examples/PIPS/antiword/src/misc.c

00001 /*
00002  * misc.c
00003  * Copyright (C) 1998-2005 A.J. van Os; Released under GNU GPL
00004  *
00005  * Description:
00006  * Miscellaneous functions
00007  */
00008 
00009 #include <stdio.h>
00010 #include <stdlib.h>
00011 #include <string.h>
00012 #include <ctype.h>
00013 #include <time.h>
00014 #if defined(__riscos)
00015 #include "DeskLib:SWI.h"
00016 #else
00017 #include <errno.h>
00018 #include <sys/types.h>
00019 #include <sys/stat.h>
00020 #endif /* __riscos */
00021 #if !defined(S_ISREG)
00022 #define S_ISREG(x)      (((x) & S_IFMT) == S_IFREG)
00023 #endif /* !S_ISREG */
00024 #include "antiword.h"
00025 #if defined(__vms)
00026 #include <unixlib.h>
00027 #endif
00028 
00029 #if !defined(__riscos)
00030 /*
00031  * szGetHomeDirectory - get the name of the home directory
00032  */
00033 const char *
00034 szGetHomeDirectory(void)
00035 {
00036         const char      *szHome;
00037 
00038 #if defined(__vms)
00039         szHome = decc$translate_vms(getenv("HOME"));
00040 #elif defined(__Plan9__)
00041         szHome = getenv("home");
00042 #else
00043         szHome = getenv("HOME");
00044 #endif /* __vms */
00045 
00046         if (szHome == NULL || szHome[0] == '\0') {
00047 #if defined(N_PLAT_NLM)
00048                 szHome = "SYS:";
00049 #elif defined(__dos)
00050                 szHome = "C:";
00051 #else
00052                 werr(0, "I can't find the name of your HOME directory");
00053                 szHome = "";
00054 #endif /* __dos */
00055         }
00056         return szHome;
00057 } /* end of szGetHomeDirectory */
00058 
00059 /*
00060  * szGetAntiwordDirectory - get the name of the Antiword directory
00061  */
00062 const char *
00063 szGetAntiwordDirectory(void)
00064 {
00065 #if defined(__vms)
00066         return decc$translate_vms(getenv("ANTIWORDHOME"));
00067 #else
00068         return getenv("ANTIWORDHOME");
00069 #endif /* __vms */
00070 } /* end of szGetAntiwordDirectory */
00071 #endif /* !__riscos */
00072 
00073 /*
00074  * Get the size of the specified file.
00075  * Returns -1 if the file does not exist or is not a proper file.
00076  */
00077 long
00078 lGetFilesize(const char *szFilename)
00079 {
00080 #if defined(__riscos)
00081         os_error        *e;
00082         int     iType, iSize;
00083 
00084         e = SWI(2, 5, SWI_OS_File | XOS_Bit,
00085                 17, szFilename,
00086                 &iType, NULL, NULL, NULL, &iSize);
00087         if (e != NULL) {
00088                 werr(0, "Get Filesize error %d: %s",
00089                         e->errnum, e->errmess);
00090                 return -1;
00091         }
00092         if (iType != 1) {
00093                 /* It's not a proper file or the file does not exist */
00094                 return -1;
00095         }
00096         return (long)iSize;
00097 #else
00098         struct stat     tBuffer;
00099 
00100         errno = 0;
00101         if (stat(szFilename, &tBuffer) != 0) {
00102                 werr(0, "Get Filesize error %d", errno);
00103                 return -1;
00104         }
00105         if (!S_ISREG(tBuffer.st_mode)) {
00106                 /* It's not a regular file */
00107                 return -1;
00108         }
00109         return (long)tBuffer.st_size;
00110 #endif /* __riscos */
00111 } /* end of lGetFilesize */
00112 
00113 #if defined(DEBUG)
00114 void
00115 vPrintBlock(const char  *szFile, int iLine,
00116                 const UCHAR *aucBlock, size_t tLength)
00117 {
00118         int i, j;
00119 
00120         fail(szFile == NULL || iLine < 0 || aucBlock == NULL);
00121 
00122         fprintf(stderr, "%s[%3d]:\n", szFile, iLine);
00123         for (i = 0; i < 32; i++) {
00124                 if (16 * i >= (int)tLength) {
00125                         return;
00126                 }
00127                 fprintf(stderr, "%03x: ", (unsigned int)(16 * i));
00128                 for (j = 0; j < 16; j++) {
00129                         if (16 * i + j < (int)tLength) {
00130                                 fprintf(stderr, "%02x ",
00131                                         (unsigned int)aucBlock[16 * i + j]);
00132                         }
00133                 }
00134                 fprintf(stderr, "\n");
00135         }
00136 } /* end of vPrintBlock */
00137 
00138 void
00139 vPrintUnicode(const char *szFile, int iLine, const UCHAR *aucUni, size_t tLen)
00140 {
00141         char    *szASCII;
00142 
00143         fail(tLen % 2 != 0);
00144 
00145         tLen /= 2;      /* Length in bytes to length in characters */
00146         szASCII = xmalloc(tLen + 1);
00147         (void)unincpy(szASCII, aucUni, tLen);
00148         szASCII[tLen] = '\0';
00149         (void)fprintf(stderr, "%s[%3d]: %.*s\n",
00150                                 szFile, iLine, (int)tLen, szASCII);
00151         szASCII = xfree(szASCII);
00152 } /* end of vPrintUnicode */
00153 
00154 BOOL
00155 bCheckDoubleLinkedList(output_type *pAnchor)
00156 {
00157         output_type     *pCurr, *pLast;
00158         int             iInList;
00159 
00160         pLast = pAnchor;
00161         iInList = 0;
00162         for (pCurr = pAnchor; pCurr != NULL; pCurr = pCurr->pNext) {
00163                 pLast = pCurr;
00164                 iInList++;
00165         }
00166         NO_DBG_DEC(iInList);
00167         for (pCurr = pLast; pCurr != NULL; pCurr = pCurr->pPrev) {
00168                 pLast = pCurr;
00169                 iInList--;
00170         }
00171         DBG_DEC_C(iInList != 0, iInList);
00172         return pAnchor == pLast && iInList == 0;
00173 } /* end of bCheckDoubleLinkedList */
00174 #endif /* DEBUG */
00175 
00176 /*
00177  * bReadBytes
00178  * This function reads the specified number of bytes from the specified file,
00179  * starting from the specified offset.
00180  * Returns TRUE when successfull, otherwise FALSE
00181  */
00182 BOOL
00183 bReadBytes(UCHAR *aucBytes, size_t tMemb, ULONG ulOffset, FILE *pFile)
00184 {
00185         fail(aucBytes == NULL || pFile == NULL || ulOffset > (ULONG)LONG_MAX);
00186 
00187         if (ulOffset > (ULONG)LONG_MAX) {
00188                 return FALSE;
00189         }
00190         if (fseek(pFile, (long)ulOffset, SEEK_SET) != 0) {
00191                 return FALSE;
00192         }
00193         if (fread(aucBytes, sizeof(UCHAR), tMemb, pFile) != tMemb) {
00194                 return FALSE;
00195         }
00196         return TRUE;
00197 } /* end of bReadBytes */
00198 
00199 /*
00200  * bReadBuffer
00201  * This function fills the specified buffer with the specified number of bytes,
00202  * starting at the specified offset within the Big/Small Block Depot.
00203  *
00204  * Returns TRUE when successful, otherwise FALSE
00205  */
00206 BOOL
00207 bReadBuffer(FILE *pFile, ULONG ulStartBlock,
00208         const ULONG *aulBlockDepot, size_t tBlockDepotLen, size_t tBlockSize,
00209         UCHAR *aucBuffer, ULONG ulOffset, size_t tToRead)
00210 {
00211         ULONG   ulBegin, ulIndex;
00212         size_t  tLen;
00213 
00214         fail(pFile == NULL);
00215         fail(ulStartBlock > MAX_BLOCKNUMBER && ulStartBlock != END_OF_CHAIN);
00216         fail(aulBlockDepot == NULL);
00217         fail(tBlockSize != BIG_BLOCK_SIZE && tBlockSize != SMALL_BLOCK_SIZE);
00218         fail(aucBuffer == NULL);
00219         fail(tToRead == 0);
00220 
00221         for (ulIndex = ulStartBlock;
00222              ulIndex != END_OF_CHAIN && tToRead != 0;
00223              ulIndex = aulBlockDepot[ulIndex]) {
00224                 if (ulIndex >= (ULONG)tBlockDepotLen) {
00225                         DBG_DEC(ulIndex);
00226                         DBG_DEC(tBlockDepotLen);
00227                         if (tBlockSize >= BIG_BLOCK_SIZE) {
00228                                 werr(1, "The Big Block Depot is damaged");
00229                         } else {
00230                                 werr(1, "The Small Block Depot is damaged");
00231                         }
00232                 }
00233                 if (ulOffset >= (ULONG)tBlockSize) {
00234                         ulOffset -= tBlockSize;
00235                         continue;
00236                 }
00237                 ulBegin = ulDepotOffset(ulIndex, tBlockSize) + ulOffset;
00238                 tLen = min(tBlockSize - (size_t)ulOffset, tToRead);
00239                 ulOffset = 0;
00240                 if (!bReadBytes(aucBuffer, tLen, ulBegin, pFile)) {
00241                         werr(0, "Read big block 0x%lx not possible", ulBegin);
00242                         return FALSE;
00243                 }
00244                 aucBuffer += tLen;
00245                 tToRead -= tLen;
00246         }
00247         DBG_DEC_C(tToRead != 0, tToRead);
00248         return tToRead == 0;
00249 } /* end of bReadBuffer */
00250 
00251 /*
00252  * Convert a Word colornumber into a true color for use in a drawfile
00253  *
00254  * Returns the true color
00255  */
00256 ULONG
00257 ulColor2Color(UCHAR ucFontColor)
00258 {
00259         static const ULONG      aulColorTable[] = {
00260                 /*  0 */        0x00000000UL,   /* Automatic */
00261                 /*  1 */        0x00000000UL,   /* Black */
00262                 /*  2 */        0xff000000UL,   /* Blue */
00263                 /*  3 */        0xffff0000UL,   /* Turquoise */
00264                 /*  4 */        0x00ff0000UL,   /* Bright Green */
00265                 /*  5 */        0xff00ff00UL,   /* Pink */
00266                 /*  6 */        0x0000ff00UL,   /* Red */
00267                 /*  7 */        0x00ffff00UL,   /* Yellow */
00268                 /*  8 */        0xffffff00UL,   /* White */
00269                 /*  9 */        0x80000000UL,   /* Dark Blue */
00270                 /* 10 */        0x80800000UL,   /* Teal */
00271                 /* 11 */        0x00800000UL,   /* Green */
00272                 /* 12 */        0x80008000UL,   /* Violet */
00273                 /* 13 */        0x00008000UL,   /* Dark Red */
00274                 /* 14 */        0x00808000UL,   /* Dark Yellow */
00275                 /* 15 */        0x80808000UL,   /* Gray 50% */
00276                 /* 16 */        0xc0c0c000UL,   /* Gray 25% */
00277         };
00278         if ((size_t)ucFontColor >= elementsof(aulColorTable)) {
00279                 return aulColorTable[0];
00280         }
00281         return aulColorTable[(int)ucFontColor];
00282 } /* end of ulColor2Color */
00283 
00284 /*
00285  * iFindSplit - find a place to split the string
00286  *
00287  * returns the index of the split character or -1 if no split found.
00288  */
00289 static int
00290 iFindSplit(const char *szString, size_t tStringLen)
00291 {
00292         size_t  tSplit;
00293 
00294         if (tStringLen == 0) {
00295                 return -1;
00296         }
00297         tSplit = tStringLen - 1;
00298         while (tSplit >= 1) {
00299                 if (szString[tSplit] == ' ' ||
00300                     (szString[tSplit] == '-' && szString[tSplit - 1] != ' ')) {
00301                         return (int)tSplit;
00302                 }
00303                 tSplit--;
00304         }
00305         return -1;
00306 } /* end of iFindSplit */
00307 
00308 /*
00309  * pSplitList - split the specified list in a printable part and a leftover part
00310  *
00311  * returns the pointer to the leftover part
00312  */
00313 output_type *
00314 pSplitList(output_type *pAnchor)
00315 {
00316         output_type     *pCurr, *pLeftOver;
00317         int             iIndex;
00318 
00319         fail(pAnchor == NULL);
00320 
00321         for (pCurr = pAnchor; pCurr->pNext != NULL; pCurr = pCurr->pNext)
00322                 ;       /* EMPTY */
00323         iIndex = -1;
00324         for (; pCurr != NULL; pCurr = pCurr->pPrev) {
00325                 iIndex = iFindSplit(pCurr->szStorage, pCurr->tNextFree);
00326                 if (iIndex >= 0) {
00327                         break;
00328                 }
00329         }
00330 
00331         if (pCurr == NULL || iIndex < 0) {
00332                 /* No split, no leftover */
00333                 return NULL;
00334         }
00335         /* Split over the iIndex-th character */
00336         NO_DBG_MSG("pLeftOver");
00337         pLeftOver = xmalloc(sizeof(*pLeftOver));
00338         fail(pCurr->tNextFree < (size_t)iIndex);
00339         pLeftOver->tStorageSize = pCurr->tNextFree - (size_t)iIndex;
00340         pLeftOver->szStorage = xmalloc(pLeftOver->tStorageSize);
00341         pLeftOver->tNextFree = pCurr->tNextFree - (size_t)iIndex - 1;
00342         (void)strncpy(pLeftOver->szStorage,
00343                 pCurr->szStorage + iIndex + 1, pLeftOver->tNextFree);
00344         pLeftOver->szStorage[pLeftOver->tNextFree] = '\0';
00345         NO_DBG_MSG(pLeftOver->szStorage);
00346         pLeftOver->ucFontColor = pCurr->ucFontColor;
00347         pLeftOver->usFontStyle = pCurr->usFontStyle;
00348         pLeftOver->tFontRef = pCurr->tFontRef;
00349         pLeftOver->usFontSize = pCurr->usFontSize;
00350         pLeftOver->lStringWidth = lComputeStringWidth(
00351                                         pLeftOver->szStorage,
00352                                         pLeftOver->tNextFree,
00353                                         pLeftOver->tFontRef,
00354                                         pLeftOver->usFontSize);
00355         pLeftOver->pPrev = NULL;
00356         pLeftOver->pNext = pCurr->pNext;
00357         if (pLeftOver->pNext != NULL) {
00358                 pLeftOver->pNext->pPrev = pLeftOver;
00359         }
00360         fail(!bCheckDoubleLinkedList(pLeftOver));
00361 
00362         NO_DBG_MSG("pAnchor");
00363         NO_DBG_HEX(pCurr->szStorage[iIndex]);
00364         while (iIndex >= 0 && isspace((int)(UCHAR)pCurr->szStorage[iIndex])) {
00365                 iIndex--;
00366         }
00367         pCurr->tNextFree = (size_t)iIndex + 1;
00368         pCurr->szStorage[pCurr->tNextFree] = '\0';
00369         NO_DBG_MSG(pCurr->szStorage);
00370         pCurr->lStringWidth = lComputeStringWidth(
00371                                         pCurr->szStorage,
00372                                         pCurr->tNextFree,
00373                                         pCurr->tFontRef,
00374                                         pCurr->usFontSize);
00375         pCurr->pNext = NULL;
00376         fail(!bCheckDoubleLinkedList(pAnchor));
00377 
00378         return pLeftOver;
00379 } /* end of pSplitList */
00380 
00381 /*
00382  * tNumber2Roman - convert a number to Roman Numerals
00383  *
00384  * returns the number of characters written
00385  */
00386 size_t
00387 tNumber2Roman(UINT uiNumber, BOOL bUpperCase, char *szOutput)
00388 {
00389         char    *outp, *p, *q;
00390         UINT    uiNextVal, uiValue;
00391 
00392         fail(szOutput == NULL);
00393 
00394         uiNumber %= 4000;       /* Very high numbers can't be represented */
00395         if (uiNumber == 0) {
00396                 szOutput[0] = '\0';
00397                 return 0;
00398         }
00399 
00400         outp = szOutput;
00401         p = bUpperCase ? "M\2D\5C\2L\5X\2V\5I" : "m\2d\5c\2l\5x\2v\5i";
00402         uiValue = 1000;
00403         for (;;) {
00404                 while (uiNumber >= uiValue) {
00405                         *outp++ = *p;
00406                         uiNumber -= uiValue;
00407                 }
00408                 if (uiNumber == 0) {
00409                         *outp = '\0';
00410                         fail(outp < szOutput);
00411                         return (size_t)(outp - szOutput);
00412                 }
00413                 q = p + 1;
00414                 uiNextVal = uiValue / (UINT)(UCHAR)*q;
00415                 if ((int)*q == 2) {             /* magic */
00416                         uiNextVal /= (UINT)(UCHAR)*(q += 2);
00417                 }
00418                 if (uiNumber + uiNextVal >= uiValue) {
00419                         *outp++ = *++q;
00420                         uiNumber += uiNextVal;
00421                 } else {
00422                         p++;
00423                         uiValue /= (UINT)(UCHAR)(*p++);
00424                 }
00425         }
00426 } /* end of tNumber2Roman */
00427 
00428 /*
00429  * iNumber2Alpha - convert a number to alphabetic "numbers"
00430  *
00431  * returns the number of characters written
00432  */
00433 size_t
00434 tNumber2Alpha(UINT uiNumber, BOOL bUpperCase, char *szOutput)
00435 {
00436         char    *outp;
00437         UINT    uiTmp;
00438 
00439         fail(szOutput == NULL);
00440 
00441         if (uiNumber == 0) {
00442                 szOutput[0] = '\0';
00443                 return 0;
00444         }
00445 
00446         outp = szOutput;
00447         uiTmp = (UINT)(bUpperCase ? 'A': 'a');
00448         if (uiNumber <= 26) {
00449                 uiNumber -= 1;
00450                 *outp++ = (char)(uiTmp + uiNumber);
00451         } else if (uiNumber <= 26U + 26U*26U) {
00452                 uiNumber -= 26 + 1;
00453                 *outp++ = (char)(uiTmp + uiNumber / 26);
00454                 *outp++ = (char)(uiTmp + uiNumber % 26);
00455         } else if (uiNumber <= 26U + 26U*26U + 26U*26U*26U) {
00456                 uiNumber -= 26 + 26*26 + 1;
00457                 *outp++ = (char)(uiTmp + uiNumber / (26*26));
00458                 *outp++ = (char)(uiTmp + uiNumber / 26 % 26);
00459                 *outp++ = (char)(uiTmp + uiNumber % 26);
00460         }
00461         *outp = '\0';
00462         fail(outp < szOutput);
00463         return (size_t)(outp - szOutput);
00464 } /* end of tNumber2Alpha */
00465 
00466 /*
00467  * unincpy - copy a counted Unicode string to an single-byte string
00468  */
00469 char *
00470 unincpy(char *s1, const UCHAR *s2, size_t n)
00471 {
00472         char    *pcDest;
00473         ULONG   ulChar;
00474         size_t  tLen;
00475         USHORT  usUni;
00476 
00477         for (pcDest = s1, tLen = 0; tLen < n; pcDest++, tLen++) {
00478                 usUni = usGetWord(tLen * 2, s2);
00479                 if (usUni == 0) {
00480                         break;
00481                 }
00482                 ulChar = ulTranslateCharacters(usUni, 0, 8,
00483                                 conversion_unknown, encoding_neutral, FALSE);
00484                 if (ulChar == IGNORE_CHARACTER) {
00485                         ulChar = (ULONG)'?';
00486                 }
00487                 *pcDest = (char)ulChar;
00488         }
00489         for (; tLen < n; tLen++) {
00490                 *pcDest++ = '\0';
00491         }
00492         return s1;
00493 } /* end of unincpy */
00494 
00495 /*
00496  * unilen - calculate the length of a Unicode string
00497  *
00498  * returns the length in bytes
00499  */
00500 size_t
00501 unilen(const UCHAR *s)
00502 {
00503         size_t  tLen;
00504         USHORT  usUni;
00505 
00506         tLen = 0;
00507         for (;;) {
00508                 usUni = usGetWord(tLen, s);
00509                 if (usUni == 0) {
00510                         return tLen;
00511                 }
00512                 tLen += 2;
00513         }
00514 } /* end of unilen */
00515 
00516 /*
00517  * szBaseName - get the basename of the specified filename
00518  */
00519 const char *
00520 szBasename(const char *szFilename)
00521 {
00522         const char      *szTmp;
00523 
00524         fail(szFilename == NULL);
00525 
00526         if (szFilename == NULL || szFilename[0] == '\0') {
00527                 return "null";
00528         }
00529 
00530         szTmp = strrchr(szFilename, FILE_SEPARATOR[0]);
00531         if (szTmp == NULL) {
00532                 return szFilename;
00533         }
00534         return ++szTmp;
00535 } /* end of szBasename */
00536 
00537 /*
00538  * lComputeLeading - compute the leading
00539  *
00540  * NOTE: the fontsize is specified in half points
00541  *
00542  * Returns the leading in drawunits
00543  */
00544 long
00545 lComputeLeading(USHORT usFontSize)
00546 {
00547         long    lLeading;
00548 
00549         lLeading = (long)usFontSize * 500L;
00550         if (usFontSize < 18) {          /* Small text: 112% */
00551                 lLeading *= 112;
00552         } else if (usFontSize < 28) {   /* Normal text: 124% */
00553                 lLeading *= 124;
00554         } else if (usFontSize < 48) {   /* Small headlines: 104% */
00555                 lLeading *= 104;
00556         } else {                        /* Large headlines: 100% */
00557                 lLeading *= 100;
00558         }
00559         lLeading = lMilliPoints2DrawUnits(lLeading);
00560         lLeading += 50;
00561         lLeading /= 100;
00562         return lLeading;
00563 } /* end of lComputeLeading */
00564 
00565 /*
00566  * Convert a UCS character to an UTF-8 string
00567  *
00568  * Returns the string length of the result
00569  */
00570 size_t
00571 tUcs2Utf8(ULONG ulChar, char *szResult, size_t tMaxResultLen)
00572 {
00573         if (szResult == NULL || tMaxResultLen == 0) {
00574                 return 0;
00575         }
00576 
00577         if (ulChar < 0x80 && tMaxResultLen >= 2) {
00578                 szResult[0] = (char)ulChar;
00579                 szResult[1] = '\0';
00580                 return 1;
00581         }
00582         if (ulChar < 0x800 && tMaxResultLen >= 3) {
00583                 szResult[0] = (char)(0xc0 | ulChar >> 6);
00584                 szResult[1] = (char)(0x80 | (ulChar & 0x3f));
00585                 szResult[2] = '\0';
00586                 return 2;
00587         }
00588         if (ulChar < 0x10000 && tMaxResultLen >= 4) {
00589                 szResult[0] = (char)(0xe0 | ulChar >> 12);
00590                 szResult[1] = (char)(0x80 | (ulChar >> 6 & 0x3f));
00591                 szResult[2] = (char)(0x80 | (ulChar & 0x3f));
00592                 szResult[3] = '\0';
00593                 return 3;
00594         }
00595         if (ulChar < 0x200000 && tMaxResultLen >= 5) {
00596                 szResult[0] = (char)(0xf0 | ulChar >> 18);
00597                 szResult[1] = (char)(0x80 | (ulChar >> 12 & 0x3f));
00598                 szResult[2] = (char)(0x80 | (ulChar >> 6 & 0x3f));
00599                 szResult[3] = (char)(0x80 | (ulChar & 0x3f));
00600                 szResult[4] = '\0';
00601                 return 4;
00602         }
00603         szResult[0] = '\0';
00604         return 0;
00605 } /* end of tUcs2Utf8 */
00606 
00607 /*
00608  * vGetBulletValue - get the bullet value for the conversing type and encoding
00609  */
00610 void
00611 vGetBulletValue(conversion_type eConversionType, encoding_type eEncoding,
00612         char *szResult, size_t tMaxResultLen)
00613 {
00614         fail(szResult == NULL);
00615         fail(tMaxResultLen < 2);
00616 
00617         if (eEncoding == encoding_utf_8) {
00618                 (void)tUcs2Utf8(UNICODE_BULLET, szResult, tMaxResultLen);
00619         } else {
00620                 szResult[0] = (char)ucGetBulletCharacter(eConversionType,
00621                                                         eEncoding);
00622                 szResult[1] = '\0';
00623         }
00624 } /* end of vGetBulletValue */
00625 
00626 /*
00627  * bAllZero - are all bytes zero?
00628  */
00629 BOOL
00630 bAllZero(const UCHAR *aucBytes, size_t tLength)
00631 {
00632         size_t  tIndex;
00633 
00634         if (aucBytes == NULL || tLength == 0) {
00635                 return TRUE;
00636         }
00637 
00638         for (tIndex = 0; tIndex < tLength; tIndex++) {
00639                 if (aucBytes[tIndex] != 0) {
00640                         return FALSE;
00641                 }
00642         }
00643         return TRUE;
00644 } /* end of bAllZero */
00645 
00646 #if !defined(__riscos)
00647 /*
00648  * GetCodesetFromLocale - get the codeset from the current locale
00649  *
00650  * Original version: Copyright (C) 1999  Bruno Haible
00651  * Syntax:
00652  * language[_territory][.codeset][@modifier][+special][,[sponsor][_revision]]
00653  *
00654  * Returns TRUE when sucessful, otherwise FALSE
00655  */
00656 static BOOL
00657 bGetCodesetFromLocale(char *szCodeset, size_t tMaxCodesetLength, BOOL *pbEuro)
00658 {
00659 #if !defined(__dos)
00660         const char      *szLocale;
00661         const char      *pcTmp;
00662         size_t          tIndex;
00663         char            szModifier[6];
00664 #endif /* __dos */
00665 
00666         if (pbEuro != NULL) {
00667                 *pbEuro = FALSE;        /* Until proven otherwise */
00668         }
00669         if (szCodeset == NULL || tMaxCodesetLength == 0) {
00670                 return FALSE;
00671         }
00672 
00673 #if defined(__dos)
00674         if (tMaxCodesetLength < 2 + sizeof(int) * 3 + 1) {
00675                 DBG_DEC(tMaxCodesetLength);
00676                 DBG_DEC(2 + sizeof(int) * 3 + 1);
00677                 return FALSE;
00678         }
00679         /* Get the active codepage from DOS */
00680         sprintf(szCodeset, "cp%d", iGetCodepage());
00681         DBG_MSG(szCodeset);
00682 #else
00683         /* Get the locale from the environment */
00684         szLocale = getenv("LC_ALL");
00685         if (szLocale == NULL || szLocale[0] == '\0') {
00686                 szLocale = getenv("LC_CTYPE");
00687                 if (szLocale == NULL || szLocale[0] == '\0') {
00688                         szLocale = getenv("LANG");
00689                 }
00690         }
00691         if (szLocale == NULL || szLocale[0] == '\0') {
00692                 /* No locale, so no codeset name and no modifier */
00693                 return FALSE;
00694         }
00695         DBG_MSG(szLocale);
00696         pcTmp = strchr(szLocale, '.');
00697         if (pcTmp == NULL) {
00698                 /* No codeset name */
00699                 szCodeset[0] = '\0';
00700         } else {
00701                 /* Copy the codeset name */
00702                 pcTmp++;
00703                 for (tIndex = 0; tIndex < tMaxCodesetLength; tIndex++) {
00704                         if (*pcTmp == '@' || *pcTmp == '+' ||
00705                             *pcTmp == ',' || *pcTmp == '_' ||
00706                             *pcTmp == '\0') {
00707                                 szCodeset[tIndex] = '\0';
00708                                 break;
00709                         }
00710                         szCodeset[tIndex] = *pcTmp;
00711                         pcTmp++;
00712                 }
00713                 szCodeset[tMaxCodesetLength - 1] = '\0';
00714         }
00715         if (pbEuro == NULL) {
00716                 /* No need to get the modifier */
00717                 return TRUE;
00718         }
00719         pcTmp = strchr(szLocale, '@');
00720         if (pcTmp != NULL) {
00721                 /* Copy the modifier */
00722                 pcTmp++;
00723                 for (tIndex = 0; tIndex < sizeof(szModifier); tIndex++) {
00724                         if (*pcTmp == '+' || *pcTmp == ',' ||
00725                             *pcTmp == '_' || *pcTmp == '\0') {
00726                                 szModifier[tIndex] = '\0';
00727                                 break;
00728                         }
00729                         szModifier[tIndex] = *pcTmp;
00730                         pcTmp++;
00731                 }
00732                 szModifier[sizeof(szModifier) - 1] = '\0';
00733                 *pbEuro = STRCEQ(szModifier, "Euro");
00734         }
00735 #endif /* __dos */
00736         return TRUE;
00737 } /* end of bGetCodesetFromLocale */
00738 
00739 /*
00740  * GetNormalizedCodeset - get the normalized codeset from the current locale
00741  *
00742  * Returns TRUE when sucessful, otherwise FALSE
00743  */
00744 BOOL
00745 bGetNormalizedCodeset(char *szCodeset, size_t tMaxCodesetLength, BOOL *pbEuro)
00746 {
00747         BOOL    bOnlyDigits;
00748         const char      *pcSrc;
00749         char    *pcDest;
00750         char    *szTmp, *szCodesetNorm;
00751 
00752         if (pbEuro != NULL) {
00753                 *pbEuro = FALSE;        /* Until proven otherwise */
00754         }
00755         if (szCodeset == NULL || tMaxCodesetLength < 4) {
00756                 return FALSE;
00757         }
00758 
00759         /* Get the codeset name */
00760         szTmp = xmalloc(tMaxCodesetLength - 3);
00761         if (!bGetCodesetFromLocale(szTmp, tMaxCodesetLength - 3, pbEuro)) {
00762                 szTmp = xfree(szTmp);
00763                 return FALSE;
00764         }
00765         /* Normalize the codeset name */
00766         szCodesetNorm = xmalloc(tMaxCodesetLength - 3);
00767         bOnlyDigits = TRUE;
00768         pcDest = szCodesetNorm;
00769         for (pcSrc = szTmp; *pcSrc != '\0'; pcSrc++) {
00770                 if (isalnum(*pcSrc)) {
00771                         *pcDest = tolower(*pcSrc);
00772                         if (!isdigit(*pcDest)) {
00773                                 bOnlyDigits = FALSE;
00774                         }
00775                         pcDest++;
00776                 }
00777         }
00778         *pcDest = '\0';
00779         DBG_MSG(szCodesetNorm);
00780         /* Add "iso" when szCodesetNorm contains all digits */
00781         if (bOnlyDigits && szCodesetNorm[0] != '\0') {
00782                 fail(strlen(szCodesetNorm) + 3 >= tMaxCodesetLength);
00783                 sprintf(szCodeset, "iso%s", szCodesetNorm);
00784         } else {
00785                 fail(strlen(szCodesetNorm) >= tMaxCodesetLength);
00786                 strncpy(szCodeset, szCodesetNorm, pcDest - szCodesetNorm + 1);
00787                 szCodeset[tMaxCodesetLength - 1] = '\0';
00788         }
00789         DBG_MSG(szCodeset);
00790         /* Clean up and leave */
00791         szCodesetNorm = xfree(szCodesetNorm);
00792         szTmp = xfree(szTmp);
00793         return TRUE;
00794 } /* end of bGetNormalizedCodeset */
00795 
00796 /*
00797  * szGetDefaultMappingFile - get the default mapping file
00798  *
00799  * Returns the basename of the default mapping file
00800  */
00801 const char *
00802 szGetDefaultMappingFile(void)
00803 {
00804         static const struct {
00805                 const char      *szCodeset;
00806                 const char      *szMappingFile;
00807         } atMappingFile[] = {
00808                 { "iso88591",   MAPPING_FILE_8859_1 },
00809                 { "iso88592",   MAPPING_FILE_8859_2 },
00810                 { "iso88593",   "8859-3.txt" },
00811                 { "iso88594",   "8859-4.txt" },
00812                 { "iso88595",   "8859-5.txt" },
00813                 { "iso88596",   MAPPING_FILE_8859_5 },
00814                 { "iso88597",   "8859-7.txt" },
00815                 { "iso88598",   "8859-8.txt" },
00816                 { "iso88599",   "8859-9.txt" },
00817                 { "iso885910",  "8859-10.txt" },
00818                 { "iso885913",  "8859-13.txt" },
00819                 { "iso885914",  "8859-14.txt" },
00820                 { "iso885915",  MAPPING_FILE_8859_15 },
00821                 { "iso885916",  "8859-16.txt" },
00822                 { "koi8r",      MAPPING_FILE_KOI8_R },
00823                 { "koi8u",      MAPPING_FILE_KOI8_U },
00824                 { "utf8",       MAPPING_FILE_UTF_8 },
00825                 { "cp437",      MAPPING_FILE_CP437 },
00826                 { "cp850",      "cp850.txt" },
00827                 { "cp852",      MAPPING_FILE_CP852 },
00828                 { "cp862",      "cp862.txt" },
00829                 { "cp864",      "cp864.txt" },
00830                 { "cp866",      MAPPING_FILE_CP866 },
00831                 { "cp1250",     MAPPING_FILE_CP1250 },
00832                 { "cp1251",     MAPPING_FILE_CP1251 },
00833                 { "cp1252",     "cp1252.txt" },
00834         };
00835         size_t  tIndex;
00836         BOOL    bEuro;
00837         char    szCodeset[20];
00838 
00839         szCodeset[0] = '\0';
00840         bEuro = FALSE;
00841         /* Get the normalized codeset name */
00842         if (!bGetNormalizedCodeset(szCodeset, sizeof(szCodeset), &bEuro)) {
00843                 return MAPPING_FILE_8859_1;
00844         }
00845         if (szCodeset[0] == '\0') {
00846                 if (bEuro) {
00847                         /* Default mapping file (with Euro sign) */
00848                         return MAPPING_FILE_8859_15;
00849                 } else {
00850                         /* Default mapping file (without Euro sign) */
00851                         return MAPPING_FILE_8859_1;
00852                 }
00853         }
00854         /* Find the name in the table */
00855         for (tIndex = 0; tIndex < elementsof(atMappingFile); tIndex++) {
00856                 if (STREQ(atMappingFile[tIndex].szCodeset, szCodeset)) {
00857                         return atMappingFile[tIndex].szMappingFile;
00858                 }
00859         }
00860         /* Default default mapping file */
00861 #if defined(__dos)
00862         return MAPPING_FILE_CP437;
00863 #else
00864         return MAPPING_FILE_8859_1;
00865 #endif /* __dos */
00866 } /* end of szGetDefaultMappingFile */
00867 #endif /* !__riscos */
00868 
00869 /*
00870  * tConvertDTTM - convert Windows Date and Time format
00871  *
00872  * returns Unix time_t or -1
00873  */
00874 time_t
00875 tConvertDTTM(ULONG ulDTTM)
00876 {
00877         struct tm       tTime;
00878         time_t          tResult;
00879 
00880         if (ulDTTM == 0) {
00881                 return (time_t)-1;
00882         }
00883         memset(&tTime, 0, sizeof(tTime));
00884         tTime.tm_min = (int)(ulDTTM & 0x0000003f);
00885         tTime.tm_hour = (int)((ulDTTM & 0x000007c0) >> 6);
00886         tTime.tm_mday = (int)((ulDTTM & 0x0000f800) >> 11);
00887         tTime.tm_mon = (int)((ulDTTM & 0x000f0000) >> 16);
00888         tTime.tm_year = (int)((ulDTTM & 0x1ff00000) >> 20);
00889         tTime.tm_isdst = -1;
00890         tTime.tm_mon--;         /* From 01-12 to 00-11 */
00891         tResult = mktime(&tTime);
00892         NO_DBG_MSG(ctime(&tResult));
00893         return tResult;
00894 } /* end of tConvertDTTM */

Generated by  doxygen 1.6.2