本文对C的正则库regex和pcre在做域名验证的场景下做评测。
验证DNS域名的正则表达式为:
"^[0-9a-zA-Z_-]+(\\.[0-9a-zA-Z_-]+)*(\\.[a-zA-Z]{2,}\\.)$"
对于正常DNS请求日志中的6177578条日志做正则验证处理。
1,pcre
评测所用的pcre的版本号是:7.8.3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
|
#include <stdio.h>
#include <string.h>
#include <pcre.h>
#define OVECCOUNT 30 /* should be a multiple of 3 */
#define EBUFLEN 128
#define BUFLEN 1024
int
main(
int
argc,
char
*argv[])
{
pcre *re;
const
char
*error;
int
erroffset;
FILE
*fd;
int
ovector[OVECCOUNT];
int
rc, i;
int
succ = 0, fail = 0;
char
src[1024];
char
pattern[] =
"^[0-9a-zA-Z_-]+(\\.[0-9a-zA-Z_-]+)*(\\.[a-zA-Z]{2,}\\.)$"
;
printf
(
"Pattern: %s\n"
, pattern);
re = pcre_compile(pattern, 0, &error, &erroffset, NULL);
if
(re == NULL) {
printf
(
"PCRE compilation failed at offset %d: %s\n"
, erroffset, error);
return
1;
}
if
((fd =
fopen
(argv[1],
"r"
)) == NULL) {
printf
(
"open file error\n"
);
return
1;
}
while
(
fgets
(src, 1024, fd)) {
rc = pcre_exec(re, NULL, src,
strlen
(src), 0, 0, ovector, OVECCOUNT);
if
(rc < 0) {
fail++;
}
else
{
succ++;
}
}
printf
(
"success:%d fail:%d\n"
, succ, fail);
fclose
(fd);
free
(re);
return
0;
}
|
处理完所有数据的耗时是:
$time ./pcre_t query_domains
Pattern: ^[0-9a-zA-Z_-]+(\.[0-9a-zA-Z_-]+)*(\.[a-zA-Z]{2,}\.)$
success:6177443 fail:135
real 0m8.257s
user 0m8.194s
sys 0m0.058s
2,regex
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
|
#include <stdio.h>
#include <string.h>
#include <regex.h>
#define SUBSLEN 100
#define EBUFLEN 1280
#define BUFLEN 1024
int
main(
int
argc,
char
*argv[])
{
size_t
len;
regex_t re;
regmatch_t subs[SUBSLEN];
char
matched[BUFLEN];
char
errbuf[EBUFLEN];
int
err, i, succ=0, fail=0;
FILE
*fd;
char
*src;
char
line[1024];
char
pattern[] =
"^[0-9a-zA-Z_-]+(\\.[0-9a-zA-Z_-]+)*(\\.[a-zA-Z]+\\.)$"
;
printf
(
"Pattern: %s\n"
, pattern);
if
(regcomp(&re, pattern, REG_EXTENDED | REG_NEWLINE)) {
len = regerror(err, &re, errbuf,
sizeof
(errbuf));
printf
(
"error: regcomp: %s\n"
, errbuf);
return
1;
}
if
((fd =
fopen
(argv[1],
"r"
)) == NULL) {
printf
(
"open file error\n"
);
return
1;
}
while
(
fgets
(line, 1024, fd)) {
err = regexec(&re, line, (
size_t
) SUBSLEN, subs, 0);
if
(err == REG_NOMATCH) {
fail++;
}
else
{
succ++;
}
}
printf
(
"success:%d, fails:%d\n"
, succ, fail);
fclose
(fd);
regfree(&re);
return
(0);
}
|
处理完所有数据耗时:
$time ./regex_t query_domains
Pattern: ^[0-9a-zA-Z_-]+(\.[0-9a-zA-Z_-]+)*(\.[a-zA-Z]+\.)$
success:6177443, fails:135
real 0m50.876s
user 0m50.783s
sys 0m0.058s
3,结论。
可以看到,对于域名验证的场景。pcre明显优于POSIX regex库。在规则已经编译好的情况下,pcre每秒大约处理74.8w条域名,而regex每秒大约处理12.1万条。