HW.C
For Daniel to look at
unknown
plain_text
3 years ago
7.4 kB
5
Indexable
#include <stdio.h> // IO functions
#include <stdlib.h> // std lib function
#include <unistd.h> // close function
#include <pthread.h> // creating, running, joining threads
#include <string.h> // string processing
#include "web.h" // our code :)
#define DEBUG 1
// -----------------------------------
// CSCI 340 - Operating Systems
// Fall 2021
// hw3.c file
//
// Homework 3
//
// -----------------------------------
// ---------------------------------------------
// Global variable shared by all threads
// This is not thread safe.
web_t* web_array;
// ---------------------------------------------
// Function prototypes to read and write url files
int read_url_file( char* file_path, web_t* web_array );
int write_url_file( char* file_path, web_t* web_array, int num_urls );
// ---------------------------------------------
// Function prototypes that is ran by a thread
void th_run( int* i );
int main( int argc, char *argv[] ) {
int i = 0;
int write_error = 0;
(void)write_error;
// malloc web_t array with 100 elements
web_array = malloc( sizeof( web_t )*100 );
// -------------------------------------------------
// Verify the correct number of arguments are provided
// when the application is executed.
if ( argc == 2 ) {
int num_urls = read_url_file( argv[1], web_array );
if ( DEBUG ) printf("Number of urls in text file = %d\n", num_urls);
// -----------------------------------------
// TODO:
// ------------------------------------------
// You complete steps 1 to 3 below.
//
// If either the thread create or join fails, you may exit the entire
// program.
//
// Hint: you will loop through an array of threads
pthread_t thread_array[num_urls];
int index[num_urls];
for ( i=0; i<num_urls; i++ ) {
if ( DEBUG ) {
printf("URL=%s\n", web_array[i].url );
printf("CNT=%d\n", web_array[i].link_cnt );
//printf("WEBPAGE=%s\n", web_array[i].webpage );
}
index[i] = i;
int results = pthread_create(&thread_array[i], NULL, (void*)th_run, &(index[i]));
if(results) {
printf("creating thread %d faild\n", i);
perror("");
exit(-1);
}
}
// -------------------------------------------------------------
// STEP 1: Create and run threads
// Hint: Each thread will execute the th_run function
// The array index value is a shared variable
//
// If one thread cannot be created (i.e. an error occurs) you may exit
// the entire program.
}
for(i=0; i< num_urls; i++ ) {
if(pthread_join(thread_array[i], NULL)) {
perror("pthread_join error");
exit(-1);
}
}
// -------------------------------------------------------------
// STEP 2: Join all threads
//
// If one thread cannot be joined (i.e. an error occurs) you may exit
// the entire program.
// -------------------------------------------------------------
// STEP 3: write results to output file
// Note: please do not change this file name (used for testing purposes)
// output file name = "make_merica_great_again.txt"
FILE * f = fopen("make_merica_great_again.txt", "w");
if(!f) {
perror("unable to open output file");
exit(-1);
}
for(i = 0; i< num_urls; i++) {
web_t info = web_array[i];
int writing_error = fprintf(f, "%s\n", info.url);
if(writing_error < 0) {
perror("error while writing url to file");
exit(-1);
}
else {
printf( "URL file is not specified!\n" );
printf( "Example usage: %s %s\n", argv[0], "url.txt" );
}
return OK;
} // end main function
// ------------------------------------
// Function ran by an individual thread
//
//
// Arguments: i = an index value (to access individual web_t structs in the web_array)
//
// Return: nothing (void)
//
void th_run( int* i ) {
// -------------------------------------
// Please do not modify this code
// -------------------------------------
if ( web_array[ (*i) ].link_cnt == OK ) {
parse( &web_array[ (*i) ] );
} else printf( "[%s] failed to retrieve webpage\n", web_array[ (*i) ].url );
pthread_exit( NULL );
} // end th_run function
// ------------------------------------
// Function that writes the identified links
// stored in the web_t links field to formated
// text file.
//
//
//
// Arguments: file_path = fully qualified path to output text file
// web_array = pointer to array of web_t structs
// num_urls = the total number of websites in the url.txt file
//
// Return: OK on success, FAIL on error
//
int write_url_file( char* file_path, web_t* web_array, int num_urls ) {
// -----------------------------------------
// TODO:
// -----------------------------------------
// The format of this file is provided below.
//
// <url>, <link>\n
//
//
// For example,
//
// usatoday.com, /story/news/politics/2018/10/23/donald-trump-looks-ahead-life-after-midterms/1732675002/
// wsj.com, /articles/trump-not-satisfied-with-saudi-arabias-explanation-for-khashoggi-death-1540237930?mod=hp_major_pos8
// .
// .
// .
//
return OK;
} // end write_url_file function
// ------------------------------------
// Function that reads the url in the input
// url text file.
//
//
// Arguments: file_path = fully qualified path to output text file
// web_array = pointer to array of web_t structs
//
// Return: Total number of urls defined in the url text file
//
int read_url_file( char* file_path, web_t* web_array ) {
// -------------------------------------
// Please do not modify this code
// -------------------------------------
int url_cnt = 0;
FILE* fhnd;
fhnd = fopen( file_path, "r" );
char line[50];
if ( fhnd != NULL ) {
while ( fgets( line, sizeof( line ), fhnd ) != NULL ) {
line[strlen(line)-1] = '\0';
web_array[url_cnt].url = malloc(sizeof(char)*100);
sprintf( web_array[url_cnt].url, "www.%s", line );
web_array[url_cnt].link_cnt = OK;
if ( FAIL == get_webpage( &web_array[url_cnt++] ) ) {
printf("Failed to load webpage for URL( %s )\n", web_array[url_cnt--].url );
web_array[url_cnt].link_cnt = FAIL;
}
}
} else url_cnt = FAIL;
fclose( fhnd );
return url_cnt;
} // end read_url_file function
Editor is loading...