1879 lines
72 KiB
C
Executable file
1879 lines
72 KiB
C
Executable file
//Wiby Web Crawler
|
|
|
|
#include </usr/include/mysql/mysql.h>
|
|
#include <stdlib.h>
|
|
#include <stdio.h>
|
|
#include <strings.h>
|
|
//#include </usr/include/curl/curl.h> //RHEL/Rocky
|
|
//#include </usr/include/curl/easy.h> //RHEL/Rocky
|
|
#include </usr/include/x86_64-linux-gnu/curl/curl.h> //ubuntu 20/22
|
|
#include </usr/include/x86_64-linux-gnu/curl/easy.h> //ubuntu 20/22
|
|
#include "htmlparse.h"
|
|
#include "urlparse.h"
|
|
#include "checkrobots.h"
|
|
#include <unistd.h>
|
|
|
|
#define url_fromlist_arraylen 102400
|
|
#define url_insert_arraylen 1024000
|
|
|
|
char /**title, *keywords, *description, *page,*/ *windexinsert, *windexupdate, *windexRandUpdate, *titlecheckinsert, /**shardinsert,*/ correctedURL[1001], urlPath_finalURL[1001], folderPath_finalURL[1001], urlPrefix_finalURL[1001], urlNPNP_finalURL[1001], strDepth[101], url_fromlist[url_fromlist_arraylen], url_insert[url_insert_arraylen], previousfail[5][1001];
|
|
|
|
FILE *shardfile;
|
|
char *shardfilestr;
|
|
|
|
void finish_with_error(MYSQL *con)
|
|
{
|
|
fprintf(stderr, "%s\n", mysql_error(con));
|
|
mysql_close(con);
|
|
exit(1);
|
|
}
|
|
int isnum(char *source){
|
|
int sourcelength = strlen(source);
|
|
for(int i=0;i < sourcelength; i++){
|
|
if(source[i] < 48 || source[i] > 57){
|
|
return 0;
|
|
}
|
|
}
|
|
return 1;
|
|
}
|
|
size_t write_data(void *ptr, size_t size, size_t nmemb, FILE *stream) {
|
|
size_t written = fwrite(ptr, size, nmemb, stream);
|
|
return written;
|
|
}
|
|
|
|
int main(int argc, char **argv)
|
|
{
|
|
int id_assigned=0;
|
|
if(argc == 2 && isnum(argv[1])==1){
|
|
if(argv[1][0] == 48){
|
|
printf("When assigning ID's, you must start at 1. Cannot set an id of 0.\n");
|
|
exit(0);
|
|
}
|
|
id_assigned=1;
|
|
}else if(argc >= 2){
|
|
printf("\nWiby Web Crawler\n\nUsage: cr Crawler_ID\n\nThe indexqueue may have each page assigned a crawler ID. The ID is assigned when you specify to the Refresh Scheduler the total number of crawlers you are running, and when you update the variable '$num_crawlers' from inside of review.php and graveyard.php (line 73) to the number of crawlers you are using. The scheduler will assign pages in round-robin order a crawler ID within the range of that total.\n\nExample: If you want two crawlers running, then you should specify the first with an ID of 1, and the second with and ID of 2. Run them in separate folders, and provide a symlinks to the 'robots' folder and 'shards' file in each. Each crawler will crawl pages in the indexqueue with its corresponding ID.\n\nYou can also not assign an ID, and in that case the crawler will ignore the ID assignments. So if you have only one crawler running, assigning an ID is optional. Don't run multiple crawlers without assigning ID's.\n\nSpecify the total number of shard tables you wish to use in the 'shards' file. The crawler will round-robin insert/update rows in these tables (ws0 to wsX) along with the main 'windex' table. The default is 0.\n\n");
|
|
exit(0);
|
|
}
|
|
|
|
long int previousID[5] = {0, 1, 2, 3, 4};
|
|
int sanity = 1;
|
|
|
|
for(int i=0;i<5;i++){
|
|
previousfail[i][0]=0;
|
|
}
|
|
|
|
//check if there are shards to include
|
|
int nShards=0,fsize=0,shardnum=0;
|
|
char numshards[20], shardnumstr[20];
|
|
memset(numshards,0,20);
|
|
memset(shardnumstr,0,20);
|
|
sprintf(shardnumstr,"0");
|
|
if(shardfile = fopen("shards", "r")){
|
|
fseek(shardfile, 0, SEEK_END);
|
|
fsize = ftell(shardfile);
|
|
fseek(shardfile, 0, SEEK_SET);
|
|
if(fsize > 0 && fsize < 11){
|
|
shardfilestr = malloc(fsize + 1);
|
|
if(fread(shardfilestr, 1, fsize, shardfile)){}
|
|
shardfilestr[fsize] = 0;
|
|
for(int i=0;i<fsize;i++){
|
|
if(shardfilestr[i] > 47 && shardfilestr[i] < 58){
|
|
numshards[i]=shardfilestr[i];
|
|
}
|
|
}
|
|
//check if its a number
|
|
if(isnum(numshards)==1){
|
|
nShards = atoi(numshards);
|
|
}else{
|
|
printf("\nThe shard file must contain a number. Indicate the number of available shards you are using or set it to 0 if you aren't.\n\n");
|
|
exit(0);
|
|
}
|
|
free(shardfilestr);
|
|
}
|
|
if(fsize>10){
|
|
printf("\nTotal number of shards is too large (10 billion???).");
|
|
}
|
|
fclose(shardfile);
|
|
}else{
|
|
printf("\nWarning: 'shards' file is missing. Create the file and indicate the number of available shards you are using or set it to 0 if you aren't.\n\n");
|
|
}
|
|
if(nShards > 0){
|
|
srand(time(NULL));
|
|
shardnum = (rand() % nShards);
|
|
memset(shardnumstr,0,20);
|
|
sprintf(shardnumstr,"%d",shardnum);
|
|
}
|
|
|
|
while(1)
|
|
{
|
|
//printf("MySQL client version: %s\n", mysql_get_client_info());
|
|
int alreadydone = 0, permitted=1;
|
|
//allocates or initialises a MYSQL object
|
|
|
|
MYSQL *con = mysql_init(NULL);
|
|
|
|
if (con == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//establish a connection to the database. We provide connection handler, host name, user name and password parameters to the function. The other four parameters are the database name, port number, unix socket and finally the client flag
|
|
if (mysql_real_connect(con, "localhost", "crawler", "seekout", "wiby", 0, NULL, 0) == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
if (mysql_query(con, "SET CHARSET utf8;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
if(id_assigned == 0){
|
|
if (mysql_query(con, "SELECT id, url, worksafe, approver, surprise, updatable, task, crawl_tree, crawl_family, crawl_depth, crawl_pages, crawl_type, crawl_repeat, force_rules FROM indexqueue limit 1;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}else{
|
|
char indexqueuequery[2001];
|
|
memset(indexqueuequery,0,2001);
|
|
strcpy(indexqueuequery,"SELECT id, url, worksafe, approver, surprise, updatable, task, crawl_tree, crawl_family, crawl_depth, crawl_pages, crawl_type, crawl_repeat, force_rules FROM indexqueue WHERE crawler_id = '");
|
|
strcat(indexqueuequery,argv[1]);
|
|
strcat(indexqueuequery,"' LIMIT 1;");
|
|
if (mysql_query(con, indexqueuequery))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
|
|
//We get the result set using the mysql_store_result() function. The MYSQL_RES is a structure for holding a result set
|
|
MYSQL_RES *result = mysql_store_result(con);
|
|
|
|
if(result == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//get the number of fields (columns) in the table
|
|
//int num_fields = mysql_num_fields(result);
|
|
|
|
//We fetch the rows and print them to the screen.
|
|
/*MYSQL_ROW row;
|
|
while (row = mysql_fetch_row(result))
|
|
{
|
|
for(int i=0; i<num_fields; i++)
|
|
{
|
|
printf("%s ", row[i] ? row[i] : "NULL");
|
|
}
|
|
printf("\n");
|
|
}*/
|
|
|
|
MYSQL_ROW row = mysql_fetch_row(result);
|
|
|
|
int empty=0;
|
|
if(row == NULL){
|
|
//printf("\nQueue is empty\n");
|
|
empty=1;
|
|
}else{
|
|
//convert shardnum to string
|
|
if(nShards > 0){
|
|
sprintf(shardnumstr,"%d",shardnum);
|
|
//itoa(shardnum,shardnumstr,10);
|
|
}
|
|
if(id_assigned == 0){
|
|
printf("-----------------------------------------------------------------------------------\nFetching:");
|
|
}else{
|
|
printf("-----------------------------------------------------------------------------------\ncr%s Fetching:",argv[1]);
|
|
}
|
|
//grab the first entry (fifo)
|
|
/*for(int i=0; i<num_fields; i++)
|
|
{
|
|
printf("%s ", row[i] ? row[i] : "NULL");
|
|
}*/
|
|
|
|
//Store data in first row into variables
|
|
char *url = row[1];
|
|
char *id = row[0];
|
|
char *worksafe = row[2];
|
|
char *approver = row[3];
|
|
char *surprise = row[4];
|
|
char *updatable = row[5];
|
|
char *task = row[6];
|
|
|
|
char *crawl_tree = row[7];
|
|
char *crawl_family = row[8];
|
|
char *crawl_depth = row[9];
|
|
char *crawl_pages = row[10];
|
|
char *crawl_type = row[11];
|
|
char *crawl_repeat = row[12];
|
|
char *force_rules = row[13];
|
|
|
|
//convert crawl depth, pages to int
|
|
int n_crawl_depth=0, n_crawl_pages=0;
|
|
if(crawl_depth!=0){
|
|
n_crawl_depth = atoi(crawl_depth);
|
|
}
|
|
if(crawl_pages!=0){
|
|
n_crawl_pages = atoi(crawl_pages);
|
|
}
|
|
|
|
printf("\nURL: %s\nID: %s | Worksafe: %s | Surprise: %s | Approver: %s | Updatable: %s | Task: %s\n", url, id, worksafe, surprise, approver, updatable, task);
|
|
printf("Tree: %s | Family: %s | Depth: %s | Pages: %s | Type: %s | Repeat: %s | Rules: %s\n",crawl_tree,crawl_family,crawl_depth,crawl_pages,crawl_type,crawl_repeat,force_rules);
|
|
//===================check if url already indexed, ====================
|
|
|
|
//find out if its http or https or http://www. or https://www.
|
|
int httpwww=0, httpswww=0, http=0, https=0;
|
|
char prefix[14];
|
|
memset(prefix,0,14);
|
|
strcpy(prefix,"http");
|
|
int urlsize = strlen(url);
|
|
|
|
if(urlsize > 4){
|
|
if(url[4]==':' && (url[3]=='p' || url[3]=='P'))
|
|
http = 7;
|
|
}
|
|
if(urlsize > 5){
|
|
if(url[5]==':' && (url[4]=='s' || url[4]=='S'))
|
|
https = 8;
|
|
}
|
|
if(urlsize > 11){
|
|
if((url[7]=='w' || url[7]=='W') && (url[8]=='w' || url[8]=='W') && ((url[9]=='w' || url[9]=='W') || url[9]=='1' || url[9]=='2' || url[9]=='3') && url[10]=='.' ){
|
|
httpwww = 11;
|
|
http = https = 0;
|
|
}
|
|
if(url[7]=='/' && (url[8]=='w' || url[8]=='W') && (url[9]=='w' || url[9]=='W') && ((url[9]=='w' || url[9]=='W') || url[9]=='1' || url[9]=='2' || url[9]=='3') && url[11]=='.' ){
|
|
httpswww = 12;
|
|
http = https = 0;
|
|
}
|
|
}
|
|
|
|
//set the prefix
|
|
|
|
if(http > 0) strcat(prefix,"://");
|
|
else if(https > 0) strcat(prefix,"s://");
|
|
else if(httpwww > 0) strcat(prefix,"://www.");
|
|
else if(httpswww > 0) strcat(prefix,"s://www.");
|
|
|
|
int prefixsize = httpswww+httpwww+https+http;
|
|
char urlnoprefix[urlsize-prefixsize+1];
|
|
char urlnopathnoprefix[urlsize-prefixsize+1];
|
|
memset(urlnoprefix,0,urlsize-prefixsize+2);
|
|
memset(urlnopathnoprefix,0,urlsize-prefixsize+2);
|
|
int urlcount=0,urlnoprefixcount=0,urlnopathnoprefix_done=0;
|
|
|
|
//store the url without prefix to urlnoprefix
|
|
while(urlcount < urlsize+1)
|
|
{
|
|
if(urlcount>prefixsize-1)
|
|
{
|
|
urlnoprefix[urlnoprefixcount]=url[urlcount];
|
|
//get urlnopath
|
|
if(url[urlcount] != '/' && urlnopathnoprefix_done==0){
|
|
urlnopathnoprefix[urlnoprefixcount]=url[urlcount];
|
|
}else{
|
|
urlnopathnoprefix_done=1;
|
|
}
|
|
urlnoprefixcount++;
|
|
}
|
|
urlcount++;
|
|
}
|
|
|
|
//check for '/' at end of url. it may be already indexed without that so we need to account for it.
|
|
//int urlnoprefixlength = strlen(urlnoprefix);
|
|
int slashfound = 0;
|
|
char urlnoprefixnoslash[urlnoprefixcount];
|
|
memset(urlnoprefixnoslash,0,urlnoprefixcount);
|
|
if(urlnoprefix[urlnoprefixcount-1] == '/')
|
|
{
|
|
strncpy(urlnoprefixnoslash,urlnoprefix,urlnoprefixcount-1);
|
|
slashfound = 1;
|
|
}
|
|
//printf("\nurlnoprefix: %s\n",urlnoprefix);
|
|
|
|
printf("Checking if page already exists in index... ");
|
|
int idexistsalready = 0, checkurlsize = urlnoprefixcount*24+1000;
|
|
char *idexistsvalue;
|
|
char checkurl[checkurlsize];
|
|
memset(checkurl,0,checkurlsize);
|
|
if(task == 0 || task[0] == '2'){//index request did not come from refresh scheduler, or is an autocrawl url
|
|
//strcpy(checkurl,"SELECT id,updatable,title,enable,fault,url FROM windex WHERE url = 'http://"); //replace this with a simple check for url_noprefix column match
|
|
strcpy(checkurl,"SELECT id,updatable,title,enable,fault,url,url_noprefix,shard FROM windex WHERE url_noprefix = '");
|
|
if(slashfound==0)
|
|
{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/");
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/index.html");
|
|
strcat(checkurl,"' OR url_noprefix = '/index.htm");
|
|
strcat(checkurl,"';");
|
|
}
|
|
else
|
|
{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefixnoslash);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.html");
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.htm");
|
|
strcat(checkurl,"';");
|
|
}
|
|
}else{
|
|
strcpy(checkurl,"SELECT id,updatable,title,enable,fault,url,url_noprefix,shard FROM windex WHERE url = '");
|
|
strcat(checkurl,url);
|
|
strcat(checkurl,"';");
|
|
}
|
|
|
|
if (mysql_query(con, checkurl))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//We get the result set using the mysql_store_result() function. The MYSQL_RES is a structure for holding a result set
|
|
MYSQL_RES *resulturlcheck = mysql_store_result(con);
|
|
|
|
if(resulturlcheck == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//grab the first entry (fifo)
|
|
printf("Found ID ");
|
|
row = mysql_fetch_row(resulturlcheck);
|
|
char updatedefault[] = "1";
|
|
char *updatableOldDBval = updatedefault;
|
|
char *enableOldDBval = updatedefault;
|
|
char *dbtitle;
|
|
char *fault;
|
|
char *dburl;
|
|
char *dburl_noprefix;
|
|
char *shard;
|
|
|
|
//Catalog the previous crawl attempts (to see if they are all for the same page - which would be a bad sign)
|
|
previousID[4] = previousID[3];
|
|
previousID[3] = previousID[2];
|
|
previousID[2] = previousID[1];
|
|
previousID[1] = previousID[0];
|
|
|
|
if(row == NULL)
|
|
{
|
|
printf("null");
|
|
previousID[0] = -1;
|
|
}else {
|
|
printf("%s",row[0]);
|
|
idexistsalready = 1;
|
|
idexistsvalue = row[0];
|
|
previousID[0] = atoi(row[0]);
|
|
updatableOldDBval = row[1];
|
|
dbtitle = row[2];
|
|
enableOldDBval = row[3];
|
|
fault = row[4];
|
|
dburl=row[5];
|
|
dburl_noprefix=row[6];
|
|
shard=row[7];
|
|
if(task != 0 && task[0]=='2')
|
|
alreadydone=1;
|
|
}
|
|
|
|
//Log duplicate rows (they shouldn't exist)
|
|
int num_rows = mysql_num_rows(resulturlcheck);
|
|
if(num_rows > 1){
|
|
FILE *duplicates = fopen("duplicates.txt", "a");
|
|
fputs (dburl,duplicates);
|
|
fputs ("\r\n",duplicates);
|
|
fclose(duplicates);
|
|
}
|
|
|
|
//check robots.txt file for this domain
|
|
urlparse(url);
|
|
permitted = checkrobots(prefix,rootdomain,urlPath);
|
|
|
|
//Does this crawl attempt, along with the last 4 have the same ID? There is possibly a duplicate db entry, or some other problem.
|
|
if(previousID[0] != -1 && alreadydone==0){
|
|
if(previousID[0] == previousID[4] && previousID[0] == previousID[3] && previousID[0] == previousID[2] && previousID[0] == previousID[1]){
|
|
sanity = 0;
|
|
printf("\nWARNING: Last 5 crawl attempts are all for the same page. Will not continue crawling in this situation. Is the same page being submitted over and over? Also, duplicate table entries of the same URL in windex can cause this behavior. Check the database, and duplicates.txt");
|
|
exit(0);
|
|
}else{
|
|
sanity = 1;
|
|
}
|
|
|
|
}else{
|
|
sanity = 1;
|
|
}
|
|
|
|
int failedcrawl=0;
|
|
if(task != 0 && task[0]=='2' && alreadydone==0 && permitted==1){
|
|
//see if url failed to crawl last time (when link crawling)
|
|
//as it might come up multiple times during crawl of website, should avoid recrawling it
|
|
//will also check the database if this check passes
|
|
for(int i=0;i<5;i++){
|
|
if(strcasecmp(previousfail[i], urlnoprefix)==0){
|
|
sanity=0;
|
|
failedcrawl=1;
|
|
break;
|
|
}
|
|
}
|
|
if(sanity==1)
|
|
sleep(1);//do link crawling slowly
|
|
}
|
|
|
|
//if crawling through hyperlinks, doublecheck that this hyperlink hasn't been crawled recently, even if it was redirected elsewhere or failed
|
|
int alreadylogged = 0;
|
|
if(failedcrawl==0 && task !=0 && task[0]=='2' && alreadydone == 0){
|
|
if (mysql_query(con, "use wibytemp"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
memset(checkurl,0,checkurlsize);
|
|
strcpy(checkurl,"SELECT id FROM crawled WHERE url_noprefix = '");
|
|
if(slashfound==0)
|
|
{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/");
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/index.html");
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '/index.htm");
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR");
|
|
}else{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefixnoslash);
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.html");
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.htm");
|
|
strcat(checkurl,"' AND time > NOW() - INTERVAL 12 HOUR");
|
|
}
|
|
//query db
|
|
if (mysql_query(con, checkurl))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
MYSQL_RES *resultcrawledurlcheck = mysql_store_result(con);
|
|
if(resultcrawledurlcheck == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//grab the first entry (fifo)
|
|
MYSQL_ROW rowCrawledURLCheck = mysql_fetch_row(resultcrawledurlcheck);
|
|
if(rowCrawledURLCheck != NULL)
|
|
{
|
|
sanity=0;
|
|
alreadylogged = 1;
|
|
printf("\nThis hyperlink was crawled recently. It cannot be crawled again for at least 12 hours.");
|
|
}
|
|
mysql_free_result(resultcrawledurlcheck);
|
|
if (mysql_query(con, "use wiby"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
|
|
//printf("\n\n%ld, %ld, %ld, %ld, %ld\n",previousID[0],previousID[1],previousID[2],previousID[3],previousID[4]);
|
|
|
|
//see if the server will accept http only connections on older browsers, change url to HTTP only:
|
|
char urlHTTP[strlen(url)+100];
|
|
memset(urlHTTP,0,strlen(url)+100);
|
|
strcpy(urlHTTP,"http");
|
|
if(http > 0 || https > 0){
|
|
strcat(urlHTTP,"://");
|
|
}else if(httpwww > 0 || httpswww > 0){
|
|
strcat(urlHTTP,"://www.");
|
|
}
|
|
strcat(urlHTTP,urlnoprefix);
|
|
|
|
if(updatableOldDBval[0] != '0' && enableOldDBval[0] != '0' && sanity == 1 && alreadydone==0 && permitted==1)
|
|
{
|
|
printf("\nAttempt HTTP connection: %s",urlHTTP);
|
|
printf("\nDownloading page... ");
|
|
//===============do the curl (download the webpage)=====================
|
|
curl_global_init(CURL_GLOBAL_DEFAULT);
|
|
CURL *curl;
|
|
FILE *fp;
|
|
CURLcode res;
|
|
char outfilename[FILENAME_MAX] = "page.out";
|
|
curl = curl_easy_init();
|
|
long size=0;
|
|
char *finalURL = NULL;
|
|
long response_code;
|
|
int finalURLsize = 0,skipurl=0;
|
|
if (curl) {
|
|
fp = fopen(outfilename,"wb");
|
|
//Get file size
|
|
//fseek(fp, 0L, SEEK_END);
|
|
//size = ftell(fp);
|
|
//set curl options
|
|
curl_easy_setopt(curl, CURLOPT_URL, urlHTTP);// set URL to get here
|
|
curl_easy_setopt(curl, CURLOPT_USERAGENT, "Mozilla/4.0 (compatible; WebCrawler; SearchEngine)");
|
|
curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_data);// send all data to this function //
|
|
curl_easy_setopt(curl, CURLOPT_WRITEDATA, fp);// write the page body to this file handle
|
|
curl_easy_setopt(curl,CURLOPT_FOLLOWLOCATION,1L);//allow redirects
|
|
curl_easy_setopt(curl, CURLOPT_TIMEOUT, 60L);
|
|
curl_easy_setopt(curl, CURLOPT_CONNECTTIMEOUT, 20L);
|
|
curl_easy_setopt(curl, CURLOPT_MAXREDIRS, 5L);//max num of redirects
|
|
curl_easy_setopt(curl, CURLOPT_MAXFILESIZE, 5000000L);//don't download if over 5MB
|
|
curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, 0L);//0 or 1 to verify ssl
|
|
//curl_easy_setopt(curl, CURLOPT_VERBOSE, 1L);//set verbose
|
|
res = curl_easy_perform(curl);// get it!
|
|
//if(res == CURLE_OK) {//get final redirect url //-- don't check for this, causes segfault if "transfer closed with outstanding read data remaining"
|
|
curl_easy_getinfo(curl, CURLINFO_EFFECTIVE_URL, &finalURL);
|
|
curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &response_code);
|
|
|
|
if(finalURL){
|
|
printf("Effective URL: %s\nResponse: %ld, ", finalURL,response_code);
|
|
finalURLsize = strlen(finalURL);
|
|
}
|
|
|
|
//curl_easy_cleanup(curl); //cleanup moved further down because finalURL is needed at insert
|
|
|
|
//get file size
|
|
fseek(fp, 0L, SEEK_END);
|
|
size = ftell(fp);
|
|
|
|
fclose(fp);
|
|
}
|
|
|
|
//if effective URL contains ':443', CURL will fail to download this page on next update. Remove :443 from finalURL.
|
|
char *ptr_substring = NULL;
|
|
int substringpos=0;
|
|
if(finalURLsize > 3)
|
|
ptr_substring = strstr(finalURL,":443");
|
|
if(ptr_substring != NULL && skipurl == 0 && finalURLsize<=500){
|
|
substringpos = ptr_substring - finalURL;
|
|
int poscount = substringpos;
|
|
memcpy(correctedURL,finalURL,substringpos);//copy before substring
|
|
while(1){//copy after substring
|
|
correctedURL[poscount] = finalURL[poscount+4];
|
|
if(finalURL[poscount+4] == 0)
|
|
break;
|
|
poscount++;
|
|
}
|
|
finalURL = correctedURL;
|
|
finalURLsize = strlen(finalURL);
|
|
printf("\nSetting final URL as: %s\n", finalURL);
|
|
}
|
|
|
|
int finalURLcount=0;
|
|
while(finalURL[finalURLcount]!=0){
|
|
if(finalURL[finalURLcount]=='\''){
|
|
skipurl=1;
|
|
printf("\nURL contains single-quote. Skipping.");
|
|
}
|
|
finalURLcount++;
|
|
}
|
|
|
|
//when crawling through hyperlinks, log that the url was accessed, use the original url, not finalURL
|
|
if(skipurl==0 && task != 0 && task[0]=='2'){
|
|
if (mysql_query(con, "use wibytemp"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
char sqlquerylogurl[2000];
|
|
memset(sqlquerylogurl,0,2000);
|
|
strcpy(sqlquerylogurl,"INSERT INTO crawled (url_noprefix) VALUES('");
|
|
strcat(sqlquerylogurl,urlnoprefix);
|
|
strcat(sqlquerylogurl,"')");
|
|
if (mysql_query(con, sqlquerylogurl))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if (mysql_query(con, "use wiby"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
|
|
if(finalURLsize>500){
|
|
skipurl=1;
|
|
printf("\nURL is too long");
|
|
}
|
|
|
|
if(task != 0 && task[0]=='2' && canCrawl(finalURLsize,finalURL)==0){
|
|
printf("\nEffective URL failed crawl rules.");
|
|
skipurl=1;
|
|
}
|
|
|
|
char finalURLnoprefix[finalURLsize-prefixsize+100];
|
|
char httpAllow[] = "0";
|
|
memset(finalURLnoprefix,0,finalURLsize-prefixsize+100);
|
|
int updatereserve=0;
|
|
char idReserve[200];
|
|
|
|
if(skipurl==0){
|
|
//see if server permitted an http connection
|
|
if(finalURL != NULL){
|
|
if(finalURL[4]==':')
|
|
httpAllow[0] = '1';
|
|
}
|
|
else if(http > 0 || httpwww > 0){
|
|
httpAllow[0] = '1';
|
|
}
|
|
|
|
//Remove the prefix from the final URL, to store into url_noprefix
|
|
//find out if its http or https or http://www. or https://www.
|
|
httpwww=httpswww=http=https=0;
|
|
|
|
if(finalURLsize > 4){
|
|
if(finalURL[4]==':')
|
|
http = 7;
|
|
if(finalURL[4]=='s' || finalURL[4]=='S')
|
|
https = 8;
|
|
}
|
|
if(finalURLsize > 11){
|
|
if((finalURL[7]=='w' || finalURL[7]=='W') && (finalURL[8]=='w' || finalURL[8]=='W') && (finalURL[9]=='w' || finalURL[9]=='W') && finalURL[10]=='.' ){
|
|
httpwww = 11;
|
|
http = https = 0;
|
|
}
|
|
if(finalURL[7]=='/' && (finalURL[8]=='w' || finalURL[8]=='W') && (finalURL[9]=='w' || finalURL[9]=='W') && (finalURL[10]=='w' || finalURL[10]=='W') && finalURL[11]=='.' ){
|
|
httpswww = 12;
|
|
http = https = 0;
|
|
}
|
|
}
|
|
|
|
int finalURL_prefixsize = httpswww+httpwww+https+http, finalurlnoprefixcount = 0;
|
|
urlcount=0;
|
|
|
|
//store the final url without prefix to finalURLnoprefix
|
|
while(finalURL[urlcount] != 0){
|
|
if(urlcount>finalURL_prefixsize-1)
|
|
{
|
|
finalURLnoprefix[finalurlnoprefixcount]=finalURL[urlcount];
|
|
finalurlnoprefixcount++;
|
|
}
|
|
urlcount++;
|
|
}
|
|
|
|
//Double check that the URL is in fact not in the DB, by also searching for the effective URL from libcurl and its url in the table
|
|
int foundindoublecheck=0;
|
|
if(idexistsalready == 0){
|
|
mysql_free_result(resulturlcheck);
|
|
char doublecheckurl[finalURLsize+100];
|
|
memset(doublecheckurl,0,finalURLsize+100);
|
|
strcpy(doublecheckurl,"SELECT id,updatable,title,enable,fault,url,url_noprefix,shard FROM windex WHERE url = '");
|
|
strcat(doublecheckurl,finalURL);
|
|
strcat(doublecheckurl,"';");
|
|
if (mysql_query(con, doublecheckurl))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
resulturlcheck = mysql_store_result(con);
|
|
if(resulturlcheck == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
row = mysql_fetch_row(resulturlcheck);
|
|
if(row != NULL)
|
|
{
|
|
printf("\nDoublechecked effective URL in windex, found ID %s\n",row[0]);
|
|
idexistsalready = 1;
|
|
idexistsvalue = row[0];
|
|
previousID[0] = atoi(row[0]);
|
|
updatableOldDBval = row[1];
|
|
dbtitle = row[2];
|
|
enableOldDBval = row[3];
|
|
fault = row[4];
|
|
dburl=row[5];
|
|
dburl_noprefix=row[6];
|
|
shard=row[7];
|
|
if((task != 0 && task[0]=='2') || updatableOldDBval[0] == '0')
|
|
alreadydone=1;
|
|
foundindoublecheck=1;
|
|
}
|
|
//Log duplicate rows (they shouldn't exist)
|
|
num_rows = mysql_num_rows(resulturlcheck);
|
|
if(num_rows > 1){
|
|
FILE *duplicates = fopen("duplicates.txt", "a");
|
|
fputs (dburl,duplicates);
|
|
fputs ("\r\n",duplicates);
|
|
fclose(duplicates);
|
|
}
|
|
//Does this crawl attempt, along with the last 4 have the same ID? There is possibly a duplicate db entry, or some other problem.
|
|
if(previousID[0] != -1){
|
|
if(previousID[0] == previousID[4] && previousID[0] == previousID[3] && previousID[0] == previousID[2] && previousID[0] == previousID[1]){
|
|
printf("\nWARNING: Last 5 crawl attempts are all for the same page. Will not continue crawling in this situation. Is the same page being submitted over and over? Also, duplicate table entries of the same URL in windex can cause this behavior. Check the database, and duplicates.txt");
|
|
exit(0);
|
|
}
|
|
}
|
|
}
|
|
|
|
//if doing an update when using multiple crawlers, reserve the id and verify the URL is still associated with it
|
|
if(alreadydone==0 && id_assigned==1 && idexistsalready==1){
|
|
if (mysql_query(con, "use wibytemp;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
memset(idReserve,0,200);
|
|
strcpy(idReserve,"INSERT into reserve_id (id, crawler_id) VALUES (");
|
|
strcat(idReserve,idexistsvalue);
|
|
strcat(idReserve,",");
|
|
strcat(idReserve,argv[1]);
|
|
strcat(idReserve,");");
|
|
if(mysql_query(con, idReserve))
|
|
{
|
|
printf("\nID is already reserved, will try again. Clearing old reservations...");
|
|
memset(idReserve,0,200);
|
|
strcpy(idReserve,"DELETE FROM reserve_id WHERE time < NOW() - INTERVAL 10 MINUTE OR crawler_id = ");
|
|
strcat(idReserve,argv[1]);
|
|
if(mysql_query(con, idReserve)){
|
|
finish_with_error(con);
|
|
}else{
|
|
printf(" Done.");
|
|
}
|
|
alreadydone=1;
|
|
}
|
|
//back to wiby database
|
|
if (mysql_query(con, "use wiby;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
updatereserve=1;
|
|
|
|
//check that the url being updated is still assigned to that ID
|
|
memset(checkurl,0,checkurlsize);
|
|
if(task != 0 && task[0] == '1'){
|
|
strcpy(checkurl,"SELECT id FROM windex WHERE url = '");
|
|
strcat(checkurl,url);
|
|
strcat(checkurl,"';");
|
|
}else{
|
|
if(foundindoublecheck==0){
|
|
strcpy(checkurl,"SELECT id FROM windex WHERE url_noprefix = '");
|
|
if(slashfound==0)
|
|
{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/");
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"/index.html");
|
|
strcat(checkurl,"' OR url_noprefix = '/index.htm");
|
|
strcat(checkurl,"';");
|
|
}else{
|
|
strcat(checkurl,urlnoprefix);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefixnoslash);
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.html");
|
|
strcat(checkurl,"' OR url_noprefix = '");
|
|
strcat(checkurl,urlnoprefix);strcat(checkurl,"index.htm");
|
|
strcat(checkurl,"';");
|
|
}
|
|
}else{
|
|
strcpy(checkurl,"SELECT id FROM windex WHERE url = '");
|
|
strcat(checkurl,finalURL);
|
|
strcat(checkurl,"';");
|
|
}
|
|
}
|
|
//query db
|
|
if (mysql_query(con, checkurl))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
MYSQL_RES *resulturlreservecheck = mysql_store_result(con);
|
|
if(resulturlcheck == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//grab the first entry (fifo)
|
|
char *URLcheckID;
|
|
MYSQL_ROW rowURLCheck = mysql_fetch_row(resulturlreservecheck);
|
|
if(rowURLCheck != NULL)
|
|
{
|
|
URLcheckID = rowURLCheck[0];
|
|
}
|
|
if(URLcheckID != 0 && atoi(URLcheckID) != atoi(idexistsvalue)){
|
|
printf("\nID was already reserved, will try again later.");
|
|
alreadydone=1;
|
|
}
|
|
mysql_free_result(resulturlreservecheck);
|
|
}
|
|
}
|
|
//=====================Extract text from HTML file=======================
|
|
if(size < 5000000 && skipurl==0 && alreadydone==0)
|
|
{
|
|
//switch on/off hyperlink collecting (if crawling through hyperlinks, or from regular refresh while crawl_repeat is on, or during manual submission when appropriate limits are set)
|
|
if((task != 0 && task[0]=='2' && (n_crawl_depth > 0 || n_crawl_depth < 0) && (n_crawl_pages > 0 || n_crawl_pages < 0)) || (task==0 && (n_crawl_depth > 0 || n_crawl_depth < 0) && (n_crawl_pages > 0 || n_crawl_pages < 0)) || (task != 0 && task[0]=='1' && crawl_repeat != 0 && crawl_repeat[0]=='1' && (n_crawl_pages > 0 || n_crawl_pages < 0))){
|
|
getURLs=1;
|
|
}else{
|
|
getURLs=0;
|
|
}
|
|
|
|
htmlparse();
|
|
|
|
//need the finalURL path info also
|
|
urlparse(finalURL);
|
|
memset(urlPath_finalURL,0,1001);
|
|
strcpy(urlPath_finalURL,urlPath);
|
|
memset(folderPath_finalURL,0,1001);
|
|
strcpy(folderPath_finalURL,folderPath);
|
|
memset(urlPrefix_finalURL,0,1001);
|
|
strcpy(urlPrefix_finalURL,prefix_fromlist);
|
|
memset(urlNPNP_finalURL,0,1001);
|
|
strcpy(urlNPNP_finalURL,urlnopathnoprefix_fromlist);
|
|
|
|
if(urlPrefix_finalURL[0]==0 || urlNPNP_finalURL[0]==0 || urlPath_finalURL[0]==0)
|
|
noindex = 1;
|
|
|
|
}else{
|
|
noindex = 1;
|
|
}
|
|
|
|
//check if rules are enforced (only for pages that are autocrawled)
|
|
if(force_rules != 0 && force_rules[0]=='1' && task != 0 && task[0]=='2' && noindex == 0 && response_code == 200){
|
|
if(num_scripts > 2 || num_stylesheets > 2){
|
|
noindex = 1;
|
|
printf("\nFailed rule check");
|
|
}
|
|
}
|
|
|
|
int skip = 0, titlechanged = 0, escape = 0, escapetotal = 0, redirected = 0;
|
|
//Check if noindex and size
|
|
//if(((noindex == 0 /*&& bodysize < 1900000*/ && bodysize > 10) || (noindex == 0 /*&& bodysize < 1900000*/ && descriptionsize > 10)) && response_code == 200 && alreadydone==0)
|
|
if((emptytitle == 0 || descriptionsize > 0 || bodysize > 0) && response_code == 200 && alreadydone==0 && noindex == 0)
|
|
{
|
|
//=================Allocate memory for the parsed text from htmlparse()
|
|
//title = (char*)calloc(titlesize+1,sizeof(char));
|
|
//keywords = (char*)calloc(keywordssize+1,sizeof(char));
|
|
//description = (char*)calloc(descriptionsize+1,sizeof(char));
|
|
//page = (char*)calloc(bodysize+1,sizeof(char));
|
|
windexinsert = (char*)calloc(finalURLsize+urlnoprefixcount+bodysize+descriptionsize+keywordssize+titlesize+1001,sizeof(char));
|
|
//shardinsert = (char*)calloc(finalURLsize+urlnoprefixcount+bodysize+descriptionsize+keywordssize+titlesize+1001,sizeof(char));
|
|
windexupdate = (char*)calloc(finalURLsize+urlnoprefixcount+bodysize+descriptionsize+keywordssize+titlesize+1001,sizeof(char));
|
|
windexRandUpdate = (char*)calloc(finalURLsize+urlnoprefixcount+bodysize+descriptionsize+keywordssize+titlesize+1001,sizeof(char));
|
|
titlecheckinsert = (char*)calloc(finalURLsize+titlesize+1001,sizeof(char));
|
|
|
|
/*if(title == NULL || keywords == NULL || description == NULL || page == NULL || windexinsert == NULL || windexupdate == NULL)
|
|
{
|
|
printf("\nError allocating memory for webpage");
|
|
//cleanup sql stuff
|
|
mysql_free_result(resulturlcheck);
|
|
mysql_free_result(result);
|
|
mysql_close(con);
|
|
exit(0);
|
|
}*/
|
|
|
|
|
|
//Check if this is a new page: check if the title found in windex is the same as the parsed title. If not, put the page back into review.
|
|
int dbtitlesize = 0,titlecheckTitleSize = 0, dbNoTitle=0,extrapos=0;
|
|
if(idexistsalready==1)
|
|
{
|
|
//going to insert the crawled title into a "titlecheck" table with the url for reference, then we're going to read back the
|
|
//title and count the number of bytes vs what was read from dbtitlesize to determine if title changed
|
|
//this is because bytes read from db must be the same charset as what is crawled to get a proper count
|
|
//unsupported charsets can end up truncating data, giving incorrect title check, this method avoids that issue
|
|
|
|
if (mysql_query(con, "use wibytemp;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//set charset based on crawled page charset tag
|
|
if (mysql_query(con, mysqlcharset))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//insert title into wibytemp for comparison
|
|
strcpy(titlecheckinsert,"INSERT INTO titlecheck (url,title) VALUES ('");
|
|
strcat(titlecheckinsert,finalURL);
|
|
strcat(titlecheckinsert,"','");
|
|
strcat(titlecheckinsert,title);
|
|
strcat(titlecheckinsert,"');");
|
|
if (mysql_query(con, titlecheckinsert))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if (mysql_query(con, "SET CHARSET utf8;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//now read back the title from the database
|
|
char checktitle[finalURLsize+dbtitlesize+1000];
|
|
memset(checktitle,0,finalURLsize+dbtitlesize+1000);
|
|
strcpy(checktitle,"SELECT title FROM titlecheck WHERE url = '");
|
|
strcat(checktitle,finalURL);strcat(checktitle,"' ORDER BY id DESC;");
|
|
//query db
|
|
if (mysql_query(con, checktitle))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
MYSQL_RES *resulttitlecheck = mysql_store_result(con);
|
|
if(resulttitlecheck == NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//grab the first entry (fifo)
|
|
MYSQL_ROW rowTitleCheck = mysql_fetch_row(resulttitlecheck);
|
|
char *titlecheckTitle;
|
|
int titlecheckTitleSize = 0;
|
|
titlecheckTitle = rowTitleCheck[0];
|
|
//printf("\n %s",rowTitleCheck[0]);
|
|
|
|
//delete the entry from the table
|
|
char titlecheckremove[finalURLsize+1000];
|
|
memset(titlecheckremove,0,finalURLsize+1000);
|
|
strcpy(titlecheckremove,"DELETE FROM titlecheck WHERE url ='");
|
|
strcat(titlecheckremove,finalURL);strcat(titlecheckremove,"';");
|
|
if (mysql_query(con, titlecheckremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//back to wiby database
|
|
if (mysql_query(con, "use wiby;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//check if original dburl is now getting redirected from finalurl (should be sent to review)
|
|
int finalURLnoprefix_size = strlen(finalURLnoprefix), dburl_noprefix_size = strlen(dburl_noprefix);
|
|
if(finalURLnoprefix_size != dburl_noprefix_size){
|
|
redirected = 1;
|
|
printf("\nIndexed page is being redirected.");
|
|
}else{
|
|
for(int i=0;i<finalURLnoprefix_size;i++){
|
|
if(dburl_noprefix[i] != finalURLnoprefix[i]){
|
|
redirected = 1;
|
|
printf("\nIndexed page is being redirected.");
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
while(titlecheckTitle[titlecheckTitleSize]!='\0')//get size of title in titlecheck
|
|
{
|
|
titlecheckTitleSize++;
|
|
}
|
|
//printf("\n%d",titlecheckTitleSize);
|
|
|
|
dbtitlesize = 0,dbNoTitle=0,extrapos=0;
|
|
|
|
while(dbtitle[dbtitlesize]!='\0')//get size of old title in db
|
|
{
|
|
dbtitlesize++;
|
|
}
|
|
//printf("\n%d",dbtitlesize);
|
|
|
|
//check if dbtitle matches url - If no title exists, URL's smaller than 111 chars will be used as titles, otherwise, "Untitled" will be used.
|
|
int URL_is_dbtitle = 1;
|
|
dbNoTitle=1;
|
|
if(dbtitlesize==finalURLsize){
|
|
for(int i=0;i<finalURLsize;i++){
|
|
if(dbtitle[i] != finalURL[i]){
|
|
URL_is_dbtitle = dbNoTitle = 0;
|
|
break;
|
|
}
|
|
}
|
|
}else{
|
|
URL_is_dbtitle = dbNoTitle = 0;
|
|
}
|
|
|
|
if(dbtitlesize == 8 && URL_is_dbtitle == 0)//check if old title in db is "Untitled"
|
|
{
|
|
if(dbtitle[0]=='U' && dbtitle[1]=='n' && dbtitle[2]=='t' && dbtitle[3]=='i' && dbtitle[4]=='t' && dbtitle[5]=='l' && dbtitle[6]=='e' && dbtitle[7]=='d')
|
|
dbNoTitle=1;
|
|
if(titlesize == 8 && emptytitle == 0){//off chance the title is actually called "Untitled".
|
|
if(title[0]=='U' && title[1]=='n' && title[2]=='t' && title[3]=='i' && title[4]=='t' && title[5]=='l' && title[6]=='e' && title[7]=='d')
|
|
dbNoTitle=0;
|
|
}
|
|
}
|
|
|
|
//if((dbNoTitle == 0 && dbtitlesize != (titlesize-extrapos)) || (dbNoTitle == 1 && titlesize > 0 && emptytitle == 0)) //previous, before db wibytemp titlecheck method
|
|
if((dbNoTitle == 0 && dbtitlesize != titlecheckTitleSize) || (dbNoTitle == 1 && titlesize > 0 && emptytitle == 0) || (URL_is_dbtitle == 1 && dbtitlesize != titlecheckTitleSize && titlesize > 0 && emptytitle == 0))
|
|
{
|
|
titlechanged = 1;
|
|
}
|
|
//printf("\n|%s|\n%d\n%d\n%d\n%d\n%d",dbtitle,titlesize,dbtitlesize,extrapos,dbNoTitle,titlechanged);
|
|
|
|
//cleanup some sql stuff
|
|
mysql_free_result(resulttitlecheck);
|
|
}
|
|
|
|
if(titlechanged == 0 && redirected == 0)
|
|
{
|
|
//====================Load the parsed text into windex!==================
|
|
|
|
if (mysql_query(con, mysqlcharset))//set charset based on page charset tag
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//strcpy(windexinsert,"INSERT INTO windex (url,title,tags,description,body,worksafe,enable,date,approver,surprise,updatable) VALUES ('");
|
|
strcpy(windexinsert,"INSERT INTO windex (url,url_noprefix,title,description,body,worksafe,enable,date,approver,surprise,http,updatable,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,shard) VALUES ('");
|
|
|
|
strcpy(windexupdate,"UPDATE windex SET url = '");
|
|
|
|
int copiedRandom = 0;
|
|
int reserveFail = 0;
|
|
char randomreserve[100];
|
|
char *randID;
|
|
char *randshard;
|
|
MYSQL_RES *resultRandID;
|
|
|
|
if(idexistsalready == 0){//Insert new entry
|
|
//For search topics to be evenly discovered by all replicas or duplicate connections assigned to a specific search section, new rows must be scattered randomly across the database insead of sequental:
|
|
//Existing rows will be randomly selected and copied (inserted) into a new row at the bottom, and the new page will take the ID number of the old one through an update.
|
|
//select id from windex where enable = 1 order by rand() limit 1;
|
|
//insert into windex (url,title,tags,description,body,surprise,http,updatable,worksafe,enable,date,updated,approver,fault) select url,title,tags,description,body,surprise,http,updatable,worksafe,enable,date,updated,approver,fault from windex where id = 1338;
|
|
//the corresponding shard table will also be updated with the same ID and contents, which can be offloaded to another replica.
|
|
|
|
|
|
printf("\nInserting into index... ");
|
|
|
|
if (mysql_query(con, "SELECT id, shard FROM windex WHERE enable = 1 ORDER BY rand() LIMIT 1;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
resultRandID = mysql_store_result(con);
|
|
if (resultRandID==NULL)
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
MYSQL_ROW row = mysql_fetch_row(resultRandID);
|
|
if(row != NULL){
|
|
randID = row[0];
|
|
idexistsvalue = row[0];
|
|
randshard = row[1];
|
|
}
|
|
|
|
//reserve the randomly selected ID when running more than one crawler
|
|
if(row != NULL && id_assigned==1){
|
|
if (mysql_query(con, "use wibytemp;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
memset(randomreserve,0,100);
|
|
strcpy(randomreserve,"INSERT into reserve_id (id) VALUES (");
|
|
strcat(randomreserve,randID);
|
|
strcat(randomreserve,");");
|
|
if (mysql_query(con, randomreserve))
|
|
{
|
|
printf("\nID is already reserved. Clearing old reservations...");
|
|
if(mysql_query(con, "DELETE FROM reserve_id WHERE time < NOW() - INTERVAL 10 MINUTE")){
|
|
finish_with_error(con);
|
|
}else{
|
|
printf(" Done.");
|
|
}
|
|
reserveFail=1;//if error: more than one crawler attempted to reserve the same randomly selected ID
|
|
}
|
|
//back to wiby database
|
|
if (mysql_query(con, "use wiby;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
|
|
if(row == NULL || reserveFail==1){//if no rows in db yet or fails to reserve an ID
|
|
strcat(windexinsert,finalURL);strcat(windexinsert,"','");
|
|
strcat(windexinsert,finalURLnoprefix);strcat(windexinsert,"','");
|
|
//strcat(windexinsert,prefix);strcat(windexinsert,"','");
|
|
if(titlesize > 0 && emptytitle == 0) {
|
|
strcat(windexinsert,title);
|
|
}
|
|
else {
|
|
if(finalURLsize < 111){
|
|
strcat(windexinsert,finalURL);
|
|
}
|
|
else{
|
|
strcat(windexinsert,"Untitled");
|
|
}
|
|
}
|
|
strcat(windexinsert,"','");
|
|
//if(tagsize > 0) {strcat(windexinsert,keywords);}
|
|
//strcat(windexinsert,"','");
|
|
if(descriptionsize > 0) {strcat(windexinsert,description);}
|
|
strcat(windexinsert,"','");
|
|
if(bodysize > 0) {strcat(windexinsert,body);}
|
|
strcat(windexinsert,"',");
|
|
strcat(windexinsert,worksafe);
|
|
strcat(windexinsert,",1,now(),'");
|
|
strcat(windexinsert,approver);
|
|
strcat(windexinsert,"',");
|
|
strcat(windexinsert,surprise);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,httpAllow);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,updatable);
|
|
if(task != 0 && task[0]=='2'){//came from link crawling
|
|
strcat(windexinsert,",'");
|
|
strcat(windexinsert,crawl_tree);
|
|
strcat(windexinsert,"','");
|
|
strcat(windexinsert,crawl_family);
|
|
strcat(windexinsert,"',");
|
|
strcat(windexinsert,crawl_pages);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,crawl_type);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,"0");
|
|
strcat(windexinsert,", force_rules = ");
|
|
strcat(windexinsert,force_rules);
|
|
}else{
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,"NULL,");
|
|
strcat(windexinsert,"NULL,");
|
|
strcat(windexinsert,crawl_pages);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,crawl_type);
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,crawl_repeat);
|
|
strcat(windexinsert,", force_rules = ");
|
|
strcat(windexinsert,force_rules);
|
|
}
|
|
strcat(windexinsert,",");
|
|
strcat(windexinsert,shardnumstr);
|
|
strcat(windexinsert,")");
|
|
if (mysql_query(con, windexinsert))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//insert into the shard table for the new row
|
|
if(nShards>0){
|
|
memset(windexinsert,0,strlen(windexinsert));
|
|
strcpy(windexinsert,"INSERT INTO ws");
|
|
strcat(windexinsert,shardnumstr);
|
|
strcat(windexinsert," (id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard) SELECT id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard FROM windex WHERE id = LAST_INSERT_ID();");
|
|
/*//get the last ID
|
|
MYSQL_RES *resultIDnum;
|
|
char *lastIDnum;
|
|
|
|
if (mysql_query(con, "SELECT LAST_INSERT_ID() FROM windex limit 1"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
MYSQL_ROW rowLastID = mysql_fetch_row(resultIDnum);
|
|
if(rowLastID != NULL){
|
|
lastIDnum = rowLastID[0];
|
|
}
|
|
|
|
strcpy(shardinsert,"INSERT INTO ws");
|
|
strcat(shardinsert,shardnumstr);
|
|
strcat(shardinsert," (id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard) SELECT id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard FROM windex WHERE id = ");
|
|
strcat(shardinsert,lastIDnum);
|
|
if (mysql_query(con, shardinsert))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
mysql_free_result(resultIDnum); */
|
|
if (mysql_query(con, windexinsert))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
else{
|
|
//copy contents of randomly selected row to a new row in windex.
|
|
strcpy(windexRandUpdate,"INSERT INTO windex (url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard) SELECT url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard FROM windex WHERE id = ");
|
|
strcat(windexRandUpdate,randID);
|
|
if (mysql_query(con, windexRandUpdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(nShards>0){//Also copy that new row into a new row of the same ID in the round-robin assigned shard table
|
|
//update the shard id in windex
|
|
memset(windexRandUpdate,0,strlen(windexRandUpdate));
|
|
strcpy(windexRandUpdate,"UPDATE windex set shard = ");
|
|
strcat(windexRandUpdate,shardnumstr);
|
|
strcat(windexRandUpdate," WHERE id = LAST_INSERT_ID()");
|
|
if (mysql_query(con, windexRandUpdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//insert that row into the next shard
|
|
memset(windexRandUpdate,0,strlen(windexRandUpdate));
|
|
strcpy(windexRandUpdate,"INSERT INTO ws");
|
|
strcat(windexRandUpdate,shardnumstr);
|
|
strcat(windexRandUpdate," (id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard) SELECT id,url,url_noprefix,title,tags,description,body,surprise,http,updatable,worksafe,crawl_tree,crawl_family,crawl_pages,crawl_type,crawl_repeat,force_rules,enable,date,updated,approver,fault,shard FROM windex WHERE id = LAST_INSERT_ID()");
|
|
if (mysql_query(con, windexRandUpdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//Overwrite the randomly selected row with the contents of the newly crawled webpage
|
|
memset(windexRandUpdate,0,strlen(windexRandUpdate));
|
|
strcpy(windexRandUpdate,"UPDATE windex SET url = '");
|
|
strcat(windexRandUpdate,finalURL);
|
|
strcat(windexRandUpdate,"', url_noprefix = '");
|
|
strcat(windexRandUpdate,finalURLnoprefix);
|
|
strcat(windexRandUpdate,"', title = '");
|
|
if(titlesize > 0 && emptytitle == 0){
|
|
strcat(windexRandUpdate,title);
|
|
}
|
|
else{
|
|
if(finalURLsize < 111){
|
|
strcat(windexRandUpdate,finalURL);
|
|
}
|
|
else{
|
|
strcat(windexRandUpdate,"Untitled");
|
|
}
|
|
}
|
|
strcat(windexRandUpdate,"', tags = NULL, description = '");
|
|
strcat(windexRandUpdate,description);
|
|
strcat(windexRandUpdate,"', body = '");
|
|
strcat(windexRandUpdate,body);
|
|
strcat(windexRandUpdate,"', worksafe = ");
|
|
strcat(windexRandUpdate,worksafe);
|
|
strcat(windexRandUpdate,", approver = '");
|
|
strcat(windexRandUpdate,approver);
|
|
strcat(windexRandUpdate,"', surprise = ");
|
|
strcat(windexRandUpdate,surprise);
|
|
strcat(windexRandUpdate,", http = ");
|
|
strcat(windexRandUpdate,httpAllow);
|
|
strcat(windexRandUpdate,", updatable = ");
|
|
strcat(windexRandUpdate,updatable);
|
|
if(task==0){//didn't come from refresh or link crawling
|
|
strcat(windexRandUpdate,", crawl_tree = NULL");
|
|
strcat(windexRandUpdate,", crawl_family = NULL");
|
|
strcat(windexRandUpdate,", crawl_pages = ");
|
|
strcat(windexRandUpdate,crawl_pages);
|
|
strcat(windexRandUpdate,", crawl_type = ");
|
|
strcat(windexRandUpdate,crawl_type);
|
|
strcat(windexRandUpdate,", crawl_repeat = ");
|
|
strcat(windexRandUpdate,crawl_repeat);
|
|
strcat(windexRandUpdate,", force_rules = ");
|
|
strcat(windexRandUpdate,force_rules);
|
|
}else if(task != 0 && task[0]=='2'){//came from link crawling
|
|
strcat(windexRandUpdate,", crawl_tree = '");
|
|
strcat(windexRandUpdate,crawl_tree);
|
|
strcat(windexRandUpdate,"', crawl_family ='");
|
|
strcat(windexRandUpdate,crawl_family);
|
|
strcat(windexRandUpdate,"', crawl_pages = ");
|
|
strcat(windexRandUpdate,crawl_pages);
|
|
strcat(windexRandUpdate,", crawl_type = ");
|
|
strcat(windexRandUpdate,crawl_type);
|
|
strcat(windexRandUpdate,", crawl_repeat = ");
|
|
strcat(windexRandUpdate,"0");
|
|
strcat(windexRandUpdate,", force_rules = ");
|
|
strcat(windexRandUpdate,force_rules);
|
|
}
|
|
strcat(windexRandUpdate,", updated = CURRENT_TIMESTAMP, date = now(), fault = 0 WHERE id = ");
|
|
strcat(windexRandUpdate,randID);
|
|
if (mysql_query(con, windexRandUpdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//Finally, update the corresponding shard table row
|
|
if(randshard != 0){
|
|
memset(windexRandUpdate,0,strlen(windexRandUpdate));
|
|
strcpy(windexRandUpdate,"UPDATE ws");
|
|
strcat(windexRandUpdate,randshard);
|
|
strcat(windexRandUpdate," SET url = '");
|
|
strcat(windexRandUpdate,finalURL);
|
|
strcat(windexRandUpdate,"', url_noprefix = '");
|
|
strcat(windexRandUpdate,finalURLnoprefix);
|
|
strcat(windexRandUpdate,"', title = '");
|
|
if(titlesize > 0 && emptytitle == 0){
|
|
strcat(windexRandUpdate,title);
|
|
}
|
|
else{
|
|
if(finalURLsize < 111){
|
|
strcat(windexRandUpdate,finalURL);
|
|
}
|
|
else{
|
|
strcat(windexRandUpdate,"Untitled");
|
|
}
|
|
}
|
|
strcat(windexRandUpdate,"', tags = NULL, description = '");
|
|
strcat(windexRandUpdate,description);
|
|
strcat(windexRandUpdate,"', body = '");
|
|
strcat(windexRandUpdate,body);
|
|
strcat(windexRandUpdate,"', worksafe = ");
|
|
strcat(windexRandUpdate,worksafe);
|
|
strcat(windexRandUpdate,", approver = '");
|
|
strcat(windexRandUpdate,approver);
|
|
strcat(windexRandUpdate,"', surprise = ");
|
|
strcat(windexRandUpdate,surprise);
|
|
strcat(windexRandUpdate,", http = ");
|
|
strcat(windexRandUpdate,httpAllow);
|
|
strcat(windexRandUpdate,", updatable = ");
|
|
strcat(windexRandUpdate,updatable);
|
|
if(task==0){//didn't come from refresh or link crawling
|
|
strcat(windexRandUpdate,", crawl_tree = NULL");
|
|
strcat(windexRandUpdate,", crawl_family = NULL");
|
|
strcat(windexRandUpdate,", crawl_pages = ");
|
|
strcat(windexRandUpdate,crawl_pages);
|
|
strcat(windexRandUpdate,", crawl_type = ");
|
|
strcat(windexRandUpdate,crawl_type);
|
|
strcat(windexRandUpdate,", crawl_repeat = ");
|
|
strcat(windexRandUpdate,crawl_repeat);
|
|
strcat(windexRandUpdate,", force_rules = ");
|
|
strcat(windexRandUpdate,force_rules);
|
|
}else if(task != 0 && task[0]=='2'){//came from link crawling
|
|
strcat(windexRandUpdate,", crawl_tree = '");
|
|
strcat(windexRandUpdate,crawl_tree);
|
|
strcat(windexRandUpdate,"', crawl_family ='");
|
|
strcat(windexRandUpdate,crawl_family);
|
|
strcat(windexRandUpdate,"', crawl_pages = ");
|
|
strcat(windexRandUpdate,crawl_pages);
|
|
strcat(windexRandUpdate,", crawl_type = ");
|
|
strcat(windexRandUpdate,crawl_type);
|
|
strcat(windexRandUpdate,", crawl_repeat = ");
|
|
strcat(windexRandUpdate,"0");
|
|
strcat(windexRandUpdate,", force_rules = ");
|
|
strcat(windexRandUpdate,force_rules);
|
|
}
|
|
strcat(windexRandUpdate,", updated = CURRENT_TIMESTAMP, date = now(), fault = 0 WHERE id = ");
|
|
strcat(windexRandUpdate,randID);
|
|
if (mysql_query(con, windexRandUpdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
copiedRandom = 1;
|
|
}
|
|
}
|
|
if(idexistsalready == 1 || (copiedRandom == 1 && nShards == 0)){ //update an existing entry or a new entry with no shard listed in row
|
|
|
|
printf("\nUpdating index... ");
|
|
strcat(windexupdate,finalURL);
|
|
strcat(windexupdate,"', url_noprefix = '");
|
|
strcat(windexupdate,finalURLnoprefix);
|
|
strcat(windexupdate,"', title = '");
|
|
if(titlesize > 0 && emptytitle == 0){
|
|
strcat(windexupdate,title);
|
|
}
|
|
else{
|
|
if(finalURLsize < 111){
|
|
strcat(windexupdate,finalURL);
|
|
}
|
|
else{
|
|
strcat(windexupdate,"Untitled");
|
|
}
|
|
}
|
|
if(copiedRandom == 0)//normal update
|
|
strcat(windexupdate,"', description = '");
|
|
else{
|
|
strcat(windexupdate,"', tags = NULL, description = '");
|
|
}
|
|
strcat(windexupdate,description);
|
|
strcat(windexupdate,"', body = '");
|
|
strcat(windexupdate,body);
|
|
strcat(windexupdate,"', worksafe = ");
|
|
strcat(windexupdate,worksafe);
|
|
strcat(windexupdate,", approver = '");
|
|
strcat(windexupdate,approver);
|
|
strcat(windexupdate,"', surprise = ");
|
|
strcat(windexupdate,surprise);
|
|
strcat(windexupdate,", http = ");
|
|
strcat(windexupdate,httpAllow);
|
|
strcat(windexupdate,", updatable = ");
|
|
strcat(windexupdate,updatable);
|
|
if(task==0){//didn't come from refresh or link crawling
|
|
strcat(windexupdate,", crawl_tree = NULL");
|
|
strcat(windexupdate,", crawl_family = NULL");
|
|
strcat(windexupdate,", crawl_pages = ");
|
|
strcat(windexupdate,crawl_pages);
|
|
strcat(windexupdate,", crawl_type = ");
|
|
strcat(windexupdate,crawl_type);
|
|
strcat(windexupdate,", crawl_repeat = ");
|
|
strcat(windexupdate,crawl_repeat);
|
|
strcat(windexupdate,", force_rules = ");
|
|
strcat(windexupdate,force_rules);
|
|
}else if(task != 0 && task[0]=='2' && idexistsalready == 0){//came from link crawling
|
|
strcat(windexupdate,", crawl_tree = '");
|
|
strcat(windexupdate,crawl_tree);
|
|
strcat(windexupdate,"', crawl_family ='");
|
|
strcat(windexupdate,crawl_family);
|
|
strcat(windexupdate,"', crawl_pages = ");
|
|
strcat(windexupdate,crawl_pages);
|
|
strcat(windexupdate,", crawl_type = ");
|
|
strcat(windexupdate,crawl_type);
|
|
strcat(windexupdate,", crawl_repeat = ");
|
|
strcat(windexupdate,"0");
|
|
strcat(windexupdate,", force_rules = ");
|
|
strcat(windexupdate,force_rules);
|
|
}
|
|
if(copiedRandom == 0)//normal update
|
|
strcat(windexupdate,", updated = CURRENT_TIMESTAMP, fault = 0 WHERE id = ");
|
|
else
|
|
strcat(windexupdate,", updated = CURRENT_TIMESTAMP, date = now(), fault = 0 WHERE id = ");
|
|
strcat(windexupdate,idexistsvalue);//will be same as randID if a new page is replacing that row
|
|
if (mysql_query(con, windexupdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
//update shard
|
|
if(nShards>0 && idexistsalready == 1 && shard != 0){
|
|
memset(windexupdate,0,strlen(windexupdate));
|
|
strcpy(windexupdate,"UPDATE ws");
|
|
strcat(windexupdate,shard);
|
|
strcat(windexupdate," SET url = '");
|
|
strcat(windexupdate,finalURL);
|
|
strcat(windexupdate,"', url_noprefix = '");
|
|
strcat(windexupdate,finalURLnoprefix);
|
|
strcat(windexupdate,"', title = '");
|
|
if(titlesize > 0 && emptytitle == 0){
|
|
strcat(windexupdate,title);
|
|
}
|
|
else{
|
|
if(finalURLsize < 111){
|
|
strcat(windexupdate,finalURL);
|
|
}
|
|
else{
|
|
strcat(windexupdate,"Untitled");
|
|
}
|
|
}
|
|
if(copiedRandom == 0)//normal update
|
|
strcat(windexupdate,"', description = '");
|
|
else{
|
|
strcat(windexupdate,"', tags = NULL, description = '");
|
|
}
|
|
strcat(windexupdate,description);
|
|
strcat(windexupdate,"', body = '");
|
|
strcat(windexupdate,body);
|
|
strcat(windexupdate,"', worksafe = ");
|
|
strcat(windexupdate,worksafe);
|
|
strcat(windexupdate,", approver = '");
|
|
strcat(windexupdate,approver);
|
|
strcat(windexupdate,"', surprise = ");
|
|
strcat(windexupdate,surprise);
|
|
strcat(windexupdate,", http = ");
|
|
strcat(windexupdate,httpAllow);
|
|
strcat(windexupdate,", updatable = ");
|
|
strcat(windexupdate,updatable);
|
|
if(task==0){//didn't come from refresh or link crawling
|
|
strcat(windexupdate,", crawl_tree = NULL");
|
|
strcat(windexupdate,", crawl_family = NULL");
|
|
strcat(windexupdate,", crawl_pages = ");
|
|
strcat(windexupdate,crawl_pages);
|
|
strcat(windexupdate,", crawl_type = ");
|
|
strcat(windexupdate,crawl_type);
|
|
strcat(windexupdate,", crawl_repeat = ");
|
|
strcat(windexupdate,crawl_repeat);
|
|
strcat(windexupdate,", force_rules = ");
|
|
strcat(windexupdate,force_rules);
|
|
}
|
|
strcat(windexupdate,", updated = CURRENT_TIMESTAMP, fault = 0 WHERE id = ");
|
|
strcat(windexupdate,idexistsvalue);//will be same as randID if a new page is replacing that row
|
|
if (mysql_query(con, windexupdate))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
|
|
//unreserve randomly selected ID
|
|
if(id_assigned==1 && idexistsalready==0 && reserveFail==0){
|
|
if (mysql_query(con, "use wibytemp;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
memset(randomreserve,0,100);
|
|
strcpy(randomreserve,"DELETE FROM reserve_id where id = ");
|
|
strcat(randomreserve,randID);
|
|
strcat(randomreserve,";");
|
|
if (mysql_query(con, randomreserve))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//back to wiby database
|
|
if (mysql_query(con, "use wiby;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
//unreserve ID if doing an update
|
|
if(id_assigned==1 && updatereserve==1){
|
|
if (mysql_query(con, "use wibytemp;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
memset(idReserve,0,200);
|
|
strcpy(idReserve,"DELETE FROM reserve_id where id = ");
|
|
strcat(idReserve,idexistsvalue);
|
|
strcat(idReserve,";");
|
|
if(mysql_query(con, idReserve))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
//back to wiby database
|
|
if (mysql_query(con, "use wiby;"))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
//free result
|
|
if(idexistsalready == 0){
|
|
mysql_free_result(resultRandID);
|
|
}
|
|
|
|
//===================remove the entry from the indexqueue===============
|
|
//printf("\nRemoving from queue...");
|
|
char sqlqueryremove[200];
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM indexqueue WHERE id=");
|
|
strcat(sqlqueryremove,id);strcat(sqlqueryremove,";");
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
|
|
printf("\n\nSuccess!");
|
|
}
|
|
//clear page from memory
|
|
free(windexinsert); free(windexupdate); free(titlecheckinsert); free(windexRandUpdate); //free(shardinsert);
|
|
}else{
|
|
skip = 1;
|
|
}
|
|
|
|
if((skip == 1 || titlechanged == 1 || redirected == 1)){
|
|
//from skip check: if(((noindex == 0 && bodysize < 1900000 && bodysize > 10) || (noindex == 0 && bodysize < 1900000 && descriptionsize > 10)) && response_code == 200 && alreadydone==0)
|
|
//printf("\nnoindex: %d\nbodysize: %ld\ndescriptionsize %ld\nresponse_code: %d\nalreadydone: %d\nskip: %d\ntitlechanged: %d\nredirected: %d",noindex,bodysize,descriptionsize,response_code,alreadydone,skip,titlechanged,redirected);
|
|
if(skip == 1){
|
|
printf("\nDoesn't want to be indexed, size too big, 404, already done, failed rules, or security issue.");
|
|
//log previous failed link crawls
|
|
strcpy(previousfail[4],previousfail[3]);
|
|
strcpy(previousfail[3],previousfail[2]);
|
|
strcpy(previousfail[2],previousfail[1]);
|
|
strcpy(previousfail[1],previousfail[0]);
|
|
strcpy(previousfail[0],urlnoprefix);
|
|
}
|
|
printf("\nRemoving from queue...");
|
|
char sqlqueryremove[200];
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM indexqueue WHERE id=");
|
|
strcat(sqlqueryremove,id);strcat(sqlqueryremove,";");
|
|
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(alreadydone==0){
|
|
if(idexistsalready == 1 && fault[0] == '1')
|
|
{
|
|
if(crawl_family != 0 && crawl_family[0] !='0'){
|
|
printf("\nPage may no longer exist. Originated from link crawling. Removing from the index.");
|
|
FILE *abandoned = fopen("abandoned.txt", "a");
|
|
fputs (url,abandoned);
|
|
fputs ("\r\n",abandoned);
|
|
fclose(abandoned);
|
|
}else{
|
|
printf("\nPage may no longer exist. Moving to review.");
|
|
}
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM windex WHERE id =");
|
|
strcat(sqlqueryremove,idexistsvalue);
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(nShards > 0 && shard != 0){
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM ws");
|
|
strcat(sqlqueryremove,shard);
|
|
strcat(sqlqueryremove," WHERE id = ");
|
|
strcat(sqlqueryremove,idexistsvalue);
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
if(crawl_family == 0 || (crawl_family != 0 && crawl_family[0] =='0')){
|
|
char sqlqueryreview[1001];
|
|
memset(sqlqueryreview,0,1001);
|
|
strcpy(sqlqueryreview,"INSERT INTO reviewqueue (url,worksafe) VALUES ('");
|
|
strcat(sqlqueryreview,url);strcat(sqlqueryreview,"',");
|
|
strcat(sqlqueryreview,worksafe);strcat(sqlqueryreview,");");
|
|
if (mysql_query(con, sqlqueryreview))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
else if(idexistsalready == 1 && fault[0] != '1')//mark that there is a fault with the page, crawler will throw it back into review if it happens again
|
|
{
|
|
printf("\nFault found. Will try again later.");
|
|
char sqlqueryfault[450];
|
|
memset(sqlqueryfault,0,450);
|
|
strcpy(sqlqueryfault,"UPDATE windex SET updated = CURRENT_TIMESTAMP, fault = 1 WHERE id = ");
|
|
strcat(sqlqueryfault,idexistsvalue);
|
|
if (mysql_query(con, sqlqueryfault))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(nShards>0 && shard != 0){
|
|
memset(sqlqueryfault,0,450);
|
|
strcpy(sqlqueryfault,"UPDATE ws");
|
|
strcat(sqlqueryfault,shard);
|
|
strcat(sqlqueryfault," SET updated = CURRENT_TIMESTAMP, fault = 1 WHERE id = ");
|
|
strcat(sqlqueryfault,idexistsvalue);
|
|
if (mysql_query(con, sqlqueryfault))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
else
|
|
{
|
|
FILE *abandoned = fopen("abandoned.txt", "a");
|
|
fputs (url,abandoned);
|
|
fputs ("\r\n",abandoned);
|
|
fclose(abandoned);
|
|
}
|
|
}
|
|
|
|
//check if link crawling is specified
|
|
//make sure duplicates don't get crawled more than once
|
|
//check db if its already indexed too - do this at beginning instead?
|
|
//crawl links if crawling through hyperlinks, or from regular refresh while crawl_repeat is on, or during manual submission when appropriate limits are set
|
|
}else if(nofollow==0 && getURLs==1 && alreadydone==0){
|
|
//cycle through url list, then construct an sql string around it, then insert it to indexqueue;
|
|
|
|
//force crawl depth of 1 during a refresh if crawl_repeat is set
|
|
if(crawl_repeat != 0 && crawl_repeat[0]=='1' && task != 0 && task[0]=='1'){
|
|
n_crawl_depth=1;
|
|
}
|
|
|
|
if(n_crawl_depth>0)//below 0 = unlimited depth
|
|
n_crawl_depth--;
|
|
|
|
memset(strDepth,0,101);
|
|
sprintf(strDepth,"%d",n_crawl_depth);
|
|
//itoa(n_crawl_depth,strDepth,10);
|
|
|
|
memset(url_fromlist,0,url_fromlist_arraylen);
|
|
memset(url_insert,0,url_insert_arraylen);
|
|
int loopcount=0,elementnum=0,urls=0;
|
|
if(id_assigned == 1){
|
|
strcpy(url_insert,"INSERT INTO indexqueue (url,worksafe,approver,surprise,task,crawl_tree,crawl_family,crawl_depth,crawl_pages,crawl_type,crawl_repeat,force_rules,crawler_id) VALUES (");
|
|
}else{
|
|
strcpy(url_insert,"INSERT INTO indexqueue (url,worksafe,approver,surprise,task,crawl_tree,crawl_family,crawl_depth,crawl_pages,crawl_type,crawl_repeat,force_rules) VALUES (");
|
|
}
|
|
while(urlListShuffled[loopcount]!=0){
|
|
switch(urlListShuffled[loopcount]){
|
|
case '\n' ://see if url can be indexed, if so, add to sql insert statement
|
|
urlparse(url_fromlist);
|
|
|
|
//check if internal or external url
|
|
int isinternal=1;
|
|
if(rootdomain[0]!=0){
|
|
isinternal=0;
|
|
}else if(url_fromlist[4]==':' || url_fromlist[5]==':'){
|
|
isinternal=0;
|
|
}else if((url_fromlist[0]=='w' || url_fromlist[0]=='W') && (url_fromlist[1]=='w' || url_fromlist[1]=='W') && (url_fromlist[2]=='w' || url_fromlist[2]=='W') && url_fromlist[3]=='.'){
|
|
isinternal=0;
|
|
}
|
|
int urlNPNP_finalURL_len=strlen(urlNPNP_finalURL);
|
|
int isabsolute=0;
|
|
if(isinternal==0 && urlNPNP_finalURL_len==strlen(urlnopathnoprefix_fromlist)){
|
|
isinternal=isabsolute=1;
|
|
for(int q=0;q<urlNPNP_finalURL_len;q++){
|
|
if(urlnopathnoprefix_fromlist[q]!=urlNPNP_finalURL[q]){
|
|
isinternal=isabsolute=0;
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
if(isinternal==1 && ((crawl_type != 0 && crawl_type[0] != '2') || crawl_type == 0)){//is internal link
|
|
if(url_fromlist[0]=='/' && url_fromlist[1] != '.'){//can't handle '..' otherwise append to insert
|
|
urls++;
|
|
if(urls>1){
|
|
strcat(url_insert,", (");
|
|
}
|
|
strcat(url_insert,"'");
|
|
strcat(url_insert,urlPrefix_finalURL);
|
|
strcat(url_insert,urlNPNP_finalURL);
|
|
strcat(url_insert,url_fromlist);
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,worksafe);
|
|
strcat(url_insert,",'");
|
|
strcat(url_insert,approver);
|
|
strcat(url_insert,"',0,2,'");
|
|
if(task==0){
|
|
strcat(url_insert,url);
|
|
}else{
|
|
strcat(url_insert,crawl_tree);
|
|
}
|
|
strcat(url_insert,"','");
|
|
strcat(url_insert,finalURL);
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,strDepth);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_pages);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_type);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,"0");
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,force_rules);
|
|
if(id_assigned == 1){
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,argv[1]);
|
|
}
|
|
strcat(url_insert,")");
|
|
}else if(url_fromlist[0] != '/' && url_fromlist[0] != '.'){
|
|
urls++;
|
|
if(urls>1){
|
|
strcat(url_insert,", (");
|
|
}
|
|
strcat(url_insert,"'");
|
|
if(isabsolute==0){
|
|
strcat(url_insert,urlPrefix_finalURL);
|
|
strcat(url_insert,urlNPNP_finalURL);
|
|
strcat(url_insert,folderPath_finalURL);
|
|
strcat(url_insert,urlcopy);//scrubed index.html
|
|
}else{
|
|
strcat(url_insert,urlcopy);
|
|
}
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,worksafe);
|
|
strcat(url_insert,",'");
|
|
strcat(url_insert,approver);
|
|
strcat(url_insert,"',0,2,'");
|
|
if(task==0){
|
|
strcat(url_insert,url);
|
|
}else{
|
|
strcat(url_insert,crawl_tree);
|
|
}
|
|
strcat(url_insert,"','");
|
|
strcat(url_insert,finalURL);
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,strDepth);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_pages);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_type);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,"0");
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,force_rules);
|
|
if(id_assigned == 1){
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,argv[1]);
|
|
}
|
|
strcat(url_insert,")");
|
|
}
|
|
}else if(isinternal==0 && crawl_type != 0 && crawl_type[0] != '0'){//is external link
|
|
if(url_fromlist[0] != '.'){
|
|
urls++;
|
|
if(urls>1){
|
|
strcat(url_insert,", (");
|
|
}
|
|
strcat(url_insert,"'");
|
|
strcat(url_insert,prefix_fromlist);
|
|
strcat(url_insert,rootdomain);
|
|
strcat(url_insert,urlPath);
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,worksafe);
|
|
strcat(url_insert,",'");
|
|
strcat(url_insert,approver);
|
|
strcat(url_insert,"',0,2,'");
|
|
if(task==0){
|
|
strcat(url_insert,url);
|
|
}else{
|
|
strcat(url_insert,crawl_tree);
|
|
}
|
|
strcat(url_insert,"','");
|
|
strcat(url_insert,finalURL);
|
|
strcat(url_insert,"',");
|
|
strcat(url_insert,strDepth);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_pages);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,crawl_type);
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,"0");
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,force_rules);
|
|
if(id_assigned == 1){
|
|
strcat(url_insert,",");
|
|
strcat(url_insert,argv[1]);
|
|
}
|
|
strcat(url_insert,")");
|
|
}
|
|
}
|
|
|
|
memset(url_fromlist,0,url_fromlist_arraylen);
|
|
elementnum=0;
|
|
loopcount++;
|
|
default :
|
|
if(loopcount<url_fromlist_arraylen){
|
|
url_fromlist[elementnum]=urlListShuffled[loopcount];
|
|
}
|
|
elementnum++;
|
|
loopcount++;
|
|
}
|
|
if(n_crawl_pages == urls || strlen(url_insert)>(url_insert_arraylen-10000))
|
|
break;
|
|
}
|
|
if(urls>0){
|
|
strcat(url_insert,";");
|
|
//insert into db
|
|
if (mysql_query(con, url_insert))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
if (curl){
|
|
curl_easy_cleanup(curl);// cleanup curl (finalURL used at inserts, thats why we cleanup and the end here
|
|
curl_global_cleanup();
|
|
}
|
|
}else{
|
|
if(alreadydone == 0){
|
|
printf("\nPage was flagged as unable to crawl or banned.");
|
|
}else if(idexistsalready==1){
|
|
printf("\nPage is already indexed.");
|
|
}
|
|
printf("\nRemoving from queue...");
|
|
char sqlqueryremove[200];
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM indexqueue WHERE id=");
|
|
strcat(sqlqueryremove,id);
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(idexistsalready==1 && permitted==0){
|
|
printf(" Removing from index...");
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM windex WHERE id=");
|
|
strcat(sqlqueryremove,idexistsvalue);
|
|
strcat(sqlqueryremove," AND updatable != '0'");
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
if(nShards>0 && shard != 0){
|
|
memset(sqlqueryremove,0,200);
|
|
strcpy(sqlqueryremove,"DELETE FROM ws");
|
|
strcat(sqlqueryremove,shard);
|
|
strcat(sqlqueryremove," WHERE id=");
|
|
strcat(sqlqueryremove,idexistsvalue);
|
|
strcat(sqlqueryremove," AND updatable != '0'");
|
|
if (mysql_query(con, sqlqueryremove))
|
|
{
|
|
finish_with_error(con);
|
|
}
|
|
}
|
|
}
|
|
FILE *abandoned = fopen("abandoned.txt", "a");
|
|
fputs (url,abandoned);
|
|
fputs ("\r\n",abandoned);
|
|
fclose(abandoned);
|
|
}
|
|
|
|
//cleanup more sql stuff
|
|
mysql_free_result(resulturlcheck);
|
|
|
|
//rotate shard for next insert
|
|
if(nShards > 0){
|
|
shardnum++;
|
|
if(shardnum == nShards)
|
|
shardnum=0;
|
|
sprintf(shardnumstr,"%d",shardnum);
|
|
}
|
|
printf(" Awaiting next page in queue...\n\n");
|
|
}
|
|
//cleanup more sql stuff
|
|
mysql_free_result(result);
|
|
mysql_close(con);
|
|
|
|
if(empty==1)
|
|
sleep(5);//sleep 5 seconds
|
|
}
|
|
exit(0);
|
|
}
|