From 93be35e038bbb19e8d64d3c1f9d1be76a9083d09 Mon Sep 17 00:00:00 2001 From: Philipp Stephani Date: Mon, 24 Oct 2016 21:54:51 +0200 Subject: Fix encoding of JSON surrogate pairs JSON requires that such pairs be treated as UTF-16 surrogate pairs, not individual code points; cf. Bug #24784. * lisp/json.el (json-read-escaped-char): Fix decoding of surrogate pairs. (json--decode-utf-16-surrogates): New defun. * test/lisp/json-tests.el (test-json-read-string): Add test for surrogate pairs. --- lisp/json.el | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'lisp/json.el') diff --git a/lisp/json.el b/lisp/json.el index 38f828e8fbb..b2ac356641b 100644 --- a/lisp/json.el +++ b/lisp/json.el @@ -363,6 +363,10 @@ representation will be parsed correctly." ;; String parsing +(defun json--decode-utf-16-surrogates (high low) + "Return the code point represented by the UTF-16 surrogates HIGH and LOW." + (+ (lsh (- high #xD800) 10) (- low #xDC00) #x10000)) + (defun json-read-escaped-char () "Read the JSON string escaped character at point." ;; Skip over the '\' @@ -372,6 +376,17 @@ representation will be parsed correctly." (cond (special (cdr special)) ((not (eq char ?u)) char) + ;; Special-case UTF-16 surrogate pairs, + ;; cf. https://tools.ietf.org/html/rfc7159#section-7. Note that + ;; this clause overlaps with the next one and therefore has to + ;; come first. + ((looking-at + (rx (group (any "Dd") (any "89ABab") (= 2 (any "0-9A-Fa-f"))) + "\\u" (group (any "Dd") (any "C-Fc-f") (= 2 (any "0-9A-Fa-f"))))) + (json-advance 10) + (json--decode-utf-16-surrogates + (string-to-number (match-string 1) 16) + (string-to-number (match-string 2) 16))) ((looking-at "[0-9A-Fa-f][0-9A-Fa-f][0-9A-Fa-f][0-9A-Fa-f]") (let ((hex (match-string 0))) (json-advance 4) -- cgit v1.2.3