chartrans.c

这是一个同样来自贝尔实验室的和UNIX有着渊源的操作系统, 其简洁的设计和实现易于我们学习和理解

			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;
	}

	DBG_HEX_C(usChar < 0x3000 || usChar >= 0xd800, ulFileOffset);
	DBG_HEX_C(usChar < 0x3000 || usChar >= 0xd800, usChar);
	DBG_MSG_C(usChar >= 0xe000 && usChar < 0xf900, "Private Use Area");

	/* 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 */