Merge pull request #146 from robertmrk/surrogate-pair-parsing-fix

Fix character skipping after a surrogate pair
nlohmann · Nov 14, 2015 · c013223 · c013223
2 parents 3948630 + ec7a1d8
commit c013223
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 4 deletions.
diff --git a/src/json.hpp b/src/json.hpp
@@ -6856,8 +6856,8 @@ class basic_json
                                 auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer>
                                                                (i + 7), 4).c_str(), nullptr, 16);
                                 result += to_unicode(codepoint, codepoint2);
-                                // skip the next 11 characters (xxxx\uyyyy)
-                                i += 11;
+                                // skip the next 10 characters (xxxx\uyyyy)
+                                i += 10;
                             }
                             else
                             {

diff --git a/src/json.hpp.re2c b/src/json.hpp.re2c
@@ -6162,8 +6162,8 @@ class basic_json
                                 auto codepoint2 = std::strtoul(std::string(reinterpret_cast<typename string_t::const_pointer>
                                                                (i + 7), 4).c_str(), nullptr, 16);
                                 result += to_unicode(codepoint, codepoint2);
-                                // skip the next 11 characters (xxxx\uyyyy)
-                                i += 11;
+                                // skip the next 10 characters (xxxx\uyyyy)
+                                i += 10;
                             }
                             else
                             {

diff --git a/test/unit.cpp b/test/unit.cpp
@@ -10205,4 +10205,9 @@ TEST_CASE("regression tests")
         j["string"] = bytes;
         CHECK(j["string"] == "\u0007\u0007");
     }
+
+    SECTION("character following a surrogate pair is skipped")
+    {
+        CHECK(json::parse("\"\\ud80c\\udc60abc\"").get<json::string_t>() == u8"\U00013060abc");
+    }
 }