uri.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

    if (uri->query) {
        segment =
            xmlURIEscapeStr(BAD_CAST uri->query, BAD_CAST ";/?:@&=+,$");
        NULLCHK(segment)
        ret = xmlStrcat(ret, BAD_CAST "?");
        ret = xmlStrcat(ret, segment);
        xmlFree(segment);
    }

    if (uri->opaque) {
        segment = xmlURIEscapeStr(BAD_CAST uri->opaque, BAD_CAST "");
        NULLCHK(segment)
        ret = xmlStrcat(ret, segment);
        xmlFree(segment);
    }

    if (uri->fragment) {
        segment = xmlURIEscapeStr(BAD_CAST uri->fragment, BAD_CAST "#");
        NULLCHK(segment)
        ret = xmlStrcat(ret, BAD_CAST "#");
        ret = xmlStrcat(ret, segment);
        xmlFree(segment);
    }

    xmlFreeURI(uri);
#undef NULLCHK

    return (ret);
}

/************************************************************************
 *									*
 *			Escaped URI parsing				*
 *									*
 ************************************************************************/

/**
 * xmlParseURIFragment:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI fragment string and fills in the appropriate fields
 * of the @uri structure.
 * 
 * fragment = *uric
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIFragment(xmlURIPtr uri, const char **str)
{
    const char *cur = *str;

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

    while (IS_URIC(cur) || IS_UNWISE(cur))
        NEXT(cur);
    if (uri != NULL) {
        if (uri->fragment != NULL)
            xmlFree(uri->fragment);
        uri->fragment = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return (0);
}

/**
 * xmlParseURIQuery:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse the query part of an URI
 * 
 * query = *uric
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIQuery(xmlURIPtr uri, const char **str)
{
    const char *cur = *str;

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

    while (IS_URIC(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))
        NEXT(cur);
    if (uri != NULL) {
        if (uri->query != NULL)
            xmlFree(uri->query);
        uri->query = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return (0);
}

/**
 * xmlParseURIScheme:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI scheme
 * 
 * scheme = alpha *( alpha | digit | "+" | "-" | "." )
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIScheme(xmlURIPtr uri, const char **str) {
    const char *cur;

    if (str == NULL)
	return(-1);
    
    cur = *str;
    if (!IS_ALPHA(*cur))
	return(2);
    cur++;
    while (IS_SCHEME(*cur)) cur++;
    if (uri != NULL) {
	if (uri->scheme != NULL) xmlFree(uri->scheme);
	/* !!! strndup */
	uri->scheme = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return(0);
}

/**
 * xmlParseURIOpaquePart:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI opaque part
 * 
 * opaque_part = uric_no_slash *uric
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIOpaquePart(xmlURIPtr uri, const char **str)
{
    const char *cur;

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

    cur = *str;
    if (!(IS_URIC_NO_SLASH(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))) {
        return (3);
    }
    NEXT(cur);
    while (IS_URIC(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))
        NEXT(cur);
    if (uri != NULL) {
        if (uri->opaque != NULL)
            xmlFree(uri->opaque);
        uri->opaque = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return (0);
}

/**
 * xmlParseURIServer:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse a server subpart of an URI, it's a finer grain analysis
 * of the authority part.
 * 
 * server        = [ [ userinfo "@" ] hostport ]
 * userinfo      = *( unreserved | escaped |
 *                       ";" | ":" | "&" | "=" | "+" | "$" | "," )
 * hostport      = host [ ":" port ]
 * host          = hostname | IPv4address
 * hostname      = *( domainlabel "." ) toplabel [ "." ]
 * domainlabel   = alphanum | alphanum *( alphanum | "-" ) alphanum
 * toplabel      = alpha | alpha *( alphanum | "-" ) alphanum
 * IPv4address   = 1*digit "." 1*digit "." 1*digit "." 1*digit
 * port          = *digit
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIServer(xmlURIPtr uri, const char **str) {
    const char *cur;
    const char *host, *tmp;
    const int IPmax = 4;
    int oct;

    if (str == NULL)
	return(-1);
    
    cur = *str;

    /*
     * is there an userinfo ?
     */
    while (IS_USERINFO(cur)) NEXT(cur);
    if (*cur == '@') {
	if (uri != NULL) {
	    if (uri->user != NULL) xmlFree(uri->user);
	    uri->user = xmlURIUnescapeString(*str, cur - *str, NULL);
	}
	cur++;
    } else {
	if (uri != NULL) {
	    if (uri->user != NULL) xmlFree(uri->user);
	    uri->user = NULL;
	}
        cur = *str;
    }
    /*
     * This can be empty in the case where there is no server
     */
    host = cur;
    if (*cur == '/') {
	if (uri != NULL) {
	    if (uri->authority != NULL) xmlFree(uri->authority);
	    uri->authority = NULL;
	    if (uri->server != NULL) xmlFree(uri->server);
	    uri->server = NULL;
	    uri->port = 0;
	}
	return(0);
    }
    /*
     * host part of hostport can derive either an IPV4 address
     * or an unresolved name. Check the IP first, it easier to detect
     * errors if wrong one
     */
    for (oct = 0; oct < IPmax; ++oct) {
        if (*cur == '.')
            return(3); /* e.g. http://.xml/ or http://18.29..30/ */
        while(IS_DIGIT(*cur)) cur++;
        if (oct == (IPmax-1))
            continue;
        if (*cur != '.')
	    break;
        cur++;
    }
    if (oct < IPmax || (*cur == '.' && cur++) || IS_ALPHA(*cur)) {
        /* maybe host_name */
        if (!IS_ALPHANUM(*cur))
            return(4); /* e.g. http://xml.$oft */
        do {
            do ++cur; while (IS_ALPHANUM(*cur));
            if (*cur == '-') {
	        --cur;
                if (*cur == '.')
                    return(5); /* e.g. http://xml.-soft */
	        ++cur;
		continue;
            }
    	    if (*cur == '.') {
	        --cur;
                if (*cur == '-')
                    return(6); /* e.g. http://xml-.soft */
                if (*cur == '.')
                    return(7); /* e.g. http://xml..soft */
	        ++cur;
		continue;
            }
	    break;
        } while (1);
        tmp = cur;
        if (tmp[-1] == '.')
            --tmp; /* e.g. http://xml.$Oft/ */
        do --tmp; while (tmp >= host && IS_ALPHANUM(*tmp));
        if ((++tmp == host || tmp[-1] == '.') && !IS_ALPHA(*tmp))
            return(8); /* e.g. http://xmlsOft.0rg/ */
    }
    if (uri != NULL) {
	if (uri->authority != NULL) xmlFree(uri->authority);
	uri->authority = NULL;
	if (uri->server != NULL) xmlFree(uri->server);
	uri->server = xmlURIUnescapeString(host, cur - host, NULL);
    }
    /*
     * finish by checking for a port presence.
     */
    if (*cur == ':') {
        cur++;
	if (IS_DIGIT(*cur)) {
	    if (uri != NULL)
	        uri->port = 0;
	    while (IS_DIGIT(*cur)) {
	        if (uri != NULL)
		    uri->port = uri->port * 10 + (*cur - '0');
		cur++;
	    }
	}
    }
    *str = cur;
    return(0);
}	

/**
 * xmlParseURIRelSegment:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI relative segment
 * 
 * rel_segment = 1*( unreserved | escaped | ";" | "@" | "&" | "=" |
 *                          "+" | "$" | "," )
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIRelSegment(xmlURIPtr uri, const char **str)
{
    const char *cur;

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

    cur = *str;
    if (!(IS_SEGMENT(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))) {
        return (3);
    }
    NEXT(cur);
    while (IS_SEGMENT(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))
        NEXT(cur);
    if (uri != NULL) {
        if (uri->path != NULL)
            xmlFree(uri->path);
        uri->path = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return (0);
}

/**
 * xmlParseURIPathSegments:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 * @slash:  should we add a leading slash
 *
 * Parse an URI set of path segments
 * 
 * path_segments = segment *( "/" segment )
 * segment       = *pchar *( ";" param )
 * param         = *pchar
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIPathSegments(xmlURIPtr uri, const char **str, int slash)
{
    const char *cur;

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

    cur = *str;

    do {
        while (IS_PCHAR(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))
            NEXT(cur);
        while (*cur == ';') {
            cur++;
            while (IS_PCHAR(cur) || ((uri != NULL) && (uri->cleanup) && (IS_UNWISE(cur))))
                NEXT(cur);
        }
        if (*cur != '/')
            break;
        cur++;
    } while (1);
    if (uri != NULL) {
        int len, len2 = 0;
        char *path;

        /*
         * Concat the set of path segments to the current path
         */
        len = cur - *str;
        if (slash)
            len++;

        if (uri->path != NULL) {
            len2 = strlen(uri->path);
            len += len2;
        }
        path = (char *) xmlMallocAtomic(len + 1);
        if (path == NULL) {
            xmlGenericError(xmlGenericErrorContext,
                            "xmlParseURIPathSegments: out of memory\n");
            *str = cur;
            return (-1);
        }
        if (uri->path != NULL)
            memcpy(path, uri->path, len2);
        if (slash) {
            path[len2] = '/';
            len2++;
        }
        path[len2] = 0;
        if (cur - *str > 0)
            xmlURIUnescapeString(*str, cur - *str, &path[len2]);
        if (uri->path != NULL)
            xmlFree(uri->path);
        uri->path = path;
    }
    *str = cur;
    return (0);
}

/**
 * xmlParseURIAuthority:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse the authority part of an URI.
 * 
 * authority = server | reg_name
 * server    = [ [ userinfo "@" ] hostport ]
 * reg_name  = 1*( unreserved | escaped | "$" | "," | ";" | ":" |
 *                        "@" | "&" | "=" | "+" )
 *
 * Note : this is completely ambiguous since reg_name is allowed to
 *        use the full set of chars in use by server:
 *
 *        3.2.1. Registry-based Naming Authority
 *
 *        The structure of a registry-based naming authority is specific
 *        to the URI scheme, but constrained to the allowed characters
 *        for an authority component.
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIAuthority(xmlURIPtr uri, const char **str) {
    const char *cur;
    int ret;

    if (str == NULL)
	return(-1);
    
    cur = *str;

    /*
     * try first to parse it as a server string.
     */
    ret = xmlParseURIServer(uri, str);
    if ((ret == 0) && (*str != NULL) &&
	((**str == 0) || (**str == '/') || (**str == '?')))
        return(0);
    *str = cur;

    /*
     * failed, fallback to reg_name
     */
    if (!IS_REG_NAME(cur)) {
	return(5);
    }
    NEXT(cur);
    while (IS_REG_NAME(cur)) NEXT(cur);
    if (uri != NULL) {
	if (uri->server != NULL) xmlFree(uri->server);
	uri->server = NULL;
	if (uri->user != NULL) xmlFree(uri->user);
	uri->user = NULL;
	if (uri->authority != NULL) xmlFree(uri->authority);
	uri->authority = xmlURIUnescapeString(*str, cur - *str, NULL);
    }
    *str = cur;
    return(0);
}

/**
 * xmlParseURIHierPart:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI hierarchical part
 * 
 * hier_part = ( net_path | abs_path ) [ "?" query ]
 * abs_path = "/"  path_segments
 * net_path = "//" authority [ abs_path ]
 *
 * Returns 0 or the error code
 */
static int
xmlParseURIHierPart(xmlURIPtr uri, const char **str) {
    int ret;
    const char *cur;

    if (str == NULL)
	return(-1);
    
    cur = *str;

    if ((cur[0] == '/') && (cur[1] == '/')) {
	cur += 2;
	ret = xmlParseURIAuthority(uri, &cur);
	if (ret != 0)
	    return(ret);
	if (cur[0] == '/') {
	    cur++;
	    ret = xmlParseURIPathSegments(uri, &cur, 1);
	}
    } else if (cur[0] == '/') {
	cur++;
	ret = xmlParseURIPathSegments(uri, &cur, 1);
    } else {
	return(4);