/*
* iCompare - compare two records
*
* Compares two records. For use by qsort(3C) and bsearch(3C).
*
* returns -1 if rec1 < rec2, 0 if rec1 == rec2, 1 if rec1 > rec2
*/
static int
iCompare(const void *pvRecord1, const void *pvRecord2)
{
USHORT usUnicode1, usUnicode2;
if (usUnicode1 < usUnicode2) {
return -1;
}
if (usUnicode1 > usUnicode2) {
return 1;
}
return 0;
} /* end of iCompare */
/*
* pGetCharTableRecord - get the character table record
*
* returns a pointer to the record when found, otherwise NULL
*/
static const char_table_type *
pGetCharTableRecord(USHORT usUnicode)
{
char_table_type tKey;
if (tNextPosFree == 0) {
return NULL;
}
tKey.usUnicode = usUnicode;
tKey.ucLocal = 0;
return (char_table_type *)bsearch(&tKey,
atCharTable,
tNextPosFree, sizeof(atCharTable[0]),
iCompare);
} /* end of pGetCharTableRecord */
/*
* ucGetBulletCharacter - get the local representation of the bullet
*/
UCHAR
ucGetBulletCharacter(conversion_type eConversionType, encoding_type eEncoding)
{
#if defined(__riscos)
return 0x8f;
#else
const char_table_type *pRec;
fail(eEncoding == encoding_utf_8);
if (eEncoding == encoding_latin_1 &&
(eConversionType == conversion_ps ||
eConversionType == conversion_pdf)) {
/* Ugly, but it makes the PostScript and PDF look better */
return (UCHAR)143;
}
if (eConversionType != conversion_text &&
eConversionType != conversion_fmt_text) {
pRec = pGetCharTableRecord(UNICODE_BULLET);
if (pRec != NULL) {
return pRec->ucLocal;
}
pRec = pGetCharTableRecord(UNICODE_BULLET_OPERATOR);
if (pRec != NULL) {
return pRec->ucLocal;
}
pRec = pGetCharTableRecord(UNICODE_MIDDLE_DOT);
if (pRec != NULL) {
return pRec->ucLocal;
}
}
return (UCHAR)'.';
#endif /* __riscos */
} /* end of ucGetBulletCharacter */
/*
* ucGetNbspCharacter - get the local representation of the non-breaking space
*/
UCHAR
ucGetNbspCharacter(void)
{
const char_table_type *pRec;
pRec = pGetCharTableRecord(0x00a0); /* Unicode non-breaking space */
if (pRec == NULL) {
DBG_MSG("Non-breaking space record not found");
/* No value found, use the best guess */
return (UCHAR)0xa0;
}
return pRec->ucLocal;
} /* end of ucGetNbspCharacter */
/*
* bReadCharacterMappingTable - read the mapping table
*
* Read the character mapping table from file and have the contents sorted
*
* returns TRUE if successful, otherwise FALSE
*/
BOOL
bReadCharacterMappingTable(FILE *pFile)
{
char *pcTmp;
ULONG ulUnicode;
UINT uiLocal;
int iFields;
char szLine[81];
if (pFile == NULL) {
return FALSE;
}
/* Clean the table first */
(void)memset(atCharTable, 0, sizeof(atCharTable));
/* Fill the table */
while (fgets(szLine, (int)sizeof(szLine), pFile)) {
if (szLine[0] == '#' ||
szLine[0] == '\r' ||
szLine[0] == '\n') {
/* Comment or empty line */
continue;
}
iFields = sscanf(szLine, "%x %lx %*s", &uiLocal, &ulUnicode);
if (iFields != 2) {
pcTmp = strchr(szLine, '\r');
if (pcTmp != NULL) {
*pcTmp = '\0';
}
pcTmp = strchr(szLine, '\n');
if (pcTmp != NULL) {
*pcTmp = '\0';
}
werr(0, "Syntax error in: '%s'", szLine);
continue;
}
if (uiLocal > 0xff || ulUnicode > 0xffff) {
werr(0, "Syntax error in: '%02x %04lx'",
uiLocal, ulUnicode);
continue;
}
/* Store only the relevant entries */
if (uiLocal != ulUnicode || uiLocal >= 0x80) {
atCharTable[tNextPosFree].ucLocal = (UCHAR)uiLocal;
atCharTable[tNextPosFree].usUnicode = (USHORT)ulUnicode;
tNextPosFree++;
}
if (tNextPosFree >= elementsof(atCharTable)) {
werr(0, "Too many entries in the character mapping "
"file. Ignoring the rest.");
break;
}
}
if (tNextPosFree != 0) {
DBG_HEX(atCharTable[0].usUnicode);
DBG_HEX(atCharTable[tNextPosFree - 1].usUnicode);
return TRUE;
} /* end of bReadCharacterMappingTable */
/*
* ulTranslateCharacters - Translate characters to local representation
*
* Translate all characters to local representation
*
* returns the translated character
*/
ULONG
ulTranslateCharacters(USHORT usChar, ULONG ulFileOffset, int iWordVersion,
conversion_type eConversionType, encoding_type eEncoding,
BOOL bUseMacCharSet)
{
const char_table_type *pTmp;
const USHORT *usCharSet;
usCharSet = NULL;
if (bUseMacCharSet) {
/* Macintosh character set */
usCharSet = usMacRoman;
} else if (iWordVersion == 0) {
/* DOS character set */
usCharSet = usCp850;
} else {
/* Windows character set */
switch (eEncoding) {
case encoding_latin_2:
usCharSet = usCp1250;
break;
case encoding_cyrillic:
usCharSet = usCp1251;
break;
case encoding_latin_1:
default:
usCharSet = usCp1252;
break;
}
}
fail(usCharSet == NULL);
if (usChar >= 0x80 && usChar <= 0x9f) {
/* Translate implementation defined characters */
usChar = usCharSet[usChar - 0x80];
} else if (iWordVersion < 8 && usChar >= 0xa0 && usChar <= 0xff) {
/* Translate old character set to Unixcode */
usChar = usCharSet[usChar - 0x80];
}
/* Microsoft Unicode to real Unicode */
if (usChar >= 0xf020 && usChar <= 0xf0ff) {
DBG_HEX_C(usPrivateArea[usChar - 0xf020] == 0x003f, usChar);
usChar = usPrivateArea[usChar - 0xf020];
}
/* Characters with a special meaning in Word */
switch (usChar) {
case IGNORE_CHARACTER:
case FOOTNOTE_SEPARATOR:
case FOOTNOTE_CONTINUATION:
case ANNOTATION:
case FRAME:
case LINE_FEED:
case WORD_SOFT_HYPHEN:
case UNICODE_HYPHENATION_POINT:
return IGNORE_CHARACTER;
case PICTURE:
case TABLE_SEPARATOR:
case TAB:
case HARD_RETURN:
case PAGE_BREAK:
case PAR_END:
case COLUMN_FEED:
return (ULONG)usChar;
case FOOTNOTE_OR_ENDNOTE:
NO_DBG_HEX(ulFileOffset);
switch (eGetNotetype(ulFileOffset)) {
case notetype_is_footnote:
return FOOTNOTE_CHAR;
case notetype_is_endnote:
return ENDNOTE_CHAR;
default:
return UNKNOWN_NOTE_CHAR;
}
case WORD_UNBREAKABLE_JOIN:
return (ULONG)OUR_UNBREAKABLE_JOIN;
default:
break;
}
if (eEncoding != encoding_utf_8) {
/* Latin characters in an oriental text */
if (usChar >= 0xff01 && usChar <= 0xff5e) {
usChar -= 0xfee0;
}
}
if (eEncoding == encoding_latin_1 &&
(eConversionType == conversion_ps ||
eConversionType == conversion_pdf)) {
/* Ugly, but it makes the PostScript and PDF look better */
switch (usChar) {
case UNICODE_ELLIPSIS:
return 140;
case UNICODE_TRADEMARK_SIGN:
return 141;
case UNICODE_PER_MILLE_SIGN:
return 142;
case UNICODE_BULLET:
case UNICODE_BULLET_OPERATOR:
case UNICODE_BLACK_CLUB_SUIT:
return 143;
case UNICODE_LEFT_SINGLE_QMARK:
return 144;
case UNICODE_RIGHT_SINGLE_QMARK:
return 145;
case UNICODE_SINGLE_LEFT_ANGLE_QMARK:
return 146;
case UNICODE_SINGLE_RIGHT_ANGLE_QMARK:
return 147;
case UNICODE_LEFT_DOUBLE_QMARK:
return 148;
case UNICODE_RIGHT_DOUBLE_QMARK:
return 149;
case UNICODE_DOUBLE_LOW_9_QMARK:
return 150;
case UNICODE_EN_DASH:
return 151;
case UNICODE_EM_DASH:
return 152;
case UNICODE_MINUS_SIGN:
return 153;
case UNICODE_CAPITAL_LIGATURE_OE:
return 154;
case UNICODE_SMALL_LIGATURE_OE:
return 155;
case UNICODE_DAGGER:
return 156;
case UNICODE_DOUBLE_DAGGER:
return 157;
case UNICODE_SMALL_LIGATURE_FI:
return 158;
case UNICODE_SMALL_LIGATURE_FL:
return 159;
default:
break;
}
}
if (eConversionType == conversion_pdf) {
if (eEncoding == encoding_latin_1) {
switch (usChar) {
case UNICODE_EURO_SIGN:
return 128;
default:
break;
}
} else if (eEncoding == encoding_latin_2) {
switch (usChar) {
case UNICODE_CAPITAL_D_WITH_STROKE:
case UNICODE_SMALL_D_WITH_STROKE:
return 0x3f;
default:
break;
}
}
}
if (usChar < 0x80) {
/* US ASCII */
if (usChar < 0x20 || usChar == 0x7f) {
/* Ignore control characters */
DBG_HEX(usChar);
DBG_FIXME();
return IGNORE_CHARACTER;
}
return (ULONG)usChar;
}
if (eEncoding == encoding_utf_8) {
/* No need to convert Unicode characters */
return (ULONG)usChar;
}
/* Unicode to local representation */
pTmp = pGetCharTableRecord(usChar);
if (pTmp != NULL) {
DBG_HEX_C(usChar >= 0x7f && usChar <= 0x9f, usChar);
return (ULONG)pTmp->ucLocal;
}
/* Fancy characters to simple US ASCII */
switch (usChar) {
case UNICODE_SMALL_F_HOOK:
return (ULONG)'f';
case UNICODE_GREEK_CAPITAL_CHI:
return (ULONG)'X';
case UNICODE_GREEK_SMALL_UPSILON:
return (ULONG)'v';
case UNICODE_MODIFIER_CIRCUMFLEX:
case UNICODE_UPWARDS_ARROW:
return (ULONG)'^';
case UNICODE_SMALL_TILDE:
case UNICODE_TILDE_OPERATOR:
return (ULONG)'~';
case UNICODE_EN_QUAD:
case UNICODE_EM_QUAD:
case UNICODE_EN_SPACE:
case UNICODE_EM_SPACE:
case UNICODE_THREE_PER_EM_SPACE:
case UNICODE_FOUR_PER_EM_SPACE:
case UNICODE_SIX_PER_EM_SPACE:
case UNICODE_FIGURE_SPACE:
case UNICODE_PUNCTUATION_SPACE:
case UNICODE_THIN_SPACE:
case UNICODE_NARROW_NO_BREAK_SPACE:
case UNICODE_LIGHT_SHADE:
case UNICODE_MEDIUM_SHADE:
case UNICODE_DARK_SHADE:
return (ULONG)' ';
case UNICODE_LEFT_DOUBLE_QMARK:
case UNICODE_RIGHT_DOUBLE_QMARK:
case UNICODE_DOUBLE_LOW_9_QMARK:
case UNICODE_DOUBLE_HIGH_REV_9_QMARK:
case UNICODE_DOUBLE_PRIME:
return (ULONG)'"';
case UNICODE_LEFT_SINGLE_QMARK:
case UNICODE_RIGHT_SINGLE_QMARK:
case UNICODE_SINGLE_LOW_9_QMARK:
case UNICODE_SINGLE_HIGH_REV_9_QMARK:
case UNICODE_PRIME:
return (ULONG)'\'';
case UNICODE_HYPHEN:
case UNICODE_NON_BREAKING_HYPHEN:
case UNICODE_FIGURE_DASH:
case UNICODE_EN_DASH:
case UNICODE_EM_DASH:
case UNICODE_HORIZONTAL_BAR:
case UNICODE_MINUS_SIGN:
case UNICODE_BD_LIGHT_HORIZONTAL:
case UNICODE_BD_DOUBLE_HORIZONTAL:
return (ULONG)'-';
case UNICODE_DOUBLE_VERTICAL_LINE:
case UNICODE_BD_LIGHT_VERTICAL:
case UNICODE_BD_DOUBLE_VERTICAL:
return (ULONG)'|';
case UNICODE_DOUBLE_LOW_LINE:
return (ULONG)'_';
case UNICODE_DAGGER:
return (ULONG)'+';
case UNICODE_DOUBLE_DAGGER:
return (ULONG)'#';
case UNICODE_BULLET:
case UNICODE_BULLET_OPERATOR:
case UNICODE_BLACK_CLUB_SUIT:
return (ULONG)ucGetBulletCharacter(eConversionType, eEncoding);
case UNICODE_ONE_DOT_LEADER:
case UNICODE_TWO_DOT_LEADER:
return (ULONG)'.';
case UNICODE_ELLIPSIS:
#if defined(__riscos)
return (ULONG)OUR_ELLIPSIS;
#else
if (ulFileOffset == 0) {
return (ULONG)OUR_ELLIPSIS;
}
return UNICODE_ELLIPSIS;
#endif /* __riscos */
case UNICODE_DOUBLE_LEFT_ANGLE_QMARK:
case UNICODE_TRIANGULAR_BULLET:
case UNICODE_SINGLE_LEFT_ANGLE_QMARK:
case UNICODE_LEFTWARDS_ARROW:
return (ULONG)'<';
case UNICODE_DOUBLE_RIGHT_ANGLE_QMARK:
case UNICODE_SINGLE_RIGHT_ANGLE_QMARK:
case UNICODE_RIGHTWARDS_ARROW:
return (ULONG)'>';
case UNICODE_UNDERTIE:
return (ULONG)'-';
case UNICODE_N_ARY_SUMMATION:
return (ULONG)'S';
case UNICODE_EURO_SIGN:
return (ULONG)'E';
case UNICODE_CIRCLE:
case UNICODE_SQUARE:
return (ULONG)'O';
case UNICODE_DIAMOND:
return (ULONG)OUR_DIAMOND;
case UNICODE_NUMERO_SIGN:
return (ULONG)'N';
case UNICODE_KELVIN_SIGN:
return (ULONG)'K';
case UNICODE_DOWNWARDS_ARROW:
return (ULONG)'v';
case UNICODE_FRACTION_SLASH:
case UNICODE_DIVISION_SLASH:
return (ULONG)'/';
case UNICODE_ASTERISK_OPERATOR:
return (ULONG)'*';
case UNICODE_RATIO:
return (ULONG)':';
case UNICODE_BD_LIGHT_DOWN_RIGHT:
case UNICODE_BD_LIGHT_DOWN_AND_LEFT:
case UNICODE_BD_LIGHT_UP_AND_RIGHT:
case UNICODE_BD_LIGHT_UP_AND_LEFT:
case UNICODE_BD_LIGHT_VERTICAL_AND_RIGHT:
case UNICODE_BD_LIGHT_VERTICAL_AND_LEFT:
case UNICODE_BD_LIGHT_DOWN_AND_HORIZONTAL:
case UNICODE_BD_LIGHT_UP_AND_HORIZONTAL:
case UNICODE_BD_LIGHT_VERTICAL_AND_HORIZONTAL:
case UNICODE_BD_DOUBLE_DOWN_AND_RIGHT:
case UNICODE_BD_DOUBLE_DOWN_AND_LEFT:
case UNICODE_BD_DOUBLE_UP_AND_RIGHT:
case UNICODE_BD_DOUBLE_UP_AND_LEFT:
case UNICODE_BD_DOUBLE_VERTICAL_AND_RIGHT:
case UNICODE_BD_DOUBLE_VERTICAL_AND_LEFT:
case UNICODE_BD_DOUBLE_DOWN_AND_HORIZONTAL:
case UNICODE_BD_DOUBLE_UP_AND_HORIZONTAL:
case UNICODE_BD_DOUBLE_VERTICAL_AND_HORIZONTAL:
case UNICODE_BLACK_SQUARE:
return (ULONG)'+';
case UNICODE_HAIR_SPACE:
case UNICODE_ZERO_WIDTH_SPACE:
case UNICODE_ZERO_WIDTH_NON_JOINER:
case UNICODE_ZERO_WIDTH_JOINER:
case UNICODE_LEFT_TO_RIGHT_MARK:
case UNICODE_RIGHT_TO_LEFT_MARK:
case UNICODE_LEFT_TO_RIGHT_EMBEDDING:
case UNICODE_RIGHT_TO_LEFT_EMBEDDING:
case UNICODE_POP_DIRECTIONAL_FORMATTING:
case UNICODE_LEFT_TO_RIGHT_OVERRIDE:
case UNICODE_RIGHT_TO_LEFT_OVERRIDE:
case UNICODE_ZERO_WIDTH_NO_BREAK_SPACE:
return IGNORE_CHARACTER;
default:
break;
}
if (usChar == UNICODE_TRADEMARK_SIGN) {
/*
* No local representation, it doesn't look like anything in
* US-ASCII and a question mark does more harm than good.
*/
return IGNORE_CHARACTER;
}
if (usChar >= 0xa0 && usChar <= 0xff) {
/* Before Word 97, Word did't use Unicode */
return (ULONG)usChar;
}
/* Untranslated Unicode character */
return 0x3f;
} /* end of ulTranslateCharacters */
/*
* ulToUpper - convert letter to upper case
*
* This function converts a letter to upper case. Unlike toupper(3) this
* function is independent from the settings of locale. This comes in handy
* for people who have to read Word documents in more than one language or
* contain more than one language.
*
* returns the converted letter, or ulChar if the conversion was not possible.
*/
ULONG
ulToUpper(ULONG ulChar)
{
if (ulChar < 0x80) {
/* US ASCII: use standard function */
return (ULONG)toupper((int)ulChar);
}
if (ulChar >= 0xe0 && ulChar <= 0xfe && ulChar != 0xf7) {
/*
* Lower case accented characters
* 0xf7 is Division sign; 0xd7 is Multiplication sign
* 0xff is y with diaeresis; 0xdf is Sharp s
*/
return ulChar & ~0x20;
}
#if defined(__STDC_ISO_10646__)
/*
* If this is ISO C99 and all locales have wchar_t = ISO 10646
* (e.g., glibc 2.2 or newer), then use standard function
*/
if (ulChar > 0xff) {
return (ULONG)towupper((wint_t)ulChar);
}
#endif /* __STDC_ISO_10646__ */
return ulChar;
} /* end of ulToUpper */
