ultragrep 0.1.0 → 0.10.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/bin/ultragrep_build_indexes +45 -0
- data/lib/ultragrep.rb +47 -61
- data/lib/ultragrep/config.rb +6 -0
- data/lib/ultragrep/log_collector.rb +67 -0
- data/lib/ultragrep/version.rb +1 -1
- data/src/Makefile +24 -0
- data/{ext/ultragrep → src}/extconf.rb +0 -0
- data/src/pcre.h +668 -0
- data/src/request.h +13 -0
- data/src/ug_build_index.c +109 -0
- data/src/ug_cat.c +188 -0
- data/src/ug_guts.c +199 -0
- data/src/ug_gzip.c +242 -0
- data/src/ug_gzip.h +8 -0
- data/src/ug_index.c +62 -0
- data/src/ug_index.h +23 -0
- data/src/ug_lua.c +119 -0
- data/src/ug_lua.h +10 -0
- metadata +25 -28
- data/ext/ultragrep/Makefile +0 -39
- data/ext/ultragrep/rails_req.c +0 -102
- data/ext/ultragrep/rails_req.h +0 -6
- data/ext/ultragrep/req_matcher.h +0 -17
- data/ext/ultragrep/request.c +0 -41
- data/ext/ultragrep/request.h +0 -22
- data/ext/ultragrep/ug_build_index.c +0 -99
- data/ext/ultragrep/ug_cat.c +0 -46
- data/ext/ultragrep/ug_guts.c +0 -138
- data/ext/ultragrep/ug_index.c +0 -83
- data/ext/ultragrep/ug_index.h +0 -27
- data/ext/ultragrep/work_req.c +0 -200
- data/ext/ultragrep/work_req.h +0 -6
- data/ext/ultragrep/zran.c +0 -291
data/ext/ultragrep/work_req.h
DELETED
data/ext/ultragrep/zran.c
DELETED
@@ -1,291 +0,0 @@
|
|
1
|
-
/* zran.c -- example of zlib/gzip stream indexing and random access
|
2
|
-
* Copyright (C) 2005 Mark Adler
|
3
|
-
* For conditions of distribution and use, see copyright notice in zlib.h
|
4
|
-
Version 1.0 29 May 2005 Mark Adler */
|
5
|
-
|
6
|
-
/* Illustrate the use of Z_BLOCK, inflatePrime(), and inflateSetDictionary()
|
7
|
-
for random access of a compressed file. A file containing a zlib or gzip
|
8
|
-
stream is provided on the command line. The compressed stream is decoded in
|
9
|
-
its entirety, and an index built with access points about every SPAN bytes
|
10
|
-
in the uncompressed output. The compressed file is left open, and can then
|
11
|
-
be read randomly, having to decompress on the average SPAN/2 uncompressed
|
12
|
-
bytes before getting to the desired block of data.
|
13
|
-
|
14
|
-
An access point can be created at the start of any deflate block, by saving
|
15
|
-
the starting file offset and bit of that block, and the 32K bytes of
|
16
|
-
uncompressed data that precede that block. Also the uncompressed offset of
|
17
|
-
that block is saved to provide a referece for locating a desired starting
|
18
|
-
point in the uncompressed stream. build_index() works by decompressing the
|
19
|
-
input zlib or gzip stream a block at a time, and at the end of each block
|
20
|
-
deciding if enough uncompressed data has gone by to justify the creation of
|
21
|
-
a new access point. If so, that point is saved in a data structure that
|
22
|
-
grows as needed to accommodate the points.
|
23
|
-
|
24
|
-
To use the index, an offset in the uncompressed data is provided, for which
|
25
|
-
the latest accees point at or preceding that offset is located in the index.
|
26
|
-
The input file is positioned to the specified location in the index, and if
|
27
|
-
necessary the first few bits of the compressed data is read from the file.
|
28
|
-
inflate is initialized with those bits and the 32K of uncompressed data, and
|
29
|
-
the decompression then proceeds until the desired offset in the file is
|
30
|
-
reached. Then the decompression continues to read the desired uncompressed
|
31
|
-
data from the file.
|
32
|
-
*/
|
33
|
-
|
34
|
-
#include <stdio.h>
|
35
|
-
#include <stdlib.h>
|
36
|
-
#include <string.h>
|
37
|
-
#include "zlib.h"
|
38
|
-
#include "ug_index.h"
|
39
|
-
|
40
|
-
#define WINSIZE 32768U /* sliding window size */
|
41
|
-
#define CHUNK 16384 /* file input buffer size */
|
42
|
-
|
43
|
-
|
44
|
-
/* Make one entire pass through the compressed stream and build an index, with
|
45
|
-
access points about every span bytes of uncompressed output -- span is
|
46
|
-
chosen to balance the speed of random access against the memory requirements
|
47
|
-
of the list, about 32K bytes per access point. Note that data after the end
|
48
|
-
of the first zlib or gzip stream in the file is ignored. build_index()
|
49
|
-
returns the number of access points on success (>= 1), Z_MEM_ERROR for out
|
50
|
-
of memory, Z_DATA_ERROR for an error in the input file, or Z_ERRNO for a
|
51
|
-
file read error. On success, *built points to the resulting index. */
|
52
|
-
|
53
|
-
int build_gz_index(build_idx_context_t *cxt)
|
54
|
-
{
|
55
|
-
int ret, last_line_size;
|
56
|
-
off_t totin;
|
57
|
-
uint64_t idx_offset;
|
58
|
-
z_stream strm;
|
59
|
-
unsigned char input[CHUNK];
|
60
|
-
unsigned char window[WINSIZE];
|
61
|
-
unsigned char *start, *p, *output, *output_ptr;
|
62
|
-
|
63
|
-
start = p = window;
|
64
|
-
output = output_ptr = NULL;
|
65
|
-
|
66
|
-
bzero(&strm, sizeof(z_stream));
|
67
|
-
ret = inflateInit2(&strm, 47); /* automatic zlib or gzip decoding */
|
68
|
-
if (ret != Z_OK)
|
69
|
-
return ret;
|
70
|
-
|
71
|
-
/* inflate the input, maintain a sliding window, and build an index -- this
|
72
|
-
also validates the integrity of the compressed data using the check
|
73
|
-
information at the end of the gzip or zlib stream */
|
74
|
-
totin = 0;
|
75
|
-
strm.avail_out = 0;
|
76
|
-
do {
|
77
|
-
/* get some compressed data from input file */
|
78
|
-
strm.avail_in = fread(input, 1, CHUNK, cxt->log);
|
79
|
-
if (ferror(cxt->log)) {
|
80
|
-
ret = Z_ERRNO;
|
81
|
-
goto build_index_error;
|
82
|
-
}
|
83
|
-
if (strm.avail_in == 0) {
|
84
|
-
ret = Z_DATA_ERROR;
|
85
|
-
goto build_index_error;
|
86
|
-
}
|
87
|
-
strm.next_in = input;
|
88
|
-
|
89
|
-
/* process all of that, or until end of stream */
|
90
|
-
do {
|
91
|
-
/* reset sliding window if necessary */
|
92
|
-
if (strm.avail_out == 0) {
|
93
|
-
strm.avail_out = WINSIZE;
|
94
|
-
strm.next_out = window;
|
95
|
-
}
|
96
|
-
|
97
|
-
/* inflate until out of input, output, or at end of block --
|
98
|
-
update the total input and output counters */
|
99
|
-
totin += strm.avail_in;
|
100
|
-
ret = inflate(&strm, Z_BLOCK); /* return at end of block */
|
101
|
-
totin -= strm.avail_in;
|
102
|
-
if (ret == Z_NEED_DICT)
|
103
|
-
ret = Z_DATA_ERROR;
|
104
|
-
if (ret == Z_MEM_ERROR || ret == Z_DATA_ERROR)
|
105
|
-
goto build_index_error;
|
106
|
-
if (ret == Z_STREAM_END)
|
107
|
-
break;
|
108
|
-
|
109
|
-
/*
|
110
|
-
* at the end of a gzip block we reset our context information, so if handle_request
|
111
|
-
* decides to add an index somewhere inside this block we can have an index to the gzip block.
|
112
|
-
*
|
113
|
-
* note that we store the bit offset in the high byte of the offset field in the index.
|
114
|
-
*
|
115
|
-
* a data_type of 64 means done with the "last block" -- we might index here. not sure.
|
116
|
-
*/
|
117
|
-
|
118
|
-
if ((strm.data_type & 128) && !(strm.data_type & 64) && strm.total_out > 0 ) {
|
119
|
-
idx_offset = (((uint64_t) strm.data_type & 7) << 56);
|
120
|
-
idx_offset |= (totin & 0x00FFFFFFFFFFFFFF);
|
121
|
-
|
122
|
-
|
123
|
-
/* if there's room left in the buffer copy from middle -> end of buffer */
|
124
|
-
if (strm.avail_out)
|
125
|
-
memcpy(cxt->data, window + WINSIZE - strm.avail_out, strm.avail_out);
|
126
|
-
|
127
|
-
/* copy from beginning -> middle of buffer if needed */
|
128
|
-
if (strm.avail_out < WINSIZE)
|
129
|
-
memcpy(cxt->data + strm.avail_out, window, WINSIZE - strm.avail_out);
|
130
|
-
|
131
|
-
cxt->data_size = WINSIZE;
|
132
|
-
}
|
133
|
-
|
134
|
-
for(;;) {
|
135
|
-
int output_len;
|
136
|
-
p = start;
|
137
|
-
|
138
|
-
while ( (*p != '\n') && ((p - window) < (WINSIZE - strm.avail_out)) )
|
139
|
-
p++;
|
140
|
-
|
141
|
-
output_len = output_ptr - output;
|
142
|
-
output = realloc(output, (p - start) + output_len + 1);
|
143
|
-
output_ptr = output + output_len;
|
144
|
-
|
145
|
-
strncpy(output_ptr, start, p - start);
|
146
|
-
output_ptr += p - start;
|
147
|
-
|
148
|
-
if ( p == (window + (WINSIZE - strm.avail_out)) ) {
|
149
|
-
/* end of buffer or available data, don't pass along to request matching, save for later */
|
150
|
-
if ( strm.avail_out == 0 ) /* wrap to start of buffer */
|
151
|
-
start = window;
|
152
|
-
else
|
153
|
-
start = window + (WINSIZE - strm.avail_out);
|
154
|
-
|
155
|
-
break;
|
156
|
-
} else {
|
157
|
-
/* p should be a newline */
|
158
|
-
*output_ptr = '\0';
|
159
|
-
puts(output);
|
160
|
-
|
161
|
-
// funcall
|
162
|
-
free(output);
|
163
|
-
output = output_ptr = NULL;
|
164
|
-
start = p + 1;
|
165
|
-
}
|
166
|
-
}
|
167
|
-
} while (strm.avail_in != 0);
|
168
|
-
} while (ret != Z_STREAM_END);
|
169
|
-
|
170
|
-
/* clean up and return index (release unused entries in list) */
|
171
|
-
(void)inflateEnd(&strm);
|
172
|
-
return 0;
|
173
|
-
|
174
|
-
/* return error */
|
175
|
-
build_index_error:
|
176
|
-
(void)inflateEnd(&strm);
|
177
|
-
return ret;
|
178
|
-
}
|
179
|
-
|
180
|
-
#if 0
|
181
|
-
/* Use the index to read len bytes from offset into buf, return bytes read or
|
182
|
-
negative for error (Z_DATA_ERROR or Z_MEM_ERROR). If data is requested past
|
183
|
-
the end of the uncompressed data, then extract() will return a value less
|
184
|
-
than len, indicating how much as actually read into buf. This function
|
185
|
-
should not return a data error unless the file was modified since the index
|
186
|
-
was generated. extract() may also return Z_ERRNO if there is an error on
|
187
|
-
reading or seeking the input file. */
|
188
|
-
int extract(FILE *in, struct access *index, off_t offset,
|
189
|
-
unsigned char *buf, int len)
|
190
|
-
{
|
191
|
-
int ret, skip;
|
192
|
-
z_stream strm;
|
193
|
-
struct point *here;
|
194
|
-
unsigned char input[CHUNK];
|
195
|
-
unsigned char discard[WINSIZE];
|
196
|
-
|
197
|
-
/* proceed only if something reasonable to do */
|
198
|
-
if (len < 0)
|
199
|
-
return 0;
|
200
|
-
|
201
|
-
/* find where in stream to start */
|
202
|
-
here = index->list;
|
203
|
-
ret = index->have;
|
204
|
-
while (--ret && here[1].out <= offset)
|
205
|
-
here++;
|
206
|
-
|
207
|
-
/* initialize file and inflate state to start there */
|
208
|
-
strm.zalloc = Z_NULL;
|
209
|
-
strm.zfree = Z_NULL;
|
210
|
-
strm.opaque = Z_NULL;
|
211
|
-
strm.avail_in = 0;
|
212
|
-
strm.next_in = Z_NULL;
|
213
|
-
ret = inflateInit2(&strm, -15); /* raw inflate */
|
214
|
-
if (ret != Z_OK)
|
215
|
-
return ret;
|
216
|
-
ret = fseeko(in, here->in - (here->bits ? 1 : 0), SEEK_SET);
|
217
|
-
if (ret == -1)
|
218
|
-
goto extract_ret;
|
219
|
-
if (here->bits) {
|
220
|
-
ret = getc(in);
|
221
|
-
if (ret == -1) {
|
222
|
-
ret = ferror(in) ? Z_ERRNO : Z_DATA_ERROR;
|
223
|
-
goto extract_ret;
|
224
|
-
}
|
225
|
-
(void)inflatePrime(&strm, here->bits, ret >> (8 - here->bits));
|
226
|
-
}
|
227
|
-
(void)inflateSetDictionary(&strm, here->window, WINSIZE);
|
228
|
-
|
229
|
-
/* skip uncompressed bytes until offset reached, then satisfy request */
|
230
|
-
offset -= here->out;
|
231
|
-
strm.avail_in = 0;
|
232
|
-
skip = 1; /* while skipping to offset */
|
233
|
-
do {
|
234
|
-
/* define where to put uncompressed data, and how much */
|
235
|
-
if (offset == 0 && skip) { /* at offset now */
|
236
|
-
strm.avail_out = len;
|
237
|
-
strm.next_out = buf;
|
238
|
-
skip = 0; /* only do this once */
|
239
|
-
}
|
240
|
-
if (offset > WINSIZE) { /* skip WINSIZE bytes */
|
241
|
-
strm.avail_out = WINSIZE;
|
242
|
-
strm.next_out = discard;
|
243
|
-
offset -= WINSIZE;
|
244
|
-
}
|
245
|
-
else if (offset != 0) { /* last skip */
|
246
|
-
strm.avail_out = (unsigned)offset;
|
247
|
-
strm.next_out = discard;
|
248
|
-
offset = 0;
|
249
|
-
}
|
250
|
-
|
251
|
-
/* uncompress until avail_out filled, or end of stream */
|
252
|
-
do {
|
253
|
-
if (strm.avail_in == 0) {
|
254
|
-
strm.avail_in = fread(input, 1, CHUNK, in);
|
255
|
-
if (ferror(in)) {
|
256
|
-
ret = Z_ERRNO;
|
257
|
-
goto extract_ret;
|
258
|
-
}
|
259
|
-
if (strm.avail_in == 0) {
|
260
|
-
ret = Z_DATA_ERROR;
|
261
|
-
goto extract_ret;
|
262
|
-
}
|
263
|
-
strm.next_in = input;
|
264
|
-
}
|
265
|
-
ret = inflate(&strm, Z_NO_FLUSH); /* normal inflate */
|
266
|
-
if (ret == Z_NEED_DICT)
|
267
|
-
ret = Z_DATA_ERROR;
|
268
|
-
if (ret == Z_MEM_ERROR || ret == Z_DATA_ERROR)
|
269
|
-
goto extract_ret;
|
270
|
-
if (ret == Z_STREAM_END)
|
271
|
-
break;
|
272
|
-
} while (strm.avail_out != 0);
|
273
|
-
|
274
|
-
/* if reach end of stream, then don't keep trying to get more */
|
275
|
-
if (ret == Z_STREAM_END)
|
276
|
-
break;
|
277
|
-
|
278
|
-
/* do until offset reached and requested data read, or stream ends */
|
279
|
-
} while (skip);
|
280
|
-
|
281
|
-
/* compute number of uncompressed bytes read after offset */
|
282
|
-
ret = skip ? 0 : len - strm.avail_out;
|
283
|
-
|
284
|
-
/* clean up and return bytes read or error */
|
285
|
-
extract_ret:
|
286
|
-
(void)inflateEnd(&strm);
|
287
|
-
return ret;
|
288
|
-
}
|
289
|
-
#endif
|
290
|
-
|
291
|
-
|