Is there a way in C to parse a piece of text and obtain values for argv and argc, as if the text had been passed to an application on the command line?
This doesn't have to work on Windows, just Linux - I also don't care about quoting of arguments.
13 Answers
This one I wrote also considers quotes (but not nested)
Feel free to contribute.
/*
Tokenize string considering also quotes.
By Zibri <zibri AT zibri DOT org>
https://github.com/Zibri/tokenize
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdbool.h>
int main(int argc, char *argv[])
{
char *str1, *token;
int j;
char *qstart = NULL;
bool quoted = false;
if (argc != 2) {
fprintf(stderr, "Usage: %s string\n", argv[0]);
exit(EXIT_FAILURE);
}
for (j = 1, str1 = argv[1];; j++, str1 = NULL) {
token = strtok(str1, " ");
if (token == NULL)
break;
if ((token[0] == 0x27) || (token[0] == 0x22)) {
qstart = token + 1;
quoted = true;
}
if ((token[strlen(token) - 1] == 0x27) || (token[strlen(token) - 1] == 0x22)) {
quoted = false;
token[strlen(token) - 1] = 0;
printf("%d: %s\n", j, qstart);
} else {
if (quoted) {
token[strlen(token)] = 0x20;
j--;
} else
printf("%d: %s\n", j, token);
}
}
if (quoted) {
fprintf(stderr, "String quoting error\n");
return EXIT_FAILURE;
} else
return EXIT_SUCCESS;
}
Example output:
$ ./tokenize "1 2 3 '4 5 6' 7 8 \"test abc\" 10 11"
1: 1
2: 2
3: 3
4: 4 5 6
5: 7
6: 8
7: test abc
8: 10
9: 11
My project requires breaking a string into argc and argv.
Found a pretty excellent code of Torek. But it alters the input buffer so I made some modifications to fit my needs.
I just put a little bit more to handle quote mixing when input in the command line so the behavior is more (not completely) like Linux Shell.
Note: This function doesn't edit the original string, so you can reuse the input buffer (error report,etc).
void remove_quote(char* input){
//Implementing yourself to remove quotes so it would be completely like Linux shell
}
size_t cmd_param_split(char *buffer, char *argv[], size_t argv_max_size)
{
char *p, *start_of_word;
int c, i;
enum states { DULL=0, IN_WORD, IN_STRING, QUOTE_DOUBLE,QUOTE_SINGLE } state = DULL;
size_t argc = 0;
int quote = 0;
for (p = buffer; argc < argv_max_size && *p != '\0'; p++) {
c = (unsigned char) *p;
printf("processing %c, state = %d\n", c,state);
switch (state) {
case DULL:
if (isspace(c)) {
continue;
}
if (c == '"' ||c == '\'') {
quote = c;
state = IN_STRING;
start_of_word = p + 1;
continue;
}
state = IN_WORD;
start_of_word = p;
continue;
case IN_STRING:
if (c == '"' || c == '\'') {
if (c!=quote)
continue;
else
quote = 0;
strncpy(argv[argc],start_of_word, p - start_of_word);
remove_quote(argv[argc]);
argc++;
state = DULL;
}
continue;
case IN_WORD:
if(quote==0 && (c == '\"' ||c == '\''))
quote = c;
else if (quote == c)
quote = 0;
if (isspace(c) && quote==0) {
strncpy(argv[argc],start_of_word, p - start_of_word);
remove_quote(argv[argc]);
argc++;
state = DULL;
}
continue;
}
}
if (state != DULL && argc < argv_max_size){
strncpy(argv[argc],start_of_word, p - start_of_word);
remove_quote(argv[argc]);
argc++;
}
if (quote){
printf("WARNING: Quote is unbalanced. This could lead to unwanted-behavior\n");
for(i = 0;i<argc;i++)
printf("arg %d = [%s]\n",i,argv[i]);
printf("Original buffer: [%s]\n",buffer);
}
return argc;
}
int main()
{
int i=0;
int argc;
char* argv[64];
for(i=0;i<64;i++){
argv[i] = malloc(256);
memset(argv[i],0x0,256);
}
char* buffer="1 2 3 \'3 4\"567\' \"bol\'obala\" 2x2=\"foo\"";
argc = cmd_param_split(buffer,argv,64);
for(i = 0;i<argc;i++)
printf("arg %d = [%s]\n",i,argv[i]);
return 0;
}
Tested with below strings
1. "1 2 3 \'3 4\"567\' \"bol\'obala\" 2x2=\"foo\""
arg 0 = [1]
arg 1 = [2]
arg 2 = [3]
arg 3 = [3 4"567]
arg 4 = [bol'obala]
arg 5 = [2x2="foo"]
2. "./foo bar=\"Hanoi HoChiMinh\" exp='foo123 \"boo111' mixquote \"hanoi \'s\""
arg 0 = [./foo]
arg 1 = [bar="Hanoi HoChiMinh"]
arg 2 = [exp='foo123 "boo111']
arg 3 = [mixquote]
arg 4 = [hanoi 's]
However, Linux shell would remove quotes, even in mixed case, as below when running from cmd line, tested in a RaspberryPi.
./foo bar="Hanoi HoChiMinh" exp='foo123 "boo111' mixquote "hanoi 's"
arg 0 = [./foo]
arg 1 = [bar=Hanoi HoChiMinh]
arg 2 = [exp=foo123 "boo111]
arg 3 = [mixquote]
arg 4 = [hanoi 's]
So if you really want to mimic the whole Linux shell's behavior, just put a little bit more effort into removing quotes remove_quote() function as I leave blank above.
Related