Merge branch 'develop' into json-view/5869-7-check-utf8

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

tests/src/unit-bson.cpp

@@ -1311,7 +1311,7 @@ TEST_CASE("BSON declared document size bounds the content")
     // records the events that a SAX parser receives
     struct recorder : nlohmann::json_sax<json>
     {
-        std::string events;
+        std::string events{};
 
         bool null() override
         {

tests/src/unit-json_pointer.cpp

@@ -1122,17 +1122,17 @@ TEST_CASE("flatten of structured values")
 
         // the same for other keys that are not valid array indices
         CHECK_THROWS_WITH_AS(json({{"/a/0", 1}, {"/a/1a", 2}}).unflatten(),
-                             "[json.exception.out_of_range.404] unresolved reference token '1a'", json::out_of_range&);
+        "[json.exception.out_of_range.404] unresolved reference token '1a'", json::out_of_range&);
         CHECK_THROWS_WITH_AS(json({{"/a/0", 1}, {"/a/-", 2}}).unflatten(),
-                             "[json.exception.out_of_range.404] unresolved reference token '-'", json::out_of_range&);
+        "[json.exception.out_of_range.404] unresolved reference token '-'", json::out_of_range&);
     }
 
     SECTION("unflatten with the empty token among array indices")
     {
         CHECK_THROWS_WITH_AS(json({{"/a/0", 1}, {"/a/", 2}}).unflatten(),
-                             "[json.exception.out_of_range.404] unresolved reference token ''", json::out_of_range&);
+        "[json.exception.out_of_range.404] unresolved reference token ''", json::out_of_range&);
         CHECK_THROWS_WITH_AS(json({{"/a/0", 1}, {"/a/1", 2}, {"/a/", 3}}).unflatten(),
-                             "[json.exception.out_of_range.404] unresolved reference token ''", json::out_of_range&);
+        "[json.exception.out_of_range.404] unresolved reference token ''", json::out_of_range&);
 
         // without an index "0", the empty token is an ordinary object key
         CHECK(json({{"/a/", 2}}).unflatten() == json({{"a", {{"", 2}}}}));

tests/src/unit-json_view.cpp

@@ -666,6 +666,61 @@ TEST_CASE("json_view")
         CHECK(small.root().materialize() == json::parse("[1,[2,3],{\"a\":\"b\\n\"}]"));
     }
 
+    SECTION("block sizes")
+    {
+        // documents whose node estimate lies around the size that is allocated
+        // with the header, and whose decoded strings need the first growth of
+        // the buffer, in all sizes around the points where the policies change
+        for (const int count :
+                {
+                    1, 2, 10, 50, 100, 255, 256, 257, 500, 1000, 1023, 1024, 1025, 2000, 5000
+                })
+        {
+            // strings with escapes (decoded text, minified)
+            std::string escaped = "[";
+            for (int i = 0; i < count; ++i)
+            {
+                escaped += (i != 0 ? ",\"a\\n" : "\"a\\n") + std::to_string(i) + "\\u00e4\"";
+            }
+            escaped += ']';
+            json_document e = json_document::parse(escaped);
+            CHECK(e.root().materialize() == json::parse(escaped));
+            e.shrink_to_fit();
+            CHECK(e.root().materialize() == json::parse(escaped));
+
+            // dense numbers (more nodes per byte than the estimate expects)
+            std::string dense = "[";
+            for (int i = 0; i < count; ++i)
+            {
+                dense += (i != 0 ? ",1" : "1");
+            }
+            dense += ']';
+            json_document n = json_document::parse(dense);
+            CHECK(n.node_count() == static_cast<std::size_t>(count) + 1);
+            CHECK(n.root().materialize() == json::parse(dense));
+            n.shrink_to_fit();
+            CHECK(n.root().size() == static_cast<std::size_t>(count));
+
+            // the same documents, pretty-printed
+            const std::string pretty = json::parse(escaped).dump(2);
+            json_document p = json_document::parse(pretty);
+            CHECK(p.root().materialize() == json::parse(escaped));
+        }
+
+        // a large document with a single escape does not need a buffer of the
+        // size of its input
+        std::string large = R"(["a\n")";
+        for (int i = 0; i < 100000; ++i)
+        {
+            large += ",1";
+        }
+        large += ']';
+        const json_document l = json_document::parse(large);
+        CHECK(l.root().size() == 100001);
+        CHECK(l.root()[0].get_string() == "a\n");
+        CHECK(l.root()[100000].get<int>() == 1);
+    }
+
     SECTION("source offsets")
     {
         const std::string text = R"(  {"key": "value", "escaped": "a\nb", "n": 42})";

tests/src/unit-json_view_from_json.cpp

@@ -41,6 +41,13 @@ using nlohmann::ordered_json_document;
     #include <optional>
 #endif
 
+// NLOHMANN_JSON_SERIALIZE_ENUM uses a static std::pair; same() compares floating-point values on purpose
+DOCTEST_CLANG_SUPPRESS_WARNING_PUSH
+DOCTEST_CLANG_SUPPRESS_WARNING("-Wexit-time-destructors")
+DOCTEST_CLANG_SUPPRESS_WARNING("-Wfloat-equal")
+DOCTEST_GCC_SUPPRESS_WARNING_PUSH
+DOCTEST_GCC_SUPPRESS_WARNING("-Wfloat-equal")
+
 // The from_json() functions of the NLOHMANN_DEFINE_TYPE_* and
 // NLOHMANN_JSON_SERIALIZE_ENUM macros read views directly; other from_json()
 // functions get the value of materialize().
@@ -61,10 +68,11 @@ struct point
 
 struct shape
 {
-    std::string name;
-    std::vector<point> points;
-    point center;
-    std::map<std::string, point> named;
+    // initialized for GCC -Weffc++
+    std::string name{};
+    std::vector<point> points{};
+    point center{};
+    std::map<std::string, point> named{};
     NLOHMANN_DEFINE_TYPE_INTRUSIVE(shape, name, points, center, named)
 };
 
@@ -79,7 +87,7 @@ NLOHMANN_DEFINE_TYPE_NON_INTRUSIVE_WITH_DEFAULT(with_defaults, a, b, c)
 struct with_names
 {
     int value = 0;
-    std::string text;
+    std::string text{};  // initialized for GCC -Weffc++
 };
 NLOHMANN_DEFINE_TYPE_NON_INTRUSIVE_WITH_NAMES(with_names, "Value", value, "Text", text)
 
@@ -447,3 +455,6 @@ TEST_CASE("json_view from_json")
     }
 #endif
 }
+
+DOCTEST_CLANG_SUPPRESS_WARNING_POP
+DOCTEST_GCC_SUPPRESS_WARNING_POP

tests/src/unit-pointer_access.cpp

@@ -141,6 +141,7 @@ TEST_CASE("pointer access")
         // check if pointers are returned correctly
         const test_type* p1 = value.get_ptr<test_type*>();
         CHECK(p1 == value.get_ptr<test_type*>());
+        // @infer-ignore NULLPTR_DEREFERENCE
         CHECK(*p1 == value.get<test_type>());
 
         const test_type* p2 = value.get_ptr<const test_type*>();
@@ -170,6 +171,7 @@ TEST_CASE("pointer access")
         // check if pointers are returned correctly
         test_type* p1 = value.get_ptr<test_type*>();
         CHECK(p1 == value.get_ptr<test_type*>());
+        // @infer-ignore NULLPTR_DEREFERENCE
         CHECK(*p1 == value.get<test_type>());
 
         const test_type* p2 = value.get_ptr<const test_type*>();
@@ -199,6 +201,7 @@ TEST_CASE("pointer access")
         // check if pointers are returned correctly
         const test_type* p1 = value.get_ptr<test_type*>();
         CHECK(p1 == value.get_ptr<test_type*>());
+        // @infer-ignore NULLPTR_DEREFERENCE
         CHECK(*p1 == value.get<test_type>());
 
         const test_type* p2 = value.get_ptr<const test_type*>();
@@ -232,6 +235,7 @@ TEST_CASE("pointer access")
 
         const test_type* p2 = value.get_ptr<const test_type*>();
         CHECK(p2 == value.get_ptr<const test_type*>());
+        // @infer-ignore NULLPTR_DEREFERENCE
         CHECK(*p2 == value.get<test_type>());
 
         const test_type* const p3 = value.get_ptr<const test_type* const>();
@@ -257,6 +261,7 @@ TEST_CASE("pointer access")
         // check if pointers are returned correctly
         const test_type* p1 = value.get_ptr<test_type*>();
         CHECK(p1 == value.get_ptr<test_type*>());
+        // @infer-ignore NULLPTR_DEREFERENCE
         CHECK(*p1 == value.get<test_type>());
 
         const test_type* p2 = value.get_ptr<const test_type*>();