Source Code (Use browser search to find items of interest.)
Class Index
kdelibs'BookmarkTokenizer (./kdelibs/kfile/booktoken.h:73)
class BookmarkTokenizer
{
public:
BookmarkTokenizer();
~BookmarkTokenizer();
void begin();
void write( const QString& );
void end();
char* nextToken();
bool hasMoreTokens();
void first()
{ curr = head; }
protected:
void reset();
void appendToken( const char *t, int len )
{
if ( len < 1 )
return;
BMToken *tok = new BMToken( t, len );
if ( head )
{
tail->setNext( tok );
tail = tok;
}
else
{
head = tail = tok;
}
if ( !curr )
curr = tok;
}
protected:
char *buffer;
char *dest;
BMToken *head;
BMToken *tail;
BMToken *curr;
// the size of buffer
int size;
// are we in a html tag
bool tag;
// are we in quotes within a html tag
bool tquote;
// To avoid multiple spaces
bool space;
// Discard line breaks immediately after tags
bool discardCR;
// Area we in a <!-- comment --> block
bool comment;
};
//-----------------------------------------------------------------------------
kdelibs'BookmarkTokenizer::BookmarkTokenizer() (./kdelibs/kfile/booktoken.cpp:42)
BookmarkTokenizer::BookmarkTokenizer()
{
head = tail = curr = 0;
buffer = 0;
}
kdelibs'BookmarkTokenizer::reset() (./kdelibs/kfile/booktoken.cpp:48)
void BookmarkTokenizer::reset()
{
while ( head )
{
curr = head->next();
delete head;
head = curr;
}
head = tail = curr = 0;
if ( buffer )
delete [] buffer;
buffer = 0;
}
kdelibs'BookmarkTokenizer::begin() (./kdelibs/kfile/booktoken.cpp:64)
void BookmarkTokenizer::begin()
{
reset();
size = 1000;
buffer = new char[ 1024 ];
dest = buffer;
tag = false;
space = false;
discardCR = false;
comment = false;
tquote = false;
}
kdelibs'BookmarkTokenizer::write() (./kdelibs/kfile/booktoken.cpp:77)
void BookmarkTokenizer::write( const QString& str )
{
KCharsets *charsets=KGlobal::charsets();
if ( str.isNull() )
return;
const char *src = str.ascii();
while ( *src != 0 )
{
// do we need to enlarge the buffer?
if ( dest - buffer > size )
{
char *newbuf = new char [ size + 1024 ];
memcpy( newbuf, buffer, dest - buffer + 1 );
dest = newbuf + ( dest - buffer );
delete [] buffer;
buffer = newbuf;
size += 1024;
}
if ( comment )
{
if ( !strncmp( src, "->", 2 ) )
{
src += 2;
comment = false;
}
else
src++;
}
else if ( *src == '&' )
{
space = false;
// Is the string long enough?
if ( *(src+1) != '\0' && *(src+2) != '\0' )
{
// Special character by number?
if ( *(src + 1) == '#' )
{
char *endptr;
int z = (int) strtol( src+2, &endptr, 10 );
// parse only ascii characters unless in quotes
if (z<128 || tquote){
*dest++ = z;
}
else{
*dest=0;
// add currend token
appendToken(buffer,dest-buffer);
// add token with the amp-sequence for further conversion
memcpy(buffer,src,endptr-src);
buffer[endptr-src]=0;
appendToken(buffer,endptr-src);
dest=buffer;
}
src = endptr;
// Skip a trailing ';' ?
if ( *src == ';' )
src++;
}
// Special character ?
else if ( isalpha( *(src + 1) ) )
{
if (!tag){
// add currend token
*dest=0;
appendToken(buffer,dest-buffer);
dest=buffer;
const char *endptr=src+1;
while(*endptr && isalpha(*endptr)) endptr++;
if (*endptr==';') endptr++;
// add token with the amp-sequence for further conversion
memcpy(buffer,src,endptr-src);
buffer[endptr-src]=0;
appendToken(buffer,endptr-src);
src=endptr;
*dest=0;
}
else{
// There is no need for font switching
// when we are in tag quotes, so amp-sequences can be
// translated here. I hope noone uses non iso-8859-1
// characters here.
int len=0;
const QChar res=charsets->fromEntity(QString::fromLatin1(src), len);
if ( len > 0 )
{
// we should check it's ascii...
*dest++ = (res.unicode() & 0xFF);
src+=len;
}
else
{
*dest++ = *src++;
}
}
}
else
*dest++ = *src++;
}
else
*dest++ = *src++;
}
else if ( *src == '<' && !tquote )
{
src++;
if ( strncmp( src, "!-", 2 ) == 0 )
{
src += 2;
comment = true;
continue;
}
space = true; // skip leading spaces
discardCR = true; // skip leading CR
tquote = false;
if ( dest > buffer )
{
*dest = 0;
appendToken( buffer, dest-buffer );
dest = buffer;
}
*dest = TAG_ESCAPE;
dest++;
*dest = '<';
dest++;
tag = true;
}
else if ( *src == '>' && tag && !tquote )
{
space = false;
discardCR = false;
*dest = '>';
*(dest+1) = 0;
// make the tag lower case
char *ptr = buffer+2;
while ( *ptr && *ptr != ' ' )
{
*ptr = tolower( *ptr );
ptr++;
}
appendToken( buffer, dest-buffer+1 );
dest = buffer;
tag = false;
src++;
}
else if ( *src == '\n' )
{
if ( !discardCR )
{
if ( tag )
{
if ( !space )
{
*dest = ' ';
dest++;
space = true;
}
}
else if ( !space )
{
*dest = 0;
appendToken( buffer, dest-buffer );
dest = buffer;
*dest = ' ';
*(dest+1) = 0;
appendToken( buffer, 1 );
dest = buffer;
space = true;
}
}
src++;
}
else if ( *src == ' ' )
{
if ( tag )
{
if ( !space )
{
*dest = ' ';
dest++;
space = true;
}
}
else if ( !space )
{
*dest = 0;
appendToken( buffer, dest-buffer );
dest = buffer;
*dest = ' ';
*(dest+1) = 0;
appendToken( buffer, 1 );
dest = buffer;
space = true;
}
src++;
}
else if ( *src == '\t' )
{
if ( tag )
{
if ( !space )
{
*dest = ' ';
dest++;
space = true;
}
}
else if ( !space )
{
*dest = 0;
appendToken( buffer, dest-buffer );
dest = buffer;
*dest = ' ';
*(dest+1) = 0;
appendToken( buffer, 1 );
dest = buffer;
space = true;
}
src++;
}
else if ( *src == 13 )
{
// discard
src++;
}
else if ( *src == '\"' || *src == '\'')
{ // we treat " & ' the same in tags
if ( tag )
{
src++;
if ( *(dest-1) == '=' )
{
tquote = true;
*dest = '\"';
dest++;
space = false;
discardCR = false;
}
else if ( tquote )
{
tquote = false;
*dest = '\"';
dest++;
*dest = ' ';
dest++;
space = true;
discardCR = true;
}
else
continue; // stray '\"'
}
else
{
space = false;
discardCR = false;
*dest = *src++;
dest++;
}
}
else if ( *src == '=' )
{
src++;
if ( tag )
{
if ( tquote )
{
space = false;
discardCR = false;
*dest = '=';
dest++;
}
else
{
// discard space before '='
if ( *(dest-1) == ' ' )
dest--;
*dest = '=';
dest++;
space = true;
discardCR = true;
}
}
else
{
space = false;
discardCR = false;
*dest = '=';
dest++;
}
}
else
{
space = false;
discardCR = false;
*dest = *src++;
dest++;
}
}
}
kdelibs'BookmarkTokenizer::end() (./kdelibs/kfile/booktoken.cpp:399)
void BookmarkTokenizer::end()
{
if ( buffer == 0 )
return;
if ( dest > buffer )
{
*dest = 0;
appendToken( buffer, dest-buffer );
}
delete [] buffer;
buffer = 0;
}
kdelibs'BookmarkTokenizer::nextToken() (./kdelibs/kfile/booktoken.cpp:414)
char* BookmarkTokenizer::nextToken()
{
char *t = curr->token();
curr = curr->next();
return t;
}
kdelibs'BookmarkTokenizer::hasMoreTokens() (./kdelibs/kfile/booktoken.cpp:422)
bool BookmarkTokenizer::hasMoreTokens()
{
return ( curr != 0 );
}
kdelibs'BookmarkTokenizer::~BookmarkTokenizer() (./kdelibs/kfile/booktoken.cpp:427)
BookmarkTokenizer::~BookmarkTokenizer()
{
reset();
}
//-----------------------------------------------------------------------------