Quellcodebibliothek Statistik Leitseite products/Sources/formale Sprachen/C/Firefox/intl/icu/source/tools/icuswap/   (Browser von der Mozilla Stiftung Version 136.0.1©)  Datei vom 10.2.2025 mit Größe 21 kB image not shown  

Quelle  icuswap.cpp   Sprache: C

 
// © 2016 and later: Unicode, Inc. and others.
// License & terms of use: http://www.unicode.org/copyright.html
/*
*******************************************************************************
*
*   Copyright (C) 2003-2014, International Business Machines
*   Corporation and others.  All Rights Reserved.
*
*******************************************************************************
*   file name:  icuswap.cpp
*   encoding:   UTF-8
*   tab size:   8 (not used)
*   indentation:4
*
*   created on: 2003aug08
*   created by: Markus W. Scherer
*
*   This tool takes an ICU data file and "swaps" it, that is, changes its
*   platform properties between big-/little-endianness and ASCII/EBCDIC charset
*   families.
*   The modified data file is written to a new file.
*   Useful as an install-time tool for shipping only one flavor of ICU data
*   and preparing data files for the target platform.
*   Will not work with data DLLs (shared libraries).
*/


#include "unicode/utypes.h"
#include "unicode/putil.h"
#include "unicode/udata.h"
#include "cmemory.h"
#include "cstring.h"
#include "uinvchar.h"
#include "uarrsort.h"
#include "ucmndata.h"
#include "udataswp.h"
#include "swapimpl.h"
#include "toolutil.h"
#include "uoptions.h"

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

/* definitions */

#define DEFAULT_PADDING_LENGTH 15

static UOption options[]={
    UOPTION_HELP_H,
    UOPTION_HELP_QUESTION_MARK,
    UOPTION_DEF("type"'t', UOPT_REQUIRES_ARG)
};

enum {
    OPT_HELP_H,
    OPT_HELP_QUESTION_MARK,
    OPT_OUT_TYPE
};

static int32_t
fileSize(FILE *f) {
    int32_t size;

    fseek(f, 0, SEEK_END);
    size = static_cast<int32_t>(ftell(f));
    fseek(f, 0, SEEK_SET);
    return size;
}

/**
 * Swap an ICU .dat package, including swapping of enclosed items.
 */

U_CFUNC int32_t U_CALLCONV
udata_swapPackage(const char *inFilename, const char *outFilename,
                  const UDataSwapper *ds,
                  const void *inData, int32_t length, void *outData,
                  UErrorCode *pErrorCode);

U_CDECL_BEGIN
static void U_CALLCONV
printError(void *context, const char *fmt, va_list args) {
    vfprintf((FILE *)context, fmt, args);
}
U_CDECL_END

static int
printUsage(const char *pname, UBool ishelp) {
    fprintf(stderr,
            "%csage: %s [ -h, -?, --help ] -tl|-tb|-te|--type=b|... infilename outfilename\n",
            ishelp ? 'U' : 'u', pname);
    if(ishelp) {
        fprintf(stderr,
              "\nOptions: -h, -?, --help print this message and exit\n"
                " Read the input file, swap its platform properties according\n"
                " to the -t or --type option, and write the result to the output file.\n"
                " -tl change to little-endian/ASCII charset family\n"
                " -tb change to big-endian/ASCII charset family\n"
                " -te change to big-endian/EBCDIC charset family\n");
    }

    return !ishelp;
}

extern int
main(int argc, char *argv[]) {
    FILE *in, *out;
    const char *pname;
    char *data;
    int32_t length;
    UBool ishelp;
    int rc;

    UDataSwapper *ds;
    const UDataInfo *pInfo;
    UErrorCode errorCode;
    uint8_t outCharset;
    UBool outIsBigEndian;

    U_MAIN_INIT_ARGS(argc, argv);

    fprintf(stderr, "Warning: icuswap is an obsolete tool and it will be removed in the next ICU release.\nPlease use the icupkg tool instead.\n");

    /* get the program basename */
    pname=strrchr(argv[0], U_FILE_SEP_CHAR);
    if(pname==nullptr) {
        pname=strrchr(argv[0], '/');
    }
    if(pname!=nullptr) {
        ++pname;
    } else {
        pname=argv[0];
    }

    argc=u_parseArgs(argc, argv, UPRV_LENGTHOF(options), options);
    ishelp=options[OPT_HELP_H].doesOccur || options[OPT_HELP_QUESTION_MARK].doesOccur;
    if(ishelp || argc!=3) {
        return printUsage(pname, ishelp);
    }

    /* parse the output type option */
    data = const_cast<char*>(options[OPT_OUT_TYPE].value);
    if(data[0]==0 || data[1]!=0) {
        /* the type must be exactly one letter */
        return printUsage(pname, false);
    }
    switch(data[0]) {
    case 'l':
        outIsBigEndian=false;
        outCharset=U_ASCII_FAMILY;
        break;
    case 'b':
        outIsBigEndian=true;
        outCharset=U_ASCII_FAMILY;
        break;
    case 'e':
        outIsBigEndian=true;
        outCharset=U_EBCDIC_FAMILY;
        break;
    default:
        return printUsage(pname, false);
    }

    in=out=nullptr;
    data=nullptr;

    /* open the input file, get its length, allocate memory for it, read the file */
    in=fopen(argv[1], "rb");
    if(in==nullptr) {
        fprintf(stderr, "%s: unable to open input file \"%s\"\n", pname, argv[1]);
        rc=2;
        goto done;
    }

    length=fileSize(in);
    if(length<DEFAULT_PADDING_LENGTH) {
        fprintf(stderr, "%s: empty input file \"%s\"\n", pname, argv[1]);
        rc=2;
        goto done;
    }

    /*
     * +15: udata_swapPackage() may need to add a few padding bytes to the
     * last item if charset swapping is done,
     * because the last item may be resorted into the middle and then needs
     * additional padding bytes
     */

    data = static_cast<char*>(malloc(length + DEFAULT_PADDING_LENGTH));
    if(data==nullptr) {
        fprintf(stderr, "%s: error allocating memory for \"%s\"\n", pname, argv[1]);
        rc=2;
        goto done;
    }

    /* set the last 15 bytes to the usual padding byte, see udata_swapPackage() */
    uprv_memset(data+length-DEFAULT_PADDING_LENGTH, 0xaa, DEFAULT_PADDING_LENGTH);

    if (length != static_cast<int32_t>(fread(data, 1, length, in))) {
        fprintf(stderr, "%s: error reading \"%s\"\n", pname, argv[1]);
        rc=3;
        goto done;
    }

    fclose(in);
    in=nullptr;

    /* swap the data in-place */
    errorCode=U_ZERO_ERROR;
    ds=udata_openSwapperForInputData(data, length, outIsBigEndian, outCharset, &errorCode);
    if(U_FAILURE(errorCode)) {
        fprintf(stderr, "%s: udata_openSwapperForInputData(\"%s\") failed - %s\n",
                pname, argv[1], u_errorName(errorCode));
        rc=4;
        goto done;
    }

    ds->printError=printError;
    ds->printErrorContext=stderr;

    /* speculative cast, protected by the following length check */
    pInfo = reinterpret_cast<const UDataInfo*>(data + 4);

    if( length>=20 &&
        pInfo->dataFormat[0]==0x43 &&   /* dataFormat="CmnD" */
        pInfo->dataFormat[1]==0x6d &&
        pInfo->dataFormat[2]==0x6e &&
        pInfo->dataFormat[3]==0x44
    ) {
        /*
         * swap the .dat package
         * udata_swapPackage() needs to rename ToC name entries from the old package
         * name to the new one.
         * We pass it the filenames, and udata_swapPackage() will extract the
         * package names.
         */

        length=udata_swapPackage(argv[1], argv[2], ds, data, length, data, &errorCode);
        udata_closeSwapper(ds);
        if(U_FAILURE(errorCode)) {
            fprintf(stderr, "%s: udata_swapPackage(\"%s\") failed - %s\n",
                    pname, argv[1], u_errorName(errorCode));
            rc=4;
            goto done;
        }
    } else {
        /* swap the data, which is not a .dat package */
        length=udata_swap(ds, data, length, data, &errorCode);
        udata_closeSwapper(ds);
        if(U_FAILURE(errorCode)) {
            fprintf(stderr, "%s: udata_swap(\"%s\") failed - %s\n",
                    pname, argv[1], u_errorName(errorCode));
            rc=4;
            goto done;
        }
    }

    out=fopen(argv[2], "wb");
    if(out==nullptr) {
        fprintf(stderr, "%s: unable to open output file \"%s\"\n", pname, argv[2]);
        rc=5;
        goto done;
    }

    if (length != static_cast<int32_t>(fwrite(data, 1, length, out))) {
        fprintf(stderr, "%s: error writing \"%s\"\n", pname, argv[2]);
        rc=6;
        goto done;
    }

    fclose(out);
    out=nullptr;

    /* all done */
    rc=0;

done:
    if(in!=nullptr) {
        fclose(in);
    }
    if(out!=nullptr) {
        fclose(out);
    }
    if(data!=nullptr) {
        free(data);
    }
    return rc;
}

/* swap .dat package files -------------------------------------------------- */

static int32_t
extractPackageName(const UDataSwapper *ds, const char *filename,
                   char pkg[], int32_t capacity,
                   UErrorCode *pErrorCode) {
    const char *basename;
    int32_t len;

    if(U_FAILURE(*pErrorCode)) {
        return 0;
    }

    basename=findBasename(filename);
    len = static_cast<int32_t>(uprv_strlen(basename)) - 4; /* -4: subtract the length of ".dat" */

    if(len<=0 || 0!=uprv_strcmp(basename+len, ".dat")) {
        udata_printError(ds, "udata_swapPackage(): \"%s\" is not recognized as a package filename (must end with .dat)\n",
                         basename);
        *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
        return 0;
    }

    if(len>=capacity) {
        udata_printError(ds, "udata_swapPackage(): the package name \"%s\" is too long (>=%ld)\n",
                         static_cast<long>(capacity));
        *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
        return 0;
    }

    uprv_memcpy(pkg, basename, len);
    pkg[len]=0;
    return len;
}

struct ToCEntry {
    uint32_t nameOffset, inOffset, outOffset, length;
};

U_CDECL_BEGIN
static int32_t U_CALLCONV
compareToCEntries(const void *context, const void *left, const void *right) {
    const char *chars=(const char *)context;
    return (int32_t)uprv_strcmp(chars+((const ToCEntry *)left)->nameOffset,
                                chars+((const ToCEntry *)right)->nameOffset);
}
U_CDECL_END

U_CFUNC int32_t U_CALLCONV
udata_swapPackage(const char *inFilename, const char *outFilename,
                  const UDataSwapper *ds,
                  const void *inData, int32_t length, void *outData,
                  UErrorCode *pErrorCode) {
    const UDataInfo *pInfo;
    int32_t headerSize;

    const uint8_t *inBytes;
    uint8_t *outBytes;

    uint32_t itemCount, offset, i;
    int32_t itemLength;

    const UDataOffsetTOCEntry *inEntries;
    UDataOffsetTOCEntry *outEntries;

    ToCEntry *table;

    char inPkgName[32], outPkgName[32];
    int32_t inPkgNameLength, outPkgNameLength;

    /* udata_swapDataHeader checks the arguments */
    headerSize=udata_swapDataHeader(ds, inData, length, outData, pErrorCode);
    if(pErrorCode==nullptr || U_FAILURE(*pErrorCode)) {
        return 0;
    }

    /* check data format and format version */
    pInfo=(const UDataInfo *)((const char *)inData+4);
    if(!(
        pInfo->dataFormat[0]==0x43 &&   /* dataFormat="CmnD" */
        pInfo->dataFormat[1]==0x6d &&
        pInfo->dataFormat[2]==0x6e &&
        pInfo->dataFormat[3]==0x44 &&
        pInfo->formatVersion[0]==1
    )) {
        udata_printError(ds, "udata_swapPackage(): data format %02x.%02x.%02x.%02x (format version %02x) is not recognized as an ICU .dat package\n",
                         pInfo->dataFormat[0], pInfo->dataFormat[1],
                         pInfo->dataFormat[2], pInfo->dataFormat[3],
                         pInfo->formatVersion[0]);
        *pErrorCode=U_UNSUPPORTED_ERROR;
        return 0;
    }

    /*
     * We need to change the ToC name entries so that they have the correct
     * package name prefix.
     * Extract the package names from the in/out filenames.
     */

    inPkgNameLength=extractPackageName(
                        ds, inFilename,
                        inPkgName, (int32_t)sizeof(inPkgName),
                        pErrorCode);
    outPkgNameLength=extractPackageName(
                        ds, outFilename,
                        outPkgName, (int32_t)sizeof(outPkgName),
                        pErrorCode);
    if(U_FAILURE(*pErrorCode)) {
        return 0;
    }

    /*
     * It is possible to work with inPkgNameLength!=outPkgNameLength,
     * but then the length of the data file would change more significantly,
     * which we are not currently prepared for.
     */

    if(inPkgNameLength!=outPkgNameLength) {
        udata_printError(ds, "udata_swapPackage(): the package names \"%s\" and \"%s\" must have the same length\n",
                         inPkgName, outPkgName);
        *pErrorCode=U_ILLEGAL_ARGUMENT_ERROR;
        return 0;
    }

    inBytes=(const uint8_t *)inData+headerSize;
    inEntries=(const UDataOffsetTOCEntry *)(inBytes+4);

    if(length<0) {
        /* preflighting */
        itemCount=ds->readUInt32(*(const uint32_t *)inBytes);
        if(itemCount==0) {
            /* no items: count only the item count and return */
            return headerSize+4;
        }

        /* read the last item's offset and preflight it */
        offset=ds->readUInt32(inEntries[itemCount-1].dataOffset);
        itemLength=udata_swap(ds, inBytes+offset, -1, nullptr, pErrorCode);

        if(U_SUCCESS(*pErrorCode)) {
            return headerSize+offset+(uint32_t)itemLength;
        } else {
            return 0;
        }
    } else {
        /* check that the itemCount fits, then the ToC table, then at least the header of the last item */
        length-=headerSize;
        if(length<4) {
            /* itemCount does not fit */
            offset=0xffffffff;
            itemCount=0; /* make compilers happy */
        } else {
            itemCount=ds->readUInt32(*(const uint32_t *)inBytes);
            if(itemCount==0) {
                offset=4;
            } else if((uint32_t)length<(4+8*itemCount)) {
                /* ToC table does not fit */
                offset=0xffffffff;
            } else {
                /* offset of the last item plus at least 20 bytes for its header */
                offset=20+ds->readUInt32(inEntries[itemCount-1].dataOffset);
            }
        }
        if((uint32_t)length<offset) {
            udata_printError(ds, "udata_swapPackage(): too few bytes (%d after header) for a .dat package\n",
                             length);
            *pErrorCode=U_INDEX_OUTOFBOUNDS_ERROR;
            return 0;
        }

        outBytes=(uint8_t *)outData+headerSize;

        /* swap the item count */
        ds->swapArray32(ds, inBytes, 4, outBytes, pErrorCode);

        if(itemCount==0) {
            /* no items: just return now */
            return headerSize+4;
        }

        /* swap the item name strings */
        offset=4+8*itemCount;
        itemLength=(int32_t)(ds->readUInt32(inEntries[0].dataOffset)-offset);
        udata_swapInvStringBlock(ds, inBytes+offset, itemLength, outBytes+offset, pErrorCode);
        if(U_FAILURE(*pErrorCode)) {
            udata_printError(ds, "udata_swapPackage() failed to swap the data item name strings\n");
            return 0;
        }
        /* keep offset and itemLength in case we allocate and copy the strings below */

        /* swap the package names into the output charset */
        if(ds->outCharset!=U_CHARSET_FAMILY) {
            UDataSwapper *ds2;
            ds2=udata_openSwapper(true, U_CHARSET_FAMILY, true, ds->outCharset, pErrorCode);
            ds2->swapInvChars(ds2, inPkgName, inPkgNameLength, inPkgName, pErrorCode);
            ds2->swapInvChars(ds2, outPkgName, outPkgNameLength, outPkgName, pErrorCode);
            udata_closeSwapper(ds2);
            if(U_FAILURE(*pErrorCode)) {
                udata_printError(ds, "udata_swapPackage() failed to swap the input/output package names\n");
            }
        }

        /* change the prefix of each ToC entry name from the old to the new package name */
        {
            char *entryName;

            for(i=0; i<itemCount; ++i) {
                entryName=(char *)inBytes+ds->readUInt32(inEntries[i].nameOffset);

                if(0==uprv_memcmp(entryName, inPkgName, inPkgNameLength)) {
                    uprv_memcpy(entryName, outPkgName, inPkgNameLength);
                } else {
                    udata_printError(ds, "udata_swapPackage() failed: ToC item %ld does not have the input package name as a prefix\n",
                                     (long)i);
                    *pErrorCode=U_INVALID_FORMAT_ERROR;
                    return 0;
                }
            }
        }

        /*
         * Allocate the ToC table and, if necessary, a temporary buffer for
         * pseudo-in-place swapping.
         *
         * We cannot swap in-place because:
         *
         * 1. If the swapping of an item fails mid-way, then in-place swapping
         * has destroyed its data.
         * Out-of-place swapping allows us to then copy its original data.
         *
         * 2. If swapping changes the charset family, then we must resort
         * not only the ToC table but also the data items themselves.
         * This requires a permutation and is best done with separate in/out
         * buffers.
         *
         * We swapped the strings above to avoid the malloc below if string swapping fails.
         */

        if(inData==outData) {
            /* +15: prepare for extra padding of a newly-last item */
            table=(ToCEntry *)uprv_malloc(itemCount*sizeof(ToCEntry)+length+DEFAULT_PADDING_LENGTH);
            if(table!=nullptr) {
                outBytes=(uint8_t *)(table+itemCount);

                /* copy the item count and the swapped strings */
                uprv_memcpy(outBytes, inBytes, 4);
                uprv_memcpy(outBytes+offset, inBytes+offset, itemLength);
            }
        } else {
            table=(ToCEntry *)uprv_malloc(itemCount*sizeof(ToCEntry));
        }
        if(table==nullptr) {
            udata_printError(ds, "udata_swapPackage(): out of memory allocating %d bytes\n",
                             inData==outData ?
                                 itemCount*sizeof(ToCEntry)+length+DEFAULT_PADDING_LENGTH :
                                 itemCount*sizeof(ToCEntry));
            *pErrorCode=U_MEMORY_ALLOCATION_ERROR;
            return 0;
        }
        outEntries=(UDataOffsetTOCEntry *)(outBytes+4);

        /* read the ToC table */
        for(i=0; i<itemCount; ++i) {
            table[i].nameOffset=ds->readUInt32(inEntries[i].nameOffset);
            table[i].inOffset=ds->readUInt32(inEntries[i].dataOffset);
            if(i>0) {
                table[i-1].length=table[i].inOffset-table[i-1].inOffset;
            }
        }
        table[itemCount-1].length=(uint32_t)length-table[itemCount-1].inOffset;

        if(ds->inCharset==ds->outCharset) {
            /* no charset swapping, no resorting: keep item offsets the same */
            for(i=0; i<itemCount; ++i) {
                table[i].outOffset=table[i].inOffset;
            }
        } else {
            /* charset swapping: resort items by their swapped names */

            /*
             * Before the actual sorting, we need to make sure that each item
             * has a length that is a multiple of 16 bytes so that all items
             * are 16-aligned.
             * Only the old last item may be missing up to 15 padding bytes.
             * Add padding bytes for it.
             * Since the icuswap main() function has already allocated enough
             * input buffer space and set the last 15 bytes there to 0xaa,
             * we only need to increase the total data length and the length
             * of the last item here.
             */

            if((length&0xf)!=0) {
                int32_t delta=16-(length&0xf);
                length+=delta;
                table[itemCount-1].length+=(uint32_t)delta;
            }

            /* Save the offset before we sort the TOC. */
            offset=table[0].inOffset;
            /* sort the TOC entries */
            uprv_sortArray(table, (int32_t)itemCount, (int32_t)sizeof(ToCEntry),
                           compareToCEntries, outBytes, false, pErrorCode);

            /*
             * Note: Before sorting, the inOffset values were in order.
             * Now the outOffset values are in order.
             */


            /* assign outOffset values */
            for(i=0; i<itemCount; ++i) {
                table[i].outOffset=offset;
                offset+=table[i].length;
            }
        }

        /* write the output ToC table */
        for(i=0; i<itemCount; ++i) {
            ds->writeUInt32(&outEntries[i].nameOffset, table[i].nameOffset);
            ds->writeUInt32(&outEntries[i].dataOffset, table[i].outOffset);
        }

        /* swap each data item */
        for(i=0; i<itemCount; ++i) {
            /* first copy the item bytes to make sure that unreachable bytes are copied */ 
            uprv_memcpy(outBytes+table[i].outOffset, inBytes+table[i].inOffset, table[i].length);

            /* swap the item */
            udata_swap(ds, inBytes+table[i].inOffset, (int32_t)table[i].length,
                          outBytes+table[i].outOffset, pErrorCode);

            if(U_FAILURE(*pErrorCode)) {
                if(ds->outCharset==U_CHARSET_FAMILY) {
                    udata_printError(ds, "warning: udata_swapPackage() failed to swap item \"%s\"\n"
                                         " at inOffset 0x%x length 0x%x - %s\n"
                                         " the data item will be copied, not swapped\n\n",
                                     (char *)outBytes+table[i].nameOffset,
                                     table[i].inOffset, table[i].length, u_errorName(*pErrorCode));
                } else {
                    udata_printError(ds, "warning: udata_swapPackage() failed to swap an item\n"
                                         " at inOffset 0x%x length 0x%x - %s\n"
                                         " the data item will be copied, not swapped\n\n",
                                     table[i].inOffset, table[i].length, u_errorName(*pErrorCode));
                }
                /* reset the error code, copy the data item, and continue */
                *pErrorCode=U_ZERO_ERROR;
                uprv_memcpy(outBytes+table[i].outOffset, inBytes+table[i].inOffset, table[i].length);
            }
        }

        if(inData==outData) {
            /* copy the data from the temporary buffer to the in-place buffer */
            uprv_memcpy((uint8_t *)outData+headerSize, outBytes, length);
        }
        uprv_free(table);

        return headerSize+length;
    }
}

/*
 * Hey, Emacs, please set the following:
 *
 * Local Variables:
 * indent-tabs-mode: nil
 * End:
 *
 */


Messung V0.5
C=92 H=86 G=88

¤ Dauer der Verarbeitung: 0.7 Sekunden  ¤

*© Formatika GbR, Deutschland






Wurzel

Suchen

Beweissystem der NASA

Beweissystem Isabelle

NIST Cobol Testsuite

Cephes Mathematical Library

Wiener Entwicklungsmethode

Haftungshinweis

Die Informationen auf dieser Webseite wurden nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit, noch Qualität der bereit gestellten Informationen zugesichert.

Bemerkung:

Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.