summaryrefslogtreecommitdiff
path: root/source3/python/py_tdbpack.c
blob: 8aa4ee316a8ea5aeb5407f065481e067faca00e9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
/* -*- c-file-style: "python"; indent-tabs-mode: nil; -*-
	 
   Python wrapper for Samba tdb pack/unpack functions
   Copyright (C) Martin Pool 2002


   NOTE PYTHON STYLE GUIDE
   http://www.python.org/peps/pep-0007.html
   
   
   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.
   
   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.
   
   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/

#include "Python.h"

/* This module is supposed to be standalone, however for portability
   it would be good to use the FUNCTION_MACRO preprocessor define. */

#include "include/config.h"

#ifdef HAVE_FUNCTION_MACRO
#define FUNCTION_MACRO  (__FUNCTION__)
#else
#define FUNCTION_MACRO  (__FILE__)
#endif

static PyObject * pytdbpack_number(char ch, PyObject *val_iter, PyObject *packed_list);
static PyObject * pytdbpack_str(char ch,
				PyObject *val_iter, PyObject *packed_list,
				const char *encoding);
static PyObject * pytdbpack_buffer(PyObject *val_iter, PyObject *packed_list);

static PyObject *pytdbunpack_item(char, char **pbuf, int *plen, PyObject *);

static PyObject *pytdbpack_data(const char *format_str,
				     PyObject *val_seq,
				     PyObject *val_list);

static PyObject *
pytdbunpack_string(char **pbuf, int *plen, const char *encoding);

static void pack_le_uint32(unsigned long val_long, unsigned char *pbuf);


static PyObject *pytdbpack_bad_type(char ch,
				    const char *expected,
				    PyObject *val_obj);

static const char * pytdbpack_docstring =
"Convert between Python values and Samba binary encodings.

This module is conceptually similar to the standard 'struct' module, but it
uses both a different binary format and a different description string.

Samba's encoding is based on that used inside DCE-RPC and SMB: a
little-endian, unpadded, non-self-describing binary format.  It is intended
that these functions be as similar as possible to the routines in Samba's
tdb/tdbutil module, with appropriate adjustments for Python datatypes.

Python strings are used to specify the format of data to be packed or
unpacked.

String encodings are implied by the database format: they may be either DOS
codepage (currently hardcoded to 850), or Unix codepage (currently hardcoded
to be the same as the default Python encoding).

tdbpack format strings:

    'f': NUL-terminated string in codepage iso8859-1
   
    'P': same as 'f'

    'F': NUL-terminated string in iso-8859-1

    'd':  4 byte little-endian unsigned number

    'w':  2 byte little-endian unsigned number

    'P': \"Pointer\" value -- in the subset of DCERPC used by Samba, this is
          really just an \"exists\" or \"does not exist\" flag.  The boolean
          value of the Python object is used.
    
    'B': 4-byte LE length, followed by that many bytes of binary data.
         Corresponds to a Python integer giving the length, followed by a byte
         string of the appropriate length.

    '$': Special flag indicating that the preceding format code should be
         repeated while data remains.  This is only supported for unpacking.

    Every code corresponds to a single Python object, except 'B' which
    corresponds to two values (length and contents), and '$', which produces
    however many make sense.
";


static char const pytdbpack_doc[] = 
"pack(format, values) -> buffer
Pack Python objects into Samba binary format according to format string.

arguments:
    format -- string of tdbpack format characters
    values -- sequence of value objects corresponding 1:1 to format characters

returns:
    buffer -- string containing packed data

raises:
    IndexError -- if there are too few values for the format
    ValueError -- if any of the format characters is illegal
    TypeError  -- if the format is not a string, or values is not a sequence,
        or any of the values is of the wrong type for the corresponding
        format character

notes:
    For historical reasons, it is not an error to pass more values than are consumed
    by the format.
";


static char const pytdbunpack_doc[] =
"unpack(format, buffer) -> (values, rest)
Unpack Samba binary data according to format string.

arguments:
    format -- string of tdbpack characters
    buffer -- string of packed binary data

returns:
    2-tuple of:
        values -- sequence of values corresponding 1:1 to format characters
        rest -- string containing data that was not decoded, or '' if the
            whole string was consumed

raises:
    IndexError -- if there is insufficient data in the buffer for the
        format (or if the data is corrupt and contains a variable-length
        field extending past the end)
    ValueError -- if any of the format characters is illegal

notes:
    Because unconsumed data is returned, you can feed it back in to the
    unpacker to extract further fields.  Alternatively, if you wish to modify
    some fields near the start of the data, you may be able to save time by
    only unpacking and repacking the necessary part.
";


const char *pytdb_dos_encoding = "cp850";

/* NULL, meaning that the Samba default encoding *must* be the same as the
   Python default encoding. */
const char *pytdb_unix_encoding = NULL;


/*
  * Pack objects to bytes.
  *
  * All objects are first individually encoded onto a list, and then the list
  * of strings is concatenated.  This is faster than concatenating strings,
  * and reasonably simple to code.
  */
static PyObject *
pytdbpack(PyObject *self,
	       PyObject *args)
{
	char *format_str;
	PyObject *val_seq, *val_iter = NULL,
		*packed_list = NULL, *packed_str = NULL,
		*empty_str = NULL;

	/* TODO: Test passing wrong types or too many arguments */
	if (!PyArg_ParseTuple(args, "sO", &format_str, &val_seq))
		return NULL;

	if (!(val_iter = PyObject_GetIter(val_seq)))
		goto out;

	/* Create list to hold strings until we're done, then join them all. */
	if (!(packed_list = PyList_New(0)))
		goto out;

	if (!pytdbpack_data(format_str, val_iter, packed_list))
		goto out;

	/* this function is not officially documented but it works */
	if (!(empty_str = PyString_InternFromString("")))
		goto out;
	
	packed_str = _PyString_Join(empty_str, packed_list);

  out:
	Py_XDECREF(empty_str);
	Py_XDECREF(val_iter);
	Py_XDECREF(packed_list);

	return packed_str;
}


/*
  Pack data according to FORMAT_STR from the elements of VAL_SEQ into
  PACKED_BUF.

  The string has already been checked out, so we know that VAL_SEQ is large
  enough to hold the packed data, and that there are enough value items.
  (However, their types may not have been thoroughly checked yet.)

  In addition, val_seq is a Python Fast sequence.

  Returns NULL for error (with exception set), or None.
*/
PyObject *
pytdbpack_data(const char *format_str,
		    PyObject *val_iter,
		    PyObject *packed_list)
{
	int format_i, val_i = 0;

	for (format_i = 0, val_i = 0; format_str[format_i]; format_i++) {
		char ch = format_str[format_i];

		switch (ch) {
			/* dispatch to the appropriate packer for this type,
			   which should pull things off the iterator, and
			   append them to the packed_list */
		case 'w':
		case 'd':
		case 'p':
			if (!(packed_list = pytdbpack_number(ch, val_iter, packed_list)))
				return NULL;
			break;

		case 'f':
		case 'P':
			if (!(packed_list = pytdbpack_str(ch, val_iter, packed_list, pytdb_unix_encoding)))
				return NULL;
			break;

		case 'B':
			if (!(packed_list = pytdbpack_buffer(val_iter, packed_list)))
				return NULL;
			break;

		default:
			PyErr_Format(PyExc_ValueError,
				     "%s: format character '%c' is not supported",
				     FUNCTION_MACRO, ch);
			return NULL;
		}
	}

	return packed_list;
}


static PyObject *
pytdbpack_number(char ch, PyObject *val_iter, PyObject *packed_list)
{
	unsigned long val_long;
	PyObject *val_obj = NULL, *long_obj = NULL, *result_obj = NULL;
	PyObject *new_list = NULL;
	unsigned char pack_buf[4];

	if (!(val_obj = PyIter_Next(val_iter)))
		goto out;

	if (!(long_obj = PyNumber_Long(val_obj))) {
		pytdbpack_bad_type(ch, "Number", val_obj);
		goto out;
	}

	val_long = PyLong_AsUnsignedLong(long_obj);
	pack_le_uint32(val_long, pack_buf);

	/* pack as 32-bit; if just packing a 'w' 16-bit word then only take
	   the first two bytes. */
	
	if (!(result_obj = PyString_FromStringAndSize(pack_buf, ch == 'w' ? 2 : 4)))
		goto out;

	if (PyList_Append(packed_list, result_obj) != -1)
		new_list = packed_list;

  out:
	Py_XDECREF(val_obj);
	Py_XDECREF(long_obj);
	Py_XDECREF(result_obj);

	return new_list;
}


/*
 * Take one string from the iterator val_iter, convert it to 8-bit, and return
 * it.
 *
 * If the input is neither a string nor Unicode, an exception is raised.
 *
 * If the input is Unicode, then it is converted to the appropriate encoding.
 *
 * If the input is a String, and encoding is not null, then it is converted to
 * Unicode using the default decoding method, and then converted to the
 * encoding.  If the encoding is NULL, then the string is written out as-is --
 * this is used when the default Python encoding is the same as the Samba
 * encoding.
 *
 * I hope this approach avoids being too fragile w.r.t. being passed either
 * Unicode or String objects.
 */
static PyObject *
pytdbpack_str(char ch,
	      PyObject *val_iter, PyObject *packed_list, const char *encoding)
{
	PyObject *val_obj = NULL;
	PyObject *unicode_obj = NULL;
	PyObject *coded_str = NULL;
	PyObject *nul_str = NULL;
	PyObject *new_list = NULL;

	if (!(val_obj = PyIter_Next(val_iter)))
		goto out;

	if (PyUnicode_Check(val_obj)) {
		if (!(coded_str = PyUnicode_AsEncodedString(val_obj, encoding, NULL)))
			goto out;
	}
	else if (PyString_Check(val_obj) && !encoding) {
		/* For efficiency, we assume that the Python interpreter has
		   the same default string encoding as Samba's native string
		   encoding.  On the PSA, both are always 8859-1. */
		coded_str = val_obj;
		Py_INCREF(coded_str);
	}
	else if (PyString_Check(val_obj)) {
		/* String, but needs to be converted */
		if (!(unicode_obj = PyString_AsDecodedObject(val_obj, NULL, NULL)))
			goto out;
		if (!(coded_str = PyUnicode_AsEncodedString(unicode_obj, encoding, NULL)))
			goto out;
	}
	else {
		pytdbpack_bad_type(ch, "String or Unicode", val_obj);
		goto out;
	}

	if (!nul_str)
		/* this is constant and often-used; hold it forever */
		if (!(nul_str = PyString_FromStringAndSize("", 1)))
			goto out;

	if ((PyList_Append(packed_list, coded_str) != -1)
	    && (PyList_Append(packed_list, nul_str) != -1))
		new_list = packed_list;

  out:
	Py_XDECREF(val_obj);
	Py_XDECREF(unicode_obj);
	Py_XDECREF(coded_str);

	return new_list;
}


/*
 * Pack (LENGTH, BUFFER) pair onto the list.
 *
 * The buffer must already be a String, not Unicode, because it contains 8-bit
 * untranslated data.  In some cases it will actually be UTF_16_LE data.
 */
static PyObject *
pytdbpack_buffer(PyObject *val_iter, PyObject *packed_list)
{
	PyObject *val_obj;
	PyObject *new_list = NULL;
	
	/* pull off integer and stick onto list */
	if (!(packed_list = pytdbpack_number('d', val_iter, packed_list)))
		return NULL;

	/* this assumes that the string is the right length; the old code did
	   the same. */
	if (!(val_obj = PyIter_Next(val_iter)))
		return NULL;

	if (!PyString_Check(val_obj)) {
		pytdbpack_bad_type('B', "String", val_obj);
		goto out;
	}
	
	if (PyList_Append(packed_list, val_obj) != -1)
		new_list = packed_list;

  out:
	Py_XDECREF(val_obj);
	return new_list;
}


static PyObject *pytdbpack_bad_type(char ch,
				    const char *expected,
				    PyObject *val_obj)
{
	PyObject *r = PyObject_Repr(val_obj);
	if (!r)
		return NULL;
	PyErr_Format(PyExc_TypeError,
		     "tdbpack: format '%c' requires %s, not %s",
		     ch, expected, PyString_AS_STRING(r));
	Py_DECREF(r);
	return val_obj;
}


/*
  XXX: glib and Samba have quicker macro for doing the endianness conversions,
  but I don't know of one in plain libc, and it's probably not a big deal.  I
  realize this is kind of dumb because we'll almost always be on x86, but
  being safe is important.
*/
static void pack_le_uint32(unsigned long val_long, unsigned char *pbuf)
{
	pbuf[0] =         val_long & 0xff;
	pbuf[1] = (val_long >> 8)  & 0xff;
	pbuf[2] = (val_long >> 16) & 0xff;
	pbuf[3] = (val_long >> 24) & 0xff;
}


static void pack_bytes(long len, const char *from,
		       unsigned char **pbuf)
{
	memcpy(*pbuf, from, len);
	(*pbuf) += len;
}



static PyObject *
pytdbunpack(PyObject *self,
		 PyObject *args)
{
	char *format_str, *packed_str, *ppacked;
	PyObject *val_list = NULL, *ret_tuple = NULL;
	PyObject *rest_string = NULL;
	int format_len, packed_len;
	char last_format = '#';	/* invalid */
	int i;
	
	/* get arguments */
	if (!PyArg_ParseTuple(args, "ss#", &format_str, &packed_str, &packed_len))
		return NULL;

	format_len = strlen(format_str);
	
	/* Allocate list to hold results.  Initially empty, and we append
	   results as we go along. */
	val_list = PyList_New(0);
	if (!val_list)
		goto failed;
	ret_tuple = PyTuple_New(2);
	if (!ret_tuple)
		goto failed;
	
	/* For every object, unpack.  */
	for (ppacked = packed_str, i = 0; i < format_len && format_str[i] != '$'; i++) {
		last_format = format_str[i];
		/* packed_len is reduced in place */
		if (!pytdbunpack_item(format_str[i], &ppacked, &packed_len, val_list))
			goto failed;
	}

	/* If the last character was '$', keep going until out of space */
	if (format_str[i] == '$') {
		if (i == 0) {
			PyErr_Format(PyExc_ValueError,
				     "%s: '$' may not be first character in format",
				     FUNCTION_MACRO);
			return NULL;
		} 
		while (packed_len > 0)
			if (!pytdbunpack_item(last_format, &ppacked, &packed_len, val_list))
				goto failed;
	}
	
	/* save leftovers for next time */
	rest_string = PyString_FromStringAndSize(ppacked, packed_len);
	if (!rest_string)
		goto failed;

	/* return (values, rest) tuple; give up references to them */
	PyTuple_SET_ITEM(ret_tuple, 0, val_list);
	val_list = NULL;
	PyTuple_SET_ITEM(ret_tuple, 1, rest_string);
	val_list = NULL;
	return ret_tuple;

  failed:
	/* handle failure: deallocate anything.  XDECREF forms handle NULL
	   pointers for objects that haven't been allocated yet. */
	Py_XDECREF(val_list);
	Py_XDECREF(ret_tuple);
	Py_XDECREF(rest_string);
	return NULL;
}


static void
pytdbunpack_err_too_short(void)
{
	PyErr_Format(PyExc_IndexError,
		     "%s: data too short for unpack format", FUNCTION_MACRO);
}


static PyObject *
pytdbunpack_uint32(char **pbuf, int *plen)
{
	unsigned long v;
	unsigned char *b;
	
	if (*plen < 4) {
		pytdbunpack_err_too_short();
		return NULL;
	}

	b = *pbuf;
	v = b[0] | b[1]<<8 | b[2]<<16 | b[3]<<24;
	
	(*pbuf) += 4;
	(*plen) -= 4;

	return PyLong_FromUnsignedLong(v);
}


static PyObject *pytdbunpack_int16(char **pbuf, int *plen)
{
	long v;
	unsigned char *b;
	
	if (*plen < 2) {
		pytdbunpack_err_too_short();
		return NULL;
	}

	b = *pbuf;
	v = b[0] | b[1]<<8;
	
	(*pbuf) += 2;
	(*plen) -= 2;

	return PyInt_FromLong(v);
}


static PyObject *
pytdbunpack_string(char **pbuf, int *plen, const char *encoding)
{
	int len;
	char *nul_ptr, *start;

	start = *pbuf;
	
	nul_ptr = memchr(start, '\0', *plen);
	if (!nul_ptr) {
		pytdbunpack_err_too_short();
		return NULL;
	}

	len = nul_ptr - start;

	*pbuf += len + 1;	/* skip \0 */
	*plen -= len + 1;

	return PyString_Decode(start, len, encoding, NULL);
}


static PyObject *
pytdbunpack_buffer(char **pbuf, int *plen, PyObject *val_list)
{
	/* first get 32-bit len */
	long slen;
	unsigned char *b;
	unsigned char *start;
	PyObject *str_obj = NULL, *len_obj = NULL;
	
	if (*plen < 4) {
		pytdbunpack_err_too_short();
		return NULL;
	}
	
	b = *pbuf;
	slen = b[0] | b[1]<<8 | b[2]<<16 | b[3]<<24;

	if (slen < 0) { /* surely you jest */
		PyErr_Format(PyExc_ValueError,
			     "%s: buffer seems to have negative length", FUNCTION_MACRO);
		return NULL;
	}

	(*pbuf) += 4;
	(*plen) -= 4;
	start = *pbuf;

	if (*plen < slen) {
		PyErr_Format(PyExc_IndexError,
			     "%s: not enough data to unpack buffer: "
			     "need %d bytes, have %d", FUNCTION_MACRO,
			     (int) slen, *plen);
		return NULL;
	}

	(*pbuf) += slen;
	(*plen) -= slen;

	if (!(len_obj = PyInt_FromLong(slen)))
		goto failed;

	if (PyList_Append(val_list, len_obj) == -1)
		goto failed;
	
	if (!(str_obj = PyString_FromStringAndSize(start, slen)))
		goto failed;
	
	if (PyList_Append(val_list, str_obj) == -1)
		goto failed;
	
	Py_DECREF(len_obj);
	Py_DECREF(str_obj);
	
	return val_list;

  failed:
	Py_XDECREF(len_obj);	/* handles NULL */
	Py_XDECREF(str_obj);
	return NULL;
}


/* Unpack a single field from packed data, according to format character CH.
   Remaining data is at *PBUF, of *PLEN.

   *PBUF is advanced, and *PLEN reduced to reflect the amount of data that has
   been consumed.

   Returns a reference to None, or NULL for failure.
*/
static PyObject *pytdbunpack_item(char ch,
				  char **pbuf,
				  int *plen,
				  PyObject *val_list)
{
	PyObject *unpacked;
	
	if (ch == 'w') {	/* 16-bit int */
		unpacked = pytdbunpack_int16(pbuf, plen);
	}
	else if (ch == 'd' || ch == 'p') { /* 32-bit int */
		/* pointers can just come through as integers */
		unpacked = pytdbunpack_uint32(pbuf, plen);
	}
	else if (ch == 'f' || ch == 'P') { /* nul-term string  */
		unpacked = pytdbunpack_string(pbuf, plen, pytdb_unix_encoding);
	}
	else if (ch == 'B') { /* length, buffer */
		return pytdbunpack_buffer(pbuf, plen, val_list);
	}
	else {
		PyErr_Format(PyExc_ValueError,
			     "%s: format character '%c' is not supported", 
                             FUNCTION_MACRO, ch);
		
		return NULL;
	}

	/* otherwise OK */
	if (!unpacked)
		return NULL;

	if (PyList_Append(val_list, unpacked) == -1)
		val_list = NULL;

	/* PyList_Append takes a new reference to the inserted object.
	   Therefore, we no longer need the original reference. */
	Py_DECREF(unpacked);
	
	return val_list;
}






static PyMethodDef pytdbpack_methods[] = {
	{ "pack", pytdbpack, METH_VARARGS, (char *) pytdbpack_doc },
	{ "unpack", pytdbunpack, METH_VARARGS, (char *) pytdbunpack_doc },
};

DL_EXPORT(void)
inittdbpack(void)
{
	Py_InitModule3("tdbpack", pytdbpack_methods,
		       (char *) pytdbpack_docstring);
}