TB9.2 Example Applications: examples/PIPS/antiword/src/chartrans.c Source File

00001 /*
-00002  * chartrans.c
-00003  * Copyright (C) 1999-2004 A.J. van Os; Released under GNU GPL
-00004  *
-00005  * Description:
-00006  * Translate Word characters to local representation
-00007  */
-00008 
-00009 #include <stdlib.h>
-00010 #include <string.h>
-00011 #include <ctype.h>
-00012 #if defined(__STDC_ISO_10646__)
-00013 #include <wctype.h>
-00014 #endif /* __STDC_ISO_10646__ */
-00015 #include "antiword.h"
-00016 
-00017 static const USHORT usCp850[] = {       /* DOS implementation of Latin1 */
-00018         0x00c7, 0x00fc, 0x00e9, 0x00e2, 0x00e4, 0x00e0, 0x00e5, 0x00e7,
-00019         0x00ea, 0x00eb, 0x00e8, 0x00ef, 0x00ee, 0x00ec, 0x00c4, 0x00c5,
-00020         0x00c9, 0x00e6, 0x00c6, 0x00f4, 0x00f6, 0x00f2, 0x00fb, 0x00f9,
-00021         0x00ff, 0x00d6, 0x00dc, 0x00f8, 0x00a3, 0x00d8, 0x00d7, 0x0192,
-00022         0x00e1, 0x00ed, 0x00f3, 0x00fa, 0x00f1, 0x00d1, 0x00aa, 0x00ba,
-00023         0x00bf, 0x00ae, 0x00ac, 0x00bd, 0x00bc, 0x00a1, 0x00ab, 0x00bb,
-00024         0x2591, 0x2592, 0x2593, 0x2502, 0x2524, 0x00c1, 0x00c2, 0x00c0,
-00025         0x00a9, 0x2563, 0x2551, 0x2557, 0x255d, 0x00a2, 0x00a5, 0x2510,
-00026         0x2514, 0x2534, 0x252c, 0x251c, 0x2500, 0x253c, 0x00e3, 0x00c3,
-00027         0x255a, 0x2554, 0x2569, 0x2566, 0x2560, 0x2550, 0x256c, 0x00a4,
-00028         0x00f0, 0x00d0, 0x00ca, 0x00cb, 0x00c8, 0x0131, 0x00cd, 0x00ce,
-00029         0x00cf, 0x2518, 0x250c, 0x2588, 0x2584, 0x00a6, 0x00cc, 0x2580,
-00030         0x00d3, 0x00df, 0x00d4, 0x00d2, 0x00f5, 0x00d5, 0x00b5, 0x00fe,
-00031         0x00de, 0x00da, 0x00db, 0x00d9, 0x00fd, 0x00dd, 0x00af, 0x00b4,
-00032         0x00ad, 0x00b1, 0x2017, 0x00be, 0x00b6, 0x00a7, 0x00f7, 0x00b8,
-00033         0x00b0, 0x00a8, 0x00b7, 0x00b9, 0x00b3, 0x00b2, 0x25a0, 0x00a0,
-00034 };
-00035 
-00036 static const USHORT usCp1250[] = {      /* Windows implementation of Latin2 */
-00037         0x20ac, 0x003f, 0x201a, 0x003f, 0x201e, 0x2026, 0x2020, 0x2021,
-00038         0x003f, 0x2030, 0x0160, 0x2039, 0x015a, 0x0164, 0x017d, 0x0179,
-00039         0x003f, 0x2018, 0x2019, 0x201c, 0x201d, 0x2022, 0x2013, 0x2014,
-00040         0x003f, 0x2122, 0x0161, 0x203a, 0x015b, 0x0165, 0x017e, 0x017a,
-00041         0x00a0, 0x02c7, 0x02d8, 0x0141, 0x00a4, 0x0104, 0x00a6, 0x00a7,
-00042         0x00a8, 0x00a9, 0x015e, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x017b,
-00043         0x00b0, 0x00b1, 0x02db, 0x0142, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
-00044         0x00b8, 0x0105, 0x015f, 0x00bb, 0x013d, 0x02dd, 0x013e, 0x017c,
-00045         0x0154, 0x00c1, 0x00c2, 0x0102, 0x00c4, 0x0139, 0x0106, 0x00c7,
-00046         0x010c, 0x00c9, 0x0118, 0x00cb, 0x011a, 0x00cd, 0x00ce, 0x010e,
-00047         0x0110, 0x0143, 0x0147, 0x00d3, 0x00d4, 0x0150, 0x00d6, 0x00d7,
-00048         0x0158, 0x016e, 0x00da, 0x0170, 0x00dc, 0x00dd, 0x0162, 0x00df,
-00049         0x0155, 0x00e1, 0x00e2, 0x0103, 0x00e4, 0x013a, 0x0107, 0x00e7,
-00050         0x010d, 0x00e9, 0x0119, 0x00eb, 0x011b, 0x00ed, 0x00ee, 0x010f,
-00051         0x0111, 0x0144, 0x0148, 0x00f3, 0x00f4, 0x0151, 0x00f6, 0x00f7,
-00052         0x0159, 0x016f, 0x00fa, 0x0171, 0x00fc, 0x00fd, 0x0163, 0x02d9,
-00053 };
-00054 
-00055 static const USHORT usCp1251[] = {      /* Windows implementation of Cyrillic */
-00056         0x0402, 0x0403, 0x201a, 0x0453, 0x201e, 0x2026, 0x2020, 0x2021,
-00057         0x20ac, 0x2030, 0x0409, 0x2039, 0x040a, 0x040c, 0x040b, 0x040f,
-00058         0x0452, 0x2018, 0x2019, 0x201c, 0x201d, 0x2022, 0x2013, 0x2014,
-00059         0x00f3, 0x2122, 0x0459, 0x203a, 0x045a, 0x045c, 0x045b, 0x045f,
-00060         0x00a0, 0x040e, 0x045e, 0x0408, 0x00a4, 0x0490, 0x00a6, 0x00a7,
-00061         0x0401, 0x00a9, 0x0404, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x0407,
-00062         0x00b0, 0x00b1, 0x0406, 0x0456, 0x0491, 0x00b5, 0x00b6, 0x00b7,
-00063         0x0451, 0x2116, 0x0454, 0x00bb, 0x0458, 0x0405, 0x0455, 0x0457,
-00064         0x0410, 0x0411, 0x0412, 0x0413, 0x0414, 0x0415, 0x0416, 0x0417,
-00065         0x0418, 0x0419, 0x041a, 0x041b, 0x041c, 0x041d, 0x041e, 0x041f,
-00066         0x0420, 0x0421, 0x0422, 0x0423, 0x0424, 0x0425, 0x0426, 0x0427,
-00067         0x0428, 0x0429, 0x042a, 0x042b, 0x042c, 0x042d, 0x042e, 0x042f,
-00068         0x0430, 0x0431, 0x0432, 0x0433, 0x0434, 0x0435, 0x0436, 0x0437,
-00069         0x0438, 0x0439, 0x043a, 0x043b, 0x043c, 0x043d, 0x043e, 0x043f,
-00070         0x0440, 0x0441, 0x0442, 0x0443, 0x0444, 0x0445, 0x0446, 0x0447,
-00071         0x0448, 0x0449, 0x044a, 0x044b, 0x044c, 0x044d, 0x044e, 0x044f,
-00072 };
-00073 
-00074 static const USHORT usCp1252[] = {      /* Windows implementation of Latin1 */
-00075         0x20ac, 0x003f, 0x201a, 0x0192, 0x201e, 0x2026, 0x2020, 0x2021,
-00076         0x02c6, 0x2030, 0x0160, 0x2039, 0x0152, 0x003f, 0x017d, 0x003f,
-00077         0x003f, 0x2018, 0x2019, 0x201c, 0x201d, 0x2022, 0x2013, 0x2014,
-00078         0x02dc, 0x2122, 0x0161, 0x203a, 0x0153, 0x003f, 0x017e, 0x0178,
-00079         0x00a0, 0x00a1, 0x00a2, 0x00a3, 0x00a4, 0x00a5, 0x00a6, 0x00a7,
-00080         0x00a8, 0x00a9, 0x00aa, 0x00ab, 0x00ac, 0x00ad, 0x00ae, 0x00af,
-00081         0x00b0, 0x00b1, 0x00b2, 0x00b3, 0x00b4, 0x00b5, 0x00b6, 0x00b7,
-00082         0x00b8, 0x00b9, 0x00ba, 0x00bb, 0x00bc, 0x00bd, 0x00be, 0x00bf,
-00083         0x00c0, 0x00c1, 0x00c2, 0x00c3, 0x00c4, 0x00c5, 0x00c6, 0x00c7,
-00084         0x00c8, 0x00c9, 0x00ca, 0x00cb, 0x00cc, 0x00cd, 0x00ce, 0x00cf,
-00085         0x00d0, 0x00d1, 0x00d2, 0x00d3, 0x00d4, 0x00d5, 0x00d6, 0x00d7,
-00086         0x00d8, 0x00d9, 0x00da, 0x00db, 0x00dc, 0x00dd, 0x00de, 0x00df,
-00087         0x00e0, 0x00e1, 0x00e2, 0x00e3, 0x00e4, 0x00e5, 0x00e6, 0x00e7,
-00088         0x00e8, 0x00e9, 0x00ea, 0x00eb, 0x00ec, 0x00ed, 0x00ee, 0x00ef,
-00089         0x00f0, 0x00f1, 0x00f2, 0x00f3, 0x00f4, 0x00f5, 0x00f6, 0x00f7,
-00090         0x00f8, 0x00f9, 0x00fa, 0x00fb, 0x00fc, 0x00fd, 0x00fe, 0x00ff,
-00091 };
-00092 
-00093 static const USHORT usMacRoman[] = {    /* Apple implementation of Latin1 */
-00094         0x00c4, 0x00c5, 0x00c7, 0x00c9, 0x00d1, 0x00d6, 0x00dc, 0x00e1,
-00095         0x00e0, 0x00e2, 0x00e4, 0x00e3, 0x00e5, 0x00e7, 0x00e9, 0x00e8,
-00096         0x00ea, 0x00eb, 0x00ed, 0x00ec, 0x00ee, 0x00ef, 0x00f1, 0x00f3,
-00097         0x00f2, 0x00f4, 0x00f6, 0x00f5, 0x00fa, 0x00f9, 0x00fb, 0x00fc,
-00098         0x2020, 0x00b0, 0x00a2, 0x00a3, 0x00a7, 0x2022, 0x00b6, 0x00df,
-00099         0x00ae, 0x00a9, 0x2122, 0x00b4, 0x00a8, 0x2260, 0x00c6, 0x00d8,
-00100         0x221e, 0x00b1, 0x2264, 0x2265, 0x00a5, 0x00b5, 0x2202, 0x2211,
-00101         0x220f, 0x03c0, 0x222b, 0x00aa, 0x00ba, 0x2126, 0x00e6, 0x00f8,
-00102         0x00bf, 0x00a1, 0x00ac, 0x221a, 0x0192, 0x2248, 0x2206, 0x00ab,
-00103         0x00bb, 0x2026, 0x00a0, 0x00c0, 0x00c3, 0x00d5, 0x0152, 0x0153,
-00104         0x2013, 0x2014, 0x201c, 0x201d, 0x2018, 0x2019, 0x00f7, 0x25ca,
-00105         0x00ff, 0x0178, 0x2044, 0x00a4, 0x2039, 0x203a, 0xfb01, 0xfb02,
-00106         0x2021, 0x00b7, 0x201a, 0x201e, 0x2030, 0x00c2, 0x00ca, 0x00c1,
-00107         0x00cb, 0x00c8, 0x00cd, 0x00ce, 0x00cf, 0x00cc, 0x00d3, 0x00d4,
-00108         0x003f, 0x00d2, 0x00da, 0x00db, 0x00d9, 0x0131, 0x02c6, 0x02dc,
-00109         0x00af, 0x02d8, 0x02d9, 0x02da, 0x00b8, 0x02dd, 0x02db, 0x02c7,
-00110 };
-00111 
-00112 static const USHORT usPrivateArea[] = {
-00113         0x0020, 0x0021, 0x2200, 0x0023, 0x2203, 0x0025, 0x0026, 0x220d,
-00114         0x0028, 0x0029, 0x2217, 0x002b, 0x002c, 0x2212, 0x002e, 0x002f,
-00115         0x0030, 0x0031, 0x0032, 0x0033, 0x0034, 0x0035, 0x0036, 0x0037,
-00116         0x0038, 0x0039, 0x003a, 0x003b, 0x003c, 0x2019, 0x003e, 0x003f,
-00117         0x201d, 0x201c, 0x0392, 0x03a7, 0x0394, 0x0395, 0x03a6, 0x0393,
-00118         0x0397, 0x0399, 0x03d1, 0x039a, 0x039b, 0x039c, 0x039d, 0x039f,
-00119         0x03a0, 0x0398, 0x03a1, 0x03a3, 0x03a4, 0x03a5, 0x03c2, 0x03a9,
-00120         0x039e, 0x03a8, 0x0396, 0x005b, 0x2234, 0x005d, 0x22a5, 0x005f,
-00121         0x003f, 0x03b1, 0x03b2, 0x03c7, 0x03b4, 0x03b5, 0x03c6, 0x03b3,
-00122         0x03b7, 0x03b9, 0x03d5, 0x03ba, 0x03bb, 0x03bc, 0x03bd, 0x03bf,
-00123         0x03c0, 0x03b8, 0x03c1, 0x03c3, 0x03c4, 0x03c5, 0x03d6, 0x03c9,
-00124         0x03be, 0x03c8, 0x03b6, 0x007b, 0x007c, 0x007d, 0x223c, 0x003f,
-00125         0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f,
-00126         0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f,
-00127         0x003f, 0x003f, 0x003f, 0x2022, 0x003f, 0x003f, 0x003f, 0x003f,
-00128         0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f, 0x003f,
-00129         0x20ac, 0x03d2, 0x2032, 0x2264, 0x2044, 0x221e, 0x0192, 0x2663,
-00130         0x2666, 0x2665, 0x2660, 0x2194, 0x2190, 0x2191, 0x2192, 0x2193,
-00131         0x00b0, 0x00b1, 0x2033, 0x2265, 0x00d7, 0x221d, 0x2202, 0x2022,
-00132         0x00f7, 0x2260, 0x2261, 0x2248, 0x2026, 0x007c, 0x23af, 0x21b5,
-00133         0x2135, 0x2111, 0x211c, 0x2118, 0x2297, 0x2295, 0x2205, 0x2229,
-00134         0x222a, 0x2283, 0x2287, 0x2284, 0x2282, 0x2286, 0x2208, 0x2209,
-00135         0x2220, 0x2207, 0x00ae, 0x00a9, 0x2122, 0x220f, 0x221a, 0x22c5,
-00136         0x00ac, 0x2227, 0x2228, 0x21d4, 0x21d0, 0x21d1, 0x21d2, 0x21d3,
-00137         0x22c4, 0x3008, 0x00ae, 0x00a9, 0x2122, 0x2211, 0x239b, 0x239c,
-00138         0x239d, 0x23a1, 0x23a2, 0x23a3, 0x23a7, 0x23a8, 0x23a9, 0x23aa,
-00139         0x003f, 0x3009, 0x222b, 0x2320, 0x23ae, 0x2321, 0x239e, 0x239f,
-00140         0x23a0, 0x23a4, 0x23a5, 0x23a6, 0x23ab, 0x23ac, 0x23ad, 0x003f,
-00141 };
-00142 
-00143 typedef struct char_table_tag {
-00144         UCHAR   ucLocal;
-00145         USHORT  usUnicode;
-00146 } char_table_type;
-00147 
-00148 static char_table_type  atCharTable[256];
-00149 static size_t           tNextPosFree = 0;
-00150 
-00151 
-00152 /*
-00153  * iCompare - compare two records
-00154  *
-00155  * Compares two records. For use by qsort(3C) and bsearch(3C).
-00156  *
-00157  * returns -1 if rec1 < rec2, 0 if rec1 == rec2, 1 if rec1 > rec2
-00158  */
-00159 static int
-00160 iCompare(const void *pvRecord1, const void *pvRecord2)
-00161 {
-00162         USHORT  usUnicode1, usUnicode2;
-00163 
-00164         usUnicode1 = ((char_table_type *)pvRecord1)->usUnicode;
-00165         usUnicode2 = ((char_table_type *)pvRecord2)->usUnicode;
-00166 
-00167         if (usUnicode1 < usUnicode2) {
-00168                 return -1;
-00169         }
-00170         if (usUnicode1 > usUnicode2) {
-00171                 return 1;
-00172         }
-00173         return 0;
-00174 } /* end of iCompare */
-00175 
-00176 /*
-00177  * pGetCharTableRecord - get the character table record
-00178  *
-00179  * returns a pointer to the record when found, otherwise NULL
-00180  */
-00181 static const char_table_type *
-00182 pGetCharTableRecord(USHORT usUnicode)
-00183 {
-00184         char_table_type tKey;
-00185 
-00186         if (tNextPosFree == 0) {
-00187                 return NULL;
-00188         }
-00189         tKey.usUnicode = usUnicode;
-00190         tKey.ucLocal = 0;
-00191         return (char_table_type *)bsearch(&tKey,
-00192                         atCharTable,
-00193                         tNextPosFree, sizeof(atCharTable[0]),
-00194                         iCompare);
-00195 } /* end of pGetCharTableRecord */
-00196 
-00197 /*
-00198  * ucGetBulletCharacter - get the local representation of the bullet
-00199  */
-00200 UCHAR
-00201 ucGetBulletCharacter(conversion_type eConversionType, encoding_type eEncoding)
-00202 {
-00203 #if defined(__riscos)
-00204         return 0x8f;
-00205 #else
-00206         const char_table_type   *pRec;
-00207 
-00208         fail(eEncoding == encoding_utf_8);
-00209 
-00210         if (eEncoding == encoding_latin_1 &&
-00211             (eConversionType == conversion_ps ||
-00212              eConversionType == conversion_pdf)) {
-00213                 /* Ugly, but it makes the PostScript and PDF look better */
-00214                 return (UCHAR)143;
-00215         }
-00216         if (eConversionType != conversion_text &&
-00217             eConversionType != conversion_fmt_text) {
-00218                 pRec = pGetCharTableRecord(UNICODE_BULLET);
-00219                 if (pRec != NULL) {
-00220                         return pRec->ucLocal;
-00221                 }
-00222                 pRec = pGetCharTableRecord(UNICODE_BULLET_OPERATOR);
-00223                 if (pRec != NULL) {
-00224                         return pRec->ucLocal;
-00225                 }
-00226                 pRec = pGetCharTableRecord(UNICODE_MIDDLE_DOT);
-00227                 if (pRec != NULL) {
-00228                         return pRec->ucLocal;
-00229                 }
-00230         }
-00231         return (UCHAR)'.';
-00232 #endif /* __riscos */
-00233 } /* end of ucGetBulletCharacter */
-00234 
-00235 /*
-00236  * ucGetNbspCharacter - get the local representation of the non-breaking space
-00237  */
-00238 UCHAR
-00239 ucGetNbspCharacter(void)
-00240 {
-00241         const char_table_type   *pRec;
-00242 
-00243         pRec = pGetCharTableRecord(0x00a0);     /* Unicode non-breaking space */
-00244         if (pRec == NULL) {
-00245                 DBG_MSG("Non-breaking space record not found");
-00246                 /* No value found, use the best guess */
-00247                 return (UCHAR)0xa0;
-00248         }
-00249         return pRec->ucLocal;
-00250 } /* end of ucGetNbspCharacter */
-00251 
-00252 /*
-00253  * bReadCharacterMappingTable - read the mapping table
-00254  *
-00255  * Read the character mapping table from file and have the contents sorted
-00256  *
-00257  * returns TRUE if successful, otherwise FALSE
-00258  */
-00259 BOOL
-00260 bReadCharacterMappingTable(FILE *pFile)
-00261 {
-00262         char    *pcTmp;
-00263         ULONG   ulUnicode;
-00264         UINT    uiLocal;
-00265         int     iFields;
-00266         char    szLine[81];
-00267 
-00268         if (pFile == NULL) {
-00269                 return FALSE;
-00270         }
-00271 
-00272         /* Clean the table first */
-00273         (void)memset(atCharTable, 0, sizeof(atCharTable));
-00274 
-00275         /* Fill the table */
-00276         while (fgets(szLine, (int)sizeof(szLine), pFile)) {
-00277                 if (szLine[0] == '#' ||
-00278                     szLine[0] == '\r' ||
-00279                     szLine[0] == '\n') {
-00280                         /* Comment or empty line */
-00281                         continue;
-00282                 }
-00283                 iFields = sscanf(szLine, "%x %lx %*s", &uiLocal, &ulUnicode);
-00284                 if (iFields != 2) {
-00285                         pcTmp = strchr(szLine, '\r');
-00286                         if (pcTmp != NULL) {
-00287                                 *pcTmp = '\0';
-00288                         }
-00289                         pcTmp = strchr(szLine, '\n');
-00290                         if (pcTmp != NULL) {
-00291                                 *pcTmp = '\0';
-00292                         }
-00293                         werr(0, "Syntax error in: '%s'", szLine);
-00294                         continue;
-00295                 }
-00296                 if (uiLocal > 0xff || ulUnicode > 0xffff) {
-00297                         werr(0, "Syntax error in: '%02x %04lx'",
-00298                                         uiLocal, ulUnicode);
-00299                         continue;
-00300                 }
-00301                 /* Store only the relevant entries */
-00302                 if (uiLocal != ulUnicode || uiLocal >= 0x80) {
-00303                         atCharTable[tNextPosFree].ucLocal = (UCHAR)uiLocal;
-00304                         atCharTable[tNextPosFree].usUnicode = (USHORT)ulUnicode;
-00305                         tNextPosFree++;
-00306                 }
-00307                 if (tNextPosFree >= elementsof(atCharTable)) {
-00308                         werr(0, "Too many entries in the character mapping "
-00309                                 "file. Ignoring the rest.");
-00310                         break;
-00311                 }
-00312         }
-00313 
-00314         if (tNextPosFree != 0) {
-00315                 DBG_HEX(atCharTable[0].usUnicode);
-00316                 DBG_HEX(atCharTable[tNextPosFree - 1].usUnicode);
-00317 
-00318                 qsort(atCharTable,
-00319                         tNextPosFree, sizeof(atCharTable[0]),
-00320                         iCompare);
-00321 
-00322                 DBG_HEX(atCharTable[0].usUnicode);
-00323                 DBG_HEX(atCharTable[tNextPosFree - 1].usUnicode);
-00324         }
-00325 
-00326         return TRUE;
-00327 } /* end of bReadCharacterMappingTable */
-00328 
-00329 /*
-00330  * ulTranslateCharacters - Translate characters to local representation
-00331  *
-00332  * Translate all characters to local representation
-00333  *
-00334  * returns the translated character
-00335  */
-00336 ULONG
-00337 ulTranslateCharacters(USHORT usChar, ULONG ulFileOffset, int iWordVersion,
-00338         conversion_type eConversionType, encoding_type eEncoding,
-00339         BOOL bUseMacCharSet)
-00340 {
-00341         const char_table_type   *pTmp;
-00342         const USHORT    *usCharSet;
-00343 
-00344         usCharSet = NULL;
-00345         if (bUseMacCharSet) {
-00346                 /* Macintosh character set */
-00347                 usCharSet = usMacRoman;
-00348         } else if (iWordVersion == 0) {
-00349                 /* DOS character set */
-00350                 usCharSet = usCp850;
-00351         } else {
-00352                 /* Windows character set */
-00353                 switch (eEncoding) {
-00354                 case encoding_latin_2:
-00355                         usCharSet = usCp1250;
-00356                         break;
-00357                 case encoding_cyrillic:
-00358                         usCharSet = usCp1251;
-00359                         break;
-00360                 case encoding_latin_1:
-00361                 default:
-00362                         usCharSet = usCp1252;
-00363                         break;
-00364                 }
-00365         }
-00366         fail(usCharSet == NULL);
-00367         if (usChar >= 0x80 && usChar <= 0x9f) {
-00368                 /* Translate implementation defined characters */
-00369                 usChar = usCharSet[usChar - 0x80];
-00370         } else if (iWordVersion < 8 && usChar >= 0xa0 && usChar <= 0xff) {
-00371                 /* Translate old character set to Unixcode */
-00372                 usChar = usCharSet[usChar - 0x80];
-00373         }
-00374 
-00375         /* Microsoft Unicode to real Unicode */
-00376         if (usChar >= 0xf020 && usChar <= 0xf0ff) {
-00377                 DBG_HEX_C(usPrivateArea[usChar - 0xf020] == 0x003f, usChar);
-00378                 usChar = usPrivateArea[usChar - 0xf020];
-00379         }
-00380 
-00381         /* Characters with a special meaning in Word */
-00382         switch (usChar) {
-00383         case IGNORE_CHARACTER:
-00384         case FOOTNOTE_SEPARATOR:
-00385         case FOOTNOTE_CONTINUATION:
-00386         case ANNOTATION:
-00387         case FRAME:
-00388         case LINE_FEED:
-00389         case WORD_SOFT_HYPHEN:
-00390         case UNICODE_HYPHENATION_POINT:
-00391                 return IGNORE_CHARACTER;
-00392         case PICTURE:
-00393         case TABLE_SEPARATOR:
-00394         case TAB:
-00395         case HARD_RETURN:
-00396         case PAGE_BREAK:
-00397         case PAR_END:
-00398         case COLUMN_FEED:
-00399                 return (ULONG)usChar;
-00400         case FOOTNOTE_OR_ENDNOTE:
-00401                 NO_DBG_HEX(ulFileOffset);
-00402                 switch (eGetNotetype(ulFileOffset)) {
-00403                 case notetype_is_footnote:
-00404                         return FOOTNOTE_CHAR;
-00405                 case notetype_is_endnote:
-00406                         return ENDNOTE_CHAR;
-00407                 default:
-00408                         return UNKNOWN_NOTE_CHAR;
-00409                 }
-00410         case WORD_UNBREAKABLE_JOIN:
-00411                 return (ULONG)OUR_UNBREAKABLE_JOIN;
-00412         default:
-00413                 break;
-00414         }
-00415 
-00416         if (eEncoding != encoding_utf_8) {
-00417                 /* Latin characters in an oriental text */
-00418                 if (usChar >= 0xff01 && usChar <= 0xff5e) {
-00419                         usChar -= 0xfee0;
-00420                 }
-00421         }
-00422 
-00423         if (eEncoding == encoding_latin_1 &&
-00424             (eConversionType == conversion_ps ||
-00425              eConversionType == conversion_pdf)) {
-00426                 /* Ugly, but it makes the PostScript and PDF look better */
-00427                 switch (usChar) {
-00428                 case UNICODE_ELLIPSIS:
-00429                         return 140;
-00430                 case UNICODE_TRADEMARK_SIGN:
-00431                         return 141;
-00432                 case UNICODE_PER_MILLE_SIGN:
-00433                         return 142;
-00434                 case UNICODE_BULLET:
-00435                 case UNICODE_BULLET_OPERATOR:
-00436                 case UNICODE_BLACK_CLUB_SUIT:
-00437                         return 143;
-00438                 case UNICODE_LEFT_SINGLE_QMARK:
-00439                         return 144;
-00440                 case UNICODE_RIGHT_SINGLE_QMARK:
-00441                         return 145;
-00442                 case UNICODE_SINGLE_LEFT_ANGLE_QMARK:
-00443                         return 146;
-00444                 case UNICODE_SINGLE_RIGHT_ANGLE_QMARK:
-00445                         return 147;
-00446                 case UNICODE_LEFT_DOUBLE_QMARK:
-00447                         return 148;
-00448                 case UNICODE_RIGHT_DOUBLE_QMARK:
-00449                         return 149;
-00450                 case UNICODE_DOUBLE_LOW_9_QMARK:
-00451                         return 150;
-00452                 case UNICODE_EN_DASH:
-00453                         return 151;
-00454                 case UNICODE_EM_DASH:
-00455                         return 152;
-00456                 case UNICODE_MINUS_SIGN:
-00457                         return 153;
-00458                 case UNICODE_CAPITAL_LIGATURE_OE:
-00459                         return 154;
-00460                 case UNICODE_SMALL_LIGATURE_OE:
-00461                         return 155;
-00462                 case UNICODE_DAGGER:
-00463                         return 156;
-00464                 case UNICODE_DOUBLE_DAGGER:
-00465                         return 157;
-00466                 case UNICODE_SMALL_LIGATURE_FI:
-00467                         return 158;
-00468                 case UNICODE_SMALL_LIGATURE_FL:
-00469                         return 159;
-00470                 default:
-00471                         break;
-00472                 }
-00473         }
-00474 
-00475         if (eConversionType == conversion_pdf) {
-00476                 if (eEncoding == encoding_latin_1) {
-00477                         switch (usChar) {
-00478                         case UNICODE_EURO_SIGN:
-00479                                 return 128;
-00480                         default:
-00481                                 break;
-00482                         }
-00483                 } else if (eEncoding == encoding_latin_2) {
-00484                         switch (usChar) {
-00485                         case UNICODE_CAPITAL_D_WITH_STROKE:
-00486                         case UNICODE_SMALL_D_WITH_STROKE:
-00487                                 return 0x3f;
-00488                         default:
-00489                                 break;
-00490                         }
-00491                 }
-00492         }
-00493 
-00494         if (usChar < 0x80) {
-00495                 /* US ASCII */
-00496                 if (usChar < 0x20 || usChar == 0x7f) {
-00497                         /* Ignore control characters */
-00498                         DBG_HEX(usChar);
-00499                         DBG_FIXME();
-00500                         return IGNORE_CHARACTER;
-00501                 }
-00502                 return (ULONG)usChar;
-00503         }
-00504 
-00505         if (eEncoding == encoding_utf_8) {
-00506                 /* No need to convert Unicode characters */
-00507                 return (ULONG)usChar;
-00508         }
-00509 
-00510         /* Unicode to local representation */
-00511         pTmp = pGetCharTableRecord(usChar);
-00512         if (pTmp != NULL) {
-00513                 DBG_HEX_C(usChar >= 0x7f && usChar <= 0x9f, usChar);
-00514                 return (ULONG)pTmp->ucLocal;
-00515         }
-00516 
-00517         /* Fancy characters to simple US ASCII */
-00518         switch (usChar) {
-00519         case UNICODE_SMALL_F_HOOK:
-00520                 return (ULONG)'f';
-00521         case UNICODE_GREEK_CAPITAL_CHI:
-00522                 return (ULONG)'X';
-00523         case UNICODE_GREEK_SMALL_UPSILON:
-00524                 return (ULONG)'v';
-00525         case UNICODE_MODIFIER_CIRCUMFLEX:
-00526         case UNICODE_UPWARDS_ARROW:
-00527                 return (ULONG)'^';
-00528         case UNICODE_SMALL_TILDE:
-00529         case UNICODE_TILDE_OPERATOR:
-00530                 return (ULONG)'~';
-00531         case UNICODE_EN_QUAD:
-00532         case UNICODE_EM_QUAD:
-00533         case UNICODE_EN_SPACE:
-00534         case UNICODE_EM_SPACE:
-00535         case UNICODE_THREE_PER_EM_SPACE:
-00536         case UNICODE_FOUR_PER_EM_SPACE:
-00537         case UNICODE_SIX_PER_EM_SPACE:
-00538         case UNICODE_FIGURE_SPACE:
-00539         case UNICODE_PUNCTUATION_SPACE:
-00540         case UNICODE_THIN_SPACE:
-00541         case UNICODE_NARROW_NO_BREAK_SPACE:
-00542         case UNICODE_LIGHT_SHADE:
-00543         case UNICODE_MEDIUM_SHADE:
-00544         case UNICODE_DARK_SHADE:
-00545                 return (ULONG)' ';
-00546         case UNICODE_LEFT_DOUBLE_QMARK:
-00547         case UNICODE_RIGHT_DOUBLE_QMARK:
-00548         case UNICODE_DOUBLE_LOW_9_QMARK:
-00549         case UNICODE_DOUBLE_HIGH_REV_9_QMARK:
-00550         case UNICODE_DOUBLE_PRIME:
-00551                 return (ULONG)'"';
-00552         case UNICODE_LEFT_SINGLE_QMARK:
-00553         case UNICODE_RIGHT_SINGLE_QMARK:
-00554         case UNICODE_SINGLE_LOW_9_QMARK:
-00555         case UNICODE_SINGLE_HIGH_REV_9_QMARK:
-00556         case UNICODE_PRIME:
-00557                 return (ULONG)'\'';
-00558         case UNICODE_HYPHEN:
-00559         case UNICODE_NON_BREAKING_HYPHEN:
-00560         case UNICODE_FIGURE_DASH:
-00561         case UNICODE_EN_DASH:
-00562         case UNICODE_EM_DASH:
-00563         case UNICODE_HORIZONTAL_BAR:
-00564         case UNICODE_MINUS_SIGN:
-00565         case UNICODE_BD_LIGHT_HORIZONTAL:
-00566         case UNICODE_BD_DOUBLE_HORIZONTAL:
-00567                 return (ULONG)'-';
-00568         case UNICODE_DOUBLE_VERTICAL_LINE:
-00569         case UNICODE_BD_LIGHT_VERTICAL:
-00570         case UNICODE_BD_DOUBLE_VERTICAL:
-00571                 return (ULONG)'|';
-00572         case UNICODE_DOUBLE_LOW_LINE:
-00573                 return (ULONG)'_';
-00574         case UNICODE_DAGGER:
-00575                 return (ULONG)'+';
-00576         case UNICODE_DOUBLE_DAGGER:
-00577                 return (ULONG)'#';
-00578         case UNICODE_BULLET:
-00579         case UNICODE_BULLET_OPERATOR:
-00580         case UNICODE_BLACK_CLUB_SUIT:
-00581                 return (ULONG)ucGetBulletCharacter(eConversionType, eEncoding);
-00582         case UNICODE_ONE_DOT_LEADER:
-00583         case UNICODE_TWO_DOT_LEADER:
-00584                 return (ULONG)'.';
-00585         case UNICODE_ELLIPSIS:
-00586 #if defined(__riscos)
-00587                 return (ULONG)OUR_ELLIPSIS;
-00588 #else
-00589                 if (ulFileOffset == 0) {
-00590                         return (ULONG)OUR_ELLIPSIS;
-00591                 }
-00592                 return UNICODE_ELLIPSIS;
-00593 #endif /* __riscos */
-00594         case UNICODE_DOUBLE_LEFT_ANGLE_QMARK:
-00595         case UNICODE_TRIANGULAR_BULLET:
-00596         case UNICODE_SINGLE_LEFT_ANGLE_QMARK:
-00597         case UNICODE_LEFTWARDS_ARROW:
-00598                 return (ULONG)'<';
-00599         case UNICODE_DOUBLE_RIGHT_ANGLE_QMARK:
-00600         case UNICODE_SINGLE_RIGHT_ANGLE_QMARK:
-00601         case UNICODE_RIGHTWARDS_ARROW:
-00602                 return (ULONG)'>';
-00603         case UNICODE_UNDERTIE:
-00604                 return (ULONG)'-';
-00605         case UNICODE_N_ARY_SUMMATION:
-00606                 return (ULONG)'S';
-00607         case UNICODE_EURO_SIGN:
-00608                 return (ULONG)'E';
-00609         case UNICODE_CIRCLE:
-00610         case UNICODE_SQUARE:
-00611                 return (ULONG)'O';
-00612         case UNICODE_DIAMOND:
-00613                 return (ULONG)OUR_DIAMOND;
-00614         case UNICODE_NUMERO_SIGN:
-00615                 return (ULONG)'N';
-00616         case UNICODE_KELVIN_SIGN:
-00617                 return (ULONG)'K';
-00618         case UNICODE_DOWNWARDS_ARROW:
-00619                 return (ULONG)'v';
-00620         case UNICODE_FRACTION_SLASH:
-00621         case UNICODE_DIVISION_SLASH:
-00622                 return (ULONG)'/';
-00623         case UNICODE_ASTERISK_OPERATOR:
-00624                 return (ULONG)'*';
-00625         case UNICODE_RATIO:
-00626                 return (ULONG)':';
-00627         case UNICODE_BD_LIGHT_DOWN_RIGHT:
-00628         case UNICODE_BD_LIGHT_DOWN_AND_LEFT:
-00629         case UNICODE_BD_LIGHT_UP_AND_RIGHT:
-00630         case UNICODE_BD_LIGHT_UP_AND_LEFT:
-00631         case UNICODE_BD_LIGHT_VERTICAL_AND_RIGHT:
-00632         case UNICODE_BD_LIGHT_VERTICAL_AND_LEFT:
-00633         case UNICODE_BD_LIGHT_DOWN_AND_HORIZONTAL:
-00634         case UNICODE_BD_LIGHT_UP_AND_HORIZONTAL:
-00635         case UNICODE_BD_LIGHT_VERTICAL_AND_HORIZONTAL:
-00636         case UNICODE_BD_DOUBLE_DOWN_AND_RIGHT:
-00637         case UNICODE_BD_DOUBLE_DOWN_AND_LEFT:
-00638         case UNICODE_BD_DOUBLE_UP_AND_RIGHT:
-00639         case UNICODE_BD_DOUBLE_UP_AND_LEFT:
-00640         case UNICODE_BD_DOUBLE_VERTICAL_AND_RIGHT:
-00641         case UNICODE_BD_DOUBLE_VERTICAL_AND_LEFT:
-00642         case UNICODE_BD_DOUBLE_DOWN_AND_HORIZONTAL:
-00643         case UNICODE_BD_DOUBLE_UP_AND_HORIZONTAL:
-00644         case UNICODE_BD_DOUBLE_VERTICAL_AND_HORIZONTAL:
-00645         case UNICODE_BLACK_SQUARE:
-00646                 return (ULONG)'+';
-00647         case UNICODE_HAIR_SPACE:
-00648         case UNICODE_ZERO_WIDTH_SPACE:
-00649         case UNICODE_ZERO_WIDTH_NON_JOINER:
-00650         case UNICODE_ZERO_WIDTH_JOINER:
-00651         case UNICODE_LEFT_TO_RIGHT_MARK:
-00652         case UNICODE_RIGHT_TO_LEFT_MARK:
-00653         case UNICODE_LEFT_TO_RIGHT_EMBEDDING:
-00654         case UNICODE_RIGHT_TO_LEFT_EMBEDDING:
-00655         case UNICODE_POP_DIRECTIONAL_FORMATTING:
-00656         case UNICODE_LEFT_TO_RIGHT_OVERRIDE:
-00657         case UNICODE_RIGHT_TO_LEFT_OVERRIDE:
-00658         case UNICODE_ZERO_WIDTH_NO_BREAK_SPACE:
-00659                 return IGNORE_CHARACTER;
-00660         default:
-00661                 break;
-00662         }
-00663 
-00664         if (usChar == UNICODE_TRADEMARK_SIGN) {
-00665                 /*
-00666                  * No local representation, it doesn't look like anything in
-00667                  * US-ASCII and a question mark does more harm than good.
-00668                  */
-00669                 return IGNORE_CHARACTER;
-00670         }
-00671 
-00672         if (usChar >= 0xa0 && usChar <= 0xff) {
-00673                 /* Before Word 97, Word did't use Unicode */
-00674                 return (ULONG)usChar;
-00675         }
-00676 
-00677         DBG_HEX_C(usChar < 0x3000 || usChar >= 0xd800, ulFileOffset);
-00678         DBG_HEX_C(usChar < 0x3000 || usChar >= 0xd800, usChar);
-00679         DBG_MSG_C(usChar >= 0xe000 && usChar < 0xf900, "Private Use Area");
-00680 
-00681         /* Untranslated Unicode character */
-00682         return 0x3f;
-00683 } /* end of ulTranslateCharacters */
-00684 
-00685 /*
-00686  * ulToUpper - convert letter to upper case
-00687  *
-00688  * This function converts a letter to upper case. Unlike toupper(3) this
-00689  * function is independent from the settings of locale. This comes in handy
-00690  * for people who have to read Word documents in more than one language or
-00691  * contain more than one language.
-00692  *
-00693  * returns the converted letter, or ulChar if the conversion was not possible.
-00694  */
-00695 ULONG
-00696 ulToUpper(ULONG ulChar)
-00697 {
-00698         if (ulChar < 0x80) {
-00699                 /* US ASCII: use standard function */
-00700                 return (ULONG)toupper((int)ulChar);
-00701         }
-00702         if (ulChar >= 0xe0 && ulChar <= 0xfe && ulChar != 0xf7) {
-00703                 /*
-00704                  * Lower case accented characters
-00705                  * 0xf7 is Division sign; 0xd7 is Multiplication sign
-00706                  * 0xff is y with diaeresis; 0xdf is Sharp s
-00707                  */
-00708                 return ulChar & ~0x20;
-00709         }
-00710 #if defined(__STDC_ISO_10646__)
-00711         /*
-00712          * If this is ISO C99 and all locales have wchar_t = ISO 10646
-00713          * (e.g., glibc 2.2 or newer), then use standard function
-00714          */
-00715         if (ulChar > 0xff) {
-00716                 return (ULONG)towupper((wint_t)ulChar);
-00717         }
-00718 #endif /* __STDC_ISO_10646__ */
-00719         return ulChar;
-00720 } /* end of ulToUpper */
-