schemagen: cache non-object $defs entries to avoid duplicate types

Extend the existing per-definition cache (`self._building`) to enum,
array, and scalar $defs, not just object definitions. This ensures
that multiple $refs to the same non-object definition reuse the same
C++ type instead of generating Role, Role2, Role3, etc.

- Cache the built (type, nullable) tuple under defname for enum,
  scalar, and array definitions.
- Pre-register array definitions before recursing into items so $ref
  cycles resolve to the same TArr instance.
- Store object definitions as (TObj, nullable) tuples so nullable object
  $defs also preserve their nullability when referenced.

Add a regression test for issue #17 covering reused enum and array-of-enum
$defs.

contrib/schemagen/CMakeLists.txt

@@ -46,6 +46,49 @@ target_link_libraries(schemagen_big PRIVATE ${PROJECT_NAME})
 target_compile_options(schemagen_big PRIVATE -Wno-switch-enum)
 add_dependencies(schemagen_big schemagen_big_h)
 
+set(NULLABLE_OBJECT_SCHEMA
+    ${CMAKE_CURRENT_SOURCE_DIR}/nullable_object.schema.json)
+set(NULLABLE_STRING_SCHEMA
+    ${CMAKE_CURRENT_SOURCE_DIR}/nullable_string.schema.json)
+set(NULLABLE_ARRAY_SCHEMA
+    ${CMAKE_CURRENT_SOURCE_DIR}/nullable_array.schema.json)
+set(NULLABLE_OBJECT_H ${CMAKE_CURRENT_BINARY_DIR}/nullable_object.h)
+set(NULLABLE_STRING_H ${CMAKE_CURRENT_BINARY_DIR}/nullable_string.h)
+set(NULLABLE_ARRAY_H ${CMAKE_CURRENT_BINARY_DIR}/nullable_array.h)
+
+add_custom_command(
+  OUTPUT ${NULLABLE_OBJECT_H}
+  COMMAND ${Python3_EXECUTABLE} ${SCHEMAGEN_SCRIPT} ${NULLABLE_OBJECT_SCHEMA} -o
+          ${NULLABLE_OBJECT_H} --namespace nullable_object
+  DEPENDS ${SCHEMAGEN_SCRIPT} ${NULLABLE_OBJECT_SCHEMA}
+  COMMENT "Generating nullable_object.h")
+
+add_custom_command(
+  OUTPUT ${NULLABLE_STRING_H}
+  COMMAND ${Python3_EXECUTABLE} ${SCHEMAGEN_SCRIPT} ${NULLABLE_STRING_SCHEMA} -o
+          ${NULLABLE_STRING_H} --namespace nullable_string
+  DEPENDS ${SCHEMAGEN_SCRIPT} ${NULLABLE_STRING_SCHEMA}
+  COMMENT "Generating nullable_string.h")
+
+add_custom_command(
+  OUTPUT ${NULLABLE_ARRAY_H}
+  COMMAND ${Python3_EXECUTABLE} ${SCHEMAGEN_SCRIPT} ${NULLABLE_ARRAY_SCHEMA} -o
+          ${NULLABLE_ARRAY_H} --namespace nullable_array
+  DEPENDS ${SCHEMAGEN_SCRIPT} ${NULLABLE_ARRAY_SCHEMA}
+  COMMENT "Generating nullable_array.h")
+
+add_custom_target(
+  schemagen_nullable_h DEPENDS ${NULLABLE_OBJECT_H} ${NULLABLE_STRING_H}
+                               ${NULLABLE_ARRAY_H})
+
+add_executable(schemagen_nullable_root
+               ${CMAKE_CURRENT_SOURCE_DIR}/test_nullable_root.cpp)
+target_include_directories(schemagen_nullable_root
+                           PRIVATE include ${CMAKE_CURRENT_BINARY_DIR})
+target_link_libraries(schemagen_nullable_root PRIVATE ${PROJECT_NAME})
+target_compile_options(schemagen_nullable_root PRIVATE -Wno-switch-enum)
+add_dependencies(schemagen_nullable_root schemagen_nullable_h)
+
 add_test(
   NAME schemagen_example
   COMMAND schemagen_example
@@ -55,3 +98,8 @@ add_test(
   NAME schemagen_big
   COMMAND schemagen_big
   WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})
+
+add_test(
+  NAME schemagen_nullable_root
+  COMMAND schemagen_nullable_root
+  WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})

contrib/schemagen/README.md

@@ -45,7 +45,7 @@ into the result, so it is non-movable.
 | `$ref` to `$defs`/`definitions`               | the referenced named struct            |
 | recursive `$ref`                              | `std::unique_ptr<T>` (cycle broken)    |
 | `additionalProperties: false`                 | unknown keys rejected                  |
-| `additionalProperties` absent / `true`        | unknown keys' values skipped           |
+| `additionalProperties` absent / `true`        | not supported (rejected at generation) |
 
 ## Schema violations (rejected at parse time)
 
@@ -55,15 +55,15 @@ into the result, so it is non-movable.
 - value not in a string `enum`
 - a number not representable in the target type (e.g. `1.5` for an `integer`)
 - duplicate object keys
-- unknown key under `additionalProperties: false`
+- unknown key in any object
 
 ## Not supported (rejected at generation time, no fallback)
 
 `oneOf` / `anyOf` / `allOf` / `not` / `if`-`then`-`else`,
 `patternProperties`, `additionalProperties` with a schema (typed map),
-`prefixItems` (tuples), `const`, `dependentSchemas`/`dependentRequired`,
-union `type` lists other than `["T", "null"]`, non-string enums, and remote
-(`$ref` to other documents).
+`additionalProperties: true`, `prefixItems` (tuples), `const`,
+`dependentSchemas`/`dependentRequired`, union `type` lists other than
+`["T", "null"]`, non-string enums, and remote (`$ref` to other documents).
 
 ## Notes
 

contrib/schemagen/nullable_array.schema.json

@@ -0,0 +1,9 @@
+{
+  "type": [
+    "array",
+    "null"
+  ],
+  "items": {
+    "type": "integer"
+  }
+}

contrib/schemagen/nullable_object.schema.json

@@ -0,0 +1,15 @@
+{
+  "type": [
+    "object",
+    "null"
+  ],
+  "additionalProperties": false,
+  "required": [
+    "x"
+  ],
+  "properties": {
+    "x": {
+      "type": "string"
+    }
+  }
+}

contrib/schemagen/nullable_string.schema.json

@@ -0,0 +1,6 @@
+{
+  "type": [
+    "string",
+    "null"
+  ]
+}

contrib/schemagen/test_nullable_root.cpp

@@ -0,0 +1,158 @@
+// Regression test for issue #13: nullable root types.
+#include <cassert>
+#include <cstdio>
+#include <string>
+
+#include "nullable_array.h"
+#include "nullable_object.h"
+#include "nullable_string.h"
+
+static int failures = 0;
+#define CHECK(cond)                                                            \
+  do {                                                                         \
+    if (!(cond)) {                                                             \
+      printf("FAIL %s:%d: %s\n", __FILE__, __LINE__, #cond);                   \
+      ++failures;                                                              \
+    }                                                                          \
+  } while (0)
+
+static WeaselJsonStatus parseStrided(nullable_object::RootBuilder &b,
+                                     std::string in) {
+  for (size_t i = 0; i < in.size(); ++i) {
+    char c = in[i];
+    WeaselJsonStatus s = b.feed(&c, 1);
+    if (s != WeaselJson_AGAIN)
+      return s;
+  }
+  return b.finish();
+}
+
+static WeaselJsonStatus parseStrided(nullable_string::RootBuilder &b,
+                                     std::string in) {
+  for (size_t i = 0; i < in.size(); ++i) {
+    char c = in[i];
+    WeaselJsonStatus s = b.feed(&c, 1);
+    if (s != WeaselJson_AGAIN)
+      return s;
+  }
+  return b.finish();
+}
+
+static WeaselJsonStatus parseStrided(nullable_array::RootBuilder &b,
+                                     std::string in) {
+  for (size_t i = 0; i < in.size(); ++i) {
+    char c = in[i];
+    WeaselJsonStatus s = b.feed(&c, 1);
+    if (s != WeaselJson_AGAIN)
+      return s;
+  }
+  return b.finish();
+}
+
+static void expectReject(nullable_object::RootBuilder &b, std::string in,
+                         const char *what) {
+  WeaselJsonStatus s = parseStrided(b, in);
+  if (s == WeaselJson_REJECT) {
+    printf("ok   reject: %s\n", what);
+  } else {
+    printf("FAIL expected reject (%s) got status %d for: %s\n", what, s,
+           in.c_str());
+    ++failures;
+  }
+}
+
+int main() {
+  // ---- nullable root object: valid document ----
+  {
+    nullable_object::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, R"({"x":"hello"})");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_object::Root r = b.take();
+      CHECK(r.has_value());
+      CHECK(r->x == "hello");
+      printf("ok   nullable root object accepts object\n");
+    }
+  }
+
+  // ---- nullable root object: null document ----
+  {
+    nullable_object::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, "null");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_object::Root r = b.take();
+      CHECK(!r.has_value());
+      printf("ok   nullable root object accepts null\n");
+    }
+  }
+
+  // ---- nullable root object: schema checks still run ----
+  {
+    nullable_object::RootBuilder b;
+    expectReject(b, R"({"x":"hello","extra":1})",
+                 "unknown key in strict nullable root object");
+  }
+  {
+    nullable_object::RootBuilder b;
+    expectReject(b, R"({})", "missing required field in nullable root object");
+  }
+
+  // ---- nullable root string: valid value ----
+  {
+    nullable_string::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, R"("hello")");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_string::Root r = b.take();
+      CHECK(r.has_value());
+      CHECK(*r == "hello");
+      printf("ok   nullable root string accepts string\n");
+    }
+  }
+
+  // ---- nullable root string: null value ----
+  {
+    nullable_string::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, "null");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_string::Root r = b.take();
+      CHECK(!r.has_value());
+      printf("ok   nullable root string accepts null\n");
+    }
+  }
+
+  // ---- nullable root array: valid value ----
+  {
+    nullable_array::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, "[1,2,3]");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_array::Root r = b.take();
+      CHECK(r.has_value());
+      CHECK(r->size() == 3);
+      CHECK((*r)[0] == 1 && (*r)[1] == 2 && (*r)[2] == 3);
+      printf("ok   nullable root array accepts array\n");
+    }
+  }
+
+  // ---- nullable root array: null value ----
+  {
+    nullable_array::RootBuilder b;
+    WeaselJsonStatus s = parseStrided(b, "null");
+    CHECK(s == WeaselJson_OK);
+    if (s == WeaselJson_OK) {
+      nullable_array::Root r = b.take();
+      CHECK(!r.has_value());
+      printf("ok   nullable root array accepts null\n");
+    }
+  }
+
+  if (failures == 0) {
+    printf("\nALL TESTS PASSED\n");
+    return 0;
+  }
+  printf("\n%d FAILURE(S)\n", failures);
+  return 1;
+}

contrib/schemagen/test_schemagen.py

@@ -3,9 +3,12 @@
 
 import json
 import os
+import re
+import shutil
 import subprocess
 import sys
 import tempfile
+import textwrap
 import unittest
 
 SCRIPT = os.path.join(os.path.dirname(__file__), "weaseljson_schemagen.py")
@@ -87,5 +90,384 @@ class SchemagenKeywordTest(unittest.TestCase):
             self.assertNotIn(f"std::optional<std::string> {kw};", stdout)
 
 
+class SchemagenCollisionTest(unittest.TestCase):
+    """Regression tests for issue #21: generated Root alias / Kind enum collisions."""
+
+    def setUp(self):
+        self.repo_root = os.path.dirname(
+            os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+        )
+        self.include_dir = os.path.join(self.repo_root, "include")
+        self.compiler = shutil.which("c++")
+
+    def generate_and_compile(self, schema):
+        """Run schemagen on schema and syntax-check the resulting header."""
+        with tempfile.TemporaryDirectory() as tmpdir:
+            schema_path = os.path.join(tmpdir, "schema.json")
+            with open(schema_path, "w") as fp:
+                json.dump(schema, fp)
+            header_path = os.path.join(tmpdir, "gen.h")
+            cmd = [
+                sys.executable,
+                SCRIPT,
+                schema_path,
+                "-o",
+                header_path,
+                "--namespace",
+                "test_schema",
+            ]
+            result = subprocess.run(cmd, capture_output=True, text=True, check=False)
+            self.assertEqual(result.returncode, 0, msg=result.stderr)
+
+            if self.compiler:
+                cpp_path = os.path.join(tmpdir, "test.cpp")
+                with open(cpp_path, "w") as fp:
+                    fp.write(
+                        '#include "gen.h"\n'
+                        "int main() {\n"
+                        "  test_schema::RootBuilder b;\n"
+                        "  test_schema::Root r = b.take();\n"
+                        "  (void)r;\n"
+                        "}\n"
+                    )
+                comp = subprocess.run(
+                    [
+                        self.compiler,
+                        "-std=c++20",
+                        "-fsyntax-only",
+                        "-I",
+                        self.include_dir,
+                        "-I",
+                        tmpdir,
+                        cpp_path,
+                    ],
+                    capture_output=True,
+                    text=True,
+                    check=False,
+                )
+                self.assertEqual(comp.returncode, 0, msg=comp.stderr)
+
+            with open(header_path) as fp:
+                return fp.read()
+
+    def test_root_alias_does_not_collide_with_user_type(self):
+        schema = {
+            "type": "array",
+            "items": {"$ref": "#/$defs/Root"},
+            "$defs": {"Root": {"enum": ["a", "b"]}},
+        }
+        out = self.generate_and_compile(schema)
+        self.assertIn("enum class Root1 : int { a, b };", out)
+        self.assertIn("using Root = std::vector<Root1>;", out)
+        self.assertNotIn("using Root = std::vector<Root>;", out)
+
+    def test_kind_enum_does_not_duplicate_arr0(self):
+        schema = {
+            "type": "object",
+            "properties": {
+                "arr": {"type": "array", "items": {"type": "string"}},
+                "obj": {"$ref": "#/$defs/Arr0"},
+            },
+            "$defs": {"Arr0": {"type": "object", "properties": {}}},
+        }
+        out = self.generate_and_compile(schema)
+        self.assertIn("struct Arr0", out)
+        m = re.search(r"enum class Kind : uint8_t \{([^}]+)\}", out)
+        self.assertIsNotNone(m)
+        enumerators = [e.strip() for e in m.group(1).split(",")]
+        self.assertIn("Arr0", enumerators)
+        self.assertIn("Arr1", enumerators)
+        self.assertEqual(len(enumerators), len(set(enumerators)))
+
+    def test_skip_user_type_is_allowed(self):
+        schema = {
+            "type": "array",
+            "items": {"$ref": "#/$defs/Skip"},
+            "$defs": {"Skip": {"type": "object", "properties": {}}},
+        }
+        out = self.generate_and_compile(schema)
+        self.assertIn("struct Skip", out)
+        self.assertNotIn("struct Skip1", out)
+        m = re.search(r"enum class Kind : uint8_t \{([^}]+)\}", out)
+        self.assertIsNotNone(m)
+        enumerators = [e.strip() for e in m.group(1).split(",")]
+        self.assertIn("Skip", enumerators)
+        self.assertIn("Arr0", enumerators)
+        self.assertEqual(len(enumerators), len(set(enumerators)))
+
+    def test_non_object_defs_reused_across_refs(self):
+        """Regression test for issue #17: enum and array $defs must be reused."""
+        schema = {
+            "type": "object",
+            "properties": {
+                "role1": {"$ref": "#/$defs/Role"},
+                "role2": {"$ref": "#/$defs/Role"},
+                "roles1": {"$ref": "#/$defs/Roles"},
+                "roles2": {"$ref": "#/$defs/Roles"},
+            },
+            "$defs": {
+                "Role": {"enum": ["admin", "user"]},
+                "Roles": {"type": "array", "items": {"$ref": "#/$defs/Role"}},
+            },
+        }
+        out = self.generate_and_compile(schema)
+        # Exactly one Role enum is generated.
+        self.assertIn("enum class Role : int { admin, user };", out)
+        self.assertNotIn("enum class Role1 : int", out)
+        self.assertNotIn("enum class Role2 : int", out)
+        # The array of enum is represented by a single kind.
+        m = re.search(r"enum class Kind : uint8_t \{([^}]+)\}", out)
+        self.assertIsNotNone(m)
+        enumerators = [e.strip() for e in m.group(1).split(",")]
+        self.assertEqual(enumerators.count("Arr0"), 1)
+        # All four fields use the same C++ types.
+        self.assertIn("std::optional<Role> role1;", out)
+        self.assertIn("std::optional<Role> role2;", out)
+        self.assertIn("std::optional<std::vector<Role>> roles1;", out)
+        self.assertIn("std::optional<std::vector<Role>> roles2;", out)
+
+
+class SchemagenAdditionalPropertiesTest(unittest.TestCase):
+    def run_schemagen(self, schema, args=None):
+        """Run schemagen on a schema dict. Returns (returncode, stdout, stderr)."""
+        with tempfile.NamedTemporaryFile("w", suffix=".json", delete=False) as fp:
+            json.dump(schema, fp)
+            schema_path = fp.name
+        try:
+            cmd = [sys.executable, SCRIPT, schema_path]
+            if args:
+                cmd.extend(args)
+            result = subprocess.run(cmd, capture_output=True, text=True, check=False)
+            return result.returncode, result.stdout, result.stderr
+        finally:
+            os.unlink(schema_path)
+
+    def test_additional_properties_true_rejected(self):
+        schema = {
+            "type": "object",
+            "additionalProperties": True,
+            "properties": {"name": {"type": "string"}},
+        }
+        rc, stdout, stderr = self.run_schemagen(schema)
+        self.assertNotEqual(rc, 0)
+        self.assertIn("additionalProperties: true is not supported", stderr)
+
+    def test_additional_properties_absent_defaults_to_strict(self):
+        schema = {
+            "type": "object",
+            "properties": {"name": {"type": "string"}},
+        }
+        rc, stdout, stderr = self.run_schemagen(schema)
+        self.assertEqual(rc, 0, msg=stderr)
+        # The generated parser should reject unknown keys. Verify the key-matching
+        # helper returns -1 for an unknown key and cbKeyData rejects it.
+        self.assertIn("int matchKey(Kind k, std::string_view key) const {", stdout)
+        self.assertNotIn("bool isStrict(Kind k) const", stdout)
+
+    def test_additional_properties_false_accepted(self):
+        schema = {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {"name": {"type": "string"}},
+        }
+        rc, stdout, stderr = self.run_schemagen(schema)
+        self.assertEqual(rc, 0, msg=stderr)
+
+
+class SchemagenIntegerBoundaryTest(unittest.TestCase):
+    """Regression tests for issue #19: integer slot parsing near int64 boundaries."""
+
+    def setUp(self):
+        self.repo_root = os.path.dirname(
+            os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+        )
+        self.include_dir = os.path.join(self.repo_root, "include")
+        self.lib_src = os.path.join(self.repo_root, "src", "lib.cpp")
+        self.compiler = shutil.which("c++")
+
+    def _compile_harness(self, tmpdir, schema, harness):
+        schema_path = os.path.join(tmpdir, "schema.json")
+        with open(schema_path, "w") as fp:
+            json.dump(schema, fp)
+        header_path = os.path.join(tmpdir, "gen.h")
+        result = subprocess.run(
+            [
+                sys.executable,
+                SCRIPT,
+                schema_path,
+                "-o",
+                header_path,
+                "--namespace",
+                "test_schema",
+            ],
+            capture_output=True,
+            text=True,
+            check=False,
+        )
+        self.assertEqual(result.returncode, 0, msg=result.stderr)
+
+        lib_obj = os.path.join(tmpdir, "lib.o")
+        comp_lib = subprocess.run(
+            [
+                self.compiler,
+                "-std=c++20",
+                "-I",
+                self.include_dir,
+                "-I",
+                os.path.join(self.repo_root, "third_party", "include"),
+                "-I",
+                os.path.join(self.repo_root, "third_party", "valgrind"),
+                "-c",
+                self.lib_src,
+                "-o",
+                lib_obj,
+            ],
+            capture_output=True,
+            text=True,
+            check=False,
+        )
+        self.assertEqual(comp_lib.returncode, 0, msg=comp_lib.stderr)
+
+        cpp_path = os.path.join(tmpdir, "test.cpp")
+        with open(cpp_path, "w") as fp:
+            fp.write(harness)
+
+        exe_path = os.path.join(tmpdir, "test")
+        comp = subprocess.run(
+            [
+                self.compiler,
+                "-std=c++20",
+                "-I",
+                self.include_dir,
+                "-I",
+                tmpdir,
+                lib_obj,
+                cpp_path,
+                "-o",
+                exe_path,
+            ],
+            capture_output=True,
+            text=True,
+            check=False,
+        )
+        self.assertEqual(comp.returncode, 0, msg=comp.stderr)
+        run = subprocess.run([exe_path], capture_output=True, text=True, check=False)
+        self.assertEqual(run.returncode, 0, msg=run.stdout + run.stderr)
+
+    def _build_cases_array(self, name, entries):
+        lines = [
+            f"  struct {name}Case {{ const char *s; WeaselJsonStatus expected; long long v; }};"
+        ]
+        lines.append(f"  {name}Case {name}_cases[] = {{")
+        for s, exp, v in entries:
+            val = f"{v}LL" if isinstance(v, int) else v
+            lines.append(f'    {{ R"({s})", {exp}, {val} }},')
+        lines.append("  };")
+        return "\n".join(lines)
+
+    def test_integer_boundary_root(self):
+        if not self.compiler:
+            self.skipTest("C++ compiler not available")
+        cases = [
+            ("9223372036854775807.0", "WeaselJson_OK", 9223372036854775807),
+            ("9223372036854775807", "WeaselJson_OK", 9223372036854775807),
+            ("9223372036854775806.0", "WeaselJson_OK", 9223372036854775806),
+            ("9223372036854775808", "WeaselJson_REJECT", 0),
+            ("-9223372036854775808", "WeaselJson_OK", "-9223372036854775807LL - 1"),
+            ("-9223372036854775808.0", "WeaselJson_OK", "-9223372036854775807LL - 1"),
+            ("-9223372036854775809", "WeaselJson_REJECT", 0),
+            ("1e3", "WeaselJson_OK", 1000),
+            ("2.0", "WeaselJson_OK", 2),
+            ("0.001", "WeaselJson_REJECT", 0),
+            ("1e-3", "WeaselJson_REJECT", 0),
+            ("1000e-3", "WeaselJson_OK", 1),
+            ("100.0e-2", "WeaselJson_OK", 1),
+            ("123.0", "WeaselJson_OK", 123),
+            ("9e18", "WeaselJson_OK", 9000000000000000000),
+            ("10e18", "WeaselJson_REJECT", 0),
+        ]
+        cases_src = self._build_cases_array("root", cases)
+        harness = textwrap.dedent(
+            f"""
+            #include "gen.h"
+            #include <cstdio>
+            #include <cstring>
+            {cases_src}
+            int main() {{
+              for (const auto \u0026c : root_cases) {{
+                test_schema::RootBuilder b;
+                char buf[512];
+                std::strncpy(buf, c.s, sizeof(buf) - 1);
+                buf[sizeof(buf) - 1] = '\\0';
+                WeaselJsonStatus st = b.feed(buf, std::strlen(buf));
+                st = b.finish();
+                if (st != c.expected) {{
+                  std::printf("root case %s expected %d got %d\\n", c.s, c.expected, st);
+                  return 1;
+                }}
+                if (st == WeaselJson_OK \u0026\u0026 b.take() != c.v) {{
+                  std::printf("root case %s value mismatch\\n", c.s);
+                  return 2;
+                }}
+              }}
+              return 0;
+            }}
+            """
+        )
+        with tempfile.TemporaryDirectory() as tmpdir:
+            self._compile_harness(tmpdir, {"type": "integer"}, harness)
+
+    def test_integer_boundary_object_field(self):
+        if not self.compiler:
+            self.skipTest("C++ compiler not available")
+        cases = [
+            ('{"age":9223372036854775807.0}', "WeaselJson_OK", 9223372036854775807),
+            ('{"age":-9223372036854775809}', "WeaselJson_REJECT", 0),
+            ('{"age":1e3}', "WeaselJson_OK", 1000),
+            ('{"age":2.0}', "WeaselJson_OK", 2),
+            ('{"age":0.001}', "WeaselJson_REJECT", 0),
+            (
+                '{"age":-9223372036854775808.0}',
+                "WeaselJson_OK",
+                "-9223372036854775807LL - 1",
+            ),
+        ]
+        cases_src = self._build_cases_array("object", cases)
+        harness = textwrap.dedent(
+            f"""
+            #include "gen.h"
+            #include <cstdio>
+            #include <cstring>
+            {cases_src}
+            int main() {{
+              for (const auto \u0026c : object_cases) {{
+                test_schema::RootBuilder b;
+                char buf[512];
+                std::strncpy(buf, c.s, sizeof(buf) - 1);
+                buf[sizeof(buf) - 1] = '\\0';
+                WeaselJsonStatus st = b.feed(buf, std::strlen(buf));
+                st = b.finish();
+                if (st != c.expected) {{
+                  std::printf("object case %s expected %d got %d\\n", c.s, c.expected, st);
+                  return 3;
+                }}
+                if (st == WeaselJson_OK \u0026\u0026 b.take().age != c.v) {{
+                  std::printf("object case %s value mismatch\\n", c.s);
+                  return 4;
+                }}
+              }}
+              return 0;
+            }}
+            """
+        )
+        schema = {
+            "type": "object",
+            "properties": {"age": {"type": "integer"}},
+            "required": ["age"],
+        }
+        with tempfile.TemporaryDirectory() as tmpdir:
+            self._compile_harness(tmpdir, schema, harness)
+
+
 if __name__ == "__main__":
     unittest.main()

contrib/schemagen/weaseljson_schemagen.py

@@ -60,7 +60,6 @@ class ObjectType:
     def __init__(self, name):
         self.name = name
         self.fields = []  # list[Field]
-        self.strict = False  # additionalProperties: false
 
 
 class EnumType:
@@ -216,12 +215,26 @@ class Builder:
         base = camel(hint)
         name = base
         i = 1
-        while name in self._used_names:
+        while self._name_taken(name):
+            candidate = f"{base}{i}"
+            # A numeric suffix can itself land on a reserved generated name
+            # (e.g. hint "Arr0" -> "Arr01").  Use an underscore separator so
+            # we never loop through the reserved block.
+            if self._is_reserved(candidate):
+                candidate = f"{base}_{i}"
+            name = candidate
             i += 1
-            name = f"{base}{i}"
         self._used_names.add(name)
         return name
 
+    def _name_taken(self, name):
+        return name in self._used_names or self._is_reserved(name)
+
+    @staticmethod
+    def _is_reserved(name):
+        """Names generated internally that must not collide with user types."""
+        return name in ("Root", "RootScalar")
+
     def ref_name(self, ref):
         if not ref.startswith("#/"):
             raise GenError(f"only local $ref supported, got: {ref}")
@@ -237,7 +250,6 @@ class Builder:
         if defname not in self.defs:
             raise GenError(f"$ref to unknown def: {defname}")
         node = self.defs[defname]
-        # For objects we must register the name before recursing into fields.
         return self.build_type(node, defname, defname=defname)
 
     def build_type(self, node, hint, defname=None):
@@ -266,6 +278,10 @@ class Builder:
         if "$ref" in node:
             return self.build_def(self.ref_name(node["$ref"]))
 
+        # Non-object $defs entries must reuse the same type for every $ref.
+        if defname is not None and defname in self._building:
+            return self._building[defname]
+
         # nullability via type lists: ["string", "null"]
         nullable = False
         typ = node.get("type")
@@ -284,18 +300,29 @@ class Builder:
                 raise GenError("only non-empty string enums are supported")
             name = defname and self.unique_name(defname) or self.unique_name(hint)
             self.enums[name] = EnumType(name, list(vals))
-            return (TEnum(name), nullable)
+            result = (TEnum(name), nullable)
+            if defname is not None:
+                self._building[defname] = result
+            return result
 
         if typ == "object" or (typ is None and "properties" in node):
-            return (self._build_object(node, hint, defname), nullable)
+            return self._build_object(node, hint, defname, nullable)
 
         if typ == "array":
             if "items" not in node or not isinstance(node["items"], dict):
                 raise GenError("arrays require a single 'items' schema")
+            # Register the array before building its items so $ref cycles back
+            # to this definition resolve to the same TArr instance.
+            t = TArr(None, False)
+            result = (t, nullable)
+            if defname is not None:
+                self._building[defname] = result
             elem, elem_nullable = self._unpack(
                 self.build_type(node["items"], hint + "Item")
             )
-            return (TArr(elem, elem_nullable), nullable)
+            t.elem = elem
+            t.elem_nullable = elem_nullable
+            return result
 
         scalar = {
             "string": "str",
@@ -304,7 +331,10 @@ class Builder:
             "boolean": "bool",
         }.get(typ)
         if scalar:
-            return (TScalar(scalar), nullable)
+            result = (TScalar(scalar), nullable)
+            if defname is not None:
+                self._building[defname] = result
+            return result
         if typ == "null":
             raise GenError("'null'-only types are not supported")
 
@@ -318,19 +348,21 @@ class Builder:
             return result
         return (result, False)
 
-    def _build_object(self, node, hint, defname):
+    def _build_object(self, node, hint, defname, nullable=False):
         name = self.unique_name(defname or hint)
         obj = ObjectType(name)
         self.objects[name] = obj
         # register for $ref cycles before building fields
+        tobj = TObj(name)
         if defname is not None:
-            self._building[defname] = TObj(name)
-        ap = node.get("additionalProperties", True)
+            self._building[defname] = (tobj, nullable)
+        ap = node.get("additionalProperties", False)
+        if ap is True:
+            raise GenError("additionalProperties: true is not supported")
         if isinstance(ap, dict):
             raise GenError(
                 "additionalProperties with a schema (typed map) is not " "supported yet"
             )
-        obj.strict = ap is False
         required = set(node.get("required", []))
         props = node.get("properties", {})
         seen_cpp = set()
@@ -344,7 +376,7 @@ class Builder:
                 cpp = f"{base}{i}"
             seen_cpp.add(cpp)
             obj.fields.append(Field(key, cpp, ty, key in required, nullable))
-        return TObj(name)
+        return tobj
 
 
 # ---------------------------------------------------------------------------
@@ -435,6 +467,7 @@ class Emitter:
         self.kind_order = []  # all Kind enumerators in declaration order
         self.root_ty = None
         self.root_nullable = False
+        self._arr_counter = 0
 
     # -- type strings -------------------------------------------------------
     def base_cpp(self, ty):
@@ -465,11 +498,19 @@ class Emitter:
     def arr_kind(self, tarr):
         sig = self.base_cpp(tarr)
         if sig not in self.arr_kinds:
-            name = f"Arr{len(self.arr_kinds)}"
+            name = self._fresh_arr_kind_name()
             self.arr_kinds[sig] = name
             self.arr_types.append((name, tarr))
+            self.b._used_names.add(name)
         return self.arr_kinds[sig]
 
+    def _fresh_arr_kind_name(self):
+        while True:
+            name = f"Arr{self._arr_counter}"
+            self._arr_counter += 1
+            if not self.b._name_taken(name):
+                return name
+
     def cat(self, ty):
         if isinstance(ty, TScalar):
             return {"str": "Str", "int": "Int", "dbl": "Dbl", "bool": "Bool"}[ty.kind]
@@ -493,6 +534,18 @@ class Emitter:
         self.root_ty, self.root_nullable = Builder._unpack(
             self.b.build_type(self.b.root_schema, "Root")
         )
+
+        # A nullable root object would otherwise produce
+        #   using Root = std::optional<Root>;
+        # which conflicts with the struct named Root.  Rename the inner struct.
+        if isinstance(self.root_ty, TObj) and self.root_nullable:
+            old_name = self.root_ty.name
+            new_name = self.b.unique_name("RootInner")
+            obj = self.b.objects.pop(old_name)
+            obj.name = new_name
+            self.b.objects[new_name] = obj
+            self.root_ty = TObj(new_name)
+
         break_cycles(self.b.objects)
 
         # register all array kinds (walk every field + root)
@@ -582,20 +635,17 @@ namespace {ns} {{"""
     def _kind_enum(self):
         kinds = list(self.b.objects.keys())
         kinds += [n for n, _ in self.arr_types]
-        root_is_container = (
-            isinstance(self.root_ty, (TObj, TArr)) and not self.root_nullable
-        )
+        root_is_container = isinstance(self.root_ty, (TObj, TArr))
         if not root_is_container:
             kinds.append("RootScalar")
-        kinds.append("Skip")
         self.kind_order = kinds
         return "  enum class Kind : uint8_t { " + ", ".join(kinds) + " };"
 
     def _root_info(self):
         """Return (root_cat, root_container_kind_or_None)."""
-        if isinstance(self.root_ty, TObj) and not self.root_nullable:
+        if isinstance(self.root_ty, TObj):
             return ("Obj", f"Kind::{self.root_ty.name}")
-        if isinstance(self.root_ty, TArr) and not self.root_nullable:
+        if isinstance(self.root_ty, TArr):
             return ("Arr", f"Kind::{self.arr_kind(self.root_ty)}")
         return (self.cat(self.root_ty), None)
 
@@ -665,6 +715,11 @@ namespace {ns} {{"""
             lines.append("    }")
         root_cat, root_kind = self._root_info()
         if root_kind is None:
+            if self.root_nullable:
+                lines.append(
+                    "    case Kind::RootScalar: { if (!result_) result_.emplace(); return &*result_; }"
+                )
+            else:
                 lines.append("    case Kind::RootScalar: return &result_;")
         lines.append("    default: return nullptr;")
         lines.append("    }")
@@ -771,17 +826,6 @@ namespace {ns} {{"""
         lines.append("  }")
         return "\n".join(lines)
 
-    def _is_strict(self):
-        strict = [n for n, o in self.b.objects.items() if o.strict]
-        if not strict:
-            return "  bool isStrict(Kind) const { return false; }"
-        cases = " ".join(f"case Kind::{n}:" for n in strict)
-        return (
-            "  bool isStrict(Kind k) const {\n"
-            f"    switch (k) {{ {cases} return true; default: return false; }}\n"
-            "  }"
-        )
-
     def _enum_name_arrays(self):
         out = []
         for e in self.b.enums.values():
@@ -807,7 +851,12 @@ namespace {ns} {{"""
         lines = ["    if (error_) return;"]
         if root_kind is not None and root_cat == event_cat:
             lines.append("    if (stack_.empty()) {")
-            lines.append(f"      stack_.push_back(Frame{{{root_kind}, &result_}});")
+            if self.root_nullable:
+                lines.append("      result_.emplace();")
+                dest = "&*result_"
+            else:
+                dest = "&result_"
+            lines.append(f"      stack_.push_back(Frame{{{root_kind}, {dest}}});")
             if event_cat == "Obj":
                 lines.append("      {")
                 lines.append(f"        int n = fieldCount({root_kind});")
@@ -818,13 +867,7 @@ namespace {ns} {{"""
         else:
             lines.append("    if (stack_.empty()) { reject(); return; }")
         lines.append("    Frame &f = stack_.back();")
-        lines.append(
-            "    if (f.kind == Kind::Skip) { stack_.push_back(Frame{Kind::Skip, nullptr}); return; }"
-        )
         lines.append("    SlotInfo si = slotInfoG(f);")
-        lines.append(
-            "    if (si.cat == Cat::Skip) { stack_.push_back(Frame{Kind::Skip, nullptr}); return; }"
-        )
         lines.append(f"    if (si.cat != Cat::{event_cat}) {{ reject(); return; }}")
         lines.append("    void *p = engage(f, true);")
         lines.append("    stack_.push_back(Frame{si.child, p});")
@@ -839,6 +882,9 @@ namespace {ns} {{"""
         root_cat, root_kind = self._root_info()
         begin_obj = self._begin_container("Obj", root_kind)
         begin_arr = self._begin_container("Arr", root_kind)
+        null_at_root = (
+            "done_ = true; return;" if self.root_nullable else "reject(); return;"
+        )
 
         return f"""
 class RootBuilder {{
@@ -872,7 +918,7 @@ public:
 
 private:
 {kind_enum}
-  enum class Cat {{ Reject, Str, Int, Dbl, Bool, Enum, Obj, Arr, Skip }};
+  enum class Cat {{ Reject, Str, Int, Dbl, Bool, Enum, Obj, Arr }};
   struct SlotInfo {{
     Cat cat;
     Kind child{{}};
@@ -883,11 +929,10 @@ private:
   struct Frame {{
     Kind kind;
     void *dest;
-    int field = -1;              // object: selected field (-1 want key, -2 skip)
+    int field = -1;              // object: selected field (-1 means waiting for key)
     std::vector<uint64_t> seen;  // populated field bitset (object frames)
   }};
   static constexpr int kWantKey = -1;
-  static constexpr int kSkip = -2;
 
 {self._enum_name_arrays()}
 
@@ -902,10 +947,93 @@ private:
 
   void reject() {{ error_ = true; }}
 
-  // Wrap the generated slotInfo() with the generic key/skip states.
+  // Parse a JSON number text as int64_t.  Accepts optional decimal point and
+  // exponent only when the mathematical value is an integer that fits in a
+  // signed 64-bit range.
+  static bool parseJsonInt64(const char *b, const char *e, int64_t &out) {{
+    const char *p = b;
+    bool neg = false;
+    if (p < e) {{
+      if (*p == '-') {{ neg = true; ++p; }}
+      else if (*p == '+') return false;
+    }}
+    const char *intStart = p;
+    while (p < e && *p >= '0' && *p <= '9') ++p;
+    const char *intEnd = p;
+    int fracDigits = 0;
+    const char *fracStart = p;
+    if (p < e && *p == '.') {{
+      ++p;
+      fracStart = p;
+      while (p < e && *p >= '0' && *p <= '9') {{ ++p; ++fracDigits; }}
+      if (fracStart == p) return false;
+    }}
+    int64_t exp = 0;
+    bool expNeg = false;
+    if (p < e && (*p == 'e' || *p == 'E')) {{
+      ++p;
+      if (p < e && (*p == '-' || *p == '+')) {{ expNeg = (*p == '-'); ++p; }}
+      if (p == e || *p < '0' || *p > '9') return false;
+      while (p < e && *p >= '0' && *p <= '9') {{
+        int digit = *p - '0';
+        if (exp <= (INT64_MAX - digit) / 10) exp = exp * 10 + digit;
+        else exp = INT64_MAX;
+        ++p;
+      }}
+      if (expNeg) exp = -exp;
+    }}
+    if (p != e) return false;
+    if (intStart == intEnd) return false;
+
+    std::string digits;
+    digits.reserve((intEnd - intStart) + fracDigits);
+    for (const char *q = intStart; q < intEnd; ++q) digits.push_back(*q);
+    for (int i = 0; i < fracDigits; ++i) digits.push_back(fracStart[i]);
+    int64_t trim = 0;
+    while (!digits.empty() && digits.back() == '0') {{
+      digits.pop_back();
+      ++trim;
+    }}
+    int64_t finalExp = exp - fracDigits + trim;
+    if (finalExp < 0) return false;
+    while (!digits.empty() && digits.front() == '0') digits.erase(digits.begin());
+    if (digits.empty()) {{ out = 0; return true; }}
+
+    constexpr uint64_t kMaxNeg = 9223372036854775808ULL;
+    constexpr uint64_t kMaxPos = 9223372036854775807ULL;
+    const uint64_t limit = neg ? kMaxNeg : kMaxPos;
+
+    if (finalExp > 19) return false;
+    int64_t maxSig = 19 - finalExp;
+    uint64_t mag = 0;
+    int64_t sigDigits = 0;
+    for (char ch : digits) {{
+      uint64_t d = static_cast<uint64_t>(ch - '0');
+      if (sigDigits >= maxSig) return false;
+      if (mag > (limit - d) / 10) return false;
+      mag = mag * 10 + d;
+      ++sigDigits;
+    }}
+    for (int64_t i = 0; i < finalExp; ++i) {{
+      if (mag > limit / 10) return false;
+      mag *= 10;
+    }}
+    if (mag > limit) return false;
+    if (neg) {{
+      if (mag == kMaxNeg) {{
+        out = INT64_MIN;
+      }} else {{
+        out = -static_cast<int64_t>(mag);
+      }}
+    }} else {{
+      out = static_cast<int64_t>(mag);
+    }}
+    return true;
+  }}
+
+  // Wrap the generated slotInfo() with the generic key state.
   SlotInfo slotInfoG(const Frame &f) {{
     if (isObjectKind(f.kind)) {{
-      if (f.field == kSkip) return SlotInfo{{Cat::Skip}};
       if (f.field < 0) return SlotInfo{{Cat::Reject}};
     }}
     return slotInfo(f);
@@ -930,11 +1058,6 @@ private:
   void cbEndObject() {{
     if (error_) return;
     Frame &f = stack_.back();
-    if (f.kind == Kind::Skip) {{
-      stack_.pop_back();
-      if (stack_.empty() || stack_.back().kind != Kind::Skip) valueComplete();
-      return;
-    }}
     if (isObjectKind(f.kind)) {{
       const auto &req = requiredMask(f.kind);
       bool missing = false;
@@ -952,11 +1075,6 @@ private:
   void cbEndArray() {{
     if (error_) return;
     Frame f = stack_.back();
-    if (f.kind == Kind::Skip) {{
-      stack_.pop_back();
-      if (stack_.empty() || stack_.back().kind != Kind::Skip) valueComplete();
-      return;
-    }}
     stack_.pop_back();
     valueComplete();
   }}
@@ -968,12 +1086,10 @@ private:
     scratch_.append(buf, len);
     if (!done) return;
     int idx = matchKey(f.kind, scratch_);
-    scratch_.clear();
     if (idx < 0) {{
-      if (isStrict(f.kind)) {{ reject(); return; }}
-      f.field = kSkip;
-      return;
+      reject(); return;  // unknown key
     }}
+    scratch_.clear();
     if (f.seen[idx >> 6] & (1ull << (idx & 63))) {{
       reject(); return;  // duplicate key
     }}
@@ -983,9 +1099,7 @@ private:
   void cbStringData(const char *buf, int len, int done) {{
     if (error_) return;
     Frame &f = stack_.back();
-    if (f.kind == Kind::Skip) return;
     SlotInfo si = slotInfoG(f);
-    if (si.cat == Cat::Skip) {{ if (done) valueComplete(); return; }}
     if (si.cat == Cat::Str) {{
       auto *s = (std::string *)engage(f, !started_);
       s->append(buf, len);
@@ -1011,9 +1125,7 @@ private:
   void cbNumberData(const char *buf, int len, int done) {{
     if (error_) return;
     Frame &f = stack_.back();
-    if (f.kind == Kind::Skip) return;
     SlotInfo si = slotInfoG(f);
-    if (si.cat == Cat::Skip) {{ if (done) valueComplete(); return; }}
     if (si.cat != Cat::Int && si.cat != Cat::Dbl) {{ reject(); return; }}
     scratch_.append(buf, len);
     started_ = true;
@@ -1028,16 +1140,11 @@ private:
         *(int64_t *)p = v;  // plain integer literal: parsed exactly
       }} else {{
         // JSON Schema "integer" accepts any number with no fractional part,
-        // including exponent/decimal forms like 1e3 or 2.0. Parse those as a
-        // double and require an integral value within int64 range.
-        double d = 0;
-        auto rd = std::from_chars(b, e, d);
-        if (rd.ec != std::errc() || rd.ptr != e || d != std::trunc(d) ||
-            d < -9223372036854775808.0 || d >= 9223372036854775808.0) {{
-          reject();
-          return;
-        }}
-        *(int64_t *)p = (int64_t)d;
+        // including exponent/decimal forms like 1e3 or 2.0. Parse those
+        // exactly as int64 when the value is integral and in range.
+        int64_t v2 = 0;
+        if (!parseJsonInt64(b, e, v2)) {{ reject(); return; }}
+        *(int64_t *)p = v2;
       }}
     }} else {{
       double v = 0;
@@ -1051,9 +1158,7 @@ private:
   void cbBool(bool value) {{
     if (error_) return;
     Frame &f = stack_.back();
-    if (f.kind == Kind::Skip) return;
     SlotInfo si = slotInfoG(f);
-    if (si.cat == Cat::Skip) {{ valueComplete(); return; }}
     if (si.cat != Cat::Bool) {{ reject(); return; }}
     *(bool *)engage(f, true) = value;
     valueComplete();
@@ -1065,10 +1170,11 @@ private:
 
   void cbNull() {{
     if (error_) return;
+    if (stack_.empty()) {{
+      {null_at_root}
+    }}
     Frame &f = stack_.back();
-    if (f.kind == Kind::Skip) return;
     SlotInfo si = slotInfoG(f);
-    if (si.cat == Cat::Skip) {{ valueComplete(); return; }}
     if (!si.nullable) {{ reject(); return; }}
     if (isArrayKind(f.kind)) appendNull(f);  // keep null array elements
     valueComplete();  // leave optional empty / pointer null
@@ -1106,8 +1212,6 @@ private:
 {self._reqmask()}
 
 {self._is_object_kind()}
-
-{self._is_strict()}
 }};
 """