REGEX. c gnu extracts and filters data, regex. cgnu

REGEX. c gnu extracts and filters data, regex. cgnu

Today, @ SVCHAO caught up with interest .. I have picked up the regular expression interest and tried notepad ++. Basically, the syntax is not forgotten. However, if it is used in an embedded solution, it still seems a little difficult.

Mark a basic syntax first.

Single Character matching

It is described in square brackets. For example, [0135678] indicates 0 1 3 5 6 7 8 can be matched through this time

Exclude using square brackets. For example, [^ azAZ] indicates that all except a z A and Z can be matched by this byte.

-Indicates the range. For example, [0-9a-zA-Z] indicates all numbers and letters.

Common expressions exist for a single character. For example, \ w represents [0-9a-zA-Z ].

Use {} to describe the number of matches. For example, \ w {2} indicates that two consecutive characters are numbers or letters.

Capture comparison results using...

 

The above content is self-developed.

This demo was found for embedded systems to adapt to application scenarios.

Slre (Github Super Light Regular Expression library)

But I don't seem to support {}, and I haven't found any greedy... So temporarily put on hold

 

Later, familiarize yourself with the regex. h In GNU.

It is found that the regex interface of python is similar to that of posix.

The machine of the potholes has not been updated for a long time. After a while, the environment has finally been tested.

#include <stdio.h>
#include <sys/types.h>
#include <regex.h>

int main(int char_c,char** char_v)
{
    char* p_str = "  Url        =  www.google.com.hk ;";
    char* p_reg = " {0,}(\\S+) {0,}= {0,}(\\S+) {0,};";

    regex_t reg;
    regmatch_t matchs[20];
    memset(matchs,0,sizeof(matchs));
    int r;

    r = regcomp(&reg, p_reg, REG_EXTENDED);

    r = regexec(&reg, p_str, 20, matchs, 0);

    return 0;
}

Because it is the code used for self-testing, I did not specifically output the code or directly break the breakpoint to see the result,

REG_EXTENDED must be added, otherwise some features will not be used,

Another strange point is that matchs [0] stores a matching range of the entire p_reg, So if you only need to extract the target result of () in our definition,

From p_reg [1.

It took a long time to explain C's string syntax... The result is \ and... However, there seems to be no more efficient solutions.

The next step is to manually add the greedy part to the SLRE, or find the GNU regex. c to crop it...

 

Let's talk about it later.

#include <stdio.h>
#include <sys/types.h>
#include <regex.h>
#include <string.h>

int main(int char_c, char** char_v) {
    char* p_str = "  Url        =  www.google.com.hk ;";
    char* p_reg = " {0,}(\\S+) {0,}= {0,}(\\S+) {0,};";

    regex_t reg;
    regmatch_t matchs[20];
    int r, i;

    r = regcomp(&reg, p_reg, REG_EXTENDED);
    if (r != 0) {
        printf("err");
    }

    r = regexec(&reg, p_str, 20, matchs, 0);
    if (r != 0) {
        printf("err");
    }

    for (i = 0; i < 20; i++) {
        int start_index = matchs[i].rm_so;
        int end_index = matchs[i].rm_eo;
        int len = end_index - start_index;

        if (start_index >= 0) {
            char dis_buffer[256];
            strncpy(dis_buffer, &p_str[start_index], len);
            dis_buffer[len]='\0';
            printf("catch:%s\n",dis_buffer);
        }
    }

    return 0;
}

	 

 

Since I have captured the output for a long time... I decided to paste the data capture code...

Use the Environment G ++ with eclipse-cdt

I don't change IDE when I get used to debugging.

catch:  Url        =  www.google.com.hk ;
catch:Url
catch:www.google.com.hk

The test output is correct...