* url-util.el (url--allowed-chars): Use upper-case for percent-encoding.

2 files changed, 6 insertions, 3 deletions

diff --git a/lisp/url/ChangeLog b/lisp/url/ChangeLog
index 3980b22d4c1..b3669a72ac3 100644
--- a/lisp/url/ChangeLog
+++ b/lisp/url/ChangeLog
@@ -3,7 +3,7 @@
         * url-util.el (url-encode-url): New function for URL quoting.
         (url-encoding-table, url-host-allowed-chars)
         (url-path-allowed-chars): New constants.
-        (url--allowed-chars): New helper function.
+        (url--allowed-chars): New helper function.  Use upper-case.
         (url-hexify-string): Use them.
 
         * url-parse.el: Improve RFC 3986 conformance.
diff --git a/lisp/url/url-util.el b/lisp/url/url-util.el
index c8016ef6cdb..4185c87918e 100644
--- a/lisp/url/url-util.el
+++ b/lisp/url/url-util.el
@@ -343,7 +343,10 @@ This is taken from RFC 3986 (section 2.3).")
 (defconst url-encoding-table
   (let ((vec (make-vector 256 nil)))
     (dotimes (byte 256)
-      (aset vec byte (format "%%%02x" byte)))
+      ;; RFC 3986 (Section 2.1): For consistency, URI producers and
+      ;; normalizers should use uppercase hexadecimal digits for all
+      ;; percent-encodings.
+      (aset vec byte (format "%%%02X" byte)))
     vec)
   "Vector translating bytes to URI-encoded %-sequences.")
 
@@ -362,7 +365,7 @@ result can be passed as the second arg to `url-hexify-string'."
 If STRING is multibyte, it is first converted to a utf-8 byte
 string.  Each byte corresponding to an allowed character is left
 as-is, while all other bytes are converted to a three-character
-string: \"%\" followed by two lowercase hex digits.
+string: \"%\" followed by two upper-case hex digits.
 
 The allowed characters are specified by ALLOWED-CHARS.  If this
 argument is nil, the list `url-unreserved-chars' determines the