proxy_util.c

Apache HTTP Server 是一个功能强大的灵活的与HTTP/1.1相兼容的web服务器.这里给出的是Apache HTTP服务器的源码。

/* Copyright 1999-2005 The Apache Software Foundation or its licensors, as
 * applicable.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

/* Utility routines for Apache proxy */
#include "mod_proxy.h"


static int proxy_match_ipaddr(struct dirconn_entry *This, request_rec *r);
static int proxy_match_domainname(struct dirconn_entry *This, request_rec *r);
static int proxy_match_hostname(struct dirconn_entry *This, request_rec *r);
static int proxy_match_word(struct dirconn_entry *This, request_rec *r);

APR_IMPLEMENT_OPTIONAL_HOOK_RUN_ALL(proxy, PROXY, int, create_req, 
                                   (request_rec *r, request_rec *pr), (r, pr),
                                   OK, DECLINED)

/* already called in the knowledge that the characters are hex digits */
PROXY_DECLARE(int) ap_proxy_hex2c(const char *x)
{
    int i, ch;

#if !APR_CHARSET_EBCDIC
    ch = x[0];
    if (apr_isdigit(ch))
	i = ch - '0';
    else if (apr_isupper(ch))
	i = ch - ('A' - 10);
    else
	i = ch - ('a' - 10);
    i <<= 4;

    ch = x[1];
    if (apr_isdigit(ch))
	i += ch - '0';
    else if (apr_isupper(ch))
	i += ch - ('A' - 10);
    else
	i += ch - ('a' - 10);
    return i;
#else /*APR_CHARSET_EBCDIC*/
    /* we assume that the hex value refers to an ASCII character
     * so convert to EBCDIC so that it makes sense locally;
     *
     * example:
     *
     * client specifies %20 in URL to refer to a space char;
     * at this point we're called with EBCDIC "20"; after turning
     * EBCDIC "20" into binary 0x20, we then need to assume that 0x20
     * represents an ASCII char and convert 0x20 to EBCDIC, yielding
     * 0x40
     */
    char buf[1];

    if (1 == sscanf(x, "%2x", &i)) {
        buf[0] = i & 0xFF;
        ap_xlate_proto_from_ascii(buf, 1);
        return buf[0];
    }
    else {
        return 0;
    }
#endif /*APR_CHARSET_EBCDIC*/
}

PROXY_DECLARE(void) ap_proxy_c2hex(int ch, char *x)
{
#if !APR_CHARSET_EBCDIC
    int i;

    x[0] = '%';
    i = (ch & 0xF0) >> 4;
    if (i >= 10)
	x[1] = ('A' - 10) + i;
    else
	x[1] = '0' + i;

    i = ch & 0x0F;
    if (i >= 10)
	x[2] = ('A' - 10) + i;
    else
	x[2] = '0' + i;
#else /*APR_CHARSET_EBCDIC*/
    static const char ntoa[] = { "0123456789ABCDEF" };
    char buf[1];

    ch &= 0xFF;

    buf[0] = ch;
    ap_xlate_proto_to_ascii(buf, 1);

    x[0] = '%';
    x[1] = ntoa[(buf[0] >> 4) & 0x0F];
    x[2] = ntoa[buf[0] & 0x0F];
    x[3] = '\0';
#endif /*APR_CHARSET_EBCDIC*/
}

/*
 * canonicalise a URL-encoded string
 */

/*
 * Convert a URL-encoded string to canonical form.
 * It decodes characters which need not be encoded,
 * and encodes those which must be encoded, and does not touch
 * those which must not be touched.
 */
PROXY_DECLARE(char *)ap_proxy_canonenc(apr_pool_t *p, const char *x, int len, enum enctype t,
	int isenc)
{
    int i, j, ch;
    char *y;
    char *allowed;	/* characters which should not be encoded */
    char *reserved;	/* characters which much not be en/de-coded */

/* N.B. in addition to :@&=, this allows ';' in an http path
 * and '?' in an ftp path -- this may be revised
 * 
 * Also, it makes a '+' character in a search string reserved, as
 * it may be form-encoded. (Although RFC 1738 doesn't allow this -
 * it only permits ; / ? : @ = & as reserved chars.)
 */
    if (t == enc_path)
	allowed = "$-_.+!*'(),;:@&=";
    else if (t == enc_search)
	allowed = "$-_.!*'(),;:@&=";
    else if (t == enc_user)
	allowed = "$-_.+!*'(),;@&=";
    else if (t == enc_fpath)
	allowed = "$-_.+!*'(),?:@&=";
    else			/* if (t == enc_parm) */
	allowed = "$-_.+!*'(),?/:@&=";

    if (t == enc_path)
	reserved = "/";
    else if (t == enc_search)
	reserved = "+";
    else
	reserved = "";

    y = apr_palloc(p, 3 * len + 1);

    for (i = 0, j = 0; i < len; i++, j++) {
/* always handle '/' first */
	ch = x[i];
	if (strchr(reserved, ch)) {
	    y[j] = ch;
	    continue;
	}
/* decode it if not already done */
	if (isenc && ch == '%') {
	    if (!apr_isxdigit(x[i + 1]) || !apr_isxdigit(x[i + 2]))
		return NULL;
	    ch = ap_proxy_hex2c(&x[i + 1]);
	    i += 2;
	    if (ch != 0 && strchr(reserved, ch)) {	/* keep it encoded */
		ap_proxy_c2hex(ch, &y[j]);
		j += 2;
		continue;
	    }
	}
/* recode it, if necessary */
	if (!apr_isalnum(ch) && !strchr(allowed, ch)) {
	    ap_proxy_c2hex(ch, &y[j]);
	    j += 2;
	}
	else
	    y[j] = ch;
    }
    y[j] = '\0';
    return y;
}

/*
 * Parses network-location.
 *    urlp           on input the URL; on output the path, after the leading /
 *    user           NULL if no user/password permitted
 *    password       holder for password
 *    host           holder for host
 *    port           port number; only set if one is supplied.
 *
 * Returns an error string.
 */
PROXY_DECLARE(char *)
     ap_proxy_canon_netloc(apr_pool_t *p, char **const urlp, char **userp,
			char **passwordp, char **hostp, apr_port_t *port)
{
    char *addr, *scope_id, *strp, *host, *url = *urlp;
    char *user = NULL, *password = NULL;
    apr_port_t tmp_port;
    apr_status_t rv;

    if (url[0] != '/' || url[1] != '/')
	return "Malformed URL";
    host = url + 2;
    url = strchr(host, '/');
    if (url == NULL)
	url = "";
    else
	*(url++) = '\0';	/* skip seperating '/' */

    /* find _last_ '@' since it might occur in user/password part */
    strp = strrchr(host, '@');

    if (strp != NULL) {
	*strp = '\0';
	user = host;
	host = strp + 1;

/* find password */
	strp = strchr(user, ':');
	if (strp != NULL) {
	    *strp = '\0';
	    password = ap_proxy_canonenc(p, strp + 1, strlen(strp + 1), enc_user, 1);
	    if (password == NULL)
		return "Bad %-escape in URL (password)";
	}

	user = ap_proxy_canonenc(p, user, strlen(user), enc_user, 1);
	if (user == NULL)
	    return "Bad %-escape in URL (username)";
    }
    if (userp != NULL) {
	*userp = user;
    }
    if (passwordp != NULL) {
	*passwordp = password;
    }

    /* Parse the host string to separate host portion from optional port.
     * Perform range checking on port.
     */
    rv = apr_parse_addr_port(&addr, &scope_id, &tmp_port, host, p);
    if (rv != APR_SUCCESS || addr == NULL || scope_id != NULL) {
        return "Invalid host/port";
    }
    if (tmp_port != 0) { /* only update caller's port if port was specified */
        *port = tmp_port;
    }

    ap_str_tolower(addr); /* DNS names are case-insensitive */

    *urlp = url;
    *hostp = addr;

    return NULL;
}

static const char * const lwday[7] =
{"Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"};

/*
 * If the date is a valid RFC 850 date or asctime() date, then it
 * is converted to the RFC 1123 format, otherwise it is not modified.
 * This routine is not very fast at doing conversions, as it uses
 * sscanf and sprintf. However, if the date is already correctly
 * formatted, then it exits very quickly.
 */
PROXY_DECLARE(const char *)
     ap_proxy_date_canon(apr_pool_t *p, const char *x1)
{
    char *x = apr_pstrdup(p, x1);
    int wk, mday, year, hour, min, sec, mon;
    char *q, month[4], zone[4], week[4];

    q = strchr(x, ',');
    /* check for RFC 850 date */
    if (q != NULL && q - x > 3 && q[1] == ' ') {
	*q = '\0';
	for (wk = 0; wk < 7; wk++)
	    if (strcmp(x, lwday[wk]) == 0)
		break;
	*q = ',';
	if (wk == 7)
	    return x;		/* not a valid date */
	if (q[4] != '-' || q[8] != '-' || q[11] != ' ' || q[14] != ':' ||
	    q[17] != ':' || strcmp(&q[20], " GMT") != 0)
	    return x;
	if (sscanf(q + 2, "%u-%3s-%u %u:%u:%u %3s", &mday, month, &year,
		   &hour, &min, &sec, zone) != 7)
	    return x;
	if (year < 70)
	    year += 2000;
	else
	    year += 1900;
    }
    else {
/* check for acstime() date */
	if (x[3] != ' ' || x[7] != ' ' || x[10] != ' ' || x[13] != ':' ||
	    x[16] != ':' || x[19] != ' ' || x[24] != '\0')
	    return x;
	if (sscanf(x, "%3s %3s %u %u:%u:%u %u", week, month, &mday, &hour,
		   &min, &sec, &year) != 7)
	    return x;
	for (wk = 0; wk < 7; wk++)
	    if (strcmp(week, apr_day_snames[wk]) == 0)
		break;
	if (wk == 7)
	    return x;
    }

/* check date */
    for (mon = 0; mon < 12; mon++)
	if (strcmp(month, apr_month_snames[mon]) == 0)
	    break;
    if (mon == 12)
	return x;

    q = apr_palloc(p, 30);
    apr_snprintf(q, 30, "%s, %.2d %s %d %.2d:%.2d:%.2d GMT", apr_day_snames[wk],
       mday, apr_month_snames[mon], year, hour, min, sec);
    return q;
}

PROXY_DECLARE(request_rec *)ap_proxy_make_fake_req(conn_rec *c, request_rec *r)
{
    request_rec *rp = apr_pcalloc(c->pool, sizeof(*r));

    rp->pool            = c->pool;
    rp->status          = HTTP_OK;

    rp->headers_in      = apr_table_make(c->pool, 50);
    rp->subprocess_env  = apr_table_make(c->pool, 50);
    rp->headers_out     = apr_table_make(c->pool, 12);
    rp->err_headers_out = apr_table_make(c->pool, 5);
    rp->notes           = apr_table_make(c->pool, 5);

    rp->server = r->server;
    rp->proxyreq = r->proxyreq;
    rp->request_time = r->request_time;
    rp->connection      = c;
    rp->output_filters  = c->output_filters;
    rp->input_filters   = c->input_filters;
    rp->proto_output_filters  = c->output_filters;
    rp->proto_input_filters   = c->input_filters;

    rp->request_config  = ap_create_request_config(c->pool);
    proxy_run_create_req(r, rp);

    return rp;
}

/*
 * Reads headers from a buffer and returns an array of headers.
 * Returns NULL on file error
 * This routine tries to deal with too long lines and continuation lines.
 *
 * Note: Currently the headers are passed through unmerged. This has to be
 * done so that headers which react badly to merging (such as Set-Cookie
 * headers, which contain commas within the date field) do not get stuffed
 * up.
 */
PROXY_DECLARE(apr_table_t *)ap_proxy_read_headers(request_rec *r, request_rec *rr, char *buffer, int size, conn_rec *c)
{
    apr_table_t *headers_out;
    int len;
    char *value, *end;
    char field[MAX_STRING_LEN];
    int saw_headers = 0;
    void *sconf = r->server->module_config;
    proxy_server_conf *psc;

    psc = (proxy_server_conf *) ap_get_module_config(sconf, &proxy_module);

    headers_out = apr_table_make(r->pool, 20);

    /*
     * Read header lines until we get the empty separator line, a read error,
     * the connection closes (EOF), or we timeout.
     */
    while ((len = ap_getline(buffer, size, rr, 1)) > 0) {

	if (!(value = strchr(buffer, ':'))) {     /* Find the colon separator */

	    /* We may encounter invalid headers, usually from buggy