schemagen: escape control characters in generated C++ string literals

Fixes #35.

Add a helper to escape C++ string literals so that JSON control characters
(\n, \r, \t, and other bytes below 0x20) are emitted as escape sequences
instead of raw bytes. Use it for:

- field comments that include the JSON property key
- object key comparison literals in matchKey()
- enum name arrays

Also add regression tests that generate and syntax-check headers for
schemas containing newlines and other control characters in property keys
and enum values.

contrib/schemagen/weaseljson_schemagen.py

@@ -18,6 +18,43 @@ import keyword
 import sys
 
 
+def _escape_cpp_string(s):
+    """Return *s* escaped for use inside a C++ double-quoted string literal.
+
+    JSON strings may contain control characters; emitting them verbatim into
+    generated C++ source breaks tokenization.  This helper escapes backslashes
+    and double quotes, maps common control characters to their short escape
+    sequences, and uses universal character names (\\u00XX) for any other
+    character below 0x20.
+    """
+    out = []
+    for ch in s:
+        cp = ord(ch)
+        if cp == 0x09:
+            out.append("\\t")
+        elif cp == 0x0A:
+            out.append("\\n")
+        elif cp == 0x0B:
+            out.append("\\v")
+        elif cp == 0x0C:
+            out.append("\\f")
+        elif cp == 0x0D:
+            out.append("\\r")
+        elif cp == 0x08:
+            out.append("\\b")
+        elif cp == 0x07:
+            out.append("\\a")
+        elif cp < 0x20:
+            out.append(f"\\u{cp:04X}")
+        elif ch == "\\":
+            out.append("\\\\")
+        elif ch == '"':
+            out.append('\\"')
+        else:
+            out.append(ch)
+    return "".join(out)
+
+
 class GenError(Exception):
     pass
 
@@ -618,7 +655,8 @@ namespace {ns} {{"""
                     and f.ty.kind in ("int", "dbl", "bool")
                 ):
                     init = " = 0" if f.ty.kind != "bool" else " = false"
-                out.append(f'  {store} {f.cpp}{init};  // "{f.key}"')
+                esc_key = _escape_cpp_string(f.key)
+                out.append(f'  {store} {f.cpp}{init};  // "{esc_key}"')
             out.append("};")
             out.append("")
         return "\n".join(out)
@@ -746,7 +784,7 @@ namespace {ns} {{"""
         for name, obj in self.b.objects.items():
             lines.append(f"    case Kind::{name}:")
             for i, fld in enumerate(obj.fields):
-                esc = fld.key.replace("\\", "\\\\").replace('"', '\\"')
+                esc = _escape_cpp_string(fld.key)
                 lines.append(f'      if (key == "{esc}") return {i};')
             lines.append("      return -1;")
         lines.append("    default: return -1;")
@@ -829,10 +867,7 @@ namespace {ns} {{"""
     def _enum_name_arrays(self):
         out = []
         for e in self.b.enums.values():
-            lits = ", ".join(
-                '"' + v.replace("\\", "\\\\").replace('"', '\\"') + '"'
-                for v in e.values
-            )
+            lits = ", ".join('"' + _escape_cpp_string(v) + '"' for v in e.values)
             out.append(
                 f"  static constexpr const char *{e.name}_names[] = {{ {lits} }};"
             )