Problem trying to convert a 1 character string to ISO-8859-1

[Donald C. Kirker]

I am working on a mobile internet browser that has to deal with entities. I
use the following functions at the bottom to lookup an entity and convert it
to the page's encoding and put it in an output string so that I can later
convert from the page's encoding to the device's encoding. This works fine
when the page's encoding is UTF-8 (or the like), but fails when the page's
encoding is ISO-8859-1.

I think it might have to do with the fact that TxtConvertEncoding only
really works with device's endcoding <-> UTF-8, UCS*. It looks as if this is
the case because I get an error: txtErrUnknownEncoding.

Is there an alternate method that I could use for converting the
"conversion" ptr (see below) to ISO-8859-1?

The code I use:

I take the entity code, such as "&amp;" and convert it to an ascii
equivalent using a lookup table, like the fragment below:

typedef struct
{
  char *tag;
  char *conversion;
}TagEntry;

static TagEntry tag_list[] =
{
 { "&amp", "&"},
  { "&apos", "\'"},
  { "&rsquo",   "\'"},
  { "&ast", "\x2a"},
  { "&bull", "." }, //"\x2022" },
  { "&cent", "\xa2" },
  { "&pound", "\xa3" },
  { "&curren", "\xa4" },
  { "&yen", "\xa5" },
  { "&brvbar", "\xa6" },
  { "&sect", "\xa7" },
  { "&uml", "\xa8" },
  { "&copy", "\xa9" },
  { "&ordf", "\xaa" },
  { "&laquo", "\xab" },
  { "&not", "\xac" },
  { "&shy", "\xad" },
  { "&macr", "\xaf" },
  { "&deg", "\xb0" },
...
}

I then take the "conversion" and pass it as sourceStr.

Char *ConvertLatinEntityToCharset(Char *sourceStr)
{
 char   dchr[maxCharBytes]; // char in device encoding
 char   uchr[maxCharBytes+1]; // char in unicode (UTF8)
 UInt16   dBytes;
 UInt16   uBytes = maxCharBytes;
 Char   *result = NULL;
 GlobalsType  *g;
 Err   err;

 FtrGet(wuAppType, ftrGlobals, (UInt32 *)&g);

 dBytes = TxtSetNextChar(dchr, 0, sourceStr[0]);

     switch (g->charSet) {
  case UTF_8:
   err = TxtConvertEncoding(true, NULL, dchr, &dBytes,
charEncodingPalmLatin,
    uchr, &uBytes, charEncodingUTF8, "?", 1);
   break;
  /*case US_ASCII:
   err = TxtConvertEncoding(true, NULL, dchr, &dBytes,
charEncodingPalmLatin,
    uchr, &uBytes, charEncodingAscii, "?", 1);
   break;*/
  // For some reason this is having issues
  case ISO_8859_1:
   err = TxtConvertEncoding(true, NULL, dchr, &dBytes,
charEncodingPalmLatin,
    uchr, &uBytes, charEncodingISO8859_1, "?", 1);
   break;
  case USC_2:
   err = TxtConvertEncoding(true, NULL, dchr, &dBytes,
charEncodingPalmLatin,
    uchr, &uBytes, charEncodingUCS2, "?", 1);
   break;
  default:
   err = TxtConvertEncoding(true, NULL, dchr, &dBytes,
charEncodingPalmLatin,
    uchr, &uBytes, charEncodingPalmLatin, "?", 1);
   break;
     }

 uchr[uBytes] = 0;

 if (uchr[0] == NULL)
  return NULL;

 result = Malloc(uBytes + 1);
 StrCopy(result, uchr);

 return result;
}


This converts the string produced after the entitiy conversion to the
device's encoding:

Char *ConvertFromTo(CharEncodingType srcEncoding, Char *sourceStr, Char
*destStr)
{
 Err  err;
 UInt16  ioDstBytes = StrLen(sourceStr);
 UInt16  ioSrcBytes = StrLen(sourceStr);
 UInt32  encoding;

     FtrGet(sysFtrCreator, sysFtrNumEncoding, &encoding); // or
charEncodingPalmLatin???

 err = TxtConvertEncoding (true, NULL, sourceStr, &ioSrcBytes,
  srcEncoding, destStr, &ioDstBytes, (CharEncodingType)encoding, "?", 1);

 if (ioDstBytes < ioSrcBytes) destStr[ioDstBytes] = 0; // string was
transformed and will be shorter...

 return (err == errNone) ? destStr : sourceStr; //if sourceStr returned than
conversion failed
}


Thanks,
Donald



-- 
For information on using the PalmSource Developer Forums, or to unsubscribe, 
please see http://www.palmos.com/dev/support/forums/