BUG/MINOR: http: make url_decode() optionally convert '+' to SP

The url_decode() function used by the url_dec converter and a few other
call points is ambiguous on its processing of the '+' character which
itself isn't stable in the spec. This one belongs to the reserved
characters for the query string but not for the path nor the scheme,
in which it must be left as-is. It's only in argument strings that
follow the application/x-www-form-urlencoded encoding that it must be
turned into a space, that is, in query strings and POST arguments.

The problem is that the function is used to process full URLs and
paths in various configs, and to process query strings from the stats
page for example.

This patch updates the function to differentiate the situation where
it's parsing a path and a query string. A new argument indicates if a
query string should be assumed, otherwise it's only assumed after seeing
a question mark.

The various locations in the code making use of this function were
updated to take care of this (most call places were using it to decode
POST arguments).

The url_dec converter is usually called on path or url samples, so it
needs to remain compatible with this and will default to parsing a path
and turning the '+' to a space only after a question mark. However in
situations where it would explicitly be extracted from a POST or a
query string, it now becomes possible to enforce the decoding by passing
a non-null value in argument.

It seems to be what was reported in issue #585. This fix may be
backported to older stable releases.

contrib/prometheus-exporter/service-prometheus.c

@@ -2292,7 +2292,7 @@ static int promex_parse_uri(struct appctx *appctx, struct stream_interface *si)
 			*(p++) = 0;
 		else if (*p == '#')
 			*p = 0;
-		len = url_decode(key);
+		len = url_decode(key, 1);
 		if (len == -1)
 			goto error;
 
@@ -2306,7 +2306,7 @@ static int promex_parse_uri(struct appctx *appctx, struct stream_interface *si)
 				*(p++) = 0;
 			else if (*p == '#')
 				*p = 0;
-			len = url_decode(value);
+			len = url_decode(value, 1);
 			if (len == -1)
 				goto error;
 		}

doc/configuration.txt

@@ -14540,9 +14540,13 @@ upper
   sample fetch function or after a transformation keyword returning a string
   type. The result is of type string.
 
-url_dec
-  Takes an url-encoded string provided as input and returns the decoded
-  version as output. The input and the output are of type string.
+url_dec([<in_form>])
+  Takes an url-encoded string provided as input and returns the decoded version
+  as output. The input and the output are of type string. If the <in_form>
+  argument is set to a non-zero integer value, the input string is assumed to
+  be part of a form or query string and the '+' character will be turned into a
+  space (' '). Otherwise this will only happen after a question mark indicating
+  a query string ('?').
 
 ungrpc(<field_number>,[<field_type>])
   This extracts the protocol buffers message field in raw mode of an input binary

include/common/standard.h

@@ -609,8 +609,12 @@ static inline const char *csv_enc(const char *str, int quote,
  * be shorter. If some forbidden characters are found, the conversion is
  * aborted, the string is truncated before the issue and non-zero is returned,
  * otherwise the operation returns non-zero indicating success.
+ * If the 'in_form' argument is non-nul the string is assumed to be part of
+ * an "application/x-www-form-urlencoded" encoded string, and the '+' will be
+ * turned to a space. If it's zero, this will only be done after a question
+ * mark ('?').
  */
-int url_decode(char *string);
+int url_decode(char *string, int in_form);
 
 /* This one is 6 times faster than strtoul() on athlon, but does
  * no check at all.

src/http_conv.c

@@ -246,6 +246,7 @@ static int sample_conv_q_preferred(const struct arg *args, struct sample *smp, v
 /* This fetch url-decode any input string. */
 static int sample_conv_url_dec(const struct arg *args, struct sample *smp, void *private)
 {
+	int in_form = 0;
 	int len;
 
 	/* If the constant flag is set or if not size is available at
@@ -262,7 +263,11 @@ static int sample_conv_url_dec(const struct arg *args, struct sample *smp, void
 
 	/* Add final \0 required by url_decode(), and convert the input string. */
 	smp->data.u.str.area[smp->data.u.str.data] = '\0';
-	len = url_decode(smp->data.u.str.area);
+
+	if (args && (args[0].type == ARGT_SINT))
+		in_form = !!args[0].data.sint;
+
+	len = url_decode(smp->data.u.str.area, in_form);
 	if (len < 0)
 		return 0;
 	smp->data.u.str.data = len;
@@ -361,7 +366,7 @@ static struct sample_conv_kw_list sample_conv_kws = {ILH, {
 	{ "language",       sample_conv_q_preferred,  ARG2(1,STR,STR),  NULL,   SMP_T_STR,  SMP_T_STR},
 	{ "capture-req",    smp_conv_req_capture,     ARG1(1,SINT),     NULL,   SMP_T_STR,  SMP_T_STR},
 	{ "capture-res",    smp_conv_res_capture,     ARG1(1,SINT),     NULL,   SMP_T_STR,  SMP_T_STR},
-	{ "url_dec",        sample_conv_url_dec,      0,                NULL,   SMP_T_STR,  SMP_T_STR},
+	{ "url_dec",        sample_conv_url_dec,      ARG1(0,SINT),     NULL,   SMP_T_STR,  SMP_T_STR},
 	{ NULL, NULL, 0, 0, 0 },
 }};
 

src/mux_fcgi.c

@@ -1306,7 +1306,7 @@ static int fcgi_set_default_param(struct fcgi_conn *fconn, struct fcgi_strm *fst
 		chunk_memcat(params->p, path.ptr, path.len);
 		path.ptr = b_tail(params->p) - path.len;
 		path.ptr[path.len] = '\0';
-		len = url_decode(path.ptr);
+		len = url_decode(path.ptr, 0);
 		if (len < 0)
 			goto error;
 		path.len = len;

src/standard.c

@@ -1749,8 +1749,12 @@ const char *csv_enc_append(const char *str, int quote, struct buffer *output)
  * be shorter. If some forbidden characters are found, the conversion is
  * aborted, the string is truncated before the issue and a negative value is
  * returned, otherwise the operation returns the length of the decoded string.
+ * If the 'in_form' argument is non-nul the string is assumed to be part of
+ * an "application/x-www-form-urlencoded" encoded string, and the '+' will be
+ * turned to a space. If it's zero, this will only be done after a question
+ * mark ('?').
  */
-int url_decode(char *string)
+int url_decode(char *string, int in_form)
 {
 	char *in, *out;
 	int ret = -1;
@@ -1760,7 +1764,7 @@ int url_decode(char *string)
 	while (*in) {
 		switch (*in) {
 		case '+' :
-			*out++ = ' ';
+			*out++ = in_form ? ' ' : *in;
 			break;
 		case '%' :
 			if (!ishex(in[1]) || !ishex(in[2]))
@@ -1768,6 +1772,9 @@ int url_decode(char *string)
 			*out++ = (hex2i(in[1]) << 4) + hex2i(in[2]);
 			in += 2;
 			break;
+		case '?':
+			in_form = 1;
+			/* fall through */
 		default:
 			*out++ = *in;
 			break;

src/stats.c

@@ -2902,7 +2902,7 @@ static int stats_process_http_post(struct stream_interface *si)
 				/* Ok, a value is found, we can mark the end of the key */
 				*value++ = '\0';
 			}
-			if (url_decode(key) < 0 || url_decode(value) < 0)
+			if (url_decode(key, 1) < 0 || url_decode(value, 1) < 0)
 				break;
 
 			/* Now we can check the key to see what to do */