path: root/tdecore/kurl.cpp

/*
    Copyright (C) 1999 Torben Weis <weis@kde.org>

    This library is free software; you can redistribute it and/or
    modify it under the terms of the GNU Library General Public
    License as published by the Free Software Foundation; either
    version 2 of the License, or (at your option) any later version.

    This library is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
    Library General Public License for more details.

    You should have received a copy of the GNU Library General Public License
    along with this library; see the file COPYING.LIB.  If not, write to
    the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
    Boston, MA 02110-1301, USA.
*/

/*
 * The currently active RFC for URL/URIs is RFC3986
 * Previous (and now deprecated) RFCs are RFC1738 and RFC2396
 */

#include "kurl.h"

// KDE_QT_ONLY is first used for dcop/client (e.g. marshalling)
#ifndef KDE_QT_ONLY
#include <kdebug.h>
#include <tdeglobal.h>
#include <kidna.h>
#include <kprotocolinfo.h>
#endif

#include <stdio.h>
#include <assert.h>
#include <ctype.h>
#include <stdlib.h>
#include <unistd.h>

#include <tqurl.h>
#include <tqdir.h>
#include <tqstringlist.h>
#include <tqregexp.h>
#include <tqstylesheet.h>
#include <tqmap.h>
#include <tqtextcodec.h>
#include <tqmutex.h>

#ifdef Q_WS_WIN
# define KURL_ROOTDIR_PATH "C:/"
#else
# define KURL_ROOTDIR_PATH "/"
#endif

static const TQString fileProt = "file";

static TQTextCodec * codecForHint( int encoding_hint /* not 0 ! */ )
{
    return TQTextCodec::codecForMib( encoding_hint );
}

// encoding_offset:
// 0 encode both @ and /
// 1 encode @ but not /
// 2 encode neither @ or /
static TQString encode( const TQString& segment, int encoding_offset, int encoding_hint, bool isRawURI = false )
{
  const char *encode_string = "/@<>#\"&?={}|^~[]\'`\\:+%";
  encode_string += encoding_offset;

  TQCString local;
  if (encoding_hint==0)
    local = segment.local8Bit();
  else
  {
      TQTextCodec * textCodec = codecForHint( encoding_hint );
      if (!textCodec)
          local = segment.local8Bit();
      else
          local = textCodec->fromUnicode( segment );
  }

  int old_length = isRawURI ? local.size() - 1 : local.length();

  if ( old_length < 1 )
    return segment.isNull() ? TQString::null : TQString(""); // differentiate null and empty

  // a worst case approximation
  TQChar *new_segment = new TQChar[ old_length * 3 + 1 ];
  int new_length = 0;

  for ( int i = 0; i < old_length; i++ )
  {
    // 'unsave' and 'reserved' characters
    // according to RFC 1738,
    // 2.2. URL Character Encoding Issues (pp. 3-4)
    // WABA: Added non-ascii
    unsigned char character = local[i];
    if ( (character <= 32) || (character >= 127) ||
         strchr(encode_string, character) )
    {
      new_segment[ new_length++ ] = '%';

      unsigned int c = character / 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_segment[ new_length++ ] = c;

      c = character % 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_segment[ new_length++ ] = c;

    }
    else
      new_segment[ new_length++ ] = (TQChar)local[i];
  }

  TQString result = TQString(new_segment, new_length);
  delete [] new_segment;
  return result;
}

static TQString encodeHost( const TQString& segment, bool encode_slash, int encoding_hint )
{
  // Hostnames are encoded differently
  // we use the IDNA transformation instead

  // Note: when merging qt-addon, use QResolver::domainToAscii here
#ifndef KDE_QT_ONLY
  Q_UNUSED( encode_slash );
  Q_UNUSED( encoding_hint );
  TQString host = KIDNA::toAscii(segment);
  if (host.isEmpty())
     return segment;
  return host;
#else
  return encode(segment, encode_slash ? 0 : 1, encoding_hint);
#endif
}

static int hex2int( unsigned int _char )
{
  if ( _char >= 'A' && _char <='F')
    return _char - 'A' + 10;
  if ( _char >= 'a' && _char <='f')
    return _char - 'a' + 10;
  if ( _char >= '0' && _char <='9')
    return _char - '0';
  return -1;
}

// WABA: The result of lazy_encode isn't usable for a URL which
// needs to satisfies RFC requirements. However, the following
// operation will make it usable again:
//      encode(decode(...))
//
// As a result one can see that url.prettyURL() does not result in
// a RFC compliant URL but that the following sequence does:
//      KURL(url.prettyURL()).url()


static TQString lazy_encode( const TQString& segment, bool encodeAt=true )
{
  int old_length = segment.length();

  if ( !old_length )
    return TQString::null;

  // a worst case approximation
  TQChar *new_segment = new TQChar[ old_length * 3 + 1 ];
  int new_length = 0;

  for ( int i = 0; i < old_length; i++ )
  {
    unsigned int character = segment[i].unicode(); // Don't use latin1()
                                                   // It returns 0 for non-latin1 values
    // Small set of really ambiguous chars
    if ((character < 32) ||  // Low ASCII
        ((character == '%') && // The escape character itself
           (i+2 < old_length) && // But only if part of a valid escape sequence!
          (hex2int(segment[i+1].unicode())!= -1) &&
          (hex2int(segment[i+2].unicode())!= -1)) ||
        (character == '?') || // Start of query delimiter
        ((character == '@') && encodeAt) || // Username delimiter
        (character == '#') || // Start of reference delimiter
        ((character == 32) && (i+1 == old_length || segment[i+1] == (TQChar)' '))) // A trailing space
    {
      new_segment[ new_length++ ] = '%';

      unsigned int c = character / 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_segment[ new_length++ ] = c;

      c = character % 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_segment[ new_length++ ] = c;
    }
    else
    new_segment[ new_length++ ] = segment[i];
  }

  TQString result = TQString(new_segment, new_length);
  delete [] new_segment;
  return result;
}

static void decode( const TQString& segment, TQString &decoded, TQString &encoded, int encoding_hint=0, bool updateDecoded = true, bool isRawURI = false )
{
  decoded = TQString::null;
  encoded = segment;

  int old_length = segment.length();
  if ( !old_length )
    return;

  TQTextCodec *textCodec = 0;
  if (encoding_hint)
      textCodec = codecForHint( encoding_hint );

  if (!textCodec)
      textCodec = TQTextCodec::codecForLocale();

  TQCString csegment = textCodec->fromUnicode(segment);
  // Check if everything went ok
  if (textCodec->toUnicode(csegment) != segment)
  {
      // Uh oh
      textCodec = codecForHint( 106 ); // Fall back to utf-8
      csegment = textCodec->fromUnicode(segment);
  }
  old_length = csegment.length();

  int new_length = 0;
  int new_length2 = 0;

  // make a copy of the old one
  char *new_segment = new char[ old_length + 1 ];
  TQChar *new_usegment = new TQChar[ old_length * 3 + 1 ];

  int i = 0;
  while( i < old_length )
  {
    bool bReencode = false;
    unsigned char character = csegment[ i++ ];
    if ((character <= ' ') || (character > 127))
       bReencode = true;

    new_usegment [ new_length2++ ] = character;
    if (character == '%' )
    {
      int a = i+1 < old_length ? hex2int( csegment[i] ) : -1;
      int b = i+1 < old_length ? hex2int( csegment[i+1] ) : -1;
      if ((a == -1) || (b == -1)) // Only replace if sequence is valid
      {
         // Contains stray %, make sure to re-encode!
         bReencode = true;
      }
      else
      {
         // Valid %xx sequence
         character = a * 16 + b; // Replace with value of %dd
         if (!isRawURI && !character && updateDecoded)
            break; // Stop at %00

         new_usegment [ new_length2++ ] = (unsigned char) csegment[i++];
         new_usegment [ new_length2++ ] = (unsigned char) csegment[i++];
      }
    }
    if (bReencode)
    {
      new_length2--;
      new_usegment [ new_length2++ ] = '%';

      unsigned int c = character / 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_usegment[ new_length2++ ] = c;

      c = character % 16;
      c += (c > 9) ? ('A' - 10) : '0';
      new_usegment[ new_length2++ ] = c;
    }

    new_segment [ new_length++ ] = character;
  }
  new_segment [ new_length ] = 0;

  encoded = TQString( new_usegment, new_length2);

  // Encoding specified
  if (updateDecoded)
  {
     decoded = textCodec->toUnicode( new_segment );
     if ( isRawURI ) {
        int length = tqstrlen( new_segment );
        while ( length < new_length ) {
            decoded += TQChar::null;
            length += 1;
            decoded += textCodec->toUnicode( new_segment + length );
            length += tqstrlen( new_segment + length );
        }
     }

     TQCString validate = textCodec->fromUnicode(decoded);

     if (strcmp(validate.data(), new_segment) != 0)
     {
        decoded = TQString::fromLocal8Bit(new_segment, new_length);
     }
  }

  delete [] new_segment;
  delete [] new_usegment;
}

static TQString decode(const TQString &segment, int encoding_hint = 0, bool isRawURI = false)
{
  TQString result;
  TQString tmp;
  decode(segment, result, tmp, encoding_hint, true, isRawURI);
  return result;
}

static TQString cleanpath(const TQString &_path, bool cleanDirSeparator, bool decodeDots)
{
  if (_path.isEmpty()) return TQString::null;

  if (TQDir::isRelativePath(_path))
     return _path; // Don't mangle mailto-style URLs

  TQString path = _path;

  int len = path.length();

  if (decodeDots)
  {
#ifndef KDE_QT_ONLY
     static const TQString &encodedDot = TDEGlobal::staticQString("%2e");
#else
     TQString encodedDot("%2e");
#endif
     if (path.find(encodedDot, 0, false) != -1)
     {
#ifndef KDE_QT_ONLY
        static const TQString &encodedDOT = TDEGlobal::staticQString("%2E"); // Uppercase!
#else
        TQString encodedDOT("%2E");
#endif
        path.replace(encodedDot, ".");
        path.replace(encodedDOT, ".");
        len = path.length();
     }
  }

  bool slash = (len && path[len-1] == '/') ||
               (len > 1 && path[len-2] == '/' && path[len-1] == '.');

  // The following code cleans up directory path much like
  // TQDir::cleanDirPath() except it can be made to ignore multiple
  // directory separators by setting the flag to false.  That fixes
  // bug# 15044, mail.altavista.com and other similar brain-dead server
  // implementations that do not follow what has been specified in
  // RFC 2396!! (dA)
  TQString result;
  int cdUp, orig_pos, pos;

  cdUp = 0;
  pos = orig_pos = len;
  while ( pos && (pos = path.findRev('/',--pos)) != -1 )
  {
    len = orig_pos - pos - 1;
    if ( len == 2 && path[pos+1] == '.' && path[pos+2] == '.' )
      cdUp++;
    else
    {
      // Ignore any occurrences of '.'
      // This includes entries that simply do not make sense like /..../
      if ( (len || !cleanDirSeparator) &&
           (len != 1 || path[pos+1] != '.' ) )
      {
          if ( !cdUp )
              result.prepend(path.mid(pos, len+1));
          else
              cdUp--;
      }
    }
    orig_pos = pos;
  }

#ifdef Q_WS_WIN // prepend drive letter if exists (js)
  if (orig_pos >= 2 && isalpha(path[0].latin1()) && path[1]==':') {
    result.prepend(TQString(path[0])+":");
  }
#endif

  if ( result.isEmpty() )
    result = KURL_ROOTDIR_PATH;
  else if ( slash && result[result.length()-1] != '/' )
       result.append('/');

  return result;
}

class KURLPrivate
{
public:
  TQString m_strInternalReferenceURL;
};

bool KURL::isRelativeURL(const TQString &_url)
{
  int len = _url.length();
  if (!len) return true; // Very short relative URL.
  const TQChar *str = _url.unicode();

  // Absolute URL must start with alpha-character
  if (!isalpha(str[0].latin1()))
     return true; // Relative URL

  for(int i = 1; i < len; i++)
  {
     char c = str[i].latin1(); // Note: non-latin1 chars return 0!
     if (c == ':')
        return false; // Absolute URL

     // Protocol part may only contain alpha, digit, + or -
     if (!isalpha(c) && !isdigit(c) && (c != '+') && (c != '-'))
        return true; // Relative URL
  }
  // URL did not contain ':'
  return true; // Relative URL
}

KURL::List::List(const KURL &url)
{
    append( url );
}

KURL::List::List(const TQStringList &list)
{
  for (TQStringList::ConstIterator it = list.begin();
       it != list.end();
       it++)
    {
      append( KURL(*it) );
    }
}

TQStringList KURL::List::toStringList() const
{
  TQStringList lst;
   for( KURL::List::ConstIterator it = begin();
        it != end();
        it++)
   {
      lst.append( (*it).url() );
   }
   return lst;
}


KURL::KURL()
{
  d = new KURLPrivate();
  reset();
}

KURL::~KURL()
{
}


KURL::KURL( const TQString &url, int encoding_hint )
{
  d = new KURLPrivate();
  reset();
  parse( url, encoding_hint );
}

KURL::KURL( const char * url, int encoding_hint )
{
  d = new KURLPrivate();
  reset();
  parse( TQString::fromLatin1(url), encoding_hint );
}

KURL::KURL( const TQCString& url, int encoding_hint )
{
  d = new KURLPrivate();
  reset();
  parse( TQString::fromLatin1(url), encoding_hint );
}

KURL::KURL( const KURL& _u )
{
  d = new KURLPrivate();
  *this = _u;
  d->m_strInternalReferenceURL = _u.d->m_strInternalReferenceURL;
}

TQDataStream & operator<< (TQDataStream & s, const KURL & a)
{
  TQString QueryForWire=a.m_strQuery_encoded;
  if (!a.m_strQuery_encoded.isNull())
    QueryForWire.prepend("?");

    s << a.m_strProtocol << a.m_strUser << a.m_strPass << a.m_strHost
      << a.m_strPath << a.m_strPath_encoded << QueryForWire << a.m_strRef_encoded
      << TQ_INT8(a.m_bIsMalformed ? 1 : 0) << a.m_iPort;
    return s;
}

TQDataStream & operator>> (TQDataStream & s, KURL & a)
{
    TQ_INT8 malf;
    TQString QueryFromWire;
    s >> a.m_strProtocol >> a.m_strUser >> a.m_strPass >> a.m_strHost
      >> a.m_strPath >> a.m_strPath_encoded >> QueryFromWire >> a.m_strRef_encoded
      >> malf >> a.m_iPort;
    a.m_bIsMalformed = (malf != 0);

    if ( QueryFromWire.isNull() )
      a.m_strQuery_encoded = TQString::null;
    else if ( QueryFromWire.length() == 1 ) // empty query
      a.m_strQuery_encoded = "";
    else
      a.m_strQuery_encoded = QueryFromWire.mid(1);

    a.m_iUriMode = KURL::uriModeForProtocol( a.m_strProtocol );

    return s;
}

#ifndef QT_NO_NETWORKPROTOCOL
KURL::KURL( const TQUrl &u )
{
  d = new KURLPrivate();
  *this = u;
}
#endif

KURL::KURL( const KURL& _u, const TQString& _rel_url, int encoding_hint )
{
  d = new KURLPrivate();
  d->m_strInternalReferenceURL = _u.d->m_strInternalReferenceURL;

  if (_u.hasSubURL()) // Operate on the last suburl, not the first
  {
    KURL::List lst = split( _u );
    KURL u(lst.last(), _rel_url, encoding_hint);
    lst.remove( lst.last() );
    lst.append( u );
    *this = join( lst );
    d->m_strInternalReferenceURL = _u.d->m_strInternalReferenceURL;
    return;
  }
  // WORKAROUND THE RFC 1606 LOOPHOLE THAT ALLOWS
  // http:/index.html AS A VALID SYNTAX FOR RELATIVE
  // URLS. ( RFC 2396 section 5.2 item # 3 )
  TQString rUrl = _rel_url;
  int len = _u.m_strProtocol.length();
  if ( !_u.m_strHost.isEmpty() && !rUrl.isEmpty() &&
       rUrl.find( _u.m_strProtocol, 0, false ) == 0 &&
       rUrl[len] == ':' && (rUrl[len+1] != '/' ||
       (rUrl[len+1] == '/' && rUrl[len+2] != '/')) )
  {
    rUrl.remove( 0, rUrl.find( ':' ) + 1 );
  }

  if ( rUrl.isEmpty() )
  {
    *this = _u;
  }
  else if ( rUrl[0] == '#' )
  {
    *this = _u;
    m_strRef_encoded = rUrl.mid(1);
    if ( m_strRef_encoded.isNull() )
        m_strRef_encoded = ""; // we know there was an (empty) html ref, we saw the '#'
  }
  else if ( isRelativeURL( rUrl) )
  {
    *this = _u;
    m_strQuery_encoded = TQString::null;
    m_strRef_encoded = TQString::null;
    if ( rUrl[0] == '/')
    {
        if ((rUrl.length() > 1) && (rUrl[1] == '/'))
        {
           m_strHost = TQString::null;
           // File protocol returns file:/// without host, strip // from rUrl
           if (_u.m_strProtocol == fileProt)
              rUrl.remove(0, 2);
        }
        m_strPath = TQString::null;
        m_strPath_encoded = TQString::null;
    }
    else if ( rUrl[0] != '?' )
    {
       int pos = m_strPath.findRev( '/' );
       if (pos >= 0)
          m_strPath.truncate(pos);
       m_strPath += '/';
       if (!m_strPath_encoded.isEmpty())
       {
          pos = m_strPath_encoded.findRev( '/' );
          if (pos >= 0)
             m_strPath_encoded.truncate(pos);
          m_strPath_encoded += '/';
       }
    }
    else
    {
       if ( m_strPath.isEmpty() )
          m_strPath = '/';
    }
    KURL tmp( url() + rUrl, encoding_hint);
    *this = tmp;
    cleanPath(false);
  }
  else
  {
    KURL tmp( rUrl, encoding_hint);
    *this = tmp;
    // Preserve userinfo if applicable.
    if (!_u.m_strUser.isEmpty() && m_strUser.isEmpty() && (_u.m_strHost == m_strHost) && (_u.m_strProtocol == m_strProtocol))
    {
       m_strUser = _u.m_strUser;
       m_strPass = _u.m_strPass;
    }
    cleanPath(false);
  }
}

void KURL::reset()
{
  m_strProtocol = TQString::null;
  m_strUser = TQString::null;
  m_strPass = TQString::null;
  m_strHost = TQString::null;
  m_strPath = TQString::null;
  m_strPath_encoded = TQString::null;
  m_strQuery_encoded = TQString::null;
  m_strRef_encoded = TQString::null;
  m_bIsMalformed = true;
  m_iPort = 0;
  m_iUriMode = Auto;
}

bool KURL::isEmpty() const
{
  return (m_strPath.isEmpty() && m_strProtocol.isEmpty());
}

void KURL::parse( const TQString& _url, int encoding_hint )
{
    if ( _url.isEmpty() || m_iUriMode == Invalid )
    {
	m_strProtocol = _url;
	m_iUriMode = Invalid;
	return;
    }

    const TQChar* buf = _url.unicode();
    const TQChar* orig = buf;
    uint len = _url.length();
    uint pos = 0;

    // Node 1: Accept alpha or slash
    TQChar x = buf[pos++];
#ifdef Q_WS_WIN
    /* win32: accept <letter>: or <letter>:/ or <letter>:\ */
    const bool alpha = isalpha((int)x);
    if (alpha && len<2)
        goto NodeErr;
    if (alpha && buf[pos]==':' && (len==2 || (len>2 && (buf[pos+1]=='/' || buf[pos+1]=='\\'))))
#else
    if ( x == (TQChar)'/' )
#endif
    {
	// A slash means we immediately proceed to parse it as a file URL.
	m_iUriMode = URL;
	m_strProtocol = fileProt;
	parseURL( _url, encoding_hint );
	return;
    }
    if ( !isalpha( (int)x ) )
	goto NodeErr;

    // Node 2: Accept any amount of (alpha|digit|'+'|'-')
    // '.' is not currently accepted, because current KURL may be confused.
    // Proceed with :// :/ or :
    while( pos < len && (isalpha((int)buf[pos]) || isdigit((int)buf[pos]) ||
			 buf[pos] == (TQChar)'+' || buf[pos] == (TQChar)'-')) pos++;

    if (pos < len && buf[pos] == (TQChar)':' )
    {
	m_strProtocol = TQString( orig, pos ).lower();
	if ( m_iUriMode == Auto )
	    m_iUriMode = uriModeForProtocol( m_strProtocol );
	// Proceed to correct parse function.
	switch ( m_iUriMode )
	{
	case RawURI:
	    parseRawURI( _url );
	    return;
	case Mailto:
	    parseMailto( _url );
	    return;
	case URL:
	    parseURL( _url, encoding_hint );
	    return;
	default:
	    // Unknown URI mode results in an invalid URI.
	    break;
	}
    }

NodeErr:
    reset();
    m_strProtocol = _url;
    m_iUriMode = Invalid;
}

void KURL::parseRawURI( const TQString& _url, int encoding_hint )
{
    uint len = _url.length();
    const TQChar* buf = _url.unicode();

    uint pos = 0;

    // Accept any amount of (alpha|digit|'+'|'-')
    // '.' is not currently accepted, because current KURL may be confused.
    // Proceed with :
    while( pos < len && (isalpha((int)buf[pos]) || isdigit((int)buf[pos]) ||
			 buf[pos] == (TQChar)'+' || buf[pos] == (TQChar)'-')) pos++;

    // Note that m_strProtocol is already set here, so we just skip over the protocol.
    if (pos < len && buf[pos] == (TQChar)':' )
	pos++;
    else { // can't happen, the caller checked all this already
	reset();
	m_strProtocol = _url;
	m_iUriMode = Invalid;
	return;
    }

    if ( pos == len ) // can't happen, the caller checked this already
	m_strPath = TQString::null;
    else
	m_strPath = decode( TQString( buf + pos, len - pos ), encoding_hint, true );

    m_bIsMalformed = false;

    return;
}

void KURL::parseMailto( const TQString& _url, int encoding_hint )
{
    parseURL( _url, encoding_hint);
    if ( m_bIsMalformed )
        return;
    TQRegExp mailre("(.+@)(.+)");
    if ( mailre.exactMatch( m_strPath ) )
    {
#ifndef KDE_QT_ONLY
	TQString host = KIDNA::toUnicode( mailre.cap( 2 ) );
	if (host.isEmpty())
	    host = TQString(mailre.cap( 2 )).lower();
#else
	TQString host = TQString(mailre.cap( 2 )).lower();
#endif
	m_strPath = mailre.cap( 1 ) + host;
  }
}

void KURL::parseURL( const TQString& _url, int encoding_hint )
{
  TQString port;
  bool badHostName = false;
  int start = 0;
  uint len = _url.length();
  const TQChar* buf = _url.unicode();

  TQChar delim;
  TQString tmp;

  uint pos = 0;

  // Node 1: Accept alpha or slash
  TQChar x = buf[pos++];
#ifdef Q_WS_WIN
  /* win32: accept <letter>: or <letter>:/ or <letter>:\ */
  const bool alpha = isalpha((int)x);
  if (alpha && len<2)
    goto NodeErr;
  if (alpha && buf[pos]==(TQChar)':' && (len==2 || (len>2 && (buf[pos+1]==(TQChar)'/' || buf[pos+1]==(TQChar)'\\'))))
#else
  if ( x == (TQChar)'/' )
#endif
    goto Node9;
  if ( !isalpha( (int)x ) )
    goto NodeErr;

  // Node 2: Accept any amount of (alpha|digit|'+'|'-')
  // '.' is not currently accepted, because current KURL may be confused.
  // Proceed with :// :/ or :
  while( pos < len && (isalpha((int)buf[pos]) || isdigit((int)buf[pos]) ||
          buf[pos] == (TQChar)'+' || buf[pos] == (TQChar)'-')) pos++;

  // Note that m_strProtocol is already set here, so we just skip over the protocol.
  if ( pos+2 < len && buf[pos] == (TQChar)':' && buf[pos+1] == (TQChar)'/' && buf[pos+2] == (TQChar)'/' )
    {
      pos += 3;
    }
  else if (pos+1 < len && buf[pos] == (TQChar)':' ) // Need to always compare length()-1 otherwise KURL passes "http:" as legal!!
    {
      pos++;
      start = pos;
      goto Node9;
    }
  else
    goto NodeErr;

  //Node 3: We need at least one character here
  if ( pos == len )
      goto NodeErr;
  start = pos;

  // Node 4: Accept any amount of characters.
  if (buf[pos] == (TQChar)'[')     // An IPv6 host follows.
      goto Node8;
  // Terminate on / or @ or ? or # or " or ; or <
  x = buf[pos];
  while( (x != (TQChar)':') && (x != (TQChar)'@') && (x != (TQChar)'/') && (x != (TQChar)'?') && (x != (TQChar)'#') )
  {
     if ((x == (TQChar)'\"') || (x == (TQChar)';') || (x == (TQChar)'<'))
        badHostName = true;
     if (++pos == len)
        break;
     x = buf[pos];
  }
  if ( pos == len )
    {
      if (badHostName)
         goto NodeErr;

      setHost(decode(TQString( buf + start, pos - start ), encoding_hint));
      goto NodeOk;
    }
  if ( x == (TQChar)'@' )
    {
      m_strUser = decode(TQString( buf + start, pos - start ), encoding_hint);
      pos++;
      goto Node7;
    }
  else if ( (x == (TQChar)'/') || (x == (TQChar)'?') || (x == (TQChar)'#'))
    {
      if (badHostName)
         goto NodeErr;

      setHost(decode(TQString( buf + start, pos - start ), encoding_hint));
      start = pos;
      goto Node9;
    }
  else if ( x != (TQChar)':' )
    goto NodeErr;
  m_strUser = decode(TQString( buf + start, pos - start ), encoding_hint);
  pos++;

  // Node 5: We need at least one character
  if ( pos == len )
    goto NodeErr;
  start = pos++;

  // Node 6: Read everything until @, /, ? or #
  while( (pos < len) &&
		(buf[pos] != (TQChar)'@') &&
		(buf[pos] != (TQChar)'/') &&
		(buf[pos] != (TQChar)'?') &&
		(buf[pos] != (TQChar)'#')) pos++;
  // If we now have a '@' the ':' seperates user and password.
  // Otherwise it seperates host and port.
  if ( (pos == len) || (buf[pos] != (TQChar)'@') )
    {
      // Ok the : was used to separate host and port
      if (badHostName)
         goto NodeErr;
      setHost(m_strUser);
      m_strUser = TQString::null;
      TQString tmp( buf + start, pos - start );
      char *endptr;
      m_iPort = (unsigned short int)strtol(tmp.ascii(), &endptr, 10);
      if ((pos == len) && (strlen(endptr) == 0))
        goto NodeOk;
      // there is more after the digits
      pos -= strlen(endptr);
      if ((buf[pos] != (TQChar)'@') &&
          (buf[pos] != (TQChar)'/') &&
          (buf[pos] != (TQChar)'?') &&
          (buf[pos] != (TQChar)'#'))
        goto NodeErr;

      start = pos;
      goto Node9;
    }
  m_strPass = decode(TQString( buf + start, pos - start), encoding_hint);
  pos++;

  // Node 7: We need at least one character
 Node7:
  if ( pos == len )
    goto NodeErr;

 Node8:
  if (buf[pos] == (TQChar)'[')
  {
    // IPv6 address
    start = ++pos; // Skip '['

    if (pos == len)
    {
       badHostName = true;
       goto NodeErr;
    }
    // Node 8a: Read everything until ] or terminate
    badHostName = false;
    x = buf[pos];
    while( (x != (TQChar)']') )
    {
       if ((x == (TQChar)'\"') || (x == (TQChar)';') || (x == (TQChar)'<'))
          badHostName = true;
       if (++pos == len)
       {
          badHostName = true;
          break;
       }
       x = buf[pos];
    }
    if (badHostName)
       goto NodeErr;
    setHost(decode(TQString( buf + start, pos - start ), encoding_hint));
    if (pos < len) pos++; // Skip ']'
    if (pos == len)
       goto NodeOk;
  }
  else
  {
    // Non IPv6 address, with a user
    start = pos;

    // Node 8b: Read everything until / : or terminate
    badHostName = false;
    x = buf[pos];
    while( (x != (TQChar)':') && (x != (TQChar)'@') && (x != (TQChar)'/') && (x != (TQChar)'?') && (x != (TQChar)'#') )
    {
       if ((x == (TQChar)'\"') || (x == (TQChar)';') || (x == (TQChar)'<'))
          badHostName = true;
       if (++pos == len)
          break;
       x = buf[pos];
    }
    if (badHostName)
       goto NodeErr;
    if ( pos == len )
    {
       setHost(decode(TQString( buf + start, pos - start ), encoding_hint));
       goto NodeOk;
    }
    setHost(decode(TQString( buf + start, pos - start ), encoding_hint));
  }
  x = buf[pos];
  if ( x == (TQChar)'/' || x == (TQChar)'#' || x == (TQChar)'?' )
    {
      start = pos;
      goto Node9;
    }
  else if ( x != (TQChar)':' )
    goto NodeErr;
  pos++;

  // Node 8c: Accept at least one digit
  if ( pos == len )
    goto NodeErr;
  start = pos;
  if ( !isdigit( buf[pos++] ) )
    goto NodeErr;

  // Node 8d: Accept any amount of digits
  while( pos < len && isdigit( buf[pos] ) ) pos++;
  port = TQString( buf + start, pos - start );
  m_iPort = port.toUShort();
  if ( pos == len )
    goto NodeOk;
  start = pos;

 Node9: // parse path until query or reference reached

  while( pos < len && buf[pos] != (TQChar)'#' && buf[pos]!=(TQChar)'?' ) pos++;

  tmp = TQString( buf + start, pos - start );
  //kdDebug(126)<<" setting encoded path to:"<<tmp<<endl;
  setEncodedPath( tmp, encoding_hint );

  if ( pos == len )
      goto NodeOk;

 //Node10: // parse query or reference depending on what comes first
  delim = (buf[pos++]==(TQChar)'#'?(TQChar)'?':(TQChar)'#');

  start = pos;

  while(pos < len && buf[pos]!=delim ) pos++;

  tmp = TQString(buf + start, pos - start);
  if (delim==(TQChar)'#')
      _setQuery(tmp, encoding_hint);
  else
      m_strRef_encoded = tmp;

  if (pos == len)
      goto NodeOk;

 //Node11: // feed the rest into the remaining variable
  tmp = TQString( buf + pos + 1, len - pos - 1);
  if (delim == (TQChar)'#')
      m_strRef_encoded = tmp;
  else
      _setQuery(tmp, encoding_hint);

 NodeOk:
  //kdDebug(126)<<"parsing finished. m_strProtocol="<<m_strProtocol<<" m_strHost="<<m_strHost<<" m_strPath="<<m_strPath<<endl;
  m_bIsMalformed = false; // Valid URL

  //kdDebug()<<"Prot="<<m_strProtocol<<"\nUser="<<m_strUser<<"\nPass="<<m_strPass<<"\nHost="<<m_strHost<<"\nPath="<<m_strPath<<"\nQuery="<<m_strQuery_encoded<<"\nRef="<<m_strRef_encoded<<"\nPort="<<m_iPort<<endl;
  if (m_strProtocol.isEmpty())
  {
    m_iUriMode = URL;
    m_strProtocol = fileProt;
  }
  return;

 NodeErr:
//  kdDebug(126) << "KURL couldn't parse URL \"" << _url << "\"" << endl;
  reset();
  m_strProtocol = _url;
  m_iUriMode = Invalid;
}

KURL& KURL::operator=( const TQString& _url )
{
  reset();
  parse( _url );

  return *this;
}

KURL& KURL::operator=( const char * _url )
{
  reset();
  parse( TQString::fromLatin1(_url) );

  return *this;
}

#ifndef QT_NO_NETWORKPROTOCOL
KURL& KURL::operator=( const TQUrl & u )
{
  m_strProtocol = u.protocol();
  m_iUriMode = Auto;
  m_strUser = u.user();
  m_strPass = u.password();
  m_strHost = u.host();
  m_strPath = u.path( false );
  m_strPath_encoded = TQString::null;
  m_strQuery_encoded = u.query();
  m_strRef_encoded = u.ref();
  m_bIsMalformed = !u.isValid();
  m_iPort = u.port();

  return *this;
}
#endif

KURL& KURL::operator=( const KURL& _u )
{
  m_strProtocol = _u.m_strProtocol;
  m_strUser = _u.m_strUser;
  m_strPass = _u.m_strPass;
  m_strHost = _u.m_strHost;
  m_strPath = _u.m_strPath;
  m_strPath_encoded = _u.m_strPath_encoded;
  m_strQuery_encoded = _u.m_strQuery_encoded;
  m_strRef_encoded = _u.m_strRef_encoded;
  m_bIsMalformed = _u.m_bIsMalformed;
  m_iPort = _u.m_iPort;
  m_iUriMode = _u.m_iUriMode;
  d->m_strInternalReferenceURL = _u.d->m_strInternalReferenceURL;

  return *this;
}

bool KURL::operator<( const KURL& _u) const
{
  int i;
  if (!_u.isValid())
  {
     if (!isValid())
     {
        i = m_strProtocol.compare(_u.m_strProtocol);
        return (i < 0);
     }
     return false;
  }
  if (!isValid())
     return true;

  i = m_strProtocol.compare(_u.m_strProtocol);
  if (i) return (i < 0);

  i = m_strHost.compare(_u.m_strHost);
  if (i) return (i < 0);

  if (m_iPort != _u.m_iPort) return (m_iPort < _u.m_iPort);

  i = m_strPath.compare(_u.m_strPath);
  if (i) return (i < 0);

  i = m_strQuery_encoded.compare(_u.m_strQuery_encoded);
  if (i) return (i < 0);

  i = m_strRef_encoded.compare(_u.m_strRef_encoded);
  if (i) return (i < 0);

  i = m_strUser.compare(_u.m_strUser);
  if (i) return (i < 0);

  i = m_strPass.compare(_u.m_strPass);
  if (i) return (i < 0);

  return false;
}

bool KURL::operator==( const KURL& _u ) const
{
  if ( !isValid() || !_u.isValid() )
    return false;

  if ( m_strProtocol == _u.m_strProtocol &&
       m_strUser == _u.m_strUser &&
       m_strPass == _u.m_strPass &&
       m_strHost == _u.m_strHost &&
       m_strPath == _u.m_strPath &&
       // The encoded path may be null, but the URLs are still equal (David)
       ( m_strPath_encoded.isNull() || _u.m_strPath_encoded.isNull() ||
         m_strPath_encoded == _u.m_strPath_encoded ) &&
       m_strQuery_encoded == _u.m_strQuery_encoded &&
       m_strRef_encoded == _u.m_strRef_encoded &&
       m_iPort == _u.m_iPort &&
       d->m_strInternalReferenceURL == _u.d->m_strInternalReferenceURL )
  {
    return true;
  }

  return false;
}

bool KURL::operator==( const TQString& _u ) const
{
  KURL u( _u );
  return ( *this == u );
}

bool KURL::cmp( const KURL &u, bool ignore_trailing ) const
{
  return equals( u, ignore_trailing );
}

bool KURL::equals( const KURL &_u, bool ignore_trailing ) const
{
  if ( !isValid() || !_u.isValid() )
    return false;

  if ( ignore_trailing )
  {
    TQString path1 = path(1);
    TQString path2 = _u.path(1);
    if ( path1 != path2 )
      return false;

    if ( m_strProtocol == _u.m_strProtocol &&
         m_strUser == _u.m_strUser &&
         m_strPass == _u.m_strPass &&
         m_strHost == _u.m_strHost &&
         m_strQuery_encoded == _u.m_strQuery_encoded &&
         m_strRef_encoded == _u.m_strRef_encoded &&
         m_iPort == _u.m_iPort )
      return true;

    return false;
  }

  return ( *this == _u );
}

bool KURL::isParentOf( const KURL& _u ) const
{
  if ( !isValid() || !_u.isValid() )
    return false;

  if ( m_strProtocol == _u.m_strProtocol &&
       m_strUser == _u.m_strUser &&
       m_strPass == _u.m_strPass &&
       m_strHost == _u.m_strHost &&
       m_strQuery_encoded == _u.m_strQuery_encoded &&
       m_strRef_encoded == _u.m_strRef_encoded &&
       m_iPort == _u.m_iPort )
  {
    if ( path().isEmpty() || _u.path().isEmpty() )
        return false; // can't work with implicit paths

    TQString p1( cleanpath( path(), true, false ) );
    if ( p1[p1.length()-1] != '/' )
        p1 += '/';
    TQString p2( cleanpath( _u.path(), true, false ) );
    if ( p2[p2.length()-1] != '/' )
        p2 += '/';

    //kdDebug(126) << "p1=" << p1 << endl;
    //kdDebug(126) << "p2=" << p2 << endl;
    //kdDebug(126) << "p1.length()=" << p1.length() << endl;
    //kdDebug(126) << "p2.left(!$)=" << p2.left( p1.length() ) << endl;
    return p2.startsWith( p1 );
  }
  return false;
}

void KURL::setFileName( const TQString& _txt )
{
  m_strRef_encoded = TQString::null;
  int i = 0;
  while( _txt[i] == (TQChar)'/' ) ++i;
  TQString tmp;
  if ( i )
    tmp = _txt.mid( i );
  else
    tmp = _txt;

  TQString path = m_strPath_encoded.isEmpty() ? m_strPath : m_strPath_encoded;
  if ( path.isEmpty() )
    path = "/";
  else
  {
    int lastSlash = path.findRev( '/' );
    if ( lastSlash == -1)
    {
      // The first character is not a '/' ???
      // This looks strange ...
      path = "/";
    }
    else if ( path.right(1) != "/" )
      path.truncate( lastSlash+1 ); // keep the "/"
  }
  if (m_strPath_encoded.isEmpty())
  {
     path += tmp;
     setPath( path );
  }
  else
  {
     path += encode_string(tmp);
     setEncodedPath( path );
  }
  cleanPath();
}

void KURL::cleanPath( bool cleanDirSeparator ) // taken from the old KURL
{
  if (m_iUriMode != URL) return;
  m_strPath = cleanpath(m_strPath, cleanDirSeparator, false);
  // WABA: Is this safe when "/../" is encoded with %?
  m_strPath_encoded = cleanpath(m_strPath_encoded, cleanDirSeparator, true);
}

static TQString trailingSlash( int _trailing, const TQString &path )
{
  TQString result = path;

  if ( _trailing == 0 )
    return result;
  else if ( _trailing == 1 )
  {
    int len = result.length();
    if ( (len == 0) || (result[ len - 1 ] != (TQChar)'/') )
      result += "/";
    return result;
  }
  else if ( _trailing == -1 )
  {
    if ( result == "/" )
      return result;
    int len = result.length();
    while (len > 1 && result[ len - 1 ] == (TQChar)'/')
    {
      len--;
    }
    result.truncate( len );
    return result;
  }
  else {
    assert( 0 );
    return TQString::null;
  }
}

void KURL::adjustPath( int _trailing )
{
  if (!m_strPath_encoded.isEmpty())
  {
     m_strPath_encoded = trailingSlash( _trailing, m_strPath_encoded );
  }
  m_strPath = trailingSlash( _trailing, m_strPath );
}


TQString KURL::encodedPathAndQuery( int _trailing, bool _no_empty_path, int encoding_hint ) const
{
  TQString tmp;
  if (!m_strPath_encoded.isEmpty() && encoding_hint == 0)
  {
     tmp = trailingSlash( _trailing, m_strPath_encoded );
  }
  else
  {
     tmp = path( _trailing );
     if ( _no_empty_path && tmp.isEmpty() )
        tmp = "/";
     if (m_iUriMode == Mailto)
     {
        tmp = encode( tmp, 2, encoding_hint );
     }
     else
     {
        tmp = encode( tmp, 1, encoding_hint );
     }
  }

  // TODO apply encoding_hint to the query
  if (!m_strQuery_encoded.isNull())
      tmp += '?' + m_strQuery_encoded;
  return tmp;
}

void KURL::setEncodedPath( const TQString& _txt, int encoding_hint )
{
  m_strPath_encoded = _txt;

  decode( m_strPath_encoded, m_strPath, m_strPath_encoded, encoding_hint );
  // Throw away encoding for local files, makes file-operations faster.
  if (m_strProtocol == fileProt)
     m_strPath_encoded = TQString::null;

  if ( m_iUriMode == Auto )
    m_iUriMode = URL;
}


void KURL::setEncodedPathAndQuery( const TQString& _txt, int encoding_hint )
{
  int pos = _txt.find( '?' );
  if ( pos == -1 )
  {
    setEncodedPath(_txt, encoding_hint);
    m_strQuery_encoded = TQString::null;
  }
  else
  {
    setEncodedPath(_txt.left( pos ), encoding_hint);
    _setQuery(_txt.right(_txt.length() - pos - 1), encoding_hint);
  }
}

TQString KURL::path( int _trailing ) const
{
  return trailingSlash( _trailing, path() );
}

bool KURL::isLocalFile() const
{
  if ( (m_strProtocol != fileProt ) || hasSubURL() )
     return false;

  if (m_strHost.isEmpty() || (m_strHost == "localhost"))
     return true;

  char hostname[ 256 ];
  hostname[ 0 ] = '\0';
  if (!gethostname( hostname, 255 ))
     hostname[sizeof(hostname)-1] = '\0';

  for(char *p = hostname; *p; p++)
     *p = tolower(*p);

  return (m_strHost == hostname);
}

void KURL::setFileEncoding(const TQString &encoding)
{
  if (!isLocalFile())
     return;

  TQString q = query();

  if (!q.isEmpty() && (q[0] == '?'))
     q = q.mid(1);

  TQStringList args = TQStringList::split('&', q);
  for(TQStringList::Iterator it = args.begin();
      it != args.end();)
  {
      TQString s = decode_string(*it);
      if (s.startsWith("charset="))
         it = args.erase(it);
      else
         ++it;
  }
  if (!encoding.isEmpty())
     args.append("charset="+encode_string(encoding));

  if (args.isEmpty())
     _setQuery(TQString::null);
  else
     _setQuery(args.join("&"));
}

TQString KURL::fileEncoding() const
{
  if (!isLocalFile())
     return TQString::null;

  TQString q = query();

  if (q.isEmpty())
     return TQString::null;

  if (q[0] == '?')
     q = q.mid(1);

  TQStringList args = TQStringList::split('&', q);
  for(TQStringList::ConstIterator it = args.begin();
      it != args.end();
      ++it)
  {
      TQString s = decode_string(*it);
      if (s.startsWith("charset="))
         return s.mid(8);
  }
  return TQString::null;
}

bool KURL::hasSubURL() const
{
  if ( m_strProtocol.isEmpty() || m_bIsMalformed )
    return false;
  if (m_strRef_encoded.isEmpty())
     return false;
  if (m_strRef_encoded.startsWith("gzip:"))
     return true;
  if (m_strRef_encoded.startsWith("bzip:"))
     return true;
  if (m_strRef_encoded.startsWith("bzip2:"))
     return true;
  if (m_strRef_encoded.startsWith("tar:"))
     return true;
  if (m_strRef_encoded.startsWith("ar:"))
     return true;
  if (m_strRef_encoded.startsWith("zip:"))
     return true;
  if ( m_strProtocol == "error" ) // anything that starts with error: has suburls
     return true;
  return false;
}

TQString KURL::url( int _trailing, int encoding_hint ) const
{
  if( m_bIsMalformed )
  {
    // Return the whole url even when the url is
    // malformed.  Under such conditions the url
    // is stored in m_strProtocol.
    return m_strProtocol;
  }

  TQString u = m_strProtocol;
  if (!u.isEmpty())
    u += ":";

  if ( hasHost() || (m_strProtocol == fileProt) )
  {
    u += "//";
    if ( hasUser() )
    {
      u += encode(m_strUser, 0, encoding_hint);
      if ( hasPass() )
      {
        u += ":";
        u += encode(m_strPass, 0, encoding_hint);
      }
      u += "@";
    }
    if ( m_iUriMode == URL )
    {
      bool IPv6 = (m_strHost.find(':') != -1);
      if (IPv6)
        u += '[' + m_strHost + ']';
      else
        u += encodeHost(m_strHost, true, encoding_hint);
      if ( m_iPort != 0 ) {
        TQString buffer;
        buffer.sprintf( ":%u", m_iPort );
        u += buffer;
      }
    }
    else
    {
      u += m_strHost;
    }
  }

  if ( m_iUriMode == URL || m_iUriMode == Mailto )
    u += encodedPathAndQuery( _trailing, false, encoding_hint );
  else
    u += encode( m_strPath, 21, encoding_hint, true );

  if ( hasRef() )
  {
    u += "#";
    u += m_strRef_encoded;
  }

  return u;
}

TQString KURL::prettyURL( int _trailing ) const
{
  if( m_bIsMalformed )
  {
    // Return the whole url even when the url is
    // malformed.  Under such conditions the url
    // is stored in m_strProtocol.
    return m_strProtocol;
  }

  TQString u = m_strProtocol;
  if (!u.isEmpty())
     u += ":";

  if ( hasHost() || (m_strProtocol == fileProt) )
  {
    u += "//";
    if ( hasUser() )
    {
      u += encode(m_strUser, 0, 0);
      // Don't show password!
      u += "@";
    }
    if ( m_iUriMode == URL )
    {
    bool IPv6 = (m_strHost.find(':') != -1);
    if (IPv6)
    {
       u += '[' + m_strHost + ']';
    }
    else
    {
       u += lazy_encode(m_strHost);
    }
    }
    else
    {
      u += lazy_encode(m_strHost);
    }
    if ( m_iPort != 0 ) {
      TQString buffer;
      buffer.sprintf( ":%u", m_iPort );
      u += buffer;
    }
  }

  if (m_iUriMode == Mailto)
  {
     u += lazy_encode( m_strPath, false );
  }
  else
  {
     u += trailingSlash( _trailing, lazy_encode( m_strPath ) );
  }

  if (!m_strQuery_encoded.isNull())
      u += '?' + m_strQuery_encoded;

  if ( hasRef() )
  {
    u += "#";
    u += m_strRef_encoded;
  }

  return u;
}

TQString KURL::prettyURL( int _trailing, AdjustementFlags _flags) const
{
  TQString u = prettyURL(_trailing);
  if (_flags & StripFileProtocol && u.startsWith("file://")) {
    u.remove(0, 7);
#ifdef Q_WS_WIN
    return TQDir::convertSeparators(u);
#endif
  }
  return u;
}

TQString KURL::pathOrURL() const
{
  if ( isLocalFile() && m_strRef_encoded.isNull() && m_strQuery_encoded.isNull() ) {
    return path();
  } else {
    return prettyURL();
  }
}

TQString KURL::htmlURL() const
{
  return TQStyleSheet::escape(prettyURL());
}

KURL::List KURL::split( const KURL& _url )
{
  TQString ref;
  KURL::List lst;
  KURL url = _url;

  while(true)
  {
     KURL u = url;
     u.m_strRef_encoded = TQString::null;
     lst.append(u);
     if (url.hasSubURL())
     {
        url = KURL(url.m_strRef_encoded);
     }
     else
     {
        ref = url.m_strRef_encoded;
        break;
     }
  }

  // Set HTML ref in all URLs.
  KURL::List::Iterator it;
  for( it = lst.begin() ; it != lst.end(); ++it )
  {
     (*it).m_strRef_encoded = ref;
  }

  return lst;
}

KURL::List KURL::split( const TQString& _url )
{
  return split(KURL(_url));
}

KURL KURL::join( const KURL::List & lst )
{
  if (lst.isEmpty()) return KURL();
  KURL tmp;

  KURL::List::ConstIterator first = lst.fromLast();
  for( KURL::List::ConstIterator it = first; it != lst.end(); --it )
  {
     KURL u(*it);
     if (it != first)
     {
        if (!u.m_strRef_encoded) u.m_strRef_encoded = tmp.url();
        else u.m_strRef_encoded += "#" + tmp.url(); // Support more than one suburl thingy
     }
     tmp = u;
  }

  return tmp;
}

TQString KURL::fileName( bool _strip_trailing_slash ) const
{
  TQString fname;
  if (hasSubURL()) { // If we have a suburl, then return the filename from there
    KURL::List list = KURL::split(*this);
    KURL::List::Iterator it = list.fromLast();
    return (*it).fileName(_strip_trailing_slash);
  }
  const TQString &path = m_strPath;

  int len = path.length();
  if ( len == 0 )
    return fname;

  if ( _strip_trailing_slash )
  {
    while ( len >= 1 && path[ len - 1 ] == TQChar('/') )
      len--;
  }
  else if ( path[ len - 1 ] == TQChar('/') )
    return fname;

  // Does the path only consist of '/' characters ?
  if ( len == 1 && path[ 0 ] == TQChar('/') )
    return fname;

  // Skip last n slashes
  int n = 1;
  if (!m_strPath_encoded.isEmpty())
  {
     // This is hairy, we need the last unencoded slash.
     // Count in the encoded string how many encoded slashes follow the last
     // unencoded one.
     int i = m_strPath_encoded.findRev( TQChar('/'), len - 1 );
     TQString fileName_encoded = m_strPath_encoded.mid(i+1);
     n += fileName_encoded.contains("%2f", false);
  }
  int i = len;
  do {
    i = path.findRev( TQChar('/'), i - 1 );
  }
  while (--n && (i > 0));

  // If ( i == -1 ) => the first character is not a '/'
  // So it's some URL like file:blah.tgz, return the whole path
  if ( i == -1 ) {
    if ( len == (int)path.length() )
      fname = path;
    else
      // Might get here if _strip_trailing_slash is true
      fname = path.left( len );
  }
  else
  {
     fname = path.mid( i + 1, len - i - 1 ); // TO CHECK
  }
  return fname;
}

void KURL::addPath( const TQString& _txt )
{
  if (hasSubURL())
  {
     KURL::List lst = split( *this );
     KURL &u = lst.last();
     u.addPath(_txt);
     *this = join( lst );
     return;
  }

  m_strPath_encoded = TQString::null;

  if ( _txt.isEmpty() )
    return;

  int i = 0;
  int len = m_strPath.length();
  // Add the trailing '/' if it is missing
  if ( _txt[0] != (TQChar)'/' && ( len == 0 || m_strPath[ len - 1 ] != (TQChar)'/' ) )
    m_strPath += "/";

  // No double '/' characters
  i = 0;
  if ( len != 0 && m_strPath[ len - 1 ] == (TQChar)'/' )
  {
    while( _txt[i] == (TQChar)'/' )
      ++i;
  }

  m_strPath += _txt.mid( i );
}

TQString KURL::directory( bool _strip_trailing_slash_from_result,
                         bool _ignore_trailing_slash_in_path ) const
{
  TQString result = m_strPath_encoded.isEmpty() ? m_strPath : m_strPath_encoded;
  if ( _ignore_trailing_slash_in_path )
    result = trailingSlash( -1, result );

  if ( result.isEmpty() || result == "/" )
    return result;

  int i = result.findRev( "/" );
  // If ( i == -1 ) => the first character is not a '/'
  // So it's some URL like file:blah.tgz, with no path
  if ( i == -1 )
    return TQString::null;

  if ( i == 0 )
  {
    result = "/";
    return result;
  }

  if ( _strip_trailing_slash_from_result )
    result = result.left( i );
  else
    result = result.left( i + 1 );

  if (!m_strPath_encoded.isEmpty())
    result = decode(result);

  return result;
}


bool KURL::cd( const TQString& _dir )
{
  if ( _dir.isEmpty() || m_bIsMalformed )
    return false;

  if (hasSubURL())
  {
     KURL::List lst = split( *this );
     KURL &u = lst.last();
     u.cd(_dir);
     *this = join( lst );
     return true;
  }

  // absolute path ?
  if ( _dir[0] == (TQChar)'/' )
  {
    m_strPath_encoded = TQString::null;
    m_strPath = _dir;
    setHTMLRef( TQString::null );
    m_strQuery_encoded = TQString::null;
    return true;
  }

  // Users home directory on the local disk ?
  if ( ( _dir[0] == (TQChar)'~' ) && ( m_strProtocol == fileProt ))
  {
    m_strPath_encoded = TQString::null;
    m_strPath = TQDir::homeDirPath();
    m_strPath += "/";
    m_strPath += _dir.right(m_strPath.length() - 1);
    setHTMLRef( TQString::null );
    m_strQuery_encoded = TQString::null;
    return true;
  }

  // relative path
  // we always work on the past of the first url.
  // Sub URLs are not touched.

  // append '/' if necessary
  TQString p = path(1);
  p += _dir;
  p = cleanpath( p, true, false );
  setPath( p );

  setHTMLRef( TQString::null );
  m_strQuery_encoded = TQString::null;

  return true;
}

KURL KURL::upURL( ) const
{
  if (!query().isEmpty())
  {
     KURL u(*this);
     u._setQuery(TQString::null);
     return u;
  };

  if (!hasSubURL())
  {
     KURL u(*this);

     u.cd("../");

     return u;
  }

  // We have a subURL.
  KURL::List lst = split( *this );
  if (lst.isEmpty())
      return KURL(); // Huh?
  while (true)
  {
     KURL &u = lst.last();
     TQString old = u.path();
     u.cd("../");
     if (u.path() != old)
         break; // Finshed.
     if (lst.count() == 1)
         break; // Finished.
     lst.remove(lst.fromLast());
  }
  return join( lst );
}

TQString KURL::htmlRef() const
{
  if ( !hasSubURL() )
  {
    return decode( ref() );
  }

  List lst = split( *this );
  return decode( (*lst.begin()).ref() );
}

TQString KURL::encodedHtmlRef() const
{
  if ( !hasSubURL() )
  {
    return ref();
  }

  List lst = split( *this );
  return (*lst.begin()).ref();
}

void KURL::setHTMLRef( const TQString& _ref )
{
  if ( !hasSubURL() )
  {
    m_strRef_encoded = encode( _ref, 0, 0 /*?*/);
    return;
  }

  List lst = split( *this );

  (*lst.begin()).setRef( encode( _ref, 0, 0 /*?*/) );

  *this = join( lst );
}

bool KURL::hasHTMLRef() const
{
  if ( !hasSubURL() )
  {
    return hasRef();
  }

  List lst = split( *this );
  return (*lst.begin()).hasRef();
}

void
KURL::setProtocol( const TQString& _txt )
{
   m_strProtocol = _txt;
   if ( m_iUriMode == Auto ) m_iUriMode = uriModeForProtocol( m_strProtocol );
   m_bIsMalformed = false;
}

void
KURL::setUser( const TQString& _txt )
{
   if ( _txt.isEmpty() )
     m_strUser = TQString::null;
   else
     m_strUser = _txt;
}

void
KURL::setPass( const TQString& _txt )
{
   if ( _txt.isEmpty() )
     m_strPass = TQString::null;
   else
     m_strPass = _txt;
}

void
KURL::setHost( const TQString& _txt )
{
  if ( m_iUriMode == Auto )
    m_iUriMode = URL;
  switch ( m_iUriMode )
  {
  case URL:
#ifndef KDE_QT_ONLY
   m_strHost = KIDNA::toUnicode(_txt);
   if (m_strHost.isEmpty())
      m_strHost = _txt.lower(); // Probably an invalid hostname, but...
#else
   m_strHost = _txt.lower();
#endif
    break;
  default:
    m_strHost = _txt;
    break;
  }
}

void
KURL::setPort( unsigned short int _p )
{
   m_iPort = _p;
}

void KURL::setPath( const TQString & path )
{
  if (isEmpty())
    m_bIsMalformed = false;
  if (m_strProtocol.isEmpty())
  {
    m_strProtocol = fileProt;
  }
  m_strPath = path;
  m_strPath_encoded = TQString::null;
  if ( m_iUriMode == Auto )
    m_iUriMode = URL;
}

void KURL::setDirectory( const TQString &dir)
{
  if ( dir.endsWith("/"))
     setPath(dir);
  else
     setPath(dir+"/");
}

void KURL::setQuery( const TQString &_txt, int encoding_hint)
{
   if (_txt[0] == (TQChar)'?')
      _setQuery( _txt.length() > 1 ? _txt.mid(1) : "" /*empty, not null*/, encoding_hint );
   else
      _setQuery( _txt, encoding_hint );
}

// This is a private function that expects a query without '?'
void KURL::_setQuery( const TQString &_txt, int encoding_hint)
{
   m_strQuery_encoded = _txt;
   if (!_txt.length())
      return;

   int l = m_strQuery_encoded.length();
   int i = 0;
   TQString result;
   while (i < l)
   {
      int s = i;
      // Re-encode. Break encoded string up according to the reserved
      // characters '&:;=/?' and re-encode part by part.
      while(i < l)
      {
         char c = m_strQuery_encoded[i].latin1();
         if ((c == '&') || (c == ':') || (c == ';') ||
             (c == '=') || (c == '/') || (c == '?'))
            break;
         i++;
      }
      if (i > s)
      {
         TQString tmp = m_strQuery_encoded.mid(s, i-s);
         TQString newTmp;
         decode( tmp, newTmp, tmp, encoding_hint, false );
         result += tmp;
      }
      if (i < l)
      {
         result += m_strQuery_encoded[i];
         i++;
      }
   }
   m_strQuery_encoded = result;
}

TQString KURL::query() const
{
    if (m_strQuery_encoded.isNull())
        return TQString::null;
    return '?'+m_strQuery_encoded;
}

TQString KURL::decode_string(const TQString &str, int encoding_hint)
{
   return decode(str, encoding_hint);
}

TQString KURL::encode_string(const TQString &str, int encoding_hint)
{
   return encode(str, 1, encoding_hint);
}

TQString KURL::encode_string_no_slash(const TQString &str, int encoding_hint)
{
   return encode(str, 0, encoding_hint);
}

bool urlcmp( const TQString& _url1, const TQString& _url2 )
{
  // Both empty ?
  if ( _url1.isEmpty() && _url2.isEmpty() )
    return true;
  // Only one empty ?
  if ( _url1.isEmpty() || _url2.isEmpty() )
    return false;

  KURL::List list1 = KURL::split( _url1 );
  KURL::List list2 = KURL::split( _url2 );

  // Malformed ?
  if ( list1.isEmpty() || list2.isEmpty() )
    return false;

  return ( list1 == list2 );
}

bool urlcmp( const TQString& _url1, const TQString& _url2, bool _ignore_trailing, bool _ignore_ref )
{
  // Both empty ?
  if ( _url1.isEmpty() && _url2.isEmpty() )
    return true;
  // Only one empty ?
  if ( _url1.isEmpty() || _url2.isEmpty() )
    return false;

  KURL::List list1 = KURL::split( _url1 );
  KURL::List list2 = KURL::split( _url2 );

  // Malformed ?
  if ( list1.isEmpty() || list2.isEmpty() )
    return false;

  unsigned int size = list1.count();
  if ( list2.count() != size )
    return false;

  if ( _ignore_ref )
  {
    (*list1.begin()).setRef(TQString::null);
    (*list2.begin()).setRef(TQString::null);
  }

  KURL::List::Iterator it1 = list1.begin();
  KURL::List::Iterator it2 = list2.begin();
  for( ; it1 != list1.end() ; ++it1, ++it2 )
    if ( !(*it1).equals( *it2, _ignore_trailing ) )
      return false;

  return true;
}

TQMap< TQString, TQString > KURL::queryItems( int options ) const {
  return queryItems(options, 0);
}

TQMap< TQString, TQString > KURL::queryItems( int options, int encoding_hint ) const {
  if ( m_strQuery_encoded.isEmpty() )
    return TQMap<TQString,TQString>();

  TQMap< TQString, TQString > result;
  TQStringList items = TQStringList::split( '&', m_strQuery_encoded );
  for ( TQStringList::const_iterator it = items.begin() ; it != items.end() ; ++it ) {
    int equal_pos = (*it).find( '=' );
    if ( equal_pos > 0 ) { // = is not the first char...
      TQString name = (*it).left( equal_pos );
      if ( options & CaseInsensitiveKeys )
	name = name.lower();
      TQString value = (*it).mid( equal_pos + 1 );
      if ( value.isEmpty() )
	result.insert( name, TQString::fromLatin1("") );
      else {
	// ### why is decoding name not necessary?
	value.replace( '+', ' ' ); // + in queries means space
	result.insert( name, decode_string( value, encoding_hint ) );
      }
    } else if ( equal_pos < 0 ) { // no =
      TQString name = (*it);
      if ( options & CaseInsensitiveKeys )
	name = name.lower();
      result.insert( name, TQString::null );
    }
  }

  return result;
}

TQString KURL::queryItem( const TQString& _item ) const
{
  return queryItem( _item, 0 );
}

TQString KURL::queryItem( const TQString& _item, int encoding_hint ) const
{
  TQString item = _item + '=';
  if ( m_strQuery_encoded.length() <= 1 )
    return TQString::null;

  TQStringList items = TQStringList::split( '&', m_strQuery_encoded );
  unsigned int _len = item.length();
  for ( TQStringList::ConstIterator it = items.begin(); it != items.end(); ++it )
  {
    if ( (*it).startsWith( item ) )
    {
      if ( (*it).length() > _len )
      {
        TQString str = (*it).mid( _len );
        str.replace( '+', ' ' ); // + in queries means space.
        return decode_string( str, encoding_hint );
      }
      else // empty value
        return TQString::fromLatin1("");
    }
  }

  return TQString::null;
}

void KURL::removeQueryItem( const TQString& _item )
{
  TQString item = _item + '=';
  if ( m_strQuery_encoded.length() <= 1 )
    return;

  TQStringList items = TQStringList::split( '&', m_strQuery_encoded );
  for ( TQStringList::Iterator it = items.begin(); it != items.end(); )
  {
    if ( (*it).startsWith( item ) || (*it == _item) )
    {
      TQStringList::Iterator deleteIt = it;
      ++it;
      items.remove(deleteIt);
    }
    else
    {
       ++it;
    }
  }
  m_strQuery_encoded = items.join( "&" );
}

void KURL::addQueryItem( const TQString& _item, const TQString& _value, int encoding_hint )
{
  TQString item = _item + '=';
  TQString value = encode( _value, 0, encoding_hint );

  if (!m_strQuery_encoded.isEmpty())
     m_strQuery_encoded += '&';
  m_strQuery_encoded += item + value;
}

// static
KURL KURL::fromPathOrURL( const TQString& text )
{
    if ( text.isEmpty() )
        return KURL();

    KURL url;
    if (!TQDir::isRelativePath(text))
        url.setPath( text );
    else
        url = text;

    return url;
}

static TQString _relativePath(const TQString &base_dir, const TQString &path, bool &isParent)
{
   TQString _base_dir(TQDir::cleanDirPath(base_dir));
   TQString _path(TQDir::cleanDirPath(path.isEmpty() || (path[0] != (TQChar)'/') ? _base_dir+"/"+path : path));

   if (_base_dir.isEmpty())
      return _path;

   if (_base_dir[_base_dir.length()-1] != '/')
      _base_dir.append('/');

   TQStringList list1 = TQStringList::split('/', _base_dir);
   TQStringList list2 = TQStringList::split('/', _path);

   // Find where they meet
   uint level = 0;
   uint maxLevel = TQMIN(list1.count(), list2.count());
   while((level < maxLevel) && (list1[level] == list2[level])) level++;

   TQString result;
   // Need to go down out of the first path to the common branch.
   for(uint i = level; i < list1.count(); i++)
      result.append("../");

   // Now up up from the common branch to the second path.
   for(uint i = level; i < list2.count(); i++)
      result.append(list2[i]).append("/");

   if ((level < list2.count()) && (path[path.length()-1] != (TQChar)'/'))
      result.truncate(result.length()-1);

   isParent = (level == list1.count());

   return result;
}

TQString KURL::relativePath(const TQString &base_dir, const TQString &path, bool *isParent)
{
   bool parent = false;
   TQString result = _relativePath(base_dir, path, parent);
   if (parent)
      result.prepend("./");

   if (isParent)
      *isParent = parent;

   return result;
}

void KURL::setInternalReferenceURL( const TQString& url ) {
    d->m_strInternalReferenceURL = url;
}

TQString KURL::internalReferenceURL( void ) const {
    return d->m_strInternalReferenceURL;
}

TQString KURL::relativeURL(const KURL &base_url, const KURL &url, int encoding_hint)
{
   if ((url.protocol() != base_url.protocol()) ||
       (url.host() != base_url.host()) ||
       (url.port() && url.port() != base_url.port()) ||
       (url.hasUser() && url.user() != base_url.user()) ||
       (url.hasPass() && url.pass() != base_url.pass()))
   {
      return url.url(0, encoding_hint);
   }

   TQString relURL;

   if ((base_url.path() != url.path()) || (base_url.query() != url.query()))
   {
      bool dummy;
      TQString basePath = base_url.directory(false, false);
      relURL = encode( _relativePath(basePath, url.path(), dummy), 1, encoding_hint);
      relURL += url.query();
   }

   if ( url.hasRef() )
   {
      relURL += "#";
      relURL += url.ref();
   }

   if ( relURL.isEmpty() )
      return "./";

   return relURL;
}

int KURL::uriMode() const
{
  return m_iUriMode;
}

KURL::URIMode KURL::uriModeForProtocol(const TQString& protocol)
{
#ifndef KDE_QT_ONLY
    KURL::URIMode mode = Auto;
    if (protocol == fileProt)
        return URL;
    if (TDEGlobal::_instance)
        mode = KProtocolInfo::uriParseMode(protocol);
    if (mode == Auto ) {
#else
        KURL::URIMode mode = Auto;
#endif
	if ( protocol == "ed2k" || protocol == "sig2dat" || protocol == "slsk" || protocol == "data" ) mode = RawURI;
	else if ( protocol == "mailto" ) mode = Mailto;
	else mode = URL;
#ifndef KDE_QT_ONLY
    }
#endif
    return mode;
}