ngx_http_parse.c

Nginx是一个高性能的HTTP和反向代理服务器

/*
 * Copyright (C) Igor Sysoev
 */


#include <ngx_config.h>
#include <ngx_core.h>
#include <ngx_http.h>


static uint32_t  usual[] = {
    0xffffdbfe, /* 1111 1111 1111 1111  1101 1011 1111 1110 */

                /* ?>=< ;:98 7654 3210  /.-, +*)( '&%$ #"!  */
    0x7fff37d6, /* 0111 1111 1111 1111  0011 0111 1101 0110 */

                /* _^]\ [ZYX WVUT SRQP  ONML KJIH GFED CBA@ */
#if (NGX_WIN32)
    0xefffffff, /* 1110 1111 1111 1111  1111 1111 1111 1111 */
#else
    0xffffffff, /* 1111 1111 1111 1111  1111 1111 1111 1111 */
#endif

                /*  ~}| {zyx wvut srqp  onml kjih gfed cba` */
    0xffffffff, /* 1111 1111 1111 1111  1111 1111 1111 1111 */

    0xffffffff, /* 1111 1111 1111 1111  1111 1111 1111 1111 */
    0xffffffff, /* 1111 1111 1111 1111  1111 1111 1111 1111 */
    0xffffffff, /* 1111 1111 1111 1111  1111 1111 1111 1111 */
    0xffffffff  /* 1111 1111 1111 1111  1111 1111 1111 1111 */
};


#if (NGX_HAVE_LITTLE_ENDIAN && NGX_HAVE_NONALIGNED)

#define ngx_str3_cmp(m, c0, c1, c2, c3)                                       \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)

#define ngx_str3Ocmp(m, c0, c1, c2, c3)                                       \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)

#define ngx_str4cmp(m, c0, c1, c2, c3)                                        \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)

#define ngx_str5cmp(m, c0, c1, c2, c3, c4)                                    \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)             \
        && m[4] == c4

#define ngx_str6cmp(m, c0, c1, c2, c3, c4, c5)                                \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)             \
        && (((uint32_t *) m)[1] & 0xffff) == ((c5 << 8) | c4)

#define ngx_str7_cmp(m, c0, c1, c2, c3, c4, c5, c6, c7)                       \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)             \
        && ((uint32_t *) m)[1] == ((c7 << 24) | (c6 << 16) | (c5 << 8) | c4)

#define ngx_str8cmp(m, c0, c1, c2, c3, c4, c5, c6, c7)                        \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)             \
        && ((uint32_t *) m)[1] == ((c7 << 24) | (c6 << 16) | (c5 << 8) | c4)

#define ngx_str9cmp(m, c0, c1, c2, c3, c4, c5, c6, c7, c8)                    \
    *(uint32_t *) m == ((c3 << 24) | (c2 << 16) | (c1 << 8) | c0)             \
        && ((uint32_t *) m)[1] == ((c7 << 24) | (c6 << 16) | (c5 << 8) | c4)  \
        && m[8] == c8

#else /* !(NGX_HAVE_LITTLE_ENDIAN && NGX_HAVE_NONALIGNED) */

#define ngx_str3_cmp(m, c0, c1, c2, c3)                                       \
    m[0] == c0 && m[1] == c1 && m[2] == c2

#define ngx_str3Ocmp(m, c0, c1, c2, c3)                                       \
    m[0] == c0 && m[2] == c2 && m[3] == c3

#define ngx_str4cmp(m, c0, c1, c2, c3)                                        \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3

#define ngx_str5cmp(m, c0, c1, c2, c3, c4)                                    \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3 && m[4] == c4

#define ngx_str6cmp(m, c0, c1, c2, c3, c4, c5)                                \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3                      \
        && m[4] == c4 && m[5] == c5

#define ngx_str7_cmp(m, c0, c1, c2, c3, c4, c5, c6, c7)                       \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3                      \
        && m[4] == c4 && m[5] == c5 && m[6] == c6

#define ngx_str8cmp(m, c0, c1, c2, c3, c4, c5, c6, c7)                        \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3                      \
        && m[4] == c4 && m[5] == c5 && m[6] == c6 && m[7] == c7

#define ngx_str9cmp(m, c0, c1, c2, c3, c4, c5, c6, c7, c8)                    \
    m[0] == c0 && m[1] == c1 && m[2] == c2 && m[3] == c3                      \
        && m[4] == c4 && m[5] == c5 && m[6] == c6 && m[7] == c7 && m[8] == c8

#endif


/* gcc, icc, msvc and others compile these switches as an jump table */

ngx_int_t
ngx_http_parse_request_line(ngx_http_request_t *r, ngx_buf_t *b)
{
    u_char  c, ch, *p, *m;
    enum {
        sw_start = 0,
        sw_method,
        sw_spaces_before_uri,
        sw_schema,
        sw_schema_slash,
        sw_schema_slash_slash,
        sw_host,
        sw_port,
        sw_after_slash_in_uri,
        sw_check_uri,
        sw_uri,
        sw_http_09,
        sw_http_H,
        sw_http_HT,
        sw_http_HTT,
        sw_http_HTTP,
        sw_first_major_digit,
        sw_major_digit,
        sw_first_minor_digit,
        sw_minor_digit,
        sw_spaces_after_digit,
        sw_almost_done
    } state;

    state = r->state;

    for (p = b->pos; p < b->last; p++) {
        ch = *p;

        switch (state) {

        /* HTTP methods: GET, HEAD, POST */
        case sw_start:
            r->request_start = p;

            if (ch == CR || ch == LF) {
                break;
            }

            if (ch < 'A' || ch > 'Z') {
                return NGX_HTTP_PARSE_INVALID_METHOD;
            }

            state = sw_method;
            break;

        case sw_method:
            if (ch == ' ') {
                r->method_end = p - 1;
                m = r->request_start;

                switch (p - m) {

                case 3:
                    if (ngx_str3_cmp(m, 'G', 'E', 'T', ' ')) {
                        r->method = NGX_HTTP_GET;
                        break;
                    }

                    if (ngx_str3_cmp(m, 'P', 'U', 'T', ' ')) {
                        r->method = NGX_HTTP_PUT;
                        break;
                    }

                    break;

                case 4:
                    if (m[1] == 'O') {

                        if (ngx_str3Ocmp(m, 'P', 'O', 'S', 'T')) {
                            r->method = NGX_HTTP_POST;
                            break;
                        }

                        if (ngx_str3Ocmp(m, 'C', 'O', 'P', 'Y')) {
                            r->method = NGX_HTTP_COPY;
                            break;
                        }

                        if (ngx_str3Ocmp(m, 'M', 'O', 'V', 'E')) {
                            r->method = NGX_HTTP_MOVE;
                            break;
                        }

                        if (ngx_str3Ocmp(m, 'L', 'O', 'C', 'K')) {
                            r->method = NGX_HTTP_LOCK;
                            break;
                        }

                    } else {

                        if (ngx_str4cmp(m, 'H', 'E', 'A', 'D')) {
                            r->method = NGX_HTTP_HEAD;
                            break;
                        }
                    }

                    break;

                case 5:
                    if (ngx_str5cmp(m, 'M', 'K', 'C', 'O', 'L')) {
                        r->method = NGX_HTTP_MKCOL;
                    }

                    if (ngx_str5cmp(m, 'T', 'R', 'A', 'C', 'E')) {
                        r->method = NGX_HTTP_TRACE;
                    }

                    break;

                case 6:
                    if (ngx_str6cmp(m, 'D', 'E', 'L', 'E', 'T', 'E')) {
                        r->method = NGX_HTTP_DELETE;
                        break;
                    }

                    if (ngx_str6cmp(m, 'U', 'N', 'L', 'O', 'C', 'K')) {
                        r->method = NGX_HTTP_UNLOCK;
                        break;
                    }

                    break;

                case 7:
                    if (ngx_str7_cmp(m, 'O', 'P', 'T', 'I', 'O', 'N', 'S', ' '))
                    {
                        r->method = NGX_HTTP_OPTIONS;
                    }

                    break;

                case 8:
                    if (ngx_str8cmp(m, 'P', 'R', 'O', 'P', 'F', 'I', 'N', 'D'))
                    {
                        r->method = NGX_HTTP_PROPFIND;
                    }

                    break;

                case 9:
                    if (ngx_str9cmp(m,
                            'P', 'R', 'O', 'P', 'P', 'A', 'T', 'C', 'H'))
                    {
                        r->method = NGX_HTTP_PROPPATCH;
                    }

                    break;
                }

                state = sw_spaces_before_uri;
                break;
            }

            if (ch < 'A' || ch > 'Z') {
                return NGX_HTTP_PARSE_INVALID_METHOD;
            }

            break;

        /* space* before URI */
        case sw_spaces_before_uri:

            if (ch == '/' ){
                r->uri_start = p;
                state = sw_after_slash_in_uri;
                break;
            }

            c = (u_char) (ch | 0x20);
            if (c >= 'a' && c <= 'z') {
                r->schema_start = p;
                state = sw_schema;
                break;
            }

            switch (ch) {
            case ' ':
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        case sw_schema:

            c = (u_char) (ch | 0x20);
            if (c >= 'a' && c <= 'z') {
                break;
            }

            switch (ch) {
            case ':':
                r->schema_end = p;
                state = sw_schema_slash;
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        case sw_schema_slash:
            switch (ch) {
            case '/':
                state = sw_schema_slash_slash;
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        case sw_schema_slash_slash:
            switch (ch) {
            case '/':
                r->host_start = p + 1;
                state = sw_host;
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        case sw_host:

            c = (u_char) (ch | 0x20);
            if (c >= 'a' && c <= 'z') {
                break;
            }

            if ((ch >= '0' && ch <= '9') || ch == '.' || ch == '-') {
                break;
            }

            r->host_end = p;

            switch (ch) {
            case ':':
                state = sw_port;
                break;
            case '/':
                r->uri_start = p;
                state = sw_after_slash_in_uri;
                break;
            case ' ':
                /*
                 * use single "/" from request line to preserve pointers,
                 * if request line will be copied to large client buffer
                 */
                r->uri_start = r->schema_end + 1;
                r->uri_end = r->schema_end + 2;
                state = sw_http_09;
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        case sw_port:
            if (ch >= '0' && ch <= '9') {
                break;
            }

            switch (ch) {
            case '/':
                r->port_end = p;
                r->uri_start = p;
                state = sw_after_slash_in_uri;
                break;
            case ' ':
                r->port_end = p;
                /*
                 * use single "/" from request line to preserve pointers,
                 * if request line will be copied to large client buffer
                 */
                r->uri_start = r->schema_end + 1;
                r->uri_end = r->schema_end + 2;
                state = sw_http_09;
                break;
            default:
                return NGX_HTTP_PARSE_INVALID_REQUEST;
            }
            break;

        /* check "/.", "//", "%", and "\" (Win32) in URI */
        case sw_after_slash_in_uri:

            if (usual[ch >> 5] & (1 << (ch & 0x1f))) {
                state = sw_check_uri;
                break;
            }

            switch (ch) {
            case ' ':
                r->uri_end = p;
                state = sw_http_09;
                break;
            case CR:
                r->uri_end = p;
                r->http_minor = 9;
                state = sw_almost_done;
                break;
            case LF:
                r->uri_end = p;
                r->http_minor = 9;
                goto done;
            case '.':
                r->complex_uri = 1;
                state = sw_uri;
                break;
            case '%':
                r->quoted_uri = 1;
                state = sw_uri;
                break;
            case '/':
                r->complex_uri = 1;
                state = sw_uri;
                break;
#if (NGX_WIN32)
            case '\\':
                r->complex_uri = 1;
                state = sw_uri;
                break;
#endif
            case '?':
                r->args_start = p + 1;
                state = sw_uri;
                break;
            case '#':
                r->complex_uri = 1;
                state = sw_uri;
                break;
            case '+':
                r->plus_in_uri = 1;
                break;
            case '\0':
                r->zero_in_uri = 1;
                break;
            default:
                state = sw_check_uri;
                break;
            }
            break;

        /* check "/", "%" and "\" (Win32) in URI */
        case sw_check_uri:

            if (usual[ch >> 5] & (1 << (ch & 0x1f))) {
                break;
            }

            switch (ch) {
            case '/':
                r->uri_ext = NULL;
                state = sw_after_slash_in_uri;
                break;
            case '.':
                r->uri_ext = p + 1;
                break;
            case ' ':
                r->uri_end = p;
                state = sw_http_09;
                break;
            case CR:
                r->uri_end = p;
                r->http_minor = 9;
                state = sw_almost_done;
                break;
            case LF:
                r->uri_end = p;
                r->http_minor = 9;
                goto done;
#if (NGX_WIN32)
            case '\\':
                r->complex_uri = 1;
                state = sw_after_slash_in_uri;
                break;
#endif
            case '%':
                r->quoted_uri = 1;
                state = sw_uri;
                break;
            case '?':
                r->args_start = p + 1;
                state = sw_uri;
                break;