// Stealing addresses from Google!!!

#include <stdio.h>
#include <stdlib.h>
#include <windows.h>
#include <wininet.h>

char *urls[5] = {	"http://finance.google.com/finance?catid=66529330",
					"http://finance.google.com/finance?start=20&num=40&catid=66529330",
					"http://finance.google.com/finance?start=40&num=60&catid=66529330",
					"http://finance.google.com/finance?start=60&num=80&catid=66529330",
					"http://finance.google.com/finance?start=80&num=100&catid=66529330"};

const int FILE_BUFFER_SIZE = 1024*1024;
char bFileBuffer[FILE_BUFFER_SIZE];

DWORD GetPage(char* url, char* buff, DWORD buffsize) {

	DWORD dReadFileCount = 0;
	char *b=buff;
	DWORD read=buffsize;

	HINTERNET hSession = InternetOpen ("TechMap",
                             INTERNET_OPEN_TYPE_PRECONFIG,
                             NULL, 
                             NULL,
                             0);

	if (hSession) {

		HINTERNET hSource = InternetOpenUrl(hSession, url, NULL, 0,
			INTERNET_FLAG_RAW_DATA | INTERNET_FLAG_NO_CACHE_WRITE | INTERNET_FLAG_DONT_CACHE | INTERNET_FLAG_KEEP_CONNECTION, 0);

		if (hSource) {

			while ( InternetReadFile( hSource, b, read, &dReadFileCount ) )
			{
				if ( dReadFileCount == 0 )
					break;
				b += dReadFileCount;
				read -= dReadFileCount;

			}
            InternetCloseHandle (hSource);
		}
		InternetCloseHandle (hSession);
	}

	return DWORD(b-buff);
}




char* FindTag(char* tag, char* buff, char* end, bool endtag = false) {
#define ENDTAG true
	char *b=buff;
	char tagbuff[1024];
	size_t len = strlen(tag);

//	printf("\tLooking for %s @ %x\n", tag, b);

	while (1) {
		while (b < end && *b != '<') {b++;};
		b++; // Skip '<'
		char *tag_start = b;
		while (b < end && (*b == ' ' || *b == '\t')) {b++;};

		char *t = tagbuff;
		while (b < end && (isalpha(*b) || (*b) == '/' || (*b) == '_')) {*t++ = *b++;};
		*t = '\0';

		if(*b == '!') {
			continue;
		}

//		printf("Found thistag %s @ %x\n", thistag, b);

		if(strcmp("br", tagbuff) == 0) {
			b++;
			continue;
		}
		if(strcmp("img", tagbuff) == 0) {
			b++;
			continue;
		}
		if(strcmp("input", tagbuff) == 0) {
			b++;
			continue;
		}
		if(strcmp("/b", tagbuff) == 0) {
			b++;
			continue;
		}
		if(strcmp("/tbody", tagbuff) == 0) {
			b++;
			continue;
		}


		if ((!endtag && strcmp(tagbuff, tag) == 0) || (endtag && strcmp(tagbuff+1, tag) == 0 && tagbuff[0] == '/')) {
			return b-len;
		} else {
			if(tagbuff[0] == '/') {
				return tag_start-1;
			}
			b = FindTag(tagbuff, b, end, ENDTAG);
		}
	}
}

int FindTagArray(char** tags, size_t* skips, int numtags, char* buff, char* end, int (*callback)(char* b, char* end)) {

	char* b = buff;
	for (int i = 0; i<numtags; i++) {
		size_t len = strlen(tags[i]);

        b = FindTag(tags[i], b, end);
		for (size_t j=0; j < skips[i]; j++) {
            b = FindTag(tags[i], b, end, ENDTAG);
			b = FindTag(tags[i], b, end);
		}

		for (size_t j=0; j<len; j++) if (tags[i][j] != b[j]) return -1;
	}
	return callback(b, end);
}


int ExtractAddress(char* buff, char* end, char* dest) {
	char *p=dest;
	char *b=buff;
	while (b < end && *b != '>') b++;
	b++;

	while (isalnum(*b) || !isprint(*b)) {
		while (b < end && *b != '<') {
			if (*b == '&') {
				while (b < end && *b != ';') b++;
				b++;
				*p++ = ' ';
			}
			if (*b == '-' || *b == '+') goto done;
			if (isprint(*b)) *p++ = *b;
			b++;
		}
		*p++ = ','; *p++ = ' ';
		while (b < end && *b != '>') b++;
		b++;
	}
done:

	*p = 0;
	return 0;
}


void GetAddress(char* url, char* strbuff) {

	DWORD size = GetPage(url, bFileBuffer, FILE_BUFFER_SIZE);

	char *end = bFileBuffer + size;
	end[0] = '\0';

	char *b = strstr(bFileBuffer, "More&nbsp;from&nbsp;Reuters&nbsp;&raquo;");

	if (!b) return;

	b = FindTag("/a", b, end);
	b = FindTag("/div", b, end);
	b = FindTag("table", b, end);
	b = FindTag("tr", b, end);
	b = FindTag("td", b, end);
	ExtractAddress(b, end, strbuff);
}








struct CompanyInfo {
	char name[80];
	char marketCap[20];
	char addr[256];
};

CompanyInfo companies[100];

int GetCompanies(char * buff, char* end) {
	char* b = buff-1;
	static int count = 0;
	for (int i = 0; i < 20; i++) {
		CompanyInfo *comp = &(companies[count++]);
		b = FindTag("tr", b, end);
		b = FindTag("td", b, end);
		b = FindTag("a", b, end);
		char *tag_start = b;
		b = FindTag("/a", b, end);
		while (tag_start < b-1 && *tag_start != '"') tag_start++;
		tag_start++;
		strcpy(comp->addr, "http://finance.google.com");
		char *str = comp->addr + strlen(comp->addr);
		while (tag_start < b-1 && *tag_start != '"') *str++ = *tag_start++;
		*str = 0;
		while (tag_start < b-1 && *tag_start != '>') tag_start++;
		tag_start++;
		str = comp->name;
		while (tag_start < b-1) *str++ = *tag_start++;
		*str = 0;
		b = FindTag("/td", b, end);
		b = FindTag("td", b, end);
		b = FindTag("/td", b, end);
		b = FindTag("td", b, end);
		b = FindTag("/td", b, end);
		b = FindTag("td", b, end);
		b = FindTag("/td", b, end);
		b = FindTag("td", b, end);
		tag_start = b;
		b = FindTag("/td", b, end);
		while (tag_start < b-2 && *tag_start != '>') tag_start++;
		tag_start++;
		while (tag_start < b-2 && !isdigit(*tag_start)) tag_start++;
		str = comp->marketCap;
		while (tag_start < b-2) *str++ = *tag_start++;
		*str = 0;
		b = FindTag("tr", b, end, ENDTAG);
		b += 4;
	}
	return 0;
}


int main(int argc, char ** argv) {

	for (int i = 0; i < 5; i++) {
		DWORD size = GetPage(urls[i], bFileBuffer, FILE_BUFFER_SIZE);

		char *b = bFileBuffer;
		char *end = bFileBuffer + size;

		const int NUM_TABS = 4;
		char *tags[NUM_TABS] = {"html", "body", "table", "tr"};
		size_t skips[NUM_TABS] = {0, 0, 2, 2};

		FindTagArray(tags, skips, NUM_TABS, b, end, GetCompanies);
	}

//	printf("<markers>\n");
	for (i = 0; i < 100; i++) {
		GetAddress(companies[i].addr, companies[i].addr);
//		printf("Company %i: %s, %s, %s\n", i, companies[i].name, companies[i].marketCap, companies[i].addr);
		printf("%s\n", companies[i].addr);
//		printf("\t<marker title=\"%s, Rank %i, Market Cap %s\" addr=\"%s\"/>\n",
//			companies[i].name, i, companies[i].marketCap, companies[i].addr);
	}
//	printf("</markers>\n");
	return 0;
}