correctly format unicode contents of pw.Json columns in pw.debug.compute_and_print (#9617)

GitOrigin-RevId: 45660af95d117c35687517ae13d6c293be207d50

Author: zxqfd555

CHANGELOG.md

@@ -11,6 +11,9 @@ This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.htm
 - It is now possible to assign priorities to sources within a connector group. When a priority is set, it ensures that at any moment, the source is not lagging behind any other source with a higher priority in terms of the tracked column.
 - Connector groups can now be used in the multiprocess runs.
 
+### Changed
+- **BREAKING**: The `__str__` and `dumps` methods in `pw.Json` no longer enforce the result to be an ASCII string. This way, the behavior of `pw.debug.compute_and_print` is now consistent with other output connectors.
+
 ## [0.27.1] - 2025-12-08
 
 ### Added

python/pathway/internals/json.py

@@ -52,7 +52,7 @@ class Json:
     _value: JsonValue
 
     def __str__(self) -> str:
-        return _json.dumps(self.value)
+        return _json.dumps(self.value, ensure_ascii=False)
 
     def __float__(self) -> float:
         return float(self.value)  # type:ignore[arg-type]
@@ -96,7 +96,7 @@ def parse(value: str | bytes | bytearray) -> Json:
 
     @staticmethod
     def dumps(obj: Any) -> str:
-        return _json.dumps(obj, cls=_JsonEncoder)
+        return _json.dumps(obj, cls=_JsonEncoder, ensure_ascii=False)
 
     def as_int(self) -> int:
         """Returns Json value as an int if possible.

python/pathway/tests/test_json.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import datetime
+import json
 import pathlib
 import re
 from typing import Any, Optional
@@ -1308,3 +1309,31 @@ class Schema(pw.Schema):
         match=".*can only be applied to JSON columns, but column has type <class 'int'>.",
     ):
         input.select(result=method(pw.this.data))
+
+
+def test_json_serialize_parse():
+    base_unicode_value = "żółć"
+    base = {"data": base_unicode_value}
+
+    initial_json = pw.Json(base)
+    serialized = pw.Json.dumps(initial_json)
+    roundtrip_json = pw.Json.parse(serialized)
+    assert roundtrip_json["data"] == pw.Json(base_unicode_value)
+
+    serialized_ensure_ascii = json.dumps(base, ensure_ascii=True)
+    roundtrip_json_old_format = pw.Json.parse(serialized_ensure_ascii)
+    assert roundtrip_json_old_format["data"] == pw.Json(base_unicode_value)
+
+
+def test_json_doesnt_enforce_ascii(capsys):
+    class InputSchema(pw.Schema):
+        data: pw.Json
+
+    rows = [
+        ({"data": "szczęśliwość"},),
+    ]
+    table = pw.debug.table_from_rows(schema=InputSchema, rows=rows)
+    pw.debug.compute_and_print(table, include_id=False)
+
+    captured = capsys.readouterr()
+    assert '{"data": "szczęśliwość"}' in captured.out