xmlstring.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

    if ((str2 == NULL) || (len == 0))
        return(xmlStrdup(str1));
    if (str1 == NULL)
        return(xmlStrndup(str2, len));

    size = xmlStrlen(str1);
    ret = (xmlChar *) xmlMalloc((size + len + 1) * sizeof(xmlChar));
    if (ret == NULL) {
        xmlErrMemory(NULL, NULL);
        return(xmlStrndup(str1, size));
    }
    memcpy(ret, str1, size * sizeof(xmlChar));
    memcpy(&ret[size], str2, len * sizeof(xmlChar));
    ret[size + len] = 0;
    return(ret);
}

/**
 * xmlStrcat:
 * @cur:  the original xmlChar * array
 * @add:  the xmlChar * array added
 *
 * a strcat for array of xmlChar's. Since they are supposed to be
 * encoded in UTF-8 or an encoding with 8bit based chars, we assume
 * a termination mark of '0'.
 *
 * Returns a new xmlChar * containing the concatenated string.
 */
xmlChar *
xmlStrcat(xmlChar *cur, const xmlChar *add) {
    const xmlChar *p = add;

    if (add == NULL) return(cur);
    if (cur == NULL) 
        return(xmlStrdup(add));

    while (*p != 0) p++; /* non input consuming */
    return(xmlStrncat(cur, add, p - add));
}

/**
 * xmlStrPrintf:
 * @buf:   the result buffer.
 * @len:   the result buffer length.
 * @msg:   the message with printf formatting.
 * @...:   extra parameters for the message.
 *
 * Formats @msg and places result into @buf.
 *
 * Returns the number of characters written to @buf or -1 if an error occurs.
 */
int 
xmlStrPrintf(xmlChar *buf, int len, const xmlChar *msg, ...) {
    va_list args;
    int ret;
    
    if((buf == NULL) || (msg == NULL)) {
        return(-1);
    }
    
    va_start(args, msg);
    ret = _vsnprintf((char *) buf, len, (const char *) msg, args);
    va_end(args);
    buf[len - 1] = 0; /* be safe ! */
    
    return(ret);
}

/**
 * xmlStrVPrintf:
 * @buf:   the result buffer.
 * @len:   the result buffer length.
 * @msg:   the message with printf formatting.
 * @ap:    extra parameters for the message.
 *
 * Formats @msg and places result into @buf.
 *
 * Returns the number of characters written to @buf or -1 if an error occurs.
 */
int 
xmlStrVPrintf(xmlChar *buf, int len, const xmlChar *msg, va_list ap) {
    int ret;
    
    if((buf == NULL) || (msg == NULL)) {
        return(-1);
    }
    
    ret = _vsnprintf((char *) buf, len, (const char *) msg, ap);
    buf[len - 1] = 0; /* be safe ! */
    
    return(ret);
}

/************************************************************************
 *                                                                      *
 *              Generic UTF8 handling routines                          *
 *                                                                      *
 * From rfc2044: encoding of the Unicode values on UTF-8:               *
 *                                                                      *
 * UCS-4 range (hex.)           UTF-8 octet sequence (binary)           *
 * 0000 0000-0000 007F   0xxxxxxx                                       *
 * 0000 0080-0000 07FF   110xxxxx 10xxxxxx                              *
 * 0000 0800-0000 FFFF   1110xxxx 10xxxxxx 10xxxxxx                     *
 *                                                                      *
 * I hope we won't use values > 0xFFFF anytime soon !                   *
 *                                                                      *
 ************************************************************************/


/**
 * xmlUTF8Size:
 * @utf: pointer to the UTF8 character
 *
 * calculates the internal size of a UTF8 character
 *
 * returns the numbers of bytes in the character, -1 on format error
 */
int
xmlUTF8Size(const xmlChar *utf) {
    xmlChar mask;
    int len;

    if (utf == NULL)
        return -1;
    if (*utf < 0x80)
        return 1;
    /* check valid UTF8 character */
    if (!(*utf & 0x40))
        return -1;
    /* determine number of bytes in char */
    len = 2;
    for (mask=0x20; mask != 0; mask>>=1) {
        if (!(*utf & mask))
            return len;
        len++;
    }
    return -1;
}

/**
 * xmlUTF8Charcmp:
 * @utf1: pointer to first UTF8 char
 * @utf2: pointer to second UTF8 char
 *
 * compares the two UCS4 values
 *
 * returns result of the compare as with xmlStrncmp
 */
int
xmlUTF8Charcmp(const xmlChar *utf1, const xmlChar *utf2) {

    if (utf1 == NULL ) {
        if (utf2 == NULL)
            return 0;
        return -1;
    }
    return xmlStrncmp(utf1, utf2, xmlUTF8Size(utf1));
}

/**
 * xmlUTF8Strlen:
 * @utf:  a sequence of UTF-8 encoded bytes
 *
 * compute the length of an UTF8 string, it doesn't do a full UTF8
 * checking of the content of the string.
 *
 * Returns the number of characters in the string or -1 in case of error
 */
int
xmlUTF8Strlen(const xmlChar *utf) {
    int ret = 0;

    if (utf == NULL)
        return(-1);

    while (*utf != 0) {
        if (utf[0] & 0x80) {
            if ((utf[1] & 0xc0) != 0x80)
                return(-1);
            if ((utf[0] & 0xe0) == 0xe0) {
                if ((utf[2] & 0xc0) != 0x80)
                    return(-1);
                if ((utf[0] & 0xf0) == 0xf0) {
                    if ((utf[0] & 0xf8) != 0xf0 || (utf[3] & 0xc0) != 0x80)
                        return(-1);
                    utf += 4;
                } else {
                    utf += 3;
                }
            } else {
                utf += 2;
            }
        } else {
            utf++;
        }
        ret++;
    }
    return(ret);
}

/**
 * xmlGetUTF8Char:
 * @utf:  a sequence of UTF-8 encoded bytes
 * @len:  a pointer to @bytes len
 *
 * Read one UTF8 Char from @utf
 *
 * Returns the char value or -1 in case of error, and updates *len with the
 *        number of bytes consumed
 */
int
xmlGetUTF8Char(const unsigned char *utf, int *len) {
    unsigned int c;

    if (utf == NULL)
        goto error;
    if (len == NULL)
        goto error;
    if (*len < 1)
        goto error;

    c = utf[0];
    if (c & 0x80) {
        if (*len < 2)
            goto error;
        if ((utf[1] & 0xc0) != 0x80)
            goto error;
        if ((c & 0xe0) == 0xe0) {
            if (*len < 3)
                goto error;
            if ((utf[2] & 0xc0) != 0x80)
                goto error;
            if ((c & 0xf0) == 0xf0) {
                if (*len < 4)
                    goto error;
                if ((c & 0xf8) != 0xf0 || (utf[3] & 0xc0) != 0x80)
                    goto error;
                *len = 4;
                /* 4-byte code */
                c = (utf[0] & 0x7) << 18;
                c |= (utf[1] & 0x3f) << 12;
                c |= (utf[2] & 0x3f) << 6;
                c |= utf[3] & 0x3f;
            } else {
              /* 3-byte code */
                *len = 3;
                c = (utf[0] & 0xf) << 12;
                c |= (utf[1] & 0x3f) << 6;
                c |= utf[2] & 0x3f;
            }
        } else {
          /* 2-byte code */
            *len = 2;
            c = (utf[0] & 0x1f) << 6;
            c |= utf[1] & 0x3f;
        }
    } else {
        /* 1-byte code */
        *len = 1;
    }
    return(c);

error:
    *len = 0;
    return(-1);
}

/**
 * xmlCheckUTF8:
 * @utf: Pointer to putative UTF-8 encoded string.
 *
 * Checks @utf for being valid UTF-8. @utf is assumed to be
 * null-terminated. This function is not super-strict, as it will
 * allow longer UTF-8 sequences than necessary. Note that Java is
 * capable of producing these sequences if provoked. Also note, this
 * routine checks for the 4-byte maximum size, but does not check for
 * 0x10ffff maximum value.
 *
 * Return value: true if @utf is valid.
 **/
int
xmlCheckUTF8(const unsigned char *utf)
{
    int ix;
    unsigned char c;

    for (ix = 0; (c = utf[ix]);) {
        if (c & 0x80) {
            if ((utf[ix + 1] & 0xc0) != 0x80)
                return(0);
            if ((c & 0xe0) == 0xe0) {
                if ((utf[ix + 2] & 0xc0) != 0x80)
                    return(0);
                if ((c & 0xf0) == 0xf0) {
                    if ((c & 0xf8) != 0xf0 || (utf[ix + 3] & 0xc0) != 0x80)
                        return(0);
                    ix += 4;
                    /* 4-byte code */
                } else
                    /* 3-byte code */
                    ix += 3;
            } else
                /* 2-byte code */
                ix += 2;
        } else
            /* 1-byte code */
            ix++;
      }
      return(1);
}

/**
 * xmlUTF8Strsize:
 * @utf:  a sequence of UTF-8 encoded bytes
 * @len:  the number of characters in the array
 *
 * storage size of an UTF8 string
 *
 * Returns the storage size of
 * the first 'len' characters of ARRAY
 *
 */

int
xmlUTF8Strsize(const xmlChar *utf, int len) {
    const xmlChar   *ptr=utf;
    xmlChar         ch;

    if (len <= 0)
        return(0);

    while ( len-- > 0) {
        if ( !*ptr )
            break;
        if ( (ch = *ptr++) & 0x80)
            while ( (ch<<=1) & 0x80 )
                ptr++;
    }
    return (ptr - utf);
}


/**
 * xmlUTF8Strndup:
 * @utf:  the input UTF8 *
 * @len:  the len of @utf (in chars)
 *
 * a strndup for array of UTF8's
 *
 * Returns a new UTF8 * or NULL
 */
xmlChar *
xmlUTF8Strndup(const xmlChar *utf, int len) {
    xmlChar *ret;
    int i;
    
    if ((utf == NULL) || (len < 0)) return(NULL);
    i = xmlUTF8Strsize(utf, len);
    ret = (xmlChar *) xmlMallocAtomic((i + 1) * sizeof(xmlChar));
    if (ret == NULL) {
        xmlGenericError(xmlGenericErrorContext,
                "malloc of %ld byte failed\n",
                (len + 1) * (long)sizeof(xmlChar));
        return(NULL);
    }
    memcpy(ret, utf, i * sizeof(xmlChar));
    ret[i] = 0;
    return(ret);
}

/**
 * xmlUTF8Strpos:
 * @utf:  the input UTF8 *
 * @pos:  the position of the desired UTF8 char (in chars)
 *
 * a function to provide the equivalent of fetching a
 * character from a string array
 *
 * Returns a pointer to the UTF8 character or NULL
 */
xmlChar *
xmlUTF8Strpos(const xmlChar *utf, int pos) {
    xmlChar ch;

    if (utf == NULL) return(NULL);
    if ( (pos < 0) || (pos >= xmlUTF8Strlen(utf)) )
        return(NULL);
    while (pos--) {
        if ((ch=*utf++) == 0) return(NULL);
        if ( ch & 0x80 ) {
            /* if not simple ascii, verify proper format */
            if ( (ch & 0xc0) != 0xc0 )
                return(NULL);
            /* then skip over remaining bytes for this char */
            while ( (ch <<= 1) & 0x80 )
                if ( (*utf++ & 0xc0) != 0x80 )
                    return(NULL);
        }
    }
    return((xmlChar *)utf);
}

/**
 * xmlUTF8Strloc:
 * @utf:  the input UTF8 *
 * @utfchar:  the UTF8 character to be found
 *
 * a function to provide the relative location of a UTF8 char
 *
 * Returns the relative character position of the desired char
 * or -1 if not found
 */
int
xmlUTF8Strloc(const xmlChar *utf, const xmlChar *utfchar) {
    int i, size;
    xmlChar ch;

    if (utf==NULL || utfchar==NULL) return -1;
    size = xmlUTF8Strsize(utfchar, 1);
        for(i=0; (ch=*utf) != 0; i++) {
            if (xmlStrncmp(utf, utfchar, size)==0)
                return(i);
            utf++;
            if ( ch & 0x80 ) {
                /* if not simple ascii, verify proper format */
                if ( (ch & 0xc0) != 0xc0 )
                    return(-1);
                /* then skip over remaining bytes for this char */
                while ( (ch <<= 1) & 0x80 )
                    if ( (*utf++ & 0xc0) != 0x80 )
                        return(-1);
            }
        }

    return(-1);
}
/**
 * xmlUTF8Strsub:
 * @utf:  a sequence of UTF-8 encoded bytes
 * @start: relative pos of first char
 * @len:   total number to copy
 *
 * Create a substring from a given UTF-8 string
 * Note:  positions are given in units of UTF-8 chars
 *
 * Returns a pointer to a newly created string
 * or NULL if any problem
 */

xmlChar *
xmlUTF8Strsub(const xmlChar *utf, int start, int len) {
    int            i;
    xmlChar ch;

    if (utf == NULL) return(NULL);
    if (start < 0) return(NULL);
    if (len < 0) return(NULL);

    /*
     * Skip over any leading chars
     */
    for (i = 0;i < start;i++) {
        if ((ch=*utf++) == 0) return(NULL);
        if ( ch & 0x80 ) {
            /* if not simple ascii, verify proper format */
            if ( (ch & 0xc0) != 0xc0 )
                return(NULL);
            /* then skip over remaining bytes for this char */
            while ( (ch <<= 1) & 0x80 )
                if ( (*utf++ & 0xc0) != 0x80 )
                    return(NULL);
        }
    }

    return(xmlUTF8Strndup(utf, len));
}