Added testing code for retrieving HTTP response

.gitignore

@@ -0,0 +1 @@
+main

Makefile

@@ -1,4 +1,4 @@
 main: main.c
-	gcc main.c -o main -L/usr/lib -lXm -lXt -lX11 -l:libXmHTML.so.1.1.10 
+	gcc main.c url_manipulation.c easysock.c -o main -L/usr/lib -lXm -lXt -lX11 -l:libXmHTML.so.1.1.10 
 
 

easysock.c

@@ -77,13 +77,36 @@ int create_local (int network, char transport, char* address, int port,struct so
 
 int create_remote (int network,char transport,char* address,int port,struct sockaddr* remote_addr_struct) {
 
+	struct addrinfo hints;  /* Used to tell getaddrinfo what kind of address we want */
+	struct addrinfo* results; /* Used by getaddrinfo to store the addresses */
+
+
+	if (check_ip_ver(address) < 0) { /* If the address is a domain name */
+		int err_code;
+		char* port_str = malloc(10 * sizeof(char));
+
+		sprintf(port_str,"%d",port); /* getaddrinfo expects a string for its port */
+
+
+		memset(&hints,'\0',sizeof(hints));
+		hints.ai_socktype = char_to_socktype(transport);
+
+		err_code = getaddrinfo(address,port_str,&hints,&results);
+		if (err_code != 0) {
+			exit(err_code);
+		}
+		remote_addr_struct = results->ai_addr;
+		network = inet_to_int(results->ai_family);
+	} else {
+	        create_addr(network,address,port,remote_addr_struct);
+	}
 
 	int socket = create_socket(network,transport);
 	if (socket < 0) {
                 exit(errno);
         }
-        create_addr(network,address,port,remote_addr_struct);
-        int addrlen;
+
+       	int addrlen;
 	if (network == 4) {
 		addrlen = sizeof(struct sockaddr_in);
 	} else if (network == 6) {
@@ -102,3 +125,45 @@ int create_remote (int network,char transport,char* address,int port,struct sock
 	}
         return socket;
 }
+
+
+int check_ip_ver(char* address) {
+	char buffer[16]; /* 16 chars - 128 bits - is enough to hold an ipv6 address */
+	if (inet_pton(AF_INET,address,buffer) == 1) {
+		return 4;
+	} else if (inet_pton(AF_INET6,address,buffer) == 1) {
+		return 6;
+	} else {
+		return -1;
+	}
+}
+
+int int_to_inet(int network) {
+	if (network == 4) {
+		return AF_INET;
+	} else if (network == 6) {
+		return AF_INET6;
+	} else {
+		exit(207);
+	}
+}
+
+int inet_to_int(int af_type) {
+	if (af_type == AF_INET) {
+		return 4;
+	} else if (af_type == AF_INET6) {
+		return 6;
+	} else {
+		exit(207);
+	}
+}
+
+int char_to_socktype(char transport) {
+	if (transport == 'T') {
+		return SOCK_STREAM;
+	} else if (transport == 'U') {
+		return SOCK_DGRAM;
+	} else {
+		exit(250);
+	}
+}

easysock.h

@@ -2,7 +2,9 @@
 #define EASYSOCK_H_
 
 #include <string.h>
+#include <sys/types.h>
 #include <sys/socket.h>
+#include <netdb.h>
 #include <arpa/inet.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -50,5 +52,24 @@ It prints the error returned by 'connect' if something went wrong, and exits wit
 
 int create_remote (int network,char transport,char* address,int port,struct sockaddr* remote_addr_struct);
 
+/* check_ip_ver - This function checks if the given string is an IPv4 address (returns 4), 
+IPv6 address (returns 6) or neither (returns -1). */
+
+int check_ip_ver(char* address);
+
+/* int_to_inet - Takes an int value (4 for IPv4, 6 for IPv6) and returns AF_INET or
+AF_INET6 respectively. */
+
+int int_to_inet(int network);
+
+/* char_to_socktype - Takes a character that represents a transport-layer protocol
+(currently only supports 'T' for TCP or 'U' for UDP - exits with error code 250 if 
+the given characters is neither of these) and return the appropriate SOCKTYPE value. */
+
+int char_to_socktype(char transport);
+
+/* inet_to_int - Takes an int value that corresponds to AF_INET or AF_INET6,
+and returns the appropriate int value. */
+int inet_to_int(int af_type);
 
 #endif

main.c

@@ -6,6 +6,7 @@
 #include <Xm/Form.h>
 #include <XmHTML/XmHTML.h>
 #include "easysock.h"
+#include "url_manipulation.h"
 
 void testFunc(Widget,XtPointer,XmPushButtonCallbackStruct *);
 
@@ -60,8 +61,35 @@ void testFunc(Widget w, XtPointer client_data, XmPushButtonCallbackStruct *callb
 	Widget* widget_list = (Widget *)client_data;
 	Widget result_widget = widget_list[1];
 	Widget text_widget = widget_list[0];
+	int port;
+
+	char* url = XmTextFieldGetString(text_widget);
+
+	char* protocol = url_to_proto(url); /* Extract the protocol string from the URL input */
+
+	if (strcmp(protocol,"http") == 0) {
+		port = 80;
+	} else if (strcmp(protocol,"https") == 0) {
+		port = 443;
+	} else {
+		exit(50); /* I haven't added support for other protocols yet */
+	}
 	
-	XmHTMLTextSetString(result_widget,"");
+	/* We call the create_remote function to create an address, bind it to a socket,
+	and make the connection for us. It returns the socket, so that the HTTP request can be made.
+	First parameter is IP version - set to -1 because we don't know yet if the address is IPv4 or IPv6. */
+
+	char* request = "GET /~fdc/sample.html HTTP/1.1\nHost: www.columbia.edu\n\n"
+	char* response = malloc(sizeof(char) * 5000);
+
+	struct sockaddr* address_struct;
+	int remote_socket = create_remote(-1,'T',url_to_hostname(url),port,address_struct);
+
+	send(remote_socket,request,strlen(request),0);
+	
+	recv(remote_socket,response,sizeof(response),0);		
+
+	XmHTMLTextSetString(result_widget,response);
 //	printf("%d\n",val);
 //	exit(2);
 }

url_manipulation.c

@@ -0,0 +1,83 @@
+#include <string.h>
+#include <stdlib.h>
+#include <stdio.h>
+
+#define DEFAULT_PATH "/"
+
+char* url_to_proto(char* url) {
+        size_t length = strlen(url);
+	int sentinel = 0;
+	char* to_return = malloc(sizeof(char) * 10);
+	for (int i=0;i<length;i++) {
+		if (*(url + (sizeof(char) * i)) == ':') { /* If the character is ':' ...*/
+			sentinel = i; /* ...Then we stop scanning the string */
+		}
+	}
+	strncpy(to_return,url,sentinel);
+	return to_return;
+}
+
+char* url_to_hostname(char* url) {
+	/* This function relies on a hack-y assumption: that, given a string
+	that represents a URL (e.g. https://www.example.com/index.html), the length of the host portion 
+	of the address is the distance between the second slash (:/'/') and the third ('/'index.html),
+	or the second slash and the end of the string, if the path isn't present. */
+
+	size_t length = strlen(url);
+	char* protocol = url_to_proto(url);
+
+        char* temp_substring = malloc(sizeof(char) * 256);
+	char* to_return = malloc(sizeof(char) * 256);
+
+	/* This is a very roundabout way of saying that, starting from a certain point in the
+	URL (length of protocol string + 3 (':','/','/'), we are copying the rest of the string into another
+	substring. */
+
+	strncpy(temp_substring,url+(strlen(protocol) + 3),(length - (strlen(protocol) + 3)));
+
+	/* Now we need to substring _this_ substring, by returning only the part before the next slash. */
+	int sentinel = 0;
+	for (int i=0;i<strlen(temp_substring);i++) {
+		if (*(temp_substring+(sizeof(char) * i)) == '/') {
+			sentinel = i;
+		}
+	}
+
+
+	if (sentinel == 0) { /* If there are no more slashes */
+		return temp_substring;
+	}
+	strncpy(to_return,temp_substring,sentinel);
+	free(temp_substring);
+	return to_return;
+}
+
+char* url_to_path(char* url) {
+	/* Assumption: The path starts from the third slash to the end of the string. */
+	int str_len = strlen(url);
+	int sentinel = 0;
+	int num_slashes = 0;
+	char* to_return = malloc(sizeof(char) * 512);
+	for (int i=0;i<strlen(url);i++) {
+		if (*(url + (sizeof(char) * i)) == '/') {
+			num_slashes++;
+			sentinel = i;
+		}
+		if (num_slashes == 3) {
+			break;
+		}
+	}
+
+	if (num_slashes < 3 || sentinel == str_len) { 	/* If we don't find the requisite number of slashes after 
+							parsing the string OR the third slash is at the end of the string
+							("e.g. www.example.com/")... */
+
+		strcpy(to_return,DEFAULT_PATH); /* ...Then return the default path - "/" */
+		return to_return;
+	}
+
+	/* Copy all bytes of the string, starting from the third slash */
+	strncpy(to_return, url+(sizeof(char) * sentinel), strlen(url)-(sizeof(char) * sentinel));
+	return to_return;
+
+}

url_manipulation.h

@@ -0,0 +1,16 @@
+#include <string.h>
+
+/* url_to_proto - This takes in a URL, and returns the protocol.
+Example - 'https://example.org' would return 'https'. */
+
+char* url_to_proto(char* url);
+
+/* url_to_hostname - This takes in a URL, and returns the hostname.
+Example - 'https://example.org' would return 'example.org'. */
+
+char* url_to_hostname(char* url);
+
+/* url_to_path - Returns the path portion of a URL.
+Example: 'https://example.org/index.html' will return '/index.html' 
+Any empty path will also return '/index.html' */
+char* url_to_path(char* url);