C - Find a word and get the next two - c

I am parsing a file in C, line by line. Here is an exemple of what I am trying to do :
I have a line for example :
word word word WORDTOFIND: word1 word2 word word
What I want to do is : When I find the word WORDTOFIND, get the two next words (word1 and word2 in this case) of the line. Is there an easy way to do that in C ? I know about the strstr function, but I don't find a way to get the next two words word1 and word2 after I found the good one.

One approach would be this:
int main(void)
{
char *str = "rated rat cat bat hat";
char *key = "rat ";
char *pointer = NULL;
char nwords = 2;
if ((pointer = strstr(str, key)) != NULL)
{
while (*pointer != ' ') pointer++;
while (nwords >= 0)
{
printf("%c", *pointer);
if (*pointer == ' ') {
nwords--;
} else if (*pointer == '\0') {
exit(0);
}
pointer++;
}
}
}

You can try an aproach like this, using strtok to parse the words at every space. This code also uses malloc and realloc to allocate space for an array of strings, and grows it when needed.
The code looks like this:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#define MAXCHAR 100
void exit_if_null(void *ptr, const char *msg);
char *stripped_word(char *word);
int
main(int argc, char const *argv[]) {
FILE *filename;
char line[MAXCHAR];
char *word, *newword;
char **allwords;
int init_size = 8, count = 0, i;
const char *key = "WORDTOFIND";
filename = fopen("files.txt", "r");
if (filename == NULL) {
fprintf(stderr, "%s\n", "Error reading file!");
exit(EXIT_FAILURE);
}
allwords = malloc(init_size * sizeof(*allwords));
exit_if_null(allwords, "Initial Allocation");
while (fgets(line, MAXCHAR, filename) != NULL) {
word = strtok(line, " \n");
while (word != NULL) {
if (count == init_size) {
init_size *= 2;
allwords = realloc(allwords, init_size * sizeof(*allwords));
}
allwords[count] = malloc(strlen(word)+1);
exit_if_null(allwords[count], "Initial Allocation");
newword = stripped_word(word);
strcpy(allwords[count], newword);
count++;
word = strtok(NULL, " \n");
free(newword);
}
}
for (i = 0; i < count; i++) {
if (strcmp(key, allwords[i]) == 0) {
printf("Next two words:\n");
printf("%s\n", allwords[i+1]);
printf("%s\n", allwords[i+2]);
}
free(allwords[i]);
allwords[i] = NULL;
}
free(allwords);
allwords = NULL;
return 0;
}
void
exit_if_null(void *ptr, const char *msg) {
if (!ptr) {
printf("Unexpected null pointer: %s\n", msg);
exit(EXIT_FAILURE);
}
}
char
*stripped_word(char *word) {
int i, pos = 0;
char *result;
result = malloc(strlen(word)+1);
exit_if_null(result, "Initial Allocation");
for (i = 0; word[i] != '\0'; i++) {
if (isalpha(word[i]) || isdigit(word[i])) {
result[pos++] = word[i];
}
}
result[pos] = '\0';
return result;
}

Related

Reading in strings without header <string.h>

I am a newbie in C and for an exercise I have to write a program, where I can read in strings. If my reserved memory (length BUFFER_SIZE) isn't enough, it should reserve memory in increments of +=BUFFER_SIZE, as long as needed to read the string. I tried to write some functions to get this done, but it doesn't work. Can somebody please help me?
My code:
#include <stddef.h>
#include <stdio.h>
#include <stdlib.h>
#define BUFFER_SIZE 10
size_t string_length(char *string)
{
size_t i = 0;
while(string[i] != '\0')
{
i++;
}
return i;
}
void string_concatenate(char *string, char *string_to_chain)
{
size_t length = string_length(string);
for(size_t i = 0; *(string_to_chain + i) != '\0'; i++, length++)
{
string[length] = string_to_chain [i];
}
string[length] = '\0';
}
char *string_search(char *string, char character)
{
do
{
if (*string == character)
{
return (char*)string;
}
} while (*string++);
return NULL;
}
char *get_line()
{
int line_size = BUFFER_SIZE;
char* line = malloc(line_size * sizeof(char));
if(line == NULL)
{
return NULL;
}
printf("Bitte Text eingeben: \n");
fgets(line, BUFFER_SIZE, stdin);
char *new_line_character = string_search(line, '\n');
while(new_line_character == NULL)
{
line_size += BUFFER_SIZE;
line = realloc(line, (line_size * sizeof(char)));
if(line == NULL)
{
return NULL;
}
char *new_line = line + BUFFER_SIZE - 1;
fgets(new_line, line_size, stdin);
new_line_character = string_search(line, '\n');
}
*new_line_character = '\0';
return line;
}
}
int main(void) {
char *string = get_line();
printf("s%\n", string);
}

C spliting strings into an array outputs just the first token

I am writing a microshell program as homework at my university.
Everything goes well besides one function that's not doing exactly what I'd wish it did.
I am quite new to C programming, always used higher level languages.
In fact the only times I worked with C was fiddling around with Arduino.
So I've got a line of what user inputs on the prompt. I'm trying to split it into an array of strings separated by space.
I initialized an array with
char **args = NULL;
args = malloc(sizeof(char *) * LINE_LENGTH);
And I'm sending it to a function parse_line(line, args)
The function looks like this:
bool parse_line(char *line, char **arr) {
size_t i = 0;
char *point;
point = strtok(line, " ");
while (point != NULL) {
arr[i] = malloc(strlen(point) + 1);
strcpy(arr[i], point);
point = strtok(NULL, " ");
i++;
}
arr[i] = NULL;
if (!arr)
return false;
return true;
}
The thing is that afterwards in the arr resides only first token from the splitted up line.
I am debugging it and though variable 'point' gets the right values, they aren't copied into my array. Why? I don't know.
Oh.. and the line is an array of chars, dynamic one.
char * line = NULL;
line = read_input_line();
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <string.h>
#include <stdbool.h>
#include <sys/types.h>
#include <sys/wait.h>
#define LINE_LENGTH 50
void clear_screen();
void display_prompt(bool clearscr);
char * read_input_line();
void print_line(char *line);
bool parse_line(char *line, char **arr);
int main() {
bool initialRun = true;
while (true) {
// display prompt on the screen
display_prompt(initialRun);
if (initialRun)
initialRun = false;
// read input line from terminal
char * line = NULL;
line = read_input_line();
// basic commands
if (line == NULL)
continue;
if (strcmp(line, "exit") == 0) {
free(line);
exit(EXIT_SUCCESS);
} else if (strcmp(line, "clear") == 0) {
clear_screen();
continue;
}
// parse line into array
char **args = NULL;
args = malloc(sizeof(char *) * LINE_LENGTH);
if (!parse_line(line, args)) {
printf("Error during parsing command \n");
continue;
}
}
return 0;
}
void clear_screen() {
printf("\e[2J\e[H");
}
void display_prompt(bool clearscr) {
if (clearscr)
clear_screen();
printf(" > ");
}
char * read_input_line() {
char * line = (char *)malloc(sizeof(char) * LINE_LENGTH);
if (!fgets(line, LINE_LENGTH, stdin))
return NULL;
size_t len = strlen(line);
if (len > 0 && line[len-1] == '\n') {
line[--len] = '\0';
}
return line;
}
void print_line(char *line) {
printf("%s \n", line);
}
bool parse_line(char *line, char **arr) {
size_t i = 0;
char *point;
point = strtok(line, " ");
while (point != NULL) {
arr[i] = (char *)malloc(strlen(point) + 1);
strcpy(arr[i], point);
point = strtok(NULL, " ");
i++;
}
arr[i] = NULL;
for (int j=0; j<i; j++) {
printf("%s\n", arr[i]);
}
if (!arr)
return false;
return true;
}
Input:
ls -l -h
arr[i] = NULL;
for (int j=0; j<i; j++) {
printf("%s\n", arr[i]);
}
You are not using j as index so you send NULL to printf(). It's undefined behavior.
I propose you an example of implementation(still not the best but for a beginner that enough):
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdbool.h>
#define LINE_LENGTH 50
char **parse_line(char *line, size_t *n);
int main(void) {
while (true) {
// display prompt on the screen
printf(" > ");
// read input line from terminal
char line[LINE_LENGTH];
if (!fgets(line, sizeof line, stdin)) {
return 1;
}
line[strcspn(line, "\n")] = '\0';
if (strcmp(line, "exit") == 0) {
exit(EXIT_SUCCESS);
} else if (strcmp(line, "clear") == 0) {
printf("\e[2J\e[H");
continue;
}
// parse line into array
size_t n;
char **args = parse_line(line, &n);
if (!args) {
printf("Error during parsing command \n");
continue;
}
for (size_t i = 0; i < n; i++) {
printf("%s\n", args[i]);
}
}
}
char **parse_line(char *line, size_t *n) {
char **arr = malloc(sizeof *arr);
size_t i = 0;
for (char *token = strtok(line, " "); token != NULL; token = strtok(NULL, " ")) {
char **tmp = realloc(arr, sizeof *tmp * (i + 2));
if (tmp == NULL) {
for (size_t j = 0; j < i; j++) {
free(arr[j]);
}
free(arr);
return NULL;
}
arr = tmp;
arr[i] = malloc(strlen(token) + 1);
if (arr[i] == NULL) {
for (size_t j = 0; j < i; j++) {
free(arr[j]);
}
free(arr);
return NULL;
}
strcpy(arr[i], token);
i++;
}
arr[i] = NULL;
*n = i;
return arr;
}

How to shift letters to the right or left?

I am given a file with a string, for example "The United States was founded in *1776*". What I cannot figure out is how to shift letters one space to the left or right and have the letters wrap around. I am able to shift the letters from an a to b but not change its location within the word.
Example of this output would be:
"heT
nitedU
tatesS
asw
oundedf
ni
1776**"
In C, strings are stored as an array of chars in memory. Unlike C++ vectors, you can not insert or remove element within the array, you can only access their value or change their value.
If you declare a C string as follows:
char *myStr = "Fred";
It will be stored in memory as a five character array with the 5th character being the zero value which terminates a C string:
myStr[0] = 'F'
myStr[1] = 'r'
myStr[2] = 'e'
myStr[3] = 'd'
myStr[4] = 0
You need to design a for loop that copies each array element to the one before, while remembering that you need to save the one you are about to overwrite. In this example, it should result in the following copy operations being performed:
len = strlen(myStr);
saveCh = myStr[0];
myStr[0] = myStr[1];
myStr[1] = myStr[2];
myStr[2] = myStr[3];
myStr[3] = saveCh;
So now your job is to create a for loop that does that for any C string of any length.
So to rotate the chars within a C string to the left, you need to copy each char in the array at index i to previous array element i-1. The tricky part is to handle the wrap around properly when i=0 (in this example, you want to copy myStr[0] to myStr[3]. Now do that with a for loop.
You need to also understand that the last character of any C string is the null character (value zero), which terminates a C string. If you modify that element in the array, then your string will break. That is why saveCh is copied to myStr[3] and not to myStr[4].
void rotateStrLeftOneChar(char *myStr) {
// Always check for error and special cases first!
// If myStr is a NULL pointer, do nothing and exit
// If myStr is less than 2 chars, nothing needs to be done too.
if ((myStr != NULL) && (strlen(myStr)>1)) {
int len = strlen(myStr);
char saveCh = myStr[0];
int i = 0;
// Copy each char at index i+1 left to index i in the array
for(i=0; i<len-2; i++)
myStr[i] = myStr[i+1];
// The last character is special and is set to saveCh
myStr[len-1] = saveCh;
}
}
If you just need to output the letters to shift to the left and don't want to change original input then you can do:
#include <stdio.h>
#include <string.h>
void shiftletters(char * input, int i);
int main () {
char input[256];
int shift;
printf("Enter input : ");
scanf("%[^\n]s", input);
printf("Number of shifts : ");
scanf("%d", &shift);
shiftletters(input, shift);
return 0;
}
void shiftletters(char * input, int numshifts)
{
char str[256] = {'\0'};
char * delim = " \t";
char * pch = NULL;
int j, k, len, shifts;
if (input == NULL)
{
printf ("Invalid input\n");
return;
}
strcpy (str, input);
pch = strtok (str, delim);
while (pch != NULL)
{
len = strlen (pch);
if ((numshifts == len) || (len == 1))
{
printf ("%s\n", pch);
pch = strtok (NULL, delim);
continue;
}
if (len < numshifts)
shifts = numshifts % len;
else
shifts = numshifts;
for(j=shifts; j<len; j++)
printf("%c", pch[j]);
for(k=0; k<shifts; k++)
printf("%c", pch[k]);
printf("\n");
pch = strtok (NULL, delim);
}
}
The output of the program:
Enter input : The United States was founded in *1776*
Number of shifts : 1
heT
nitedU
tatesS
asw
oundedf
ni
1776**
like this
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
enum { L = -1, R = 1};
char *rotate(char word[], int dir){
size_t len = strlen(word);
char *temp = malloc(len + 1);
if(!temp){
perror("malloc");
exit(EXIT_FAILURE);
}
strcpy(temp, word);
for(char *src = temp; *src; ++src, ++dir){//or use memmove
word[(dir+len)%len] = *src;
}
free(temp);
return word;
}
int main(int argc, char *argv[]) {
FILE *fp = fopen("data.txt", "r");
if(fp == NULL){
perror("fopen");
exit(EXIT_FAILURE);
}
if(argc < 2){
fprintf(stderr, "Usage %s L|R...\n", argv[0]);
exit(EXIT_FAILURE);
}
char word[64];
while(fscanf(fp, "%63s", word)==1){
for(char *shift = argv[1]; *shift; ++shift){
int dir = *shift == 'L' ? L : R;
rotate(word, dir);
}
printf("%s\n", word);
}
fclose(fp);
}
using memmove version
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
enum { L = -1, R = 1};
char *rotate1(char word[], int dir){
size_t len = strlen(word);
if(len > 2){
char temp;
if(dir == L){
temp = word[0];
memmove(word, word+1, len-1);
word[len-1] = temp;
} else if(dir == R){
temp = word[len-1];
memmove(word+1, word, len-1);
word[0] = temp;
}
}
return word;
}
int main(int argc, char *argv[]) {
FILE *fp = fopen("data.txt", "r");
if(fp == NULL){
perror("fopen");
exit(EXIT_FAILURE);
}
if(argc < 2){
fprintf(stderr, "Usage %s L|R...\n", argv[0]);
exit(EXIT_FAILURE);
}
char word[64];
while(fscanf(fp, "%63s", word)==1){
for(char *shift = argv[1]; *shift; ++shift){
int dir = *shift == 'L' ? L : R;
rotate1(word, dir);
}
printf("%s\n", word);
}
}

How to print only the first sentence in string [C]?

Example:
void stringEvaluation(char *name){
if (strstr(name, "Tall") != NULL)
--here I would like to print only "John Doe"--
}
int main{
char name[160 + 1];
scanf("%[^\n]%*c", name);
stringEvaluation(name);
return 0;
}
and this is the input
"John Doe (45). Black. Tall. Brown eyes"
Your request is unclear. For example, do as follows.
#include <stdio.h>
#include <string.h>
#include <ctype.h>
void stringEvaluation(const char *s){
const char *p;
if(s == NULL || *s == '\0'){
puts("invalid format.");
return ;
}
(p = strchr(s, '(')) || (p = strchr(s, '.'));
if(p == NULL){
puts("invalid format.");
return ;
}
//find end
while(!isalpha((unsigned char)p[-1]))
--p;
while(s < p)
putchar(*s++);
}
int main(void){
char name[160 + 1] = "";
scanf("%160[^\n]%*c", name);
stringEvaluation(name);
return 0;
}
I am using strtok() to get the name. Please note, this will work only if you have "()" in your string.
void stringEvaluation(char *name){
if (strstr(name, "Tall") != NULL)
{
char *nameEnd = strtok(name, "("); //Returns address of '('
if(NULL != nameEnd)
{
int nameLength = nameEnd - name - 1; // 1 for space after name.
char *onlyName = malloc((nameLength + 1) * sizeof(char)); // 1 for terminating '\0'
if(NULL != onlyName)
{
//Copy name part.
strncpy(onlyName, name, nameLength);
onlyName[nameLength] = '\0'; //Make onlyName a string
printf("Name: [%s]\n", onlyName);
free(onlyName);
onlyName = NULL;
}
}
}
}
I have assumed that name will be at first place in your input string. Then following small logic will work for you.
void stringEvaluation(char *name){
char *p;
char OutPut[50]={0};
if (strstr(name, "Tall") != NULL)
{
p = strstr(name," ");
p++;
p = strstr(p," ");
strncpy(OutPut,name,(p-name));
printf("name=%s\n",OutPut);
}
}

String parse in C

Is there a better of parsing the below string instead of doing a strtok() to get each field.
"subject=what&cc=bose#yahoo.com&server=smtp.yahoo.com:8000"
Basically I want to retrieve the value for each field into another char buf's.
Here is my code. Just wanted to know if there is any other better way of doing it (any better string parsing algos)
#include <stdio.h>
#include <stdint.h>
#include <string.h>
#include <stdlib.h>
#define SUBJECT "subject="
#define CC_LIST "cc="
#define SERVER "server="
static void
get_value (const char *tok, char **rval_buf, size_t field_len)
{
size_t val_size = 0;
if (!tok || !rval_buf)
return;
val_size = strlen(tok + field_len) + 1;
*rval_buf = calloc(1, val_size);
if (*rval_buf) {
strlcpy(*rval_buf, tok + field_len, val_size);
}
}
int
main (int argc, char **argv)
{
/* hard coded buf for testing */
char buf[] = "subject=what&cc=bose#yahoo.com&server=smtp.yahoo.com:8000";
char *subject_text = NULL;
char *cc_list = NULL;
char *server_addr = NULL;
char *tok = NULL;
int field_len = 0;
int val_len = 0;
tok = strtok(buf, "&");
while(tok) {
/*
* Handle the token
*/
/* check if it is subject */
if (strstr(tok, SUBJECT)) {
get_value(tok, &subject_text, strlen(SUBJECT));
} else if (strstr(tok, CC_LIST)) { /* check if it is CC */
get_value(tok, &cc_list, strlen(CC_LIST));
} else if (strstr(tok, SERVER)) { /* check if it is server */
get_value(tok, &server_addr, strlen(SERVER));
}
tok = strtok(NULL, "&");
}
/* dump data */
fprintf(stdout, "\nSUBJECT: \"%s\"\nCC_LIST: \"%s\"\nSERVER: \"%s\" \n\n",
subject_text, cc_list, server_addr);
return EXIT_SUCCESS;
}
strstr searches for one string ("the needle") inside another ("the haystack"), but you really only want to know whether the needle is the beginning of the haystack.
Here's a small suggestion: (requires #include <stdbool> or change the booleans to ints. I like bools.)
static bool
getval(const char* haystack, const char** res, const char* needle, size_t len) {
if (haystack && 0 == strncmp(haystack, needle, len)) {
*res = strdup(haystack + len);
return true;
}
return false;
}
and later:
for (tok = strtok(buf, "&"); tok; tok = strtok(NULL, "&")) {
getval(tok, &subject_text, SUBJECT, strlen(SUBJECT)) ||
getval(tok, &cc_list, CC_LIST, strlen(CC_LIST)) ||
getval(tok, &server_addr, SERVER, strlen(SERVER));
}
You can actually get away with doing the strlen inside of getval, which cuts down a lot on the noise, because most modern compilers are clever enough to inline getval and constant-fold the length of a constant string.
Use strtok()
char *strtok(char *str, const char *delim)
You can put '&' as a delimeter
I wrote a quick-n-dirty splitter for you:
int split(char* input, char delim, char*** parts)
{
int count = 1;
char** result;
char* t = input;
while(*t != '\0')
{
if (*t++ == delim)
{
count++;
}
}
result = (char**)malloc(count * sizeof(char*));
t = input;
int i = 0;
result[i] = input;
while(*t != '\0')
{
if (*t == delim)
{
*t = '\0';
result[++i] = ++t;
}
else
{
t++;
}
}
*parts = result;
return count;
}
int main()
{
char raw[] = "subject=\"some text\"&cc=abcd&server=acd.com";
char* str = _strdup(raw);
char** parts;
char** keyval;
int cnt = split(str, '&', &parts);
for(int i=0; i<cnt; ++i)
{
split(parts[i], '=', &keyval);
printf("[%d]: %s <--> %s\n", i, keyval[0], keyval[1]);
free(keyval);
}
free(parts);
getchar();
return 0;
}
Output
[0]: subject <--> "some text"
[1]: cc <--> abcd
[2]: server <--> acd.com

Resources