summaryrefslogtreecommitdiff
path: root/src/hashcomp.cpp
blob: 45a216e698149747156dc0828ba277f784f76fb3 (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
/*       +------------------------------------+
 *       | Inspire Internet Relay Chat Daemon |
 *       +------------------------------------+
 *
 *  InspIRCd is copyright (C) 2002-2006 ChatSpike-Dev.
 *		       E-mail:
 *		<brain@chatspike.net>
 *		<Craig@chatspike.net>
 *
 * Written by Craig Edwards, Craig McLure, and others.
 * This program is free but copyrighted software; see
 *	    the file COPYING for details.
 *
 * ---------------------------------------------------
 */

using namespace std;

#include "inspircd.h"
#include "hashcomp.h"
#include <ext/hash_map>
#define nspace __gnu_cxx

/******************************************************
 *
 * The hash functions of InspIRCd are the centrepoint
 * of the entire system. If these functions are
 * inefficient or wasteful, the whole program suffers
 * as a result. A lot of C programmers in the ircd
 * scene spend a lot of time debating (arguing) about
 * the best way to write hash functions to hash irc
 * nicknames, channels etc.
 * We are lucky as C++ developers as hash_map does
 * a lot of this for us. It does intellegent memory
 * requests, bucketing, search functions, insertion
 * and deletion etc. All we have to do is write some
 * overloaded comparison and hash value operators which
 * cause it to act in an irc-like way. The features we
 * add to the standard hash_map are:
 *
 * Case insensitivity: The hash_map will be case
 * insensitive.
 *
 * Scandanavian Comparisons: The characters [, ], \ will
 * be considered the lowercase of {, } and |.
 *
 * This file also contains hashing methods for hashing
 * insp_inaddr structs, we use this if we want to cache IP
 * addresses.
 *
 ******************************************************/

using namespace std;
using namespace irc::sockets;

/* convert a string to lowercase. Note following special circumstances
 * taken from RFC 1459. Many "official" server branches still hold to this
 * rule so i will too;
 *
 *  Because of IRC's scandanavian origin, the characters {}| are
 *  considered to be the lower case equivalents of the characters []\,
 *  respectively. This is a critical issue when determining the
 *  equivalence of two nicknames.
 */
void nspace::strlower(char *n)
{
	if (n)
	{
		for (char* t = n; *t; t++)
			*t = lowermap[(unsigned char)*t];
	}
}

size_t nspace::hash<insp_inaddr>::operator()(const insp_inaddr &a) const
{
	size_t q;
	memcpy(&q,&a,sizeof(size_t));
	return q;
}

size_t nspace::hash<string>::operator()(const string &s) const
{
	/* XXX: NO DATA COPIES! :)
	 * The hash function here is practically
	 * a copy of the one in STL's hash_fun.h,
	 * only with *x replaced with lowermap[*x].
	 * This avoids a copy to use hash<const char*>
	 */
	register size_t t = 0;
	for (std::string::const_iterator x = s.begin(); x != s.end(); ++x) /* ++x not x++, as its faster */
		t = 5 * t + lowermap[(unsigned char)*x];
	return t;
}

bool irc::StrHashComp::operator()(const std::string& s1, const std::string& s2) const
{
	unsigned char* n1 = (unsigned char*)s1.c_str();
	unsigned char* n2 = (unsigned char*)s2.c_str();
	for (; *n1 && *n2; n1++, n2++)
		if (lowermap[*n1] != lowermap[*n2])
			return false;
	return (lowermap[*n1] == lowermap[*n2]);
}

bool irc::InAddr_HashComp::operator()(const insp_inaddr &s1, const insp_inaddr &s2) const
{
#ifdef IPV6
	for (int n = 0; n < 16; n++)
		if (s2.s6_addr[n] != s1.s6_addr[n])
			return false;
	return true;
#else
	return (s1.s_addr == s1.s_addr);
#endif
}

/******************************************************
 *
 * This is the implementation of our special irc::string
 * class which is a case-insensitive equivalent to
 * std::string which is not only case-insensitive but
 * can also do scandanavian comparisons, e.g. { = [, etc.
 *
 * This class depends on the const array 'lowermap'.
 *
 ******************************************************/

bool irc::irc_char_traits::eq(char c1st, char c2nd)
{
	return lowermap[(unsigned char)c1st] == lowermap[(unsigned char)c2nd];
}

bool irc::irc_char_traits::ne(char c1st, char c2nd)
{
	return lowermap[(unsigned char)c1st] != lowermap[(unsigned char)c2nd];
}

bool irc::irc_char_traits::lt(char c1st, char c2nd)
{
	return lowermap[(unsigned char)c1st] < lowermap[(unsigned char)c2nd];
}

int irc::irc_char_traits::compare(const char* str1, const char* str2, size_t n)
{
	for(unsigned int i = 0; i < n; i++)
	{
		if(lowermap[(unsigned char)*str1] > lowermap[(unsigned char)*str2])
       			return 1;

		if(lowermap[(unsigned char)*str1] < lowermap[(unsigned char)*str2])
		       	return -1;

		if(*str1 == 0 || *str2 == 0)
		      	return 0;

	       	str1++;
		str2++;
	}
	return 0;
}

std::string operator+ (std::string& leftval, irc::string& rightval)
{
	return leftval + std::string(rightval.c_str());
}

irc::string operator+ (irc::string& leftval, std::string& rightval)
{
	return leftval + irc::string(rightval.c_str());
}

bool operator== (std::string& leftval, irc::string& rightval)
{
	return (leftval == std::string(rightval.c_str()));
}

bool operator== (irc::string& leftval, std::string& rightval)
{
	return (rightval == std::string(leftval.c_str()));
}

const char* irc::irc_char_traits::find(const char* s1, int  n, char c)
{
	while(n-- > 0 && lowermap[(unsigned char)*s1] != lowermap[(unsigned char)c])
		s1++;
	return s1;
}

/* See hashcomp.h if you care about these... */
std::ostream& operator<<(std::ostream &os, const irc::string &str)
{
	return os << str.c_str();
}

std::istream& operator>>(std::istream &is, irc::string &str)
{
	std::string tmp;
	is >> tmp;
	str = tmp.c_str();
	return is;
}

irc::tokenstream::tokenstream(const std::string &source) : tokens(source), last_pushed(false)
{
	/* Record starting position and current position */
	last_starting_position = tokens.begin();
	n = tokens.begin();
}

irc::tokenstream::~tokenstream()
{
}

const std::string irc::tokenstream::GetToken()
{
	std::string::iterator lsp = last_starting_position;

	while (n != tokens.end())
	{
		/** Skip multi space, converting "  " into " "
		 */
		while ((n+1 != tokens.end()) && (*n == ' ') && (*(n+1) == ' '))
			n++;

		if ((last_pushed) && (*n == ':'))
		{
			/* If we find a token thats not the first and starts with :,
			 * this is the last token on the line
			 */
			std::string::iterator curr = ++n;
			n = tokens.end();
			return std::string(curr, tokens.end());
		}

		last_pushed = false;

		if ((*n == ' ') || (n+1 == tokens.end()))
		{
			/* If we find a space, or end of string, this is the end of a token.
			 */
			last_starting_position = n+1;
			last_pushed = true;

			std::string strip(lsp, n+1 == tokens.end() ? n+1  : n++);
			while ((strip.length()) && (strip.find_last_of(' ') == strip.length() - 1))
				strip.erase(strip.end() - 1);

			return strip;
		}

		n++;
	}
	return "";
}

irc::sepstream::sepstream(const std::string &source, char seperator) : tokens(source), sep(seperator)
{
	last_starting_position = tokens.begin();
	n = tokens.begin();
}

const std::string irc::sepstream::GetToken()
{
	std::string::iterator lsp = last_starting_position;

	while (n != tokens.end())
	{
		if ((*n == sep) || (n+1 == tokens.end()))
		{
			last_starting_position = n+1;
			std::string strip = std::string(lsp, n+1 == tokens.end() ? n+1  : n++);

			while ((strip.length()) && (strip.find_last_of(sep) == strip.length() - 1))
				strip.erase(strip.end() - 1);

			return strip;
		}

		n++;
	}

	return "";
}

irc::sepstream::~sepstream()
{
}

std::string irc::hex(const unsigned char *raw, size_t rawsz)
{
	if (!rawsz)
		return "";

	/* EWW! This used to be using sprintf, which is WAY inefficient. -Special */
	
	const char *hex = "0123456789abcdef";
	
	char buf[rawsz*2+1];

	size_t i, j;
	for (i = 0, j = 0; j < rawsz; ++j)
	{
		buf[i++] = hex[raw[j] / 16];
		buf[i++] = hex[raw[j] % 16];
	}
	buf[i] = '\0';

	return buf;
}

const char* irc::Spacify(char* n)
{
	static char x[MAXBUF];
	strlcpy(x,n,MAXBUF);
	for (char* y = x; *y; y++)
		if (*y == '_')
			*y = ' ';
	return x;
}


irc::modestacker::modestacker(bool add) : adding(add)
{
	sequence.clear();
	sequence.push_back("");
}

void irc::modestacker::Push(char modeletter, const std::string &parameter)
{
	*(sequence.begin()) += modeletter;
	sequence.push_back(parameter);
}

void irc::modestacker::Push(char modeletter)
{
	this->Push(modeletter,"");
}

void irc::modestacker::PushPlus()
{
	this->Push('+',"");
}

void irc::modestacker::PushMinus()
{
	this->Push('-',"");
}

int irc::modestacker::GetStackedLine(std::deque<std::string> &result)
{
	int n = 0;
	result.clear();
	result.push_back(adding ? "+" : "-");

	while (!sequence[0].empty() && (sequence.size() > 1) && (result.size() < MAXMODES+1))
	{
		result[0] += *(sequence[0].begin());
		if (!sequence[1].empty())
			result.push_back(sequence[1]);
		sequence[0].erase(sequence[0].begin());
		sequence.erase(sequence.begin() + 1);
		n++;
	}

	return n;
}

irc::stringjoiner::stringjoiner(const std::string &seperator, const std::vector<std::string> &sequence, int begin, int end)
{
	for (int v = begin; v < end; v++)
		joined.append(sequence[v]).append(seperator);
	joined.append(sequence[end]);
}

irc::stringjoiner::stringjoiner(const std::string &seperator, const std::deque<std::string> &sequence, int begin, int end)
{
	for (int v = begin; v < end; v++)
		joined.append(sequence[v]).append(seperator);
	joined.append(sequence[end]);
}

irc::stringjoiner::stringjoiner(const std::string &seperator, const char** sequence, int begin, int end)
{
	for (int v = begin; v < end; v++)
		joined.append(sequence[v]).append(seperator);
	joined.append(sequence[end]);
}

std::string& irc::stringjoiner::GetJoined()
{
	return joined;
}

irc::portparser::portparser(const std::string &source, bool allow_overlapped) : in_range(0), range_begin(0), range_end(0), overlapped(allow_overlapped)
{
	sep = new irc::commasepstream(source);
	overlap_set.clear();
}

irc::portparser::~portparser()
{
	delete sep;
}

bool irc::portparser::Overlaps(long val)
{
	if (!overlapped)
		return false;

	if (overlap_set.find(val) == overlap_set.end())
	{
		overlap_set[val] = true;
		return false;
	}
	else
		return true;
}

long irc::portparser::GetToken()
{
	if (in_range > 0)
	{
		in_range++;
		if (in_range <= range_end)
		{
			if (!Overlaps(in_range))
			{
				return in_range;
			}
			else
			{
				while (((Overlaps(in_range)) && (in_range <= range_end)))
					in_range++;
				
				if (in_range <= range_end)
					return in_range;
			}
		}
		else
			in_range = 0;
	}

	std::string x = sep->GetToken();

	if (x == "")
		return 0;

	while (Overlaps(atoi(x.c_str())))
	{
		x = sep->GetToken();

		if (x == "")
			return 0;
	}

	std::string::size_type dash = x.rfind('-');
	if (dash != std::string::npos)
	{
		std::string sbegin = x.substr(0, dash);
		std::string send = x.substr(dash+1, x.length());
		range_begin = atoi(sbegin.c_str());
		range_end = atoi(send.c_str());

		if ((range_begin > 0) && (range_end > 0) && (range_begin < 65536) && (range_end < 65536) && (range_begin < range_end))
		{
			in_range = range_begin;
			return in_range;
		}
		else
		{
			/* Assume its just the one port */
			return atoi(sbegin.c_str());
		}
	}
	else
	{
		return atoi(x.c_str());
	}
}

irc::dynamicbitmask::dynamicbitmask() : bits_size(4)
{
	/* We start with 4 bytes allocated which is room
	 * for 4 items. Something makes me doubt its worth
	 * allocating less than 4 bytes.
	 */
	bits = new unsigned char[bits_size];
	memset(bits, 0, bits_size);
}

irc::dynamicbitmask::~dynamicbitmask()
{
	/* Tidy up the entire used memory on delete */
	delete[] bits;
}
		          
irc::bitfield irc::dynamicbitmask::Allocate()
{
	/* Yeah, this isnt too efficient, however a module or the core
	 * should only be allocating bitfields on load, the Toggle and
	 * Get methods are O(1) as these are called much more often.
	 */
	unsigned char* freebits = this->GetFreeBits();
	for (unsigned char i = 0; i < bits_size; i++)
	{
		/* Yes, this is right. You'll notice we terminate the  loop when !current_pos,
		 * this is because we logic shift our bit off the end of unsigned char, and its
		 * lost, making the loop counter 0 when we're done.
		 */
		for (unsigned char current_pos = 1; current_pos; current_pos = current_pos << 1)
		{
			if (!(freebits[i] & current_pos))
			{
				freebits[i] |= current_pos;
				return std::make_pair(i, current_pos);
			}
		}
	}
	/* We dont have any free space left, increase by one */

	if (bits_size == 255)
		/* Oh dear, cant grow it any further */
		throw std::bad_alloc();

	unsigned char old_bits_size = bits_size;
	bits_size++;
	/* Allocate new bitfield space */
	unsigned char* temp_bits = new unsigned char[bits_size];
	unsigned char* temp_freebits = new unsigned char[bits_size];
	/* Copy the old data in */
	memcpy(temp_bits, bits, old_bits_size);
	memcpy(temp_freebits, freebits, old_bits_size);
	/* Delete the old data pointers */
	delete[] bits;
	delete[] freebits;
	/* Swap the pointers over so now the new 
	 * pointers point to our member values
	 */
	bits = temp_bits;
	freebits = temp_freebits;
	this->SetFreeBits(freebits);
	/* Initialize the new byte on the end of
	 * the bitfields, pre-allocate the one bit
	 * for this allocation
	 */
	bits[old_bits_size] = 0;
	freebits[old_bits_size] = 1;
	/* We already know where we just allocated
	 * the bitfield, so no loop needed
	 */
	return std::make_pair(old_bits_size, 1);
}

bool irc::dynamicbitmask::Deallocate(irc::bitfield &pos)
{
	/* We dont bother to shrink the bitfield
	 * on deallocation, the most we could do
	 * is save one byte (!) and this would cost
	 * us a loop (ugly O(n) stuff) so we just
	 * clear the bit and leave the memory
	 * claimed -- nobody will care about one
	 * byte.
	 */
	if (pos.first < bits_size)
	{
		this->GetFreeBits()[pos.first] &= ~pos.second;
		return true;
	}
	/* They gave a bitfield outside of the
	 * length of our array. BAD programmer.
	 */
	return false;
}

void irc::dynamicbitmask::Toggle(irc::bitfield &pos, bool state)
{
	/* Range check the value */
	if (pos.first < bits_size)
	{
		if (state)
			/* Set state, OR the state in */
			bits[pos.first] |= pos.second;
		else
			/* Clear state, AND the !state out */
			bits[pos.first] &= ~pos.second;
	}
}

bool irc::dynamicbitmask::Get(irc::bitfield &pos)
{
	/* Range check the value */
	if (pos.first < bits_size)
		return (bits[pos.first] & pos.second);
	else
		/* We can't return false, otherwise we can't
		 * distinguish between failure and a cleared bit!
		 * Our only sensible choice is to throw (ew).
		 */
		throw ModuleException("irc::dynamicbitmask::Get(): Invalid bitfield, out of range");
}

unsigned char irc::dynamicbitmask::GetSize()
{
	return bits_size;
}