap(const char *s,
				  Py_ssize_t size,
				  PyObject *mapping,
				  const char *errors)
{
    const char *starts = s;
    Py_ssize_t startinpos;
    Py_ssize_t endinpos;
    Py_ssize_t outpos;
    const char *e;
    PyUnicodeObject *v;
    Py_UNICODE *p;
    Py_ssize_t extrachars = 0;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    Py_UNICODE *mapstring = NULL;
    Py_ssize_t maplen = 0;

    /* Default to Latin-1 */
    if (mapping == NULL)
	return PyUnicode_DecodeLatin1(s, size, errors);

    v = _PyUnicode_New(size);
    if (v == NULL)
	goto onError;
    if (size == 0)
	return (PyObject *)v;
    p = PyUnicode_AS_UNICODE(v);
    e = s + size;
    if (PyUnicode_CheckExact(mapping)) {
	mapstring = PyUnicode_AS_UNICODE(mapping);
	maplen = PyUnicode_GET_SIZE(mapping);
	while (s < e) {
	    unsigned char ch = *s;
	    Py_UNICODE x = 0xfffe; /* illegal value */

	    if (ch < maplen)
		x = mapstring[ch];

	    if (x == 0xfffe) {
		/* undefined mapping */
		outpos = p-PyUnicode_AS_UNICODE(v);
		startinpos = s-starts;
		endinpos = startinpos+1;
		if (unicode_decode_call_errorhandler(
		     errors, &errorHandler,
		     "charmap", "character maps to <undefined>",
		     starts, size, &startinpos, &endinpos, &exc, &s,
		     (PyObject **)&v, &outpos, &p)) {
		    goto onError;
		}
		continue;
	    }
	    *p++ = x;
	    ++s;
	}
    }
    else {
	while (s < e) {
	    unsigned char ch = *s;
	    PyObject *w, *x;

	    /* Get mapping (char ordinal -> integer, Unicode char or None) */
	    w = PyInt_FromLong((long)ch);
	    if (w == NULL)
		goto onError;
	    x = PyObject_GetItem(mapping, w);
	    Py_DECREF(w);
	    if (x == NULL) {
		if (PyErr_ExceptionMatches(PyExc_LookupError)) {
		    /* No mapping found means: mapping is undefined. */
		    PyErr_Clear();
		    x = Py_None;
		    Py_INCREF(x);
		} else
		    goto onError;
	    }
    
	    /* Apply mapping */
	    if (PyInt_Check(x)) {
		long value = PyInt_AS_LONG(x);
		if (value < 0 || value > 65535) {
		    PyErr_SetString(PyExc_TypeError,
				    "character mapping must be in range(65536)");
		    Py_DECREF(x);
		    goto onError;
		}
		*p++ = (Py_UNICODE)value;
	    }
	    else if (x == Py_None) {
		/* undefined mapping */
		outpos = p-PyUnicode_AS_UNICODE(v);
		startinpos = s-starts;
		endinpos = startinpos+1;
		if (unicode_decode_call_errorhandler(
		     errors, &errorHandler,
		     "charmap", "character maps to <undefined>",
		     starts, size, &startinpos, &endinpos, &exc, &s,
		     (PyObject **)&v, &outpos, &p)) {
		    Py_DECREF(x);
		    goto onError;
		}
		Py_DECREF(x);
		continue;
	    }
	    else if (PyUnicode_Check(x)) {
		Py_ssize_t targetsize = PyUnicode_GET_SIZE(x);
    
		if (targetsize == 1)
		    /* 1-1 mapping */
		    *p++ = *PyUnicode_AS_UNICODE(x);
    
		else if (targetsize > 1) {
		    /* 1-n mapping */
		    if (targetsize > extrachars) {
			/* resize first */
			Py_ssize_t oldpos = p - PyUnicode_AS_UNICODE(v);
			Py_ssize_t needed = (targetsize - extrachars) + \
				     (targetsize << 2);
			extrachars += needed;
			/* XXX overflow detection missing */
			if (_PyUnicode_Resize(&v,
					     PyUnicode_GET_SIZE(v) + needed) < 0) {
			    Py_DECREF(x);
			    goto onError;
			}
			p = PyUnicode_AS_UNICODE(v) + oldpos;
		    }
		    Py_UNICODE_COPY(p,
				    PyUnicode_AS_UNICODE(x),
				    targetsize);
		    p += targetsize;
		    extrachars -= targetsize;
		}
		/* 1-0 mapping: skip the character */
	    }
	    else {
		/* wrong return value */
		PyErr_SetString(PyExc_TypeError,
		      "character mapping must return integer, None or unicode");
		Py_DECREF(x);
		goto onError;
	    }
	    Py_DECREF(x);
	    ++s;
	}
    }
    if (p - PyUnicode_AS_UNICODE(v) < PyUnicode_GET_SIZE(v))
	if (_PyUnicode_Resize(&v, p - PyUnicode_AS_UNICODE(v)) < 0)
	    goto onError;
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    return (PyObject *)v;

 onError:
    Py_XDECREF(errorHandler);
    Py_XDECREF(exc);
    Py_XDECREF(v);
    return NULL;
}

/* Charmap encoding: the lookup table */

struct encoding_map{
  PyObject_HEAD
  unsigned char level1[32];
  int count2, count3;
  unsigned char level23[1];
};

static PyObject*
encoding_map_size(PyObject *obj, PyObject* args)
{
    struct encoding_map *map = (struct encoding_map*)obj;
    return PyInt_FromLong(sizeof(*map) - 1 + 16*map->count2 + 
                          128*map->count3);
}

static PyMethodDef encoding_map_methods[] = {
	{"size", encoding_map_size, METH_NOARGS, 
         PyDoc_STR("Return the size (in bytes) of this object") },
        { 0 }
};

static void
encoding_map_dealloc(PyObject* o)
{
	PyObject_FREE(o);
}

static PyTypeObject EncodingMapType = {
	PyVarObject_HEAD_INIT(NULL, 0)
        "EncodingMap",          /*tp_name*/
        sizeof(struct encoding_map),   /*tp_basicsize*/
        0,                      /*tp_itemsize*/
        /* methods */
        encoding_map_dealloc,   /*tp_dealloc*/
        0,                      /*tp_print*/
        0,                      /*tp_getattr*/
        0,                      /*tp_setattr*/
        0,                      /*tp_compare*/
        0,                      /*tp_repr*/
        0,                      /*tp_as_number*/
        0,                      /*tp_as_sequence*/
        0,                      /*tp_as_mapping*/
        0,                      /*tp_hash*/
        0,                      /*tp_call*/
        0,                      /*tp_str*/
        0,                      /*tp_getattro*/
        0,                      /*tp_setattro*/
        0,                      /*tp_as_buffer*/
        Py_TPFLAGS_DEFAULT,     /*tp_flags*/
        0,                      /*tp_doc*/
        0,                      /*tp_traverse*/
        0,                      /*tp_clear*/
        0,                      /*tp_richcompare*/
        0,                      /*tp_weaklistoffset*/
        0,                      /*tp_iter*/
        0,                      /*tp_iternext*/
        encoding_map_methods,   /*tp_methods*/
        0,                      /*tp_members*/
        0,                      /*tp_getset*/
        0,                      /*tp_base*/
        0,                      /*tp_dict*/
        0,                      /*tp_descr_get*/
        0,                      /*tp_descr_set*/
        0,                      /*tp_dictoffset*/
        0,                      /*tp_init*/
        0,                      /*tp_alloc*/
        0,                      /*tp_new*/
        0,                      /*tp_free*/
        0,                      /*tp_is_gc*/
};

PyObject*
PyUnicode_BuildEncodingMap(PyObject* string)
{
    Py_UNICODE *decode;
    PyObject *result;
    struct encoding_map *mresult;
    int i;
    int need_dict = 0;
    unsigned char level1[32];
    unsigned char level2[512];
    unsigned char *mlevel1, *mlevel2, *mlevel3;
    int count2 = 0, count3 = 0;

    if (!PyUnicode_Check(string) || PyUnicode_GetSize(string) != 256) {
        PyErr_BadArgument();
        return NULL;
    }
    decode = PyUnicode_AS_UNICODE(string);
    memset(level1, 0xFF, sizeof level1);
    memset(level2, 0xFF, sizeof level2);

    /* If there isn't a one-to-one mapping of NULL to \0,
       or if there are non-BMP characters, we need to use
       a mapping dictionary. */
    if (decode[0] != 0)
        need_dict = 1;
    for (i = 1; i < 256; i++) {
        int l1, l2;
        if (decode[i] == 0
            #ifdef Py_UNICODE_WIDE
            || decode[i] > 0xFFFF
            #endif
        ) {
            need_dict = 1;
            break;
        }
        if (decode[i] == 0xFFFE)
            /* unmapped character */
            continue;
        l1 = decode[i] >> 11;
        l2 = decode[i] >> 7;
        if (level1[l1] == 0xFF)
            level1[l1] = count2++;
        if (level2[l2] == 0xFF)
            level2[l2] = count3++; 
    }

    if (count2 >= 0xFF || count3 >= 0xFF)
        need_dict = 1;

    if (need_dict) {
        PyObject *result = PyDict_New();
        PyObject *key, *value;
        if (!result)
            return NULL;
        for (i = 0; i < 256; i++) {
            key = value = NULL;
            key = PyInt_FromLong(decode[i]);
            value = PyInt_FromLong(i);
            if (!key || !value)
                goto failed1;
            if (PyDict_SetItem(result, key, value) == -1)
                goto failed1;
            Py_DECREF(key);
            Py_DECREF(value);
        }
        return result;
      failed1:
        Py_XDECREF(key);
        Py_XDECREF(value);
        Py_DECREF(result);
        return NULL;
    }

    /* Create a three-level trie */
    result = PyObject_MALLOC(sizeof(struct encoding_map) +
                             16*count2 + 128*count3 - 1);
    if (!result)
        return PyErr_NoMemory();
    PyObject_Init(result, &EncodingMapType);
    mresult = (struct encoding_map*)result;
    mresult->count2 = count2;
    mresult->count3 = count3;
    mlevel1 = mresult->level1;
    mlevel2 = mresult->level23;
    mlevel3 = mresult->level23 + 16*count2;
    memcpy(mlevel1, level1, 32);
    memset(mlevel2, 0xFF, 16*count2);
    memset(mlevel3, 0, 128*count3);
    count3 = 0;
    for (i = 1; i < 256; i++) {
        int o1, o2, o3, i2, i3;
        if (decode[i] == 0xFFFE)
            /* unmapped character */
            continue;
        o1 = decode[i]>>11;
        o2 = (decode[i]>>7) & 0xF;
        i2 = 16*mlevel1[o1] + o2;
        if (mlevel2[i2] == 0xFF)
            mlevel2[i2] = count3++;
        o3 = decode[i] & 0x7F;
        i3 = 128*mlevel2[i2] + o3;
        mlevel3[i3] = i;
    }
    return result;
}

static int
encoding_map_lookup(Py_UNICODE c, PyObject *mapping)
{
    struct encoding_map *map = (struct encoding_map*)mapping;
    int l1 = c>>11;
    int l2 = (c>>7) & 0xF;
    int l3 = c & 0x7F;
    int i;

#ifdef Py_UNICODE_WIDE
    if (c > 0xFFFF) {
	return -1;
    }
#endif
    if (c == 0)
        return 0;
    /* level 1*/
    i = map->level1[l1];
    if (i == 0xFF) {
        return -1;
    }
    /* level 2*/
    i = map->level23[16*i+l2];
    if (i == 0xFF) {
        return -1;
    }
    /* level 3 */
    i = map->level23[16*map->count2 + 128*i + l3];
    if (i == 0) {
        return -1;
    }
    return i;
}

/* Lookup the character ch in the mapping. If the character
   can't be found, Py_None is returned (or NULL, if another
   error occurred). */
static PyObject *charmapencode_lookup(Py_UNICODE c, PyObject *mapping)
{
    PyObject *w = PyInt_FromLong((long)c);
    PyObject *x;

    if (w == NULL)
	 return NULL;
    x = PyObject_GetItem(mapping, w);
    Py_DECREF(w);
    if (x == NULL) {
	if (PyErr_ExceptionMatches(PyExc_LookupError)) {
	    /* No mapping found means: mapping is undefined. */
	    PyErr_Clear();
	    x = Py_None;
	    Py_INCREF(x);
	    return x;
	} else
	    return NULL;
    }
    else if (x == Py_None)
	return x;
    else if (PyInt_Check(x)) {
	long value = PyInt_AS_LONG(x);
	if (value < 0 || value > 255) {
	    PyErr_SetString(PyExc_TypeError,
			     "character mapping must be in range(256)");
	    Py_DECREF(x);
	    return NULL;
	}
	return x;
    }
    else if (PyString_Check(x))
	return x;
    else {
	/* wrong return value */
	PyErr_SetString(PyExc_TypeError,
	      "character mapping must return integer, None or str");
	Py_DECREF(x);
	return NULL;
    }
}

static int
charmapencode_resize(PyObject **outobj, Py_ssize_t *outpos, Py_ssize_t requiredsize)
{
	Py_ssize_t outsize = PyString_GET_SIZE(*outobj);
	/* exponentially overallocate to minimize reallocations */
	if (requiredsize < 2*outsize)
	    requiredsize = 2*outsize;
	if (_PyString_Resize(outobj, requiredsize)) {
	    return 0;
	}
	return 1;
}

typedef enum charmapencode_result { 
  enc_SUCCESS, enc_FAILED, enc_EXCEPTION 
}charmapencode_result;
/* lookup the character, put the result in the output string and adjust
   various state variables. Reallocate the output string if not enough
   space is available. Return a new reference to the object that
   was put in the output buffer, or Py_None, if the mapping was undefined
   (in which case no character was written) or NULL, if a
   reallocation error occurred. The caller must decref the result */
static
charmapencode_result charmapencode_output(Py_UNICODE c, PyObject *mapping,
    PyObject **outobj, Py_ssize_t *outpos)
{
    PyObject *rep;
    char *outstart;
    Py_ssize_t outsize = PyString_GET_SIZE(*outobj);

    if (Py_TYPE(mapping) == &EncodingMapType) {
        int res = encoding_map_lookup(c, mapping);
	Py_ssize_t requiredsize = *outpos+1;
        if (res == -1)
            return enc_FAILED;
	if (outsize<requiredsize) 
	    if (!charmapencode_resize(outobj, outpos, requiredsize))
		return enc_EXCEPTION;
        outstart = PyString_AS_STRING(*outobj);
	outstart[(*outpos)++] = (char)res;
	return enc_SUCCESS;
    }

    rep = charmapencode_lookup(c, mapping);
    if (rep==NULL)
	return enc_EXCEPTION;
    else if (rep==Py_None) {
	Py_DECREF(rep);
	return enc_FAILED;
    } else {
	if (PyInt_Check(rep)) {
	    Py_ssize_t requiredsize = *outpos+1;
	    if (outsize<requiredsize)
		if (!charmapencode_resize(outobj, outpos, requiredsize)) {
		    Py_DECREF(rep);
		    return enc_EXCEPTION;
		}
            outstart = PyString_AS_STRING(*outobj);
	    outstart[(*outpos)++] = (char)PyInt_AS_LONG(rep);
	}
	else {
	    const char *repchars = PyString_AS_STRING(rep);
	    Py_ssize_t repsize = PyString_GET_SIZE(rep);
	    Py_ssize_t requiredsize = *outpos+repsize;
	    if (outsize<requiredsize)
		if (!charmapencode_resize(outobj, outpos, requiredsize)) {
		    Py_DECREF(rep);
		    return enc_EXCEPTION;
		}
            outstart = PyString_AS_STRING(*outobj);
	    memcpy(outstart + *outpos, repchars, repsize);
	    *outpos += repsize;
	}
    }
    Py_DECREF(rep);
    return enc_SUCCESS;
}

/* handle an error in PyUnicode_EncodeCharmap
   Return 0 on success, -1 on error */
static
int charmap_encoding_error(
    const Py_UNICODE *p, Py_ssize_t size, Py_ssize_t *inpos, PyObject *mapping,
    PyObject **exceptionObject,
    int *known_errorHandler, PyObject **errorHandler, const char *errors,
    PyObject **res, Py_ssize_t *respos)
{
    PyObject *repunicode = NULL; /* initialize to prevent gcc warning */
    Py_ssize_t repsize;
    Py_ssize_t newpos;
    Py_UNICODE *uni2;
    /* startpos for collecting unencodable chars */
    Py_ssize_t collstartpos = *inpos;
    Py_ssize_t collendpos = *inpos+1;
    Py_ssize_t collpos;
    char *encoding = "charmap";
    char *reason = "character maps to <undefined>";
    charmapencode_result x;

    /* find all unencodable characters */
    while (collendpos < size) {
        PyObject *rep;
        if (Py_TYPE(mapping) == &EncodingMapType) {
	    int res = encoding_map_lookup(p[collendpos], mapping);
	    if (res != -1)
		break;
	    ++collendpos;
	    continue;
	}
            
	rep = charmapencode_lookup(p[collendpos], mapping);
	if (rep==NULL)
	    return -1;
	else if (rep!=Py_None) {
	    Py_DECREF(rep);
	    break;
	}
	Py_DECREF(rep);
	++collendpos;
    }
    /* cache callback name lookup
     * (if not done yet, i.e. it's the first error) */
    if (*known_errorHandler==-1) {
	if ((errors==NULL) || (!strcmp(errors, "strict")))
	    *known_errorHandler = 1;
	else if (!strcmp(errors, "replace"))
	    *known_errorHandler = 2;
	else if (!strcmp(errors, "ignore"))
	    *known_errorHandler = 3;
	else if (!strcmp(errors, "xmlcharrefreplace"))
	    *known_errorHandler = 4;
	else
	    *known_errorHandler = 0;
    }
    switch (*known_errorHandler) {
	case 1: /* strict */
	    raise_encode_exception(exceptionObject, encoding, p, size, collstartpos, collendpos, reason);
	    return -1;
	case 2: /* replace */
	    for (collpos = collstartpos; collpos<collendpos; ++collpos) {
		x = charmapencode_output('?', mapping, res, respos);
		if (x==enc_EXCEPTION) {
		    return -1;
		}
		else if (x==enc_FAILED) {
		    raise_encode_exception(exceptionObject, encoding, p, size, collstartpos, collendpos, reason);
		    return -1;
		}
	    }
	    /* fall through */
	case 3: /* ignore */
	    *inpos = collendpos;
	    break;
	case 4: /* xmlcharrefreplace */
	    /* generate replacement (temporarily (mis)uses p) */
	    for (collpos = collstartpos; collpos < collendpos; ++collpos) {
		char buffer[2+29+1+1];
		char *cp;
		sprintf(buffer, "&#%d;", (int)p[collpos]);
		for (cp = buffer; *cp; ++cp) {
		    x = charmapencode_output(*cp, mapping, res, respos);
		    if (x==enc_EXCEPTION)
			return -1;
		    else if (x==enc_FAILED) {
			raise_encode_exception(exceptionObject, encoding, p, size, collstartpos, collendpos, reason);
			return -1;
		    }
		}
	    }
	    *inpos = collendpos;
	    break;
	default:
	    repunicode = unicode_encode_call_errorhandler(errors, errorHandler,
		encoding, reason, p, size, exceptionObject,
		collstartpos, collendpos, &newpos);
	    if (repunicode == NULL)
		return -1;
	    /* generate replacement  */
	    repsize = PyUnicode_GET_SIZE(repunicode);
	    for (uni2 = PyUnicode_AS_UNICODE(repunicode); repsize-->0; ++uni2) {
		x = charmapencode_output(*uni2, mapping, res, respos);
		if (x==enc_EXCEPTION) {
		    return -1;
		}
		else if (x==enc_FAILED) {
		    Py_DECREF(repunicode);
		    raise_encode_exception(exceptionObject, encoding, p, size, collstartpos, collendpos, reason);
		    return -1;
		}
	    }
	    *inpos = newpos;
	    Py_DECREF(repunicode);
    }
    return 0;
}

PyObject *PyUnicode_EncodeCharmap(const Py_UNICODE *p,
				  Py_ssize_t size,
				  PyObject *mapping,
				  const char *errors)
{
    /* output object */
    PyObject *res = NULL;
    /* current input position */
    Py_ssize_t inpos = 0;
    /* current output position */
    Py_ssize_t respos = 0;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    /* the following variable is used for caching string comparisons
     * -1=not initialized, 0=unknown, 1=strict, 2=replace,
     * 3=ignore, 4=xmlcharrefreplace */
    int known_errorHandler = -1;

    /* Default to Latin-1 */
    if (mapping == NULL)
	return PyUnicode_EncodeLatin1(p, size, errors);

    /* allocate enough for a simple encoding without
       replacements, if we need more, we'll resize */
    res = PyString_FromStringAndSize(NULL, size);
    if (res == NULL)
        goto onError;
    if (size == 0)
	return res;

    while (inpos<size) {
	/* try to encode it */
	charmapencode_result x = charmapencode_output(p[inpos], mapping, &res, &respos);
	if (x==enc_EXCEPTION) /* error */
	    goto onError;
	if (x==enc_FAILED) { /* unencodable character */
	    if (charmap_encoding_error(p, size, &inpos, mapping,
		&exc,
		&known_errorHandler, &errorHandler, errors,
		&res, &respos)) {
		goto onError;
	    }
	}
	else
	    /* done with this character => adjust input position */
	    ++inpos;
    }

    /* Resize if we allocated to much */
    if (respos<PyString_GET_SIZE(res)) {
	if (_PyString_Resize(&res, respos))
	    goto onError;
    }
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return res;

    onError:
    Py_XDECREF(res);
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return NULL;
}

PyObject *PyUnicode_AsCharmapString(PyObject *unicode,
				    PyObject *mapping)
{
    if (!PyUnicode_Check(unicode) || mapping == NULL) {
	PyErr_BadArgument();
	return NULL;
    }
    return PyUnicode_EncodeCharmap(PyUnicode_AS_UNICODE(unicode),
				   PyUnicode_GET_SIZE(unicode),
				   mapping,
				   NULL);
}

/* create or adjust a UnicodeTranslateError */
static void make_translate_exception(PyObject **exceptionObject,
    const Py_UNICODE *unicode, Py_ssize_t size,
    Py_ssize_t startpos, Py_ssize_t endpos,
    const char *reason)
{
    if (*exceptionObject == NULL) {
    	*exceptionObject = PyUnicodeTranslateError_Create(
	    unicode, size, startpos, endpos, reason);
    }
    else {
	if (PyUnicodeTranslateError_SetStart(*exceptionObject, startpos))
	    goto onError;
	if (PyUnicodeTranslateError_SetEnd(*exceptionObject, endpos))
	    goto onError;
	if (PyUnicodeTranslateError_SetReason(*exceptionObject, reason))
	    goto onError;
	return;
	onError:
	Py_DECREF(*exceptionObject);
	*exceptionObject = NULL;
    }
}

/* raises a UnicodeTranslateError */
static void raise_translate_exception(PyObject **exceptionObject,
    const Py_UNICODE *unicode, Py_ssize_t size,
    Py_ssize_t startpos, Py_ssize_t endpos,
    const char *reason)
{
    make_translate_exception(exceptionObject,
	unicode, size, startpos, endpos, reason);
    if (*exceptionObject != NULL)
	PyCodec_StrictErrors(*exceptionObject);
}

/* error handling callback helper:
   build arguments, call the callback and check the arguments,
   put the result into newpos and return the replacement string, which
   has to be freed by the caller */
static PyObject *unicode_translate_call_errorhandler(const char *errors,
    PyObject **errorHandler,
    const char *reason,
    const Py_UNICODE *unicode, Py_ssize_t size, PyObject **exceptionObject,
    Py_ssize_t startpos, Py_ssize_t endpos,
    Py_ssize_t *newpos)
{
    static char *argparse = "O!n;translating error handler must return (unicode, int) tuple";

    Py_ssize_t i_newpos;
    PyObject *restuple;
    PyObject *resunicode;

    if (*errorHandler == NULL) {
	*errorHandler = PyCodec_LookupError(errors);
        if (*errorHandler == NULL)
	    return NULL;
    }

    make_translate_exception(exceptionObject,
	unicode, size, startpos, endpos, reason);
    if (*exceptionObject == NULL)
	return NULL;

    restuple = PyObject_CallFunctionObjArgs(
	*errorHandler, *exceptionObject, NULL);
    if (restuple == NULL)
	return NULL;
    if (!PyTuple_Check(restuple)) {
	PyErr_Format(PyExc_TypeError, &argparse[4]);
	Py_DECREF(restuple);
	return NULL;
    }
    if (!PyArg_ParseTuple(restuple, argparse, &PyUnicode_Type,
	&resunicode, &i_newpos)) {
	Py_DECREF(restuple);
	return NULL;
    }
    if (i_newpos<0)
	*newpos = size+i_newpos;
    else
        *newpos = i_newpos;
    if (*newpos<0 || *newpos>size) {
	PyErr_Format(PyExc_IndexError, "position %zd from error handler out of bounds", *newpos);
	Py_DECREF(restuple);
	return NULL;
    }
    Py_INCREF(resunicode);
    Py_DECREF(restuple);
    return resunicode;
}

/* Lookup the character ch in the mapping and put the result in result,
   which must be decrefed by the caller.
   Return 0 on success, -1 on error */
static
int charmaptranslate_lookup(Py_UNICODE c, PyObject *mapping, PyObject **result)
{
    PyObject *w = PyInt_FromLong((long)c);
    PyObject *x;

    if (w == NULL)
	 return -1;
    x = PyObject_GetItem(mapping, w);
    Py_DECREF(w);
    if (x == NULL) {
	if (PyErr_ExceptionMatches(PyExc_LookupError)) {
	    /* No mapping found means: use 1:1 mapping. */
	    PyErr_Clear();
	    *result = NULL;
	    return 0;
	} else
	    return -1;
    }
    else if (x == Py_None) {
	*result = x;
	return 0;
    }
    else if (PyInt_Check(x)) {
	long value = PyInt_AS_LONG(x);
	long max = PyUnicode_GetMax();
	if (value < 0 || value > max) {
	    PyErr_Format(PyExc_TypeError,
			     "character mapping must be in range(0x%lx)", max+1);
	    Py_DECREF(x);
	    return -1;
	}
	*result = x;
	return 0;
    }
    else if (PyUnicode_Check(x)) {
	*result = x;
	return 0;
    }
    else {
	/* wrong return value */
	PyErr_SetString(PyExc_TypeError,
	      "character mapping must return integer, None or unicode");
	Py_DECREF(x);
	return -1;
    }
}
/* ensure that *outobj is at least requiredsize characters long,
if not reallocate and adjust various state variables.
Return 0 on success, -1 on error */
static
int charmaptranslate_makespace(PyObject **outobj, Py_UNICODE **outp,
    Py_ssize_t requiredsize)
{
    Py_ssize_t oldsize = PyUnicode_GET_SIZE(*outobj);
    if (requiredsize > oldsize) {
	/* remember old output position */
	Py_ssize_t outpos = *outp-PyUnicode_AS_UNICODE(*outobj);
	/* exponentially overallocate to minimize reallocations */
	if (requiredsize < 2 * oldsize)
	    requiredsize = 2 * oldsize;
	if (_PyUnicode_Resize(outobj, requiredsize) < 0)
	    return -1;
	*outp = PyUnicode_AS_UNICODE(*outobj) + outpos;
    }
    return 0;
}
/* lookup the character, put the result in the output string and adjust
   various state variables. Return a new reference to the object that
   was put in the output buffer in *result, or Py_None, if the mapping was
   undefined (in which case no character was written).
   The called must decref result.
   Return 0 on success, -1 on error. */
static
int charmaptranslate_output(const Py_UNICODE *startinp, const Py_UNICODE *curinp,
    Py_ssize_t insize, PyObject *mapping, PyObject **outobj, Py_UNICODE **outp,
    PyObject **res)
{
    if (charmaptranslate_lookup(*curinp, mapping, res))
	return -1;
    if (*res==NULL) {
	/* not found => default to 1:1 mapping */
	*(*outp)++ = *curinp;
    }
    else if (*res==Py_None)
	;
    else if (PyInt_Check(*res)) {
	/* no overflow check, because we know that the space is enough */
	*(*outp)++ = (Py_UNICODE)PyInt_AS_LONG(*res);
    }
    else if (PyUnicode_Check(*res)) {
	Py_ssize_t repsize = PyUnicode_GET_SIZE(*res);
	if (repsize==1) {
	    /* no overflow check, because we know that the space is enough */
	    *(*outp)++ = *PyUnicode_AS_UNICODE(*res);
	}
	else if (repsize!=0) {
	    /* more than one character */
	    Py_ssize_t requiredsize = (*outp-PyUnicode_AS_UNICODE(*outobj)) +
		(insize - (curinp-startinp)) +
		repsize - 1;
	    if (charmaptranslate_makespace(outobj, outp, requiredsize))
		return -1;
	    memcpy(*outp, PyUnicode_AS_UNICODE(*res), sizeof(Py_UNICODE)*repsize);
	    *outp += repsize;
	}
    }
    else
	return -1;
    return 0;
}

PyObject *PyUnicode_TranslateCharmap(const Py_UNICODE *p,
				     Py_ssize_t size,
				     PyObject *mapping,
				     const char *errors)
{
    /* output object */
    PyObject *res = NULL;
    /* pointers to the beginning and end+1 of input */
    const Py_UNICODE *startp = p;
    const Py_UNICODE *endp = p + size;
    /* pointer into the output */
    Py_UNICODE *str;
    /* current output position */
    Py_ssize_t respos = 0;
    char *reason = "character maps to <undefined>";
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    /* the following variable is used for caching string comparisons
     * -1=not initialized, 0=unknown, 1=strict, 2=replace,
     * 3=ignore, 4=xmlcharrefreplace */
    int known_errorHandler = -1;

    if (mapping == NULL) {
	PyErr_BadArgument();
	return NULL;
    }

    /* allocate enough for a simple 1:1 translation without
       replacements, if we need more, we'll resize */
    res = PyUnicode_FromUnicode(NULL, size);
    if (res == NULL)
	goto onError;
    if (size == 0)
	return res;
    str = PyUnicode_AS_UNICODE(res);

    while (p<endp) {
	/* try to encode it */
	PyObject *x = NULL;
	if (charmaptranslate_output(startp, p, size, mapping, &res, &str, &x)) {
	    Py_XDECREF(x);
	    goto onError;
	}
	Py_XDECREF(x);
	if (x!=Py_None) /* it worked => adjust input pointer */
	    ++p;
	else { /* untranslatable character */
	    PyObject *repunicode = NULL; /* initialize to prevent gcc warning */
	    Py_ssize_t repsize;
	    Py_ssize_t newpos;
	    Py_UNICODE *uni2;
	    /* startpos for collecting untranslatable chars */
	    const Py_UNICODE *collstart = p;
	    const Py_UNICODE *collend = p+1;
	    const Py_UNICODE *coll;

	    /* find all untranslatable characters */
	    while (collend < endp) {
		if (charmaptranslate_lookup(*collend, mapping, &x))
		    goto onError;
		Py_XDECREF(x);
		if (x!=Py_None)
		    break;
		++collend;
	    }
	    /* cache callback name lookup
	     * (if not done yet, i.e. it's the first error) */
	    if (known_errorHandler==-1) {
		if ((errors==NULL) || (!strcmp(errors, "strict")))
		    known_errorHandler = 1;
		else if (!strcmp(errors, "replace"))
		    known_errorHandler = 2;
		else if (!strcmp(errors, "ignore"))
		    known_errorHandler = 3;
		else if (!strcmp(errors, "xmlcharrefreplace"))
		    known_errorHandler = 4;
		else
		    known_errorHandler = 0;
	    }
	    switch (known_errorHandler) {
		case 1: /* strict */
		    raise_translate_exception(&exc, startp, size, collstart-startp, collend-startp, reason);
		    goto onError;
		case 2: /* replace */
		    /* No need to check for space, this is a 1:1 replacement */
		    for (coll = collstart; coll<collend; ++coll)
			*str++ = '?';
		    /* fall through */
		case 3: /* ignore */
		    p = collend;
		    break;
		case 4: /* xmlcharrefreplace */
		    /* generate replacement (temporarily (mis)uses p) */
		    for (p = collstart; p < collend; ++p) {
			char buffer[2+29+1+1];
			char *cp;
			sprintf(buffer, "&#%d;", (int)*p);
			if (charmaptranslate_makespace(&res, &str,
			    (str-PyUnicode_AS_UNICODE(res))+strlen(buffer)+(endp-collend)))
			    goto onError;
			for (cp = buffer; *cp; ++cp)
			    *str++ = *cp;
		    }
		    p = collend;
		    break;
		default:
		    repunicode = unicode_translate_call_errorhandler(errors, &errorHandler,
			reason, startp, size, &exc,
			collstart-startp, collend-startp, &newpos);
		    if (repunicode == NULL)
			goto onError;
		    /* generate replacement  */
		    repsize = PyUnicode_GET_SIZE(repunicode);
		    if (charmaptranslate_makespace(&res, &str,
			(str-PyUnicode_AS_UNICODE(res))+repsize+(endp-collend))) {
			Py_DECREF(repunicode);
			goto onError;
		    }
		    for (uni2 = PyUnicode_AS_UNICODE(repunicode); repsize-->0; ++uni2)
			*str++ = *uni2;
		    p = startp + newpos;
		    Py_DECREF(repunicode);
	    }
	}
    }
    /* Resize if we allocated to much */
    respos = str-PyUnicode_AS_UNICODE(res);
    if (respos<PyUnicode_GET_SIZE(res)) {
	if (_PyUnicode_Resize(&res, respos) < 0)
	    goto onError;
    }
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return res;

    onError:
    Py_XDECREF(res);
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return NULL;
}

PyObject *PyUnicode_Translate(PyObject *str,
			      PyObject *mapping,
			      const char *errors)
{
    PyObject *result;

    str = PyUnicode_FromObject(str);
    if (str == NULL)
	goto onError;
    result = PyUnicode_TranslateCharmap(PyUnicode_AS_UNICODE(str),
					PyUnicode_GET_SIZE(str),
					mapping,
					errors);
    Py_DECREF(str);
    return result;

 onError:
    Py_XDECREF(str);
    return NULL;
}

/* --- Decimal Encoder ---------------------------------------------------- */

int PyUnicode_EncodeDecimal(Py_UNICODE *s,
			    Py_ssize_t length,
			    char *output,
			    const char *errors)
{
    Py_UNICODE *p, *end;
    PyObject *errorHandler = NULL;
    PyObject *exc = NULL;
    const char *encoding = "decimal";
    const char *reason = "invalid decimal Unicode string";
    /* the following variable is used for caching string comparisons
     * -1=not initialized, 0=unknown, 1=strict, 2=replace, 3=ignore, 4=xmlcharrefreplace */
    int known_errorHandler = -1;

    if (output == NULL) {
	PyErr_BadArgument();
	return -1;
    }

    p = s;
    end = s + length;
    while (p < end) {
	register Py_UNICODE ch = *p;
	int decimal;
	PyObject *repunicode;
	Py_ssize_t repsize;
	Py_ssize_t newpos;
	Py_UNICODE *uni2;
	Py_UNICODE *collstart;
	Py_UNICODE *collend;

	if (Py_UNICODE_ISSPACE(ch)) {
	    *output++ = ' ';
	    ++p;
	    continue;
	}
	decimal = Py_UNICODE_TODECIMAL(ch);
	if (decimal >= 0) {
	    *output++ = '0' + decimal;
	    ++p;
	    continue;
	}
	if (0 < ch && ch < 256) {
	    *output++ = (char)ch;
	    ++p;
	    continue;
	}
	/* All other characters are considered unencodable */
	collstart = p;
	collend = p+1;
	while (collend < end) {
	    if ((0 < *collend && *collend < 256) ||
	        !Py_UNICODE_ISSPACE(*collend) ||
	        Py_UNICODE_TODECIMAL(*collend))
		break;
	}
	/* cache callback name lookup
	 * (if not done yet, i.e. it's the first error) */
	if (known_errorHandler==-1) {
	    if ((errors==NULL) || (!strcmp(errors, "strict")))
		known_errorHandler = 1;
	    else if (!strcmp(errors, "replace"))
		known_errorHandler = 2;
	    else if (!strcmp(errors, "ignore"))
		known_errorHandler = 3;
	    else if (!strcmp(errors, "xmlcharrefreplace"))
		known_errorHandler = 4;
	    else
		known_errorHandler = 0;
	}
	switch (known_errorHandler) {
	    case 1: /* strict */
		raise_encode_exception(&exc, encoding, s, length, collstart-s, collend-s, reason);
		goto onError;
	    case 2: /* replace */
		for (p = collstart; p < collend; ++p)
		    *output++ = '?';
		/* fall through */
	    case 3: /* ignore */
		p = collend;
		break;
	    case 4: /* xmlcharrefreplace */
		/* generate replacement (temporarily (mis)uses p) */
		for (p = collstart; p < collend; ++p)
		    output += sprintf(output, "&#%d;", (int)*p);
		p = collend;
		break;
	    default:
		repunicode = unicode_encode_call_errorhandler(errors, &errorHandler,
		    encoding, reason, s, length, &exc,
		    collstart-s, collend-s, &newpos);
		if (repunicode == NULL)
		    goto onError;
		/* generate replacement  */
		repsize = PyUnicode_GET_SIZE(repunicode);
		for (uni2 = PyUnicode_AS_UNICODE(repunicode); repsize-->0; ++uni2) {
		    Py_UNICODE ch = *uni2;
		    if (Py_UNICODE_ISSPACE(ch))
			*output++ = ' ';
		    else {
			decimal = Py_UNICODE_TODECIMAL(ch);
			if (decimal >= 0)
			    *output++ = '0' + decimal;
			else if (0 < ch && ch < 256)
			    *output++ = (char)ch;
			else {
			    Py_DECREF(repunicode);
			    raise_encode_exception(&exc, encoding,
				s, length, collstart-s, collend-s, reason);
			    goto onError;
			}
		    }
		}
		p = s + newpos;
		Py_DECREF(repunicode);
	}
    }
    /* 0-terminate the output string */
    *output++ = '\0';
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return 0;

 onError:
    Py_XDECREF(exc);
    Py_XDECREF(errorHandler);
    return -1;
}

/* --- Helpers ------------------------------------------------------------ */

#include "stringlib/unicodedefs.h"

#define FROM_UNICODE

#include "stringlib/fastsearch.h"

#include "stringlib/count.h"
#include "stringlib/find.h"
#include "stringlib/partition.h"

/* helper macro to fixup start/end slice values */
#define FIX_START_END(obj)                      \
    if (start < 0)                              \
        start += (obj)->length;                 \
    if (start < 0)                              \
        start = 0;                              \
    if (end > (obj)->length)                    \
        end = (obj)->length;                    \
    if (end < 0)                                \
        end += (obj)->length;                   \
    if (end < 0)                                \
        end = 0;

Py_ssize_t PyUnicode_Count(PyObject *str,
                           PyObject *substr,
                           Py_ssize_t start,
                           Py_ssize_t end)
{
    Py_ssize_t result;
    PyUnicodeObject* str_obj;
    PyUnicodeObject* sub_obj;

    str_obj = (PyUnicodeObject*) PyUnicode_FromObject(str);
    if (!str_obj)
	return -1;