MINOR: http: add a new function http_validate_scheme() to validate a scheme

While http_parse_scheme() extracts a scheme from a URI by extracting
exactly the valid characters and stopping on delimiters, this new
function performs the same on a fixed-size string.

(cherry picked from commit d3d8d03d98c6624c8b15b81f7b80dd22fc3f9fd2)
[wt: context adj]
Signed-off-by: Willy Tarreau <w@1wt.eu>
(cherry picked from commit a12054991079cfb6c0f8b7484a8b8209be785cd2)
[wt: context adj]
Signed-off-by: Willy Tarreau <w@1wt.eu>

diff --git a/include/haproxy/http.h b/include/haproxy/http.h
index fb47abd..7dea530 100644 (file)
--- a/include/haproxy/http.h
+++ b/include/haproxy/http.h
@@ -36,6 +36,7 @@ extern const uint8_t http_char_classes[256];
 enum http_meth_t find_http_meth(const char *str, const int len);
 int http_get_status_idx(unsigned int status);
 const char *http_get_reason(unsigned int status);
+int http_validate_scheme(const struct ist schm);
 struct ist http_get_authority(const struct ist uri, int no_userinfo);
 struct ist http_get_path(const struct ist uri);
 int http_header_match2(const char *hdr, const char *end,

diff --git a/src/http.c b/src/http.c
index 22a5a14..6f78164 100644 (file)
--- a/src/http.c
+++ b/src/http.c
@@ -458,6 +458,29 @@ const char *http_get_reason(unsigned int status)
        }
 }
 
+/* Returns non-zero if the scheme <schm> is syntactically correct according to
+ * RFC3986#3.1, otherwise zero. It expects only the scheme and nothing else
+ * (particularly not the following "://").
+ *     Scheme = alpha *(alpha|digit|'+'|'-'|'.')
+ */
+int http_validate_scheme(const struct ist schm)
+{
+       size_t i;
+
+       for (i = 0; i < schm.len; i++) {
+               if (likely((schm.ptr[i] >= 'a' && schm.ptr[i] <= 'z') ||
+                          (schm.ptr[i] >= 'A' && schm.ptr[i] <= 'Z')))
+                       continue;
+               if (unlikely(!i)) // first char must be alpha
+                       return 0;
+               if ((schm.ptr[i] >= '0' && schm.ptr[i] <= '9') ||
+                   schm.ptr[i] == '+' || schm.ptr[i] == '-' || schm.ptr[i] == '.')
+                       continue;
+               return 0;
+       }
+       return !!i;
+}
+
 /* Parse the uri and looks for the authority, between the scheme and the
  * path. if no_userinfo is not zero, the part before the '@' (including it) is
  * skipped. If not found, an empty ist is returned. Otherwise, the ist pointing