C++ sarrayGetString函数代码示例

/*
 *  captureProtoSignature()
 *
 *      Input:  sa (output from cpp, by line)
 *              start (starting index to search; never a comment line)
 *              stop (index of line on which pattern is completed)
 *              charindex (char index of completing ')' character)
 *      Return: cleanstr (prototype string), or NULL on error
 *
 *  Notes:
 *      (1) Return all characters, ending with a ';' after the ')'
 */
static char *
captureProtoSignature(SARRAY  *sa,
                      l_int32  start,
                      l_int32  stop,
                      l_int32  charindex)
{
char    *str, *newstr, *protostr, *cleanstr;
SARRAY  *sap;
l_int32  i;

    PROCNAME("captureProtoSignature");

    if (!sa)
        return (char *)ERROR_PTR("sa not defined", procName, NULL);

    sap = sarrayCreate(0);
    for (i = start; i < stop; i++) {
        str = sarrayGetString(sa, i, L_COPY);
        sarrayAddString(sap, str, L_INSERT);
    }
    str = sarrayGetString(sa, stop, L_COPY);
    str[charindex + 1] = '\0';
    newstr = stringJoin(str, ";");
    sarrayAddString(sap, newstr, L_INSERT);
    LEPT_FREE(str);
    protostr = sarrayToString(sap, 2);
    sarrayDestroy(&sap);
    cleanstr = cleanProtoSignature(protostr);
    LEPT_FREE(protostr);

    return cleanstr;
}

/*!
 * \brief   l_asetCreateFromSarray()
 *
 * \param[in]    sa
 * \return  set using a string hash into a uint32 as the key
 */
L_ASET *
l_asetCreateFromSarray(SARRAY  *sa)
{
char     *str;
l_int32   i, n;
l_uint64  hash;
L_ASET   *set;
RB_TYPE   key;

    PROCNAME("l_asetCreateFromSarray");

    if (!sa)
        return (L_ASET *)ERROR_PTR("sa not defined", procName, NULL);

    set = l_asetCreate(L_UINT_TYPE);
    n = sarrayGetCount(sa);
    for (i = 0; i < n; i++) {
        str = sarrayGetString(sa, i, L_NOCOPY);
        l_hashStringToUint64(str, &hash);
        key.utype = hash;
        l_asetInsert(set, key);
    }

    return set;
}

/*!
 * \brief   recogAddCharstrLabels()
 *
 * \param[in]    recog
 * \return  0 if OK, 1 on error
 */
static l_int32
recogAddCharstrLabels(L_RECOG  *recog)
{
char    *text;
l_int32  i, j, n1, n2;
PIX     *pix;
PIXA    *pixa;
PIXAA   *paa;

    PROCNAME("recogAddCharstrLabels");

    if (!recog)
        return ERROR_INT("recog not defined", procName, 1);

        /* Add the labels to each unscaled pix */
    paa = recog->pixaa_u;
    n1 = pixaaGetCount(paa, NULL);
    for (i = 0; i < n1; i++) {
        pixa = pixaaGetPixa(paa, i, L_CLONE);
        text = sarrayGetString(recog->sa_text, i, L_NOCOPY);
        n2 = pixaGetCount(pixa);
        for (j = 0; j < n2; j++) {
             pix = pixaGetPix(pixa, j, L_CLONE);
             pixSetText(pix, text);
             pixDestroy(&pix);
        }
        pixaDestroy(&pixa);
    }

    return 0;
}

/*!
 * \brief   sarrayRemoveDupsByAset()
 *
 * \param[in]    sas
 * \return  sad with duplicates removed, or NULL on error
 *
 * <pre>
 * Notes:
 *      (1) This is O(nlogn), considerably slower than
 *          sarrayRemoveDupsByHash() for large string arrays.
 *      (2) The key for each string is a 64-bit hash.
 *      (3) Build a set, using hashed strings as keys.  As the set is
 *          built, first do a find; if not found, add the key to the
 *          set and add the string to the output sarray.
 * </pre>
 */
SARRAY *
sarrayRemoveDupsByAset(SARRAY  *sas)
{
char     *str;
l_int32   i, n;
l_uint64  hash;
L_ASET   *set;
RB_TYPE   key;
SARRAY   *sad;

    PROCNAME("sarrayRemoveDupsByAset");

    if (!sas)
        return (SARRAY *)ERROR_PTR("sas not defined", procName, NULL);

    set = l_asetCreate(L_UINT_TYPE);
    sad = sarrayCreate(0);
    n = sarrayGetCount(sas);
    for (i = 0; i < n; i++) {
        str = sarrayGetString(sas, i, L_NOCOPY);
        l_hashStringToUint64(str, &hash);
        key.utype = hash;
        if (!l_asetFind(set, key)) {
            sarrayAddString(sad, str, L_COPY);
            l_asetInsert(set, key);
        }
    }

    l_asetDestroy(&set);
    return sad;
}

/*!
 * \brief   sarraySortByIndex()
 *
 * \param[in]    sain
 * \param[in]    naindex na that maps from the new sarray to the input sarray
 * \return  saout sorted, or NULL on error
 */
SARRAY *
sarraySortByIndex(SARRAY  *sain,
                  NUMA    *naindex)
{
char    *str;
l_int32  i, n, index;
SARRAY  *saout;

    PROCNAME("sarraySortByIndex");

    if (!sain)
        return (SARRAY *)ERROR_PTR("sain not defined", procName, NULL);
    if (!naindex)
        return (SARRAY *)ERROR_PTR("naindex not defined", procName, NULL);

    n = sarrayGetCount(sain);
    saout = sarrayCreate(n);
    for (i = 0; i < n; i++) {
        numaGetIValue(naindex, i, &index);
        str = sarrayGetString(sain, index, L_COPY);
        sarrayAddString(saout, str, L_INSERT);
    }

    return saout;
}

/*!
 *  recogAddAllSamples()
 *
 *      Input:  recog
 *              paa (pixaa from previously trained recog)
 *              debug
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) This is used with the serialization routine recogRead(),
 *          where each pixa in the pixaa represents a set of characters
 *          in a different class.  Two different pixa may represent
 *          characters with the same label.  Before calling this
 *          function, we verify that the number of character classes,
 *          given by the setsize field in recog, equals the number of
 *          pixa in the paa.  The character labels for each set are
 *          in the sa_text field.
 */
static l_int32
recogAddAllSamples(L_RECOG  *recog,
                   PIXAA    *paa,
                   l_int32   debug)
{
char    *text;
l_int32  i, j, nc, ns;
PIX     *pix;
PIXA    *pixa;

    PROCNAME("recogAddAllSamples");

    if (!recog)
        return ERROR_INT("recog not defined", procName, 1);
    if (!paa)
        return ERROR_INT("paa not defined", procName, 1);

    nc = pixaaGetCount(paa, NULL);
    for (i = 0; i < nc; i++) {
        pixa = pixaaGetPixa(paa, i, L_CLONE);
        ns = pixaGetCount(pixa);
        text = sarrayGetString(recog->sa_text, i, L_NOCOPY);
        for (j = 0; j < ns; j++) {
            pix = pixaGetPix(pixa, j, L_CLONE);
            if (debug) {
                fprintf(stderr, "pix[%d,%d]: text = %s\n", i, j, text);
            }
            pixaaAddPix(recog->pixaa_u, i, pix, NULL, L_INSERT);
        }
        pixaDestroy(&pixa);
    }

    recogTrainingFinished(recog, debug);
    return 0;
}

/*!
 *  sarrayAppendRange()
 *
 *      Input:  sa1  (to be added to)
 *              sa2  (append specified range of strings in sa2 to sa1)
 *              start (index of first string of sa2 to append)
 *              end (index of last string of sa2 to append)
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) Copies of the strings in sarray2 are added to sarray1.
 *      (2) The [start ... end] range is truncated if necessary.
 */
l_int32
sarrayAppendRange(SARRAY  *sa1,
                  SARRAY  *sa2,
		  l_int32  start,
		  l_int32  end)
{
char    *str;
l_int32  n, i;

    PROCNAME("sarrayAppendRange");

    if (!sa1)
        return ERROR_INT("sa1 not defined", procName, 1);
    if (!sa2)
        return ERROR_INT("sa2 not defined", procName, 1);
    if (start < 0)
        start = 0;
    n = sarrayGetCount(sa2);
    if (end >= n)
        end = n - 1;
    if (start > end)
        return ERROR_INT("start > end", procName, 1);

    for (i = start; i <= end; i++) {
        str = sarrayGetString(sa2, i, L_NOCOPY);
        sarrayAddString(sa1, str, L_COPY);
    }

    return 0;
}

/*
 *  getNextNonDoubleSlashLine()
 *
 *      Input:  sa (output from cpp, by line)
 *              start (starting index to search)
 *              &next (<return> index of first uncommented line after
 *                     the start line)
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) Skips over all consecutive '//' lines, beginning at 'start'
 *      (2) If all lines to the end start with '//', return next = -1
 */
static l_int32
getNextNonDoubleSlashLine(SARRAY  *sa,
                          l_int32  start,
                          l_int32 *pnext)
{
char    *str;
l_int32  i, n, len;

    PROCNAME("getNextNonDoubleSlashLine");

    if (!sa)
        return ERROR_INT("sa not defined", procName, 1);
    if (!pnext)
        return ERROR_INT("&pnext not defined", procName, 1);

        /* Init for situation where this line and all following
         * start with '//' */
    *pnext = -1;

    n = sarrayGetCount(sa);
    for (i = start; i < n; i++) {
        if ((str = sarrayGetString(sa, i, L_NOCOPY)) == NULL)
            return ERROR_INT("str not returned; shouldn't happen", procName, 1);
        len = strlen(str);
        if (len < 2 || str[0] != '/' || str[1] != '/') {
            *pnext = i;
            return 0;
        }
    }

    return 0;
}

/*
 *  getNextNonBlankLine()
 *
 *      Input:  sa (output from cpp, by line)
 *              start (starting index to search)
 *              &next (<return> index of first nonblank line after
 *                     the start line)
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) Skips over all consecutive blank lines, beginning at 'start'
 *      (2) A blank line has only whitespace characters (' ', '\t', '\n', '\r')
 *      (3) If all lines to the end are blank, return next = -1
 */
static l_int32
getNextNonBlankLine(SARRAY  *sa,
                    l_int32  start,
                    l_int32 *pnext)
{
char    *str;
l_int32  i, j, n, len;

    PROCNAME("getNextNonBlankLine");

    if (!sa)
        return ERROR_INT("sa not defined", procName, 1);
    if (!pnext)
        return ERROR_INT("&pnext not defined", procName, 1);

        /* Init for situation where this line and all following are blank */
    *pnext = -1;

    n = sarrayGetCount(sa);
    for (i = start; i < n; i++) {
        if ((str = sarrayGetString(sa, i, L_NOCOPY)) == NULL)
            return ERROR_INT("str not returned; shouldn't happen", procName, 1);
        len = strlen(str);
        for (j = 0; j < len; j++) {
            if (str[j] != ' ' && str[j] != '\t'
                && str[j] != '\n' && str[j] != '\r') {  /* non-blank */
                *pnext = i;
                return 0;
            }
        }
    }

    return 0;
}

/*!
 *  pixReadIndexed()
 *
 *      Input:  sarray (of full pathnames)
 *              index (into pathname array)
 *      Return: pix if OK; null if not found
 *
 *  Notes:
 *      (1) This function is useful for selecting image files from a
 *          directory, where the integer @index is embedded into
 *          the file name.
 *      (2) This is typically done by generating the sarray using
 *          getNumberedPathnamesInDirectory(), so that the @index
 *          pathname would have the number @index in it.  The size
 *          of the sarray should be the largest number (plus 1) appearing
 *          in the file names, respecting the constraints in the
 *          call to getNumberedPathnamesInDirectory().
 *      (3) Consequently, for some indices into the sarray, there may
 *          be no pathnames in the directory containing that number.
 *          By convention, we place empty C strings ("") in those
 *          locations in the sarray, and it is not an error if such
 *          a string is encountered and no pix is returned.
 *          Therefore, the caller must verify that a pix is returned.
 *      (4) See convertSegmentedPagesToPS() in src/psio1.c for an
 *          example of usage.
 */
PIX *
pixReadIndexed(SARRAY  *sa,
               l_int32  index)
{
char    *fname;
l_int32  n;
PIX     *pix;

    PROCNAME("pixReadIndexed");

    if (!sa)
        return (PIX *)ERROR_PTR("sa not defined", procName, NULL);
    n = sarrayGetCount(sa);
    if (index < 0 || index >= n)
        return (PIX *)ERROR_PTR("index out of bounds", procName, NULL);

    fname = sarrayGetString(sa, index, L_NOCOPY);
    if (fname[0] == '\0')
        return NULL;

    if ((pix = pixRead(fname)) == NULL) {
        L_ERROR("pix not read from file %s\n", procName, fname);
        return NULL;
    }

    return pix;
}

/*
 *  getNextNonCommentLine()
 *
 *      Input:  sa (output from cpp, by line)
 *              start (starting index to search)
 *              &next (<return> index of first uncommented line after
 *                     the start line)
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) Skips over all consecutive comment lines, beginning at 'start'
 *      (2) If all lines to the end are '#' comments, return next = -1
 */
static l_int32
getNextNonCommentLine(SARRAY  *sa,
                      l_int32  start,
                      l_int32 *pnext)
{
char    *str;
l_int32  i, n;

    PROCNAME("getNextNonCommentLine");

    if (!sa)
        return ERROR_INT("sa not defined", procName, 1);
    if (!pnext)
        return ERROR_INT("&pnext not defined", procName, 1);

        /* Init for situation where this line and all following are comments */
    *pnext = -1;

    n = sarrayGetCount(sa);
    for (i = start; i < n; i++) {
        if ((str = sarrayGetString(sa, i, L_NOCOPY)) == NULL)
            return ERROR_INT("str not returned; shouldn't happen", procName, 1);
        if (str[0] != '#') {
            *pnext = i;
            return 0;
        }
    }

    return 0;
}

/*!
 *  sarraySelectBySubstring()
 *
 *      Input:  sain (input sarray)
 *              substr (<optional> substring for matching; can be NULL)
 *      Return: saout (output sarray, filtered with substring) or null on error
 *
 *  Notes:
 *      (1) This selects all strings in sain that have substr as a substring.
 *          Note that we can't use strncmp() because we're looking for
 *          a match to the substring anywhere within each filename.
 *      (2) If substr == NULL, returns a copy of the sarray.
 */
SARRAY *
sarraySelectBySubstring(SARRAY      *sain,
                        const char  *substr)
{
char    *str;
l_int32  n, i, offset, found;
SARRAY  *saout;

    PROCNAME("sarraySelectBySubstring");

    if (!sain)
        return (SARRAY *)ERROR_PTR("sain not defined", procName, NULL);

    n = sarrayGetCount(sain);
    if (!substr || n == 0)
        return sarrayCopy(sain);

    saout = sarrayCreate(n);
    for (i = 0; i < n; i++) {
        str = sarrayGetString(sain, i, L_NOCOPY);
        arrayFindSequence((l_uint8 *)str, strlen(str), (l_uint8 *)substr,
                          strlen(substr), &offset, &found);
        if (found)
            sarrayAddString(saout, str, L_COPY);
    }

    return saout;
}

/*!
 *  pixaReadFilesSA()
 *
 *      Input:  sarray (full pathnames for all files)
 *      Return: pixa, or null on error
 */
PIXA *
pixaReadFilesSA(SARRAY  *sa)
{
char    *str;
l_int32  i, n;
PIX     *pix;
PIXA    *pixa;

    PROCNAME("pixaReadFilesSA");

    if (!sa)
        return (PIXA *)ERROR_PTR("sa not defined", procName, NULL);

    n = sarrayGetCount(sa);
    pixa = pixaCreate(n);
    for (i = 0; i < n; i++) {
        str = sarrayGetString(sa, i, L_NOCOPY);
        if ((pix = pixRead(str)) == NULL) {
            L_WARNING("pix not read from file %s\n", procName, str);
            continue;
        }
        pixaAddPix(pixa, pix, L_INSERT);
    }

    return pixa;
}

/*
 *  sarrayConvertFilesToPS()
 *
 *      Input:  sarray (of full path names)
 *              res (typ. 300 or 600 ppi)
 *              fileout (output ps file)
 *      Return: 0 if OK, 1 on error
 *
 *  Notes:
 *      (1) See convertFilesToPS()
 */
l_int32
sarrayConvertFilesToPS(SARRAY      *sa,
                       l_int32      res,
                       const char  *fileout)
{
char    *fname;
l_int32  i, nfiles, index, firstfile, ret, format;

    PROCNAME("sarrayConvertFilesToPS");

    if (!sa)
        return ERROR_INT("sa not defined", procName, 1);
    if (!fileout)
        return ERROR_INT("fileout not defined", procName, 1);
    if (res <= 0) {
        L_INFO("setting res to 300 ppi", procName);
        res = 300;
    }
    if (res < 10 || res > 4000)
        L_WARNING("res is typically in the range 300-600 ppi", procName);

    nfiles = sarrayGetCount(sa);
    firstfile = TRUE;
    for (i = 0, index = 0; i < nfiles; i++) {
        fname = sarrayGetString(sa, i, L_NOCOPY);
        ret = pixReadHeader(fname, &format, NULL, NULL, NULL, NULL, NULL);
        if (ret) continue;
        if (format == IFF_UNKNOWN)
            continue;

        writeImageCompressedToPSFile(fname, fileout, res, &firstfile, &index);
    }

    return 0;
}

/*!
 * \brief   l_dnaHashCreateFromSarray()
 *
 * \param[in]    sa
 * \return  dahash, or NULL on error
 */
L_DNAHASH *
l_dnaHashCreateFromSarray(SARRAY  *sa)
{
char       *str;
l_int32     i, n;
l_uint32    nsize;
l_uint64    key;
L_DNAHASH  *dahash;

        /* Build up dnaHash of indices, hashed by a 64-bit key that
         * should randomize the lower bits used in bucket selection.
         * Having about 20 pts in each bucket is roughly optimal. */
    n = sarrayGetCount(sa);
    findNextLargerPrime(n / 20, &nsize);  /* buckets in hash table */
/*    fprintf(stderr, "Prime used: %d\n", nsize); */

        /* Add each string, using the hash as key and the index into %sa
         * as the value.  Storing the index enables operations that check
         * for duplicates.  */
    dahash = l_dnaHashCreate(nsize, 8);
    for (i = 0; i < n; i++) {
        str = sarrayGetString(sa, i, L_NOCOPY);
        l_hashStringToUint64(str, &key);
        l_dnaHashAdd(dahash, key, (l_float64)i);
    }

    return dahash;
}