diff options
| author | Sergey Ponomarev <stokito@gmail.com> | 2020-08-09 01:23:31 +0300 |
|---|---|---|
| committer | Denys Vlasenko <vda.linux@googlemail.com> | 2020-08-15 22:51:14 +0200 |
| commit | b414cdf5b4a3950ac09b630d0d8b7d2844a7cf81 (patch) | |
| tree | aec42c9167a55c46dafcd55376499ba585f5cd8c | |
| parent | 1a5d6fcbb5e606ab4acdf22afa26361a25f1d43b (diff) | |
| download | busybox-b414cdf5b4a3950ac09b630d0d8b7d2844a7cf81.tar.gz | |
httpd: Update to HTTP/1.1
HTTP v1.1 was released in 1999 year and it's time to update BB HTTPD.
Browsers may behave badly with HTTP/1.0
E.g. Chrome does not send the If-None-Match header with ETag.
Signed-off-by: Sergey Ponomarev <stokito@gmail.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
| -rw-r--r-- | networking/httpd.c | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/networking/httpd.c b/networking/httpd.c index f4e95768f..9141442c8 100644 --- a/networking/httpd.c +++ b/networking/httpd.c @@ -269,7 +269,7 @@ static const char DEFAULT_PATH_HTTPD_CONF[] ALIGN1 = "/etc"; static const char HTTPD_CONF[] ALIGN1 = "httpd.conf"; -static const char HTTP_200[] ALIGN1 = "HTTP/1.0 200 OK\r\n"; +static const char HTTP_200[] ALIGN1 = "HTTP/1.1 200 OK\r\n"; static const char index_html[] ALIGN1 = "index.html"; typedef struct has_next_ptr { @@ -1074,7 +1074,7 @@ static void send_headers(unsigned responseNum) strftime(date_str, sizeof(date_str), RFC1123FMT, gmtime_r(&timer, &tm)); /* ^^^ using gmtime_r() instead of gmtime() to not use static data */ len = sprintf(iobuf, - "HTTP/1.0 %u %s\r\n" + "HTTP/1.1 %u %s\r\n" "Date: %s\r\n" "Connection: close\r\n", responseNum, responseString, @@ -1099,7 +1099,7 @@ static void send_headers(unsigned responseNum) #endif if (responseNum == HTTP_MOVED_TEMPORARILY) { /* Responding to "GET /dir" with - * "HTTP/1.0 302 Found" "Location: /dir/" + * "HTTP/1.1 302 Found" "Location: /dir/" * - IOW, asking them to repeat with a slash. * Here, overflow IS possible, can't use sprintf: * mkdir test @@ -1409,7 +1409,7 @@ static NOINLINE void cgi_io_loop_and_exit(int fromCgi_rd, int toCgi_wr, int post count = safe_read(fromCgi_rd, rbuf + out_cnt, IOBUF_SIZE - 8); if (count <= 0) { /* eof (or error) and there was no "HTTP", - * send "HTTP/1.0 200 OK\r\n", then send received data */ + * send "HTTP/1.1 200 OK\r\n", then send received data */ if (out_cnt) { full_write(STDOUT_FILENO, HTTP_200, sizeof(HTTP_200)-1); full_write(STDOUT_FILENO, rbuf, out_cnt); @@ -1420,10 +1420,10 @@ static NOINLINE void cgi_io_loop_and_exit(int fromCgi_rd, int toCgi_wr, int post count = 0; /* "Status" header format is: "Status: 302 Redirected\r\n" */ if (out_cnt >= 8 && memcmp(rbuf, "Status: ", 8) == 0) { - /* send "HTTP/1.0 " */ + /* send "HTTP/1.1 " */ if (full_write(STDOUT_FILENO, HTTP_200, 9) != 9) break; - /* skip "Status: " (including space, sending "HTTP/1.0 NNN" is wrong) */ + /* skip "Status: " (including space, sending "HTTP/1.1 NNN" is wrong) */ rbuf += 8; count = out_cnt - 8; out_cnt = -1; /* buffering off */ @@ -1439,7 +1439,7 @@ static NOINLINE void cgi_io_loop_and_exit(int fromCgi_rd, int toCgi_wr, int post full_write(s, "Content-type: text/plain\r\n\r\n", 28); } * Counter-example of valid CGI without Content-type: - * echo -en "HTTP/1.0 302 Found\r\n" + * echo -en "HTTP/1.1 302 Found\r\n" * echo -en "Location: http://www.busybox.net\r\n" * echo -en "\r\n" */ @@ -1546,7 +1546,7 @@ static void send_cgi_and_exit( /* (Older versions of bbox seem to do some decoding) */ setenv1("QUERY_STRING", g_query); putenv((char*)"SERVER_SOFTWARE=busybox httpd/"BB_VER); - putenv((char*)"SERVER_PROTOCOL=HTTP/1.0"); + putenv((char*)"SERVER_PROTOCOL=HTTP/1.1"); putenv((char*)"GATEWAY_INTERFACE=CGI/1.1"); /* Having _separate_ variables for IP and port defeats * the purpose of having socket abstraction. Which "port" |