yang686526

oddbtextiterator returns wrong value
oddbtextiterator returns wrong value
with the drawing attached, on the third line of asian characters down (the line starting with ( a' ), when i use the following code, the first character gives me the wrong value, the lead character value is correct, but when retreiving the next character, the value is 0x0000, however when viewing the textptr->textstring().c_str() the value is correct? what must i be doing wrong?
my code that i use is as follows, and the drawing is attached for you.
cheers
jason
odcodepageid id = m_dwgdb->getdwgcodepage();
odgitextstyle *pstyle = new odgitextstyle();
oddbtextiteratorptr txtitr = oddbtextiterator::createobject(textptr->textstring().c_str(), strlen(textptr->textstring().c_str()), false, cp_ascii, pstyle);
do {
oduint16 curchar = 0x0000;
curchar = txtitr->nextchar();
if( odcharmapper::isleadbyte(curchar, id) ) {
curchar = curchar << 8;
oduint16 nextcurchar = 0x0000;
nextcurchar = txtitr->nextchar();
curchar = curchar | nextcurchar;
}
// now convert it to unicode.
oduint16 uchar;
odcharmapper::codepagetounicode(curchar, id, uchar);
pmystring->addchar(uchar);
} while(txtitr->currproperties().blastchar == false );
attached files
quote:
originally posted by janderssen
odcodepageid id = m_dwgdb->getdwgcodepage();
odgitextstyle *pstyle = new odgitextstyle();
oddbtextiteratorptr txtitr = oddbtextiterator::createobject(textptr->textstring().c_str(), strlen(textptr->textstring().c_str()), false, cp_ascii, pstyle);
1. you need to do next:
code:
odcodepageid id = m_dwgdb->getdwgcodepage();
odgitextstyle style;
gifromdbtextstyle(textptr->textstyle(), style);
oddbtextiteratorptr txtitr = oddbtextiterator::createobject(textptr->textstring().c_str(), strlen(textptr->textstring().c_str()), false, id, pstyle);
text style must correspond with text and parameter codepage is equal id ( no cp_ascii ).
2. nextchar() method return value depend of font coding.(for example: it is unicode for unicode shx & ttf fonts and multibyte for bigfont).
so you no need make conversion in most case.
--
best regards,
sergey z.
sergey,
all works great except for the following text inside of this drawing as well.
\m+100bd\m+100c4\m+100db\m+100b0\m+100b8
as an example, \m+100bd gets converted to bd, where it should be a totally different value.
also note that this is the only line of text that gets converted wrong. is there something else i am missing. i have played around with lots of the odcharmapper functions to try and figure it out, but to no success.
cheers
jason
quote:
originally posted by sergey z.
1. you need to do next:
code:
odcodepageid id = m_dwgdb->getdwgcodepage();
odgitextstyle style;
gifromdbtextstyle(textptr->textstyle(), style);
oddbtextiteratorptr txtitr = oddbtextiterator::createobject(textptr->textstring().c_str(), strlen(textptr->textstring().c_str()), false, id, pstyle);
text style must correspond with text and parameter codepage is equal id ( no cp_ascii ).
2. nextchar() method return value depend of font coding.(for example: it is unicode for unicode shx & ttf fonts and multibyte for bigfont).
so you no need make conversion in most case.
--
best regards,
sergey z.

hi cheers,
string \m+100bd\m+100c4\m+100db\m+100b0\m+100b8 is rendered by bigfont.shx. coding bigfont is multibyte. so text iterator return multibyte value from this string ( it is 0xbd, 0xc4 etc.).
acad help:
"\m+nxxxx
multibyte shape number. the n is a digit identifying the originating multibyte code page id. the xxxx is the hexadecimal value of the multibyte character."
--
best regards,
sergey z.
sergey,
than why does the rest of the drawings text get converted to the correct unicode charcters i need, because the rest of the asian characters are encoded in the same \m+1xxxx format.
it is only this one single line that seems to be wrong?
also note that if i check txtitr->currproperties().binbigfont it returns false?
also, how does one read the geometry of a bigfont.shx. i do all my vectorization myself manually, and do not go through the opendwg vectorization frame work. but i cannot seem to see how one can get the geometry for a single character via opendwg?
cheers
jason
quote:
originally posted by sergey z.
hi cheers,
string \m+100bd\m+100c4\m+100db\m+100b0\m+100b8 is rendered by bigfont.shx. coding bigfont is multibyte. so text iterator return multibyte value from this string ( it is 0xbd, 0xc4 etc.).
acad help:
"\m+nxxxx
multibyte shape number. the n is a digit identifying the originating multibyte code page id. the xxxx is the hexadecimal value of the multibyte character."
--
best regards,
sergey z.
quote:
originally posted by janderssen
than why does the rest of the drawings text get converted to the correct unicode charcters i need, because the rest of the asian characters are encoded in the same \m+1xxxx format.
it is only this one single line that seems to be wrong?
also note that if i check txtitr->currproperties().binbigfont it returns false?
it is strange. i am not able reproduce this on my computer. in case bigfont.shx is accessed in your computer (and dwgdirect has found it) all symbols like \m+1xxxx in attached file have to be in multibyte coding. does odamfcapp example show correctly attached file?
note: about support fonts see dwgdirect reference topic "font handling".
quote:
originally posted by janderssen
also, how does one read the geometry of a bigfont.shx. i do all my vectorization myself manually, and do not go through the opendwg vectorization frame work. but i cannot seem to see how one can get the geometry for a single character via opendwg?
i afraid you can't get geometry for a single character via opendwg directly.
see topic "text with shx font by polylines"
--
best regards,
sergey