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.

    if ((ref->authority != NULL) || (ref->server != NULL)) {
	if (ref->authority != NULL)
	    res->authority = xmlMemStrdup(ref->authority);
	else {
	    res->server = xmlMemStrdup(ref->server);
	    if (ref->user != NULL)
		res->user = xmlMemStrdup(ref->user);
            res->port = ref->port;		
	}
	if (ref->path != NULL)
	    res->path = xmlMemStrdup(ref->path);
	goto step_7;
    }
    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;		
    }

    /*
     * 5) If the path component begins with a slash character ("/"), then
     *    the reference is an absolute-path and we skip to step 7.
     */
    if ((ref->path != NULL) && (ref->path[0] == '/')) {
	res->path = xmlMemStrdup(ref->path);
	goto step_7;
    }


    /*
     * 6) If this step is reached, then we are resolving a relative-path
     *    reference.  The relative path needs to be merged with the base
     *    URI's path.  Although there are many ways to do this, we will
     *    describe a simple method using a separate string buffer.
     *
     * Allocate a buffer large enough for the result string.
     */
    len = 2; /* extra / and 0 */
    if (ref->path != NULL)
	len += strlen(ref->path);
    if (bas->path != NULL)
	len += strlen(bas->path);
    res->path = (char *) xmlMallocAtomic(len);
    if (res->path == NULL) {
	xmlGenericError(xmlGenericErrorContext,
		"xmlBuildURI: out of memory\n");
	goto done;
    }
    res->path[0] = 0;

    /*
     * a) All but the last segment of the base URI's path component is
     *    copied to the buffer.  In other words, any characters after the
     *    last (right-most) slash character, if any, are excluded.
     */
    cur = 0;
    out = 0;
    if (bas->path != NULL) {
	while (bas->path[cur] != 0) {
	    while ((bas->path[cur] != 0) && (bas->path[cur] != '/'))
		cur++;
	    if (bas->path[cur] == 0)
		break;

	    cur++;
	    while (out < cur) {
		res->path[out] = bas->path[out];
		out++;
	    }
	}
    }
    res->path[out] = 0;

    /*
     * b) The reference's path component is appended to the buffer
     *    string.
     */
    if (ref->path != NULL && ref->path[0] != 0) {
	indx = 0;
	/*
	 * Ensure the path includes a '/'
	 */
	if ((out == 0) && (bas->server != NULL))
	    res->path[out++] = '/';
	while (ref->path[indx] != 0) {
	    res->path[out++] = ref->path[indx++];
	}
    }
    res->path[out] = 0;

    /*
     * Steps c) to h) are really path normalization steps
     */
    xmlNormalizeURIPath(res->path);

step_7:

    /*
     * 7) The resulting URI components, including any inherited from the
     *    base URI, are recombined to give the absolute form of the URI
     *    reference.
     */
    val = xmlSaveUri(res);

done:
    if (ref != NULL)
	xmlFreeURI(ref);
    if (bas != NULL)
	xmlFreeURI(bas);
    if (res != NULL)
	xmlFreeURI(res);
    return(val);
}

/**
 * xmlBuildRelativeURI:
 * @URI:  the URI reference under consideration
 * @base:  the base value
 *
 * Expresses the URI of the reference in terms relative to the
 * base.  Some examples of this operation include:
 *     base = "http://site1.com/docs/book1.html"
 *        URI input                        URI returned
 *     docs/pic1.gif                    pic1.gif
 *     docs/img/pic1.gif                img/pic1.gif
 *     img/pic1.gif                     ../img/pic1.gif
 *     http://site1.com/docs/pic1.gif   pic1.gif
 *     http://site2.com/docs/pic1.gif   http://site2.com/docs/pic1.gif
 *
 *     base = "docs/book1.html"
 *        URI input                        URI returned
 *     docs/pic1.gif                    pic1.gif
 *     docs/img/pic1.gif                img/pic1.gif
 *     img/pic1.gif                     ../img/pic1.gif
 *     http://site1.com/docs/pic1.gif   http://site1.com/docs/pic1.gif
 *
 *
 * Note: if the URI reference is really wierd or complicated, it may be
 *       worthwhile to first convert it into a "nice" one by calling
 *       xmlBuildURI (using 'base') before calling this routine,
 *       since this routine (for reasonable efficiency) assumes URI has
 *       already been through some validation.
 *
 * Returns a new URI string (to be freed by the caller) or NULL in case
 * error.
 */
xmlChar *
xmlBuildRelativeURI (const xmlChar * URI, const xmlChar * base)
{
    xmlChar *val = NULL;
    int ret;
    int ix;
    int pos = 0;
    int nbslash = 0;
    int len;
    xmlURIPtr ref = NULL;
    xmlURIPtr bas = NULL;
    xmlChar *bptr, *uptr, *vptr;
    int remove_path = 0;

    if ((URI == NULL) || (*URI == 0))
	return NULL;

    /*
     * First parse URI into a standard form
     */
    ref = xmlCreateURI ();
    if (ref == NULL)
	return NULL;
    /* If URI not already in "relative" form */
    if (URI[0] != '.') {
	ret = xmlParseURIReference (ref, (const char *) URI);
	if (ret != 0)
	    goto done;		/* Error in URI, return NULL */
    } else
	ref->path = (char *)xmlStrdup(URI);

    /*
     * Next parse base into the same standard form
     */
    if ((base == NULL) || (*base == 0)) {
	val = xmlStrdup (URI);
	goto done;
    }
    bas = xmlCreateURI ();
    if (bas == NULL)
	goto done;
    if (base[0] != '.') {
	ret = xmlParseURIReference (bas, (const char *) base);
	if (ret != 0)
	    goto done;		/* Error in base, return NULL */
    } else
	bas->path = (char *)xmlStrdup(base);

    /*
     * If the scheme / server on the URI differs from the base,
     * just return the URI
     */
    if ((ref->scheme != NULL) &&
	((bas->scheme == NULL) ||
	 (xmlStrcmp ((xmlChar *)bas->scheme, (xmlChar *)ref->scheme)) ||
	 (xmlStrcmp ((xmlChar *)bas->server, (xmlChar *)ref->server)))) {
	val = xmlStrdup (URI);
	goto done;
    }
    if (xmlStrEqual((xmlChar *)bas->path, (xmlChar *)ref->path)) {
	val = xmlStrdup(BAD_CAST "");
	goto done;
    }
    if (bas->path == NULL) {
	val = xmlStrdup((xmlChar *)ref->path);
	goto done;
    }
    if (ref->path == NULL) {
        ref->path = (char *) "/";
	remove_path = 1;
    }

    /*
     * At this point (at last!) we can compare the two paths
     *
     * First we take care of the special case where either of the
     * two path components may be missing (bug 316224)
     */
    if (bas->path == NULL) {
	if (ref->path != NULL) {
	    uptr = (xmlChar *) ref->path;
	    if (*uptr == '/')
		uptr++;
	    val = xmlStrdup(uptr);
	}
	goto done;
    }
    bptr = (xmlChar *)bas->path;
    if (ref->path == NULL) {
	for (ix = 0; bptr[ix] != 0; ix++) {
	    if (bptr[ix] == '/')
		nbslash++;
	}
	uptr = NULL;
	len = 1;	/* this is for a string terminator only */
    } else {
    /*
     * Next we compare the two strings and find where they first differ
     */
	if ((ref->path[pos] == '.') && (ref->path[pos+1] == '/'))
            pos += 2;
	if ((*bptr == '.') && (bptr[1] == '/'))
            bptr += 2;
	else if ((*bptr == '/') && (ref->path[pos] != '/'))
	    bptr++;
	while ((bptr[pos] == ref->path[pos]) && (bptr[pos] != 0))
	    pos++;

	if (bptr[pos] == ref->path[pos]) {
	    val = xmlStrdup(BAD_CAST "");
	    goto done;		/* (I can't imagine why anyone would do this) */
	}

	/*
	 * In URI, "back up" to the last '/' encountered.  This will be the
	 * beginning of the "unique" suffix of URI
	 */
	ix = pos;
	if ((ref->path[ix] == '/') && (ix > 0))
	    ix--;
	else if ((ref->path[ix] == 0) && (ix > 1) && (ref->path[ix - 1] == '/'))
	    ix -= 2;
	for (; ix > 0; ix--) {
	    if (ref->path[ix] == '/')
		break;
	}
	if (ix == 0) {
	    uptr = (xmlChar *)ref->path;
	} else {
	    ix++;
	    uptr = (xmlChar *)&ref->path[ix];
	}

	/*
	 * In base, count the number of '/' from the differing point
	 */
	if (bptr[pos] != ref->path[pos]) {/* check for trivial URI == base */
	    for (; bptr[ix] != 0; ix++) {
		if (bptr[ix] == '/')
		    nbslash++;
	    }
	}
	len = xmlStrlen (uptr) + 1;
    }
    
    if (nbslash == 0) {
	if (uptr != NULL)
	    val = xmlStrdup (uptr);
	goto done;
    }

    /*
     * Allocate just enough space for the returned string -
     * length of the remainder of the URI, plus enough space
     * for the "../" groups, plus one for the terminator
     */
    val = (xmlChar *) xmlMalloc (len + 3 * nbslash);
    if (val == NULL) {
	xmlGenericError(xmlGenericErrorContext,
		"xmlBuildRelativeURI: out of memory\n");
	goto done;
    }
    vptr = val;
    /*
     * Put in as many "../" as needed
     */
    for (; nbslash>0; nbslash--) {
	*vptr++ = '.';
	*vptr++ = '.';
	*vptr++ = '/';
    }
    /*
     * Finish up with the end of the URI
     */
    if (uptr != NULL) {
        if ((vptr > val) && (len > 0) &&
	    (uptr[0] == '/') && (vptr[-1] == '/')) {
	    memcpy (vptr, uptr + 1, len - 1);
	    vptr[len - 2] = 0;
	} else {
	    memcpy (vptr, uptr, len);
	    vptr[len - 1] = 0;
	}
    } else {
	vptr[len - 1] = 0;
    }

done:
    /*
     * Free the working variables
     */
    if (remove_path != 0)
        ref->path = NULL;
    if (ref != NULL)
	xmlFreeURI (ref);
    if (bas != NULL)
	xmlFreeURI (bas);

    return val;
}

/**
 * xmlCanonicPath:
 * @path:  the resource locator in a filesystem notation
 *
 * Constructs a canonic path from the specified path. 
 *
 * Returns a new canonic path, or a duplicate of the path parameter if the 
 * construction fails. The caller is responsible for freeing the memory occupied
 * by the returned string. If there is insufficient memory available, or the 
 * argument is NULL, the function returns NULL.
 */
#define IS_WINDOWS_PATH(p) 					\
	((p != NULL) &&						\
	 (((p[0] >= 'a') && (p[0] <= 'z')) ||			\
	  ((p[0] >= 'A') && (p[0] <= 'Z'))) &&			\
	 (p[1] == ':') && ((p[2] == '/') || (p[2] == '\\')))
xmlChar *
xmlCanonicPath(const xmlChar *path)
{
/*
 * For Windows implementations, additional work needs to be done to
 * replace backslashes in pathnames with "forward slashes"
 */
#if defined(_WIN32) && !defined(__CYGWIN__)    
    int len = 0;
    int i = 0;
    xmlChar *p = NULL;
#endif
    xmlURIPtr uri;
    xmlChar *ret;
    const xmlChar *absuri;

    if (path == NULL)
	return(NULL);
    if ((uri = xmlParseURI((const char *) path)) != NULL) {
	xmlFreeURI(uri);
	return xmlStrdup(path);
    }

    /* Check if this is an "absolute uri" */
    absuri = xmlStrstr(path, BAD_CAST "://");
    if (absuri != NULL) {
        int l, j;
	unsigned char c;
	xmlChar *escURI;

        /*
	 * this looks like an URI where some parts have not been
	 * escaped leading to a parsing problem.  Check that the first
	 * part matches a protocol.
	 */
	l = absuri - path;
	/* Bypass if first part (part before the '://') is > 20 chars */
	if ((l <= 0) || (l > 20))
	    goto path_processing;
	/* Bypass if any non-alpha characters are present in first part */
	for (j = 0;j < l;j++) {
	    c = path[j];
	    if (!(((c >= 'a') && (c <= 'z')) || ((c >= 'A') && (c <= 'Z'))))
	        goto path_processing;
	}

	/* Escape all except the characters specified in the supplied path */
        escURI = xmlURIEscapeStr(path, BAD_CAST ":/?_.#&;=");
	if (escURI != NULL) {
	    /* Try parsing the escaped path */
	    uri = xmlParseURI((const char *) escURI);
	    /* If successful, return the escaped string */
	    if (uri != NULL) {
	        xmlFreeURI(uri);
		return escURI;
	    }
	}
    }

path_processing:
/* For Windows implementations, replace backslashes with 'forward slashes' */
#if defined(_WIN32) && !defined(__CYGWIN__)    
    /*
     * Create a URI structure
     */
    uri = xmlCreateURI();
    if (uri == NULL) {		/* Guard against 'out of memory' */
        return(NULL);
    }

    len = xmlStrlen(path);
    if ((len > 2) && IS_WINDOWS_PATH(path)) {
        /* make the scheme 'file' */
	uri->scheme = xmlStrdup(BAD_CAST "file");
	/* allocate space for leading '/' + path + string terminator */
	uri->path = xmlMallocAtomic(len + 2);
	if (uri->path == NULL) {
	    xmlFreeURI(uri);	/* Guard agains 'out of memory' */
	    return(NULL);
	}
	/* Put in leading '/' plus path */
	uri->path[0] = '/';
	p = uri->path + 1;
	strncpy(p, path, len + 1);
    } else {
	uri->path = xmlStrdup(path);
	if (uri->path == NULL) {
	    xmlFreeURI(uri);
	    return(NULL);
	}
	p = uri->path;
    }
    /* Now change all occurences of '\' to '/' */
    while (*p != '\0') {
	if (*p == '\\')
	    *p = '/';
	p++;
    }

    if (uri->scheme == NULL) {
	ret = xmlStrdup((const xmlChar *) uri->path);
    } else {
	ret = xmlSaveUri(uri);
    }

    xmlFreeURI(uri);
#else
    ret = xmlStrdup((const xmlChar *) path);
#endif
    return(ret);
}

/**
 * xmlPathToURI:
 * @path:  the resource locator in a filesystem notation
 *
 * Constructs an URI expressing the existing path
 *
 * Returns a new URI, or a duplicate of the path parameter if the 
 * construction fails. The caller is responsible for freeing the memory
 * occupied by the returned string. If there is insufficient memory available,
 * or the argument is NULL, the function returns NULL.
 */
xmlChar *
xmlPathToURI(const xmlChar *path)
{
    xmlURIPtr uri;
    xmlURI temp;
    xmlChar *ret, *cal;

    if (path == NULL)
        return(NULL);

    if ((uri = xmlParseURI((const char *) path)) != NULL) {
	xmlFreeURI(uri);
	return xmlStrdup(path);
    }
    cal = xmlCanonicPath(path);
    if (cal