URL Class Reference

#include <url.h>

Collaboration diagram for URL:

Public Types
typedef std::map< SWBuf, SWBuf >	ParameterMap

Public Member Functions
const char *	getHostName () const
const ParameterMap &	getParameters () const
const char *	getParameterValue (const char *name) const
const char *	getPath () const
const char *	getProtocol () const
	URL (const char *url)

Static Public Member Functions
static const SWBuf	decode (const char *encodedText)
static const SWBuf	encode (const char *urlText)

Private Member Functions
void	parse ()

Private Attributes
SWBuf	hostname
ParameterMap	parameterMap
SWBuf	path
SWBuf	protocol
SWBuf	url

Detailed Description

URL provides URL parsing The URL class provides an interface to work on the data of an URL like "http://www.crosswire.org/index.jsp?page=test&user=nobody"

Definition at line 34 of file url.h.

Member Typedef Documentation

typedef std::map<SWBuf, SWBuf> URL::ParameterMap

Definition at line 36 of file url.h.

Constructor & Destructor Documentation

URL::URL

(

const char *

url

)

Constructor.

Parameters

url	The url string which should be parsed into protocol, hostname, path and paramters

Constructors/Destructors

Definition at line 61 of file url.cpp.

    :   url(""),
        protocol(""),
        hostname(""),
        path("")
{
    if (url && strlen(url)) {
        this->url = url;
        parse();
    }
}

Member Function Documentation

const SWBuf URL::decode ( const char *  encodedText )

static

Definition at line 247 of file url.cpp.

                                           {
    /*static*/ SWBuf text;
    text = encoded; 

    SWBuf decoded;  
    const long length = text.length();
    int i = 0;
    
    while (i < length) {
        char a = text[i];
        
        if ( a == '+' ) { //handle special cases
            decoded.append(' ');
        }       
        else if ( (a == '%') && (i+2 < length)) { //decode the %ab  hex encoded char
            const char b = toupper( text[i+1] );
            const char c = toupper( text[i+2] );
            
            if (isxdigit(b) && isxdigit(c)) { //valid %ab part
                unsigned int dec = 16 * ((b >= 'A' && b <= 'F') ? (b - 'A' + 10) : (b - '0')); //dec value of the most left digit (b)
                dec += (c >= 'A' && c <= 'F') ? (c - 'A' + 10) : (c - '0'); //dec value of the right digit (c)
                
                decoded.append((char)dec); //append the decoded char
                
                i += 2; //we jump over the %ab part; we have to leave out three, but the while  loop adds one, too
            }
        }
        else { //just append the char
            decoded.append(a);
        }
        
        i++;
    }
    
    if (decoded.length()) {
        text = decoded;
    }
    return text;
}

const SWBuf URL::encode ( const char *  urlText )

static

Encodes and URL Encodes a string into a valid URL, e.g. changes http://www.crosswire.org/test.jsp?force=1&help=1 into http://www.crosswire.org/test.jsp?force=1&help=1 This function works on the data of the buf parameter.

WARNING: It doesn't check if the URL is encoded already, so http://www.crosswire.org/test.jsp?force=1&help=1 becomes http://www.crosswire.org/test.jsp?force=1&help=1

Definition at line 231 of file url.cpp.

                                           {
    /*static*/ SWBuf url;
    url = urlText;
    
    SWBuf buf;
    const long length = url.length();
    for (long i = 0; i < length; i++) { //fill "buf"
        const char& c = url[i];
        buf.append( ((m[c].length()) ? m[c] : SWBuf(c)) );
    }

    url = buf;
    return url;
}

const char * URL::getHostName

(

)

const

Get the hostname

Returns

The hostname, e.g. "www.crosswire.org" for an url like "http://www.crosswire.org/index.jsp?page=help"

Definition at line 79 of file url.cpp.

                                    {
    return hostname.c_str();
}

const URL::ParameterMap & URL::getParameters

(

)

const

All available paramters

Returns

The map which contains the parameters and their values

Definition at line 89 of file url.cpp.

                                                {
    return parameterMap;
}

const char * URL::getParameterValue

(

const char *

name

)

const

Returns the value of an URL parameter. For the URL "http://www.crosswire.org/index.jsp?page=test&user=nobody" the value of the parameter "page" would be "test". If the parameter is not set an empty string is returned.

Parameters

name	The name of the paramter.

Returns

The value of the given paramter of an empty string if the name could not be found in the list of available paramters

Returns the value of an URL parameter. For the URL "http://www.crosswire.org/index.jsp?page=test&user=nobody" the value of the parameter "page" would be "test". If the parameter is not set an empty string is returned.

Definition at line 98 of file url.cpp.

                                                         {
    static SWBuf emptyStr("");

    ParameterMap::const_iterator it = parameterMap.find(name);
    static SWBuf retVal;

    if (it != parameterMap.end())
        retVal = it->second.c_str();
        else
        retVal = emptyStr.c_str();

    return retVal.c_str();
}

const char * URL::getPath

(

)

const

Get the path

Returns

The path, e.g. "/index.jsp" for an url like "http://www.crosswire.org/index.jsp?page=help"

Definition at line 84 of file url.cpp.

                               {
    return path.c_str();
}

const char * URL::getProtocol

(

)

const

Get the protocol.

Returns

The protocol, e.g. "http" for an url like "http://www.crosswire.org/index.jsp?page=help"

Definition at line 74 of file url.cpp.

                                   {
    return protocol.c_str();
}

void URL::parse ( )

private

Parse Parse the URL into protocol, hostname, path, page and paramters

Parse the URL. Parse the URL into the protocol, the hostname, the path and the paramters with their values

Definition at line 117 of file url.cpp.

                {
    /* format example       protocol://hostname/path/path/path.pl?param1=value1&param2=value2
    * we include the script name in the path, so the path would be /path/path/path.pl in this example
    *  & could also be &
    */

    //1. Init
    const char *urlPtr = url.c_str();
     
    protocol = "";
    hostname = "";
    path     = "";
    parameterMap.clear();
     
     // 2. Get the protocol, which is from the begining to the first ://
    const char *end = strchr( urlPtr, ':' );
    if (end) { //protocol was found
        protocol.append(urlPtr, end-urlPtr);
        urlPtr = end + 1;
    
        //find the end of the protocol separator (e.g. "://")
        for (; (*urlPtr == ':') || (*urlPtr == '/'); urlPtr++);
    }

 //3.Get the hostname part. This is the part from pos up to the first slash
    bool checkPath   = true;
    bool checkParams = true;
    bool checkAnchor = true;

    end = strchr(urlPtr, '/');
    if (!end) {
        checkPath = false;
        end = strchr(urlPtr, '?');
    }
    if (!end) {
        checkParams = false;
        end = strchr(urlPtr, '#');
    }
    if (!end) {
        checkAnchor = false;
        end = urlPtr+strlen(urlPtr);
    }
     
    hostname.append(urlPtr, end-urlPtr);
        
    urlPtr = end + ((*end)? 1 : 0);

    if (checkPath) { 
        end = strchr(urlPtr, '?');
        if (!end) {
            checkParams = false;
            end = strchr(urlPtr, '#');
        }
        if (!end) {
            checkAnchor = false;
            end = urlPtr+strlen(urlPtr);
        }

        path.append(urlPtr, end-urlPtr);
        
        urlPtr = end + ((*end)? 1 : 0);
     }

    if (checkParams) {
        //5. Fill the map with the parameters and their values
        SWBuf paramName;
        SWBuf paramValue;
                
        if (checkAnchor) checkAnchor = false;
/*
        end = strchr(urlPtr, '#');
        if (!end) {
            checkAnchor = false;
            end = urlPtr+strlen(urlPtr);
        }
*/
        //end = (start && strchr(start, '?')) ? strchr(start, '?')+1 :0;
        end = urlPtr;
        while (end) {
            paramName = "";
            paramValue = "";
            
            //search for the equal sign to find the value part
            const char *valueStart = strchr(end, '=');      
            if (valueStart) {
                const char* valueEnd = strstr(valueStart, "&") ? strstr(valueStart, "&") : strstr(valueStart, "&"); //try to find a new paramter part
                
                if (valueEnd) {
                    paramName.append(end, valueStart-end);
                    paramValue.append(valueStart+1, valueEnd-(valueStart+1));
                }
                else { //this is the last paramter of the URL
                    paramName.append(end, valueStart-end);
                    paramValue.append(valueStart+1);
                }
                
                if (paramName.length() && paramValue.length()) {//insert the param into the map if it's valid
                    paramName = decode(paramName.c_str());
                    paramValue = decode(paramValue.c_str());
                    
                    parameterMap[ paramName ] = paramValue;
                }
            }
            else {
                break; //no valid parameter in the url
            }
            
            const char *start = end+1;
            end = strstr(start, "&") ? strstr(start, "&")+5 : (strstr(start, "&") ? strstr(start, "&")+1 : 0); //try to find a new paramter part
        }
    }
}

Member Data Documentation

SWBuf URL::hostname

private

Definition at line 87 of file url.h.

ParameterMap URL::parameterMap

private

Definition at line 89 of file url.h.

SWBuf URL::path

private

Definition at line 88 of file url.h.

SWBuf URL::protocol

private

Definition at line 86 of file url.h.

SWBuf URL::url

private

Definition at line 85 of file url.h.

---

The documentation for this class was generated from the following files:

• include/url.h
• src/utilfuns/url.cpp