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。

                *x = '%';
            }
            else {
                char decoded;
                decoded = x2c(y + 1);
                if (decoded == '\0') {
                    badpath = 1;
                }
                else {
                    *x = decoded;
                    y += 2;
                }
            }
        }
    }
    *x = '\0';
    if (badesc) {
        return HTTP_BAD_REQUEST;
    }
    else if (badpath) {
        return HTTP_NOT_FOUND;
    }
    else {
        return OK;
    }
}

AP_DECLARE(char *) ap_construct_server(apr_pool_t *p, const char *hostname,
                                       apr_port_t port, const request_rec *r)
{
    if (ap_is_default_port(port, r)) {
        return apr_pstrdup(p, hostname);
    }
    else {
        return apr_psprintf(p, "%s:%u", hostname, port);
    }
}

/* c2x takes an unsigned, and expects the caller has guaranteed that
 * 0 <= what < 256... which usually means that you have to cast to
 * unsigned char first, because (unsigned)(char)(x) first goes through
 * signed extension to an int before the unsigned cast.
 *
 * The reason for this assumption is to assist gcc code generation --
 * the unsigned char -> unsigned extension is already done earlier in
 * both uses of this code, so there's no need to waste time doing it
 * again.
 */
static const char c2x_table[] = "0123456789abcdef";

static APR_INLINE unsigned char *c2x(unsigned what, unsigned char *where)
{
#if APR_CHARSET_EBCDIC
    what = apr_xlate_conv_byte(ap_hdrs_to_ascii, (unsigned char)what);
#endif /*APR_CHARSET_EBCDIC*/
    *where++ = '%';
    *where++ = c2x_table[what >> 4];
    *where++ = c2x_table[what & 0xf];
    return where;
}

/*
 * escape_path_segment() escapes a path segment, as defined in RFC 1808. This
 * routine is (should be) OS independent.
 *
 * os_escape_path() converts an OS path to a URL, in an OS dependent way. In all
 * cases if a ':' occurs before the first '/' in the URL, the URL should be
 * prefixed with "./" (or the ':' escaped). In the case of Unix, this means
 * leaving '/' alone, but otherwise doing what escape_path_segment() does. For
 * efficiency reasons, we don't use escape_path_segment(), which is provided for
 * reference. Again, RFC 1808 is where this stuff is defined.
 *
 * If partial is set, os_escape_path() assumes that the path will be appended to
 * something with a '/' in it (and thus does not prefix "./").
 */

AP_DECLARE(char *) ap_escape_path_segment(apr_pool_t *p, const char *segment)
{
    char *copy = apr_palloc(p, 3 * strlen(segment) + 1);
    const unsigned char *s = (const unsigned char *)segment;
    unsigned char *d = (unsigned char *)copy;
    unsigned c;

    while ((c = *s)) {
        if (TEST_CHAR(c, T_ESCAPE_PATH_SEGMENT)) {
            d = c2x(c, d);
        }
        else {
            *d++ = c;
        }
        ++s;
    }
    *d = '\0';
    return copy;
}

AP_DECLARE(char *) ap_os_escape_path(apr_pool_t *p, const char *path, int partial)
{
    char *copy = apr_palloc(p, 3 * strlen(path) + 3);
    const unsigned char *s = (const unsigned char *)path;
    unsigned char *d = (unsigned char *)copy;
    unsigned c;

    if (!partial) {
        const char *colon = ap_strchr_c(path, ':');
        const char *slash = ap_strchr_c(path, '/');

        if (colon && (!slash || colon < slash)) {
            *d++ = '.';
            *d++ = '/';
        }
    }
    while ((c = *s)) {
        if (TEST_CHAR(c, T_OS_ESCAPE_PATH)) {
            d = c2x(c, d);
        }
        else {
            *d++ = c;
        }
        ++s;
    }
    *d = '\0';
    return copy;
}

/* ap_escape_uri is now a macro for os_escape_path */

AP_DECLARE(char *) ap_escape_html(apr_pool_t *p, const char *s)
{
    int i, j;
    char *x;

    /* first, count the number of extra characters */
    for (i = 0, j = 0; s[i] != '\0'; i++)
        if (s[i] == '<' || s[i] == '>')
            j += 3;
        else if (s[i] == '&')
            j += 4;

    if (j == 0)
        return apr_pstrmemdup(p, s, i);

    x = apr_palloc(p, i + j + 1);
    for (i = 0, j = 0; s[i] != '\0'; i++, j++)
        if (s[i] == '<') {
            memcpy(&x[j], "&lt;", 4);
            j += 3;
        }
        else if (s[i] == '>') {
            memcpy(&x[j], "&gt;", 4);
            j += 3;
        }
        else if (s[i] == '&') {
            memcpy(&x[j], "&amp;", 5);
            j += 4;
        }
        else
            x[j] = s[i];

    x[j] = '\0';
    return x;
}

AP_DECLARE(char *) ap_escape_logitem(apr_pool_t *p, const char *str)
{
    char *ret;
    unsigned char *d;
    const unsigned char *s;

    if (!str) {
        return NULL;
    }

    ret = apr_palloc(p, 4 * strlen(str) + 1); /* Be safe */
    d = (unsigned char *)ret;
    s = (const unsigned char *)str;
    for (; *s; ++s) {

        if (TEST_CHAR(*s, T_ESCAPE_LOGITEM)) {
            *d++ = '\\';
            switch(*s) {
            case '\b':
                *d++ = 'b';
                break;
            case '\n':
                *d++ = 'n';
                break;
            case '\r':
                *d++ = 'r';
                break;
            case '\t':
                *d++ = 't';
                break;
            case '\v':
                *d++ = 'v';
                break;
            case '\\':
            case '"':
                *d++ = *s;
                break;
            default:
                c2x(*s, d);
                *d = 'x';
                d += 3;
            }
        }
        else {
            *d++ = *s;
        }
    }
    *d = '\0';

    return ret;
}

AP_DECLARE(apr_size_t) ap_escape_errorlog_item(char *dest, const char *source,
                                               apr_size_t buflen)
{
    unsigned char *d, *ep;
    const unsigned char *s;

    if (!source || !buflen) { /* be safe */
        return 0;
    }

    d = (unsigned char *)dest;
    s = (const unsigned char *)source;
    ep = d + buflen - 1;

    for (; d < ep && *s; ++s) {

        if (TEST_CHAR(*s, T_ESCAPE_LOGITEM)) {
            *d++ = '\\';
            if (d >= ep) {
                --d;
                break;
            }

            switch(*s) {
            case '\b':
                *d++ = 'b';
                break;
            case '\n':
                *d++ = 'n';
                break;
            case '\r':
                *d++ = 'r';
                break;
            case '\t':
                *d++ = 't';
                break;
            case '\v':
                *d++ = 'v';
                break;
            case '\\':
                *d++ = *s;
                break;
            case '"': /* no need for this in error log */
                d[-1] = *s;
                break;
            default:
                if (d >= ep - 2) {
                    ep = --d; /* break the for loop as well */
                    break;
                }
                c2x(*s, d);
                *d = 'x';
                d += 3;
            }
        }
        else {
            *d++ = *s;
        }
    }
    *d = '\0';

    return (d - (unsigned char *)dest);
}

AP_DECLARE(int) ap_is_directory(apr_pool_t *p, const char *path)
{
    apr_finfo_t finfo;

    if (apr_stat(&finfo, path, APR_FINFO_TYPE, p) != APR_SUCCESS)
        return 0;                /* in error condition, just return no */

    return (finfo.filetype == APR_DIR);
}

AP_DECLARE(int) ap_is_rdirectory(apr_pool_t *p, const char *path)
{
    apr_finfo_t finfo;

    if (apr_lstat(&finfo, path, APR_FINFO_TYPE, p) != APR_SUCCESS)
        return 0;                /* in error condition, just return no */

    return (finfo.filetype == APR_DIR);
}

AP_DECLARE(char *) ap_make_full_path(apr_pool_t *a, const char *src1,
                                  const char *src2)
{
    apr_size_t len1, len2;
    char *path;

    len1 = strlen(src1);
    len2 = strlen(src2);
     /* allocate +3 for '/' delimiter, trailing NULL and overallocate
      * one extra byte to allow the caller to add a trailing '/'
      */
    path = (char *)apr_palloc(a, len1 + len2 + 3);
    if (len1 == 0) {
        *path = '/';
        memcpy(path + 1, src2, len2 + 1);
    }
    else {
        char *next;
        memcpy(path, src1, len1);
        next = path + len1;
        if (next[-1] != '/') {
            *next++ = '/';
        }
        memcpy(next, src2, len2 + 1);
    }
    return path;
}

/*
 * Check for an absoluteURI syntax (see section 3.2 in RFC2068).
 */
AP_DECLARE(int) ap_is_url(const char *u)
{
    register int x;

    for (x = 0; u[x] != ':'; x++) {
        if ((!u[x]) ||
            ((!apr_isalpha(u[x])) && (!apr_isdigit(u[x])) &&
             (u[x] != '+') && (u[x] != '-') && (u[x] != '.'))) {
            return 0;
        }
    }

    return (x ? 1 : 0);                /* If the first character is ':', it's broken, too */
}

AP_DECLARE(int) ap_ind(const char *s, char c)
{
    const char *p = ap_strchr_c(s, c);

    if (p == NULL)
        return -1;
    return p - s;
}

AP_DECLARE(int) ap_rind(const char *s, char c)
{
    const char *p = ap_strrchr_c(s, c);

    if (p == NULL)
        return -1;
    return p - s;
}

AP_DECLARE(void) ap_str_tolower(char *str)
{
    while (*str) {
        *str = apr_tolower(*str);
        ++str;
    }
}

static char *find_fqdn(apr_pool_t *a, struct hostent *p)
{
    int x;

    if (!strchr(p->h_name, '.')) {
        if (p->h_aliases) {
            for (x = 0; p->h_aliases[x]; ++x) {
                if (strchr(p->h_aliases[x], '.') &&
                    (!strncasecmp(p->h_aliases[x], p->h_name,
                                  strlen(p->h_name))))
                    return apr_pstrdup(a, p->h_aliases[x]);
            }
        }
        return NULL;
    }
    return apr_pstrdup(a, (void *) p->h_name);
}

char *ap_get_local_host(apr_pool_t *a)
{
#ifndef MAXHOSTNAMELEN
#define MAXHOSTNAMELEN 256
#endif
    char str[MAXHOSTNAMELEN + 1];
    char *server_hostname = NULL;
    struct hostent *p;

#ifdef BEOS_R5
    if (gethostname(str, sizeof(str) - 1) == 0)
#else
    if (gethostname(str, sizeof(str) - 1) != 0)
#endif
    {
        ap_log_perror(APLOG_MARK, APLOG_STARTUP | APLOG_WARNING, 0, a,
                     "%s: gethostname() failed to determine ServerName",
                     ap_server_argv0);
    }
    else 
    {
        str[sizeof(str) - 1] = '\0';
        /* TODO: Screaming for APR-ization */
        if ((!(p = gethostbyname(str))) 
            || (!(server_hostname = find_fqdn(a, p)))) {
            /* Recovery - return the default servername by IP: */
            if (p && p->h_addr_list[0]) {
                apr_snprintf(str, sizeof(str), "%pA", p->h_addr_list[0]);
                server_hostname = apr_pstrdup(a, str);
                /* We will drop through to report the IP-named server */
            }
        }
        else {
            /* Since we found a fdqn, return it with no logged message. */
            return server_hostname;
        }
    }

    if (!server_hostname) 
        server_hostname = apr_pstrdup(a, "127.0.0.1");

    ap_log_perror(APLOG_MARK, APLOG_ALERT|APLOG_STARTUP, 0, a,
                 "%s: Could not determine the server's fully qualified "
                 "domain name, using %s for ServerName",
                 ap_server_argv0, server_hostname);
             
    return server_hostname;
}

/* simple 'pool' alloc()ing glue to apr_base64.c
 */
AP_DECLARE(char *) ap_pbase64decode(apr_pool_t *p, const char *bufcoded)
{
    char *decoded;
    int l;

    decoded = (char *) apr_palloc(p, 1 + apr_base64_decode_len(bufcoded));
    l = apr_base64_decode(decoded, bufcoded);
    decoded[l] = '\0'; /* make binary sequence into string */

    return decoded;
}

AP_DECLARE(char *) ap_pbase64encode(apr_pool_t *p, char *string) 
{ 
    char *encoded;
    int l = strlen(string);

    encoded = (char *) apr_palloc(p, 1 + apr_base64_encode_len(l));
    l = apr_base64_encode(encoded, string, l);
    encoded[l] = '\0'; /* make binary sequence into string */

    return encoded;
}

/* we want to downcase the type/subtype for comparison purposes
 * but nothing else because ;parameter=foo values are case sensitive.
 * XXX: in truth we want to downcase parameter names... but really,
 * apache has never handled parameters and such correctly.  You
 * also need to compress spaces and such to be able to compare
 * properly. -djg
 */
AP_DECLARE(void) ap_content_type_tolower(char *str)
{
    char *semi;

    semi = strchr(str, ';');
    if (semi) {
        *semi = '\0';
    }
    while (*str) {
        *str = apr_tolower(*str);
        ++str;
    }
    if (semi) {
        *semi = ';';
    }
}

/*
 * Given a string, replace any bare " with \" .
 */
AP_DECLARE(char *) ap_escape_quotes(apr_pool_t *p, const char *instring)
{
    int newlen = 0;
    const char *inchr = instring;
    char *outchr, *outstring;

    /*
     * Look through the input string, jogging the length of the output
     * string up by an extra byte each time we find an unescaped ".
     */
    while (*inchr != '\0') {
        newlen++;
        if (*inchr == '"') {
            newlen++;
        }
        /*
         * If we find a slosh, and it's not the last byte in the string,
         * it's escaping something - advance past both bytes.
         */
        if ((*inchr == '\\') && (inchr[1] != '\0')) {
            inchr++;
            newlen++;
        }
        inchr++;
    }
    outstring = apr_palloc(p, newlen + 1);
    inchr = instring;
    outchr = outstring;
    /*
     * Now copy the input string to the output string, inserting a slosh
     * in front of every " that doesn't already have one.
     */
    while (*inchr != '\0') {
        if ((*inchr == '\\') && (inchr[1] != '\0')) {
            *outchr++ = *inchr++;
            *outchr++ = *inchr++;
        }
        if (*inchr == '"') {
            *outchr++ = '\\';
        }
        if (*inchr != '\0') {
            *outchr++ = *inchr++;
        }
    }
    *outchr = '\0';
    return outstring;
}