uri.c

xml开源解析代码.版本为libxml2-2.6.29,可支持GB3212.网络消息发送XML时很有用.

        /*
         * 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) {
	    if (uri->cleanup & 2) {
	        memcpy(&path[len2], *str, cur - *str);
		path[len2 + (cur - *str)] = 0;
	    } else
		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);
	if (uri->cleanup & 2)
	    uri->authority = STRNDUP(*str, cur - *str);
	else
	    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);
    }
    if (ret != 0)
	return(ret);
    if (*cur == '?') {
	cur++;
	ret = xmlParseURIQuery(uri, &cur);
	if (ret != 0)
	    return(ret);
    }
    *str = cur;
    return(0);
}

/**
 * xmlParseAbsoluteURI:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an URI reference string and fills in the appropriate fields
 * of the @uri structure
 * 
 * absoluteURI   = scheme ":" ( hier_part | opaque_part )
 *
 * Returns 0 or the error code
 */
static int
xmlParseAbsoluteURI(xmlURIPtr uri, const char **str) {
    int ret;
    const char *cur;

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

    ret = xmlParseURIScheme(uri, str);
    if (ret != 0) return(ret);
    if (**str != ':') {
	*str = cur;
	return(1);
    }
    (*str)++;
    if (**str == '/')
	return(xmlParseURIHierPart(uri, str));
    return(xmlParseURIOpaquePart(uri, str));
}

/**
 * xmlParseRelativeURI:
 * @uri:  pointer to an URI structure
 * @str:  pointer to the string to analyze
 *
 * Parse an relative URI string and fills in the appropriate fields
 * of the @uri structure
 * 
 * relativeURI = ( net_path | abs_path | rel_path ) [ "?" query ]
 * abs_path = "/"  path_segments
 * net_path = "//" authority [ abs_path ]
 * rel_path = rel_segment [ abs_path ]
 *
 * Returns 0 or the error code
 */
static int
xmlParseRelativeURI(xmlURIPtr uri, const char **str) {
    int ret = 0;
    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 if (cur[0] != '#' && cur[0] != '?') {
	ret = xmlParseURIRelSegment(uri, &cur);
	if (ret != 0)
	    return(ret);
	if (cur[0] == '/') {
	    cur++;
	    ret = xmlParseURIPathSegments(uri, &cur, 1);
	}
    }
    if (ret != 0)
	return(ret);
    if (*cur == '?') {
	cur++;
	ret = xmlParseURIQuery(uri, &cur);
	if (ret != 0)
	    return(ret);
    }
    *str = cur;
    return(ret);
}

/**
 * xmlParseURIReference:
 * @uri:  pointer to an URI structure
 * @str:  the string to analyze
 *
 * Parse an URI reference string and fills in the appropriate fields
 * of the @uri structure
 * 
 * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
 *
 * Returns 0 or the error code
 */
int
xmlParseURIReference(xmlURIPtr uri, const char *str) {
    int ret;
    const char *tmp = str;

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

    /*
     * Try first to parse absolute refs, then fallback to relative if
     * it fails.
     */
    ret = xmlParseAbsoluteURI(uri, &str);
    if (ret != 0) {
	xmlCleanURI(uri);
	str = tmp;
        ret = xmlParseRelativeURI(uri, &str);
    }
    if (ret != 0) {
	xmlCleanURI(uri);
	return(ret);
    }

    if (*str == '#') {
	str++;
	ret = xmlParseURIFragment(uri, &str);
	if (ret != 0) return(ret);
    }
    if (*str != 0) {
	xmlCleanURI(uri);
	return(1);
    }
    return(0);
}

/**
 * xmlParseURI:
 * @str:  the URI string to analyze
 *
 * Parse an URI 
 * 
 * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
 *
 * Returns a newly built xmlURIPtr or NULL in case of error
 */
xmlURIPtr
xmlParseURI(const char *str) {
    xmlURIPtr uri;
    int ret;

    if (str == NULL)
	return(NULL);
    uri = xmlCreateURI();
    if (uri != NULL) {
	ret = xmlParseURIReference(uri, str);
        if (ret) {
	    xmlFreeURI(uri);
	    return(NULL);
	}
    }
    return(uri);
}

/**
 * xmlParseURIRaw:
 * @str:  the URI string to analyze
 * @raw:  if 1 unescaping of URI pieces are disabled
 *
 * Parse an URI but allows to keep intact the original fragments.
 * 
 * URI-reference = [ absoluteURI | relativeURI ] [ "#" fragment ]
 *
 * Returns a newly built xmlURIPtr or NULL in case of error
 */
xmlURIPtr
xmlParseURIRaw(const char *str, int raw) {
    xmlURIPtr uri;
    int ret;

    if (str == NULL)
	return(NULL);
    uri = xmlCreateURI();
    if (uri != NULL) {
        if (raw) {
	    uri->cleanup |= 2;
	}
	ret = xmlParseURIReference(uri, str);
        if (ret) {
	    xmlFreeURI(uri);
	    return(NULL);
	}
    }
    return(uri);
}

/************************************************************************
 *									*
 *			Public functions				*
 *									*
 ************************************************************************/

/**
 * xmlBuildURI:
 * @URI:  the URI instance found in the document
 * @base:  the base value
 *
 * Computes he final URI of the reference done by checking that
 * the given URI is valid, and building the final URI using the
 * base URI. This is processed according to section 5.2 of the 
 * RFC 2396
 *
 * 5.2. Resolving Relative References to Absolute Form
 *
 * Returns a new URI string (to be freed by the caller) or NULL in case
 *         of error.
 */
xmlChar *
xmlBuildURI(const xmlChar *URI, const xmlChar *base) {
    xmlChar *val = NULL;
    int ret, len, indx, cur, out;
    xmlURIPtr ref = NULL;
    xmlURIPtr bas = NULL;
    xmlURIPtr res = NULL;

    /*
     * 1) The URI reference is parsed into the potential four components and
     *    fragment identifier, as described in Section 4.3.
     *
     *    NOTE that a completely empty URI is treated by modern browsers
     *    as a reference to "." rather than as a synonym for the current
     *    URI.  Should we do that here?
     */
    if (URI == NULL) 
	ret = -1;
    else {
	if (*URI) {
	    ref = xmlCreateURI();
	    if (ref == NULL)
		goto done;
	    ret = xmlParseURIReference(ref, (const char *) URI);
	}
	else
	    ret = 0;
    }
    if (ret != 0)
	goto done;
    if ((ref != NULL) && (ref->scheme != NULL)) {
	/*
	 * The URI is absolute don't modify.
	 */
	val = xmlStrdup(URI);
	goto done;
    }
    if (base == NULL)
	ret = -1;
    else {
	bas = xmlCreateURI();
	if (bas == NULL)
	    goto done;
	ret = xmlParseURIReference(bas, (const char *) base);
    }
    if (ret != 0) {
	if (ref)
	    val = xmlSaveUri(ref);
	goto done;
    }
    if (ref == NULL) {
	/*
	 * the base fragment must be ignored
	 */
	if (bas->fragment != NULL) {
	    xmlFree(bas->fragment);
	    bas->fragment = NULL;
	}
	val = xmlSaveUri(bas);
	goto done;
    }

    /*
     * 2) If the path component is empty and the scheme, authority, and
     *    query components are undefined, then it is a reference to the
     *    current document and we are done.  Otherwise, the reference URI's
     *    query and fragment components are defined as found (or not found)
     *    within the URI reference and not inherited from the base URI.
     *
     *    NOTE that in modern browsers, the parsing differs from the above
     *    in the following aspect:  the query component is allowed to be
     *    defined while still treating this as a reference to the current
     *    document.
     */
    res = xmlCreateURI();
    if (res == NULL)
	goto done;
    if ((ref->scheme == NULL) && (ref->path == NULL) &&
	((ref->authority == NULL) && (ref->server == NULL))) {
	if (bas->scheme != NULL)
	    res->scheme = xmlMemStrdup(bas->scheme);
	if (bas->authority != NULL)
	    res->authority = xmlMemStrdup(bas->authority);
	else if (bas->server != NULL) {
	    res->server = xmlMemStrdup(bas->server);
	    if (bas->user != NULL)
		res->user = xmlMemStrdup(bas->user);
	    res->port = bas->port;		
	}
	if (bas->path != NULL)
	    res->path = xmlMemStrdup(bas->path);
	if (ref->query_raw != NULL)
	    res->query_raw = xmlMemStrdup (ref->query_raw);
	else if (ref->query != NULL)
	    res->query = xmlMemStrdup(ref->query);
	else if (bas->query_raw != NULL)
	    res->query_raw = xmlMemStrdup(bas->query_raw);
	else if (bas->query != NULL)
	    res->query = xmlMemStrdup(bas->query);
	if (ref->fragment != NULL)
	    res->fragment = xmlMemStrdup(ref->fragment);
	goto step_7;
    }

    /*
     * 3) If the scheme component is defined, indicating that the reference
     *    starts with a scheme name, then the reference is interpreted as an
     *    absolute URI and we are done.  Otherwise, the reference URI's
     *    scheme is inherited from the base URI's scheme component.
     */
    if (ref->scheme != NULL) {
	val = xmlSaveUri(ref);
	goto done;
    }
    if (bas->scheme != NULL)
	res->scheme = xmlMemStrdup(bas->scheme);
 
    if (ref->query_raw != NULL)
	res->query_raw = xmlMemStrdup(ref->query_raw);
    else if (ref->query != NULL)
	res->query = xmlMemStrdup(ref->query);
    if (ref->fragment != NULL)
	res->fragment = xmlMemStrdup(ref->fragment);

    /*
     * 4) If the authority component is defined, then the reference is a
     *    network-path and we skip to step 7.  Otherwise, the reference
     *    URI's authority is inherited from the base URI's authority
     *    component, which will also be undefined if the URI scheme does not
     *    use an authority component.
     */