From 715a22ff71097300bdaadb53fdf237b600c161eb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tom=C3=A1s=20Far=C3=ADas=20Santana?= <tomas@tomasfarias.dev>
Date: Tue, 31 Dec 2024 11:25:10 +0100
Subject: [PATCH 1/3] fix(batch-exports): Remove zero unicode from error
 message

---
 posthog/temporal/batch_exports/batch_exports.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/posthog/temporal/batch_exports/batch_exports.py b/posthog/temporal/batch_exports/batch_exports.py
index 1b8faf19e482e..8d85d1174fb74 100644
--- a/posthog/temporal/batch_exports/batch_exports.py
+++ b/posthog/temporal/batch_exports/batch_exports.py
@@ -902,6 +902,14 @@ async def finish_batch_export_run(inputs: FinishBatchExportRunInputs) -> None:
         for key, value in dataclasses.asdict(inputs).items()
         if key not in not_model_params and value is not None
     }
+
+    latest_error = update_params.get("latest_error", None)
+    if latest_error is not None and isinstance(latest_error, str):
+        # NUL (\x00) bytes are not allowed in PostgreSQL, so we replace them from
+        # the free text field `latest_error`.
+        latest_error = latest_error.replace("\x00", "")
+        update_params["latest_error"] = latest_error
+
     batch_export_run = await database_sync_to_async(update_batch_export_run)(
         run_id=uuid.UUID(inputs.id),
         finished_at=dt.datetime.now(dt.UTC),

From 976a6ca56b8daa352ac760f9a3d73e306e3bf999 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tom=C3=A1s=20Far=C3=ADas=20Santana?= <tomas@tomasfarias.dev>
Date: Tue, 31 Dec 2024 11:32:28 +0100
Subject: [PATCH 2/3] test(batch-exports): Add test for nul bytes in
 finish_batch_export_run

---
 .../temporal/batch_exports/batch_exports.py   |  2 +-
 .../tests/batch_exports/test_run_updates.py   | 35 +++++++++++++++++++
 2 files changed, 36 insertions(+), 1 deletion(-)

diff --git a/posthog/temporal/batch_exports/batch_exports.py b/posthog/temporal/batch_exports/batch_exports.py
index 8d85d1174fb74..e9f9288688dc2 100644
--- a/posthog/temporal/batch_exports/batch_exports.py
+++ b/posthog/temporal/batch_exports/batch_exports.py
@@ -905,7 +905,7 @@ async def finish_batch_export_run(inputs: FinishBatchExportRunInputs) -> None:
 
     latest_error = update_params.get("latest_error", None)
     if latest_error is not None and isinstance(latest_error, str):
-        # NUL (\x00) bytes are not allowed in PostgreSQL, so we replace them from
+        # NUL (\x00) bytes are not allowed in PostgreSQL, so we replace them in
         # the free text field `latest_error`.
         latest_error = latest_error.replace("\x00", "")
         update_params["latest_error"] = latest_error
diff --git a/posthog/temporal/tests/batch_exports/test_run_updates.py b/posthog/temporal/tests/batch_exports/test_run_updates.py
index 649585f52836b..b719ca455f7db 100644
--- a/posthog/temporal/tests/batch_exports/test_run_updates.py
+++ b/posthog/temporal/tests/batch_exports/test_run_updates.py
@@ -214,3 +214,38 @@ async def test_finish_batch_export_run_never_pauses_with_small_check_window(acti
     await sync_to_async(batch_export.refresh_from_db)()
 
     assert batch_export.paused is False
+
+
+@pytest.mark.django_db(transaction=True)
+@pytest.mark.asyncio
+async def test_finish_batch_export_run_handles_nul_bytes(activity_environment, team, batch_export):
+    """Test if 'finish_batch_export_run' will not fail in the prescence of a NUL byte."""
+    start = dt.datetime(2023, 4, 24, tzinfo=dt.UTC)
+    end = dt.datetime(2023, 4, 25, tzinfo=dt.UTC)
+
+    inputs = StartBatchExportRunInputs(
+        team_id=team.id,
+        batch_export_id=str(batch_export.id),
+        data_interval_start=start.isoformat(),
+        data_interval_end=end.isoformat(),
+    )
+
+    batch_export_id = str(batch_export.id)
+
+    run_id = await activity_environment.run(start_batch_export_run, inputs)
+
+    finish_inputs = FinishBatchExportRunInputs(
+        id=str(run_id),
+        batch_export_id=batch_export_id,
+        status=BatchExportRun.Status.FAILED,
+        team_id=inputs.team_id,
+        latest_error="Oh No a NUL byte: \x00!",
+    )
+
+    await activity_environment.run(finish_batch_export_run, finish_inputs)
+
+    runs = BatchExportRun.objects.filter(id=run_id)
+    run = await sync_to_async(runs.first)()  # type:ignore
+    assert run is not None
+    assert run.status == "Failed"
+    assert run.latest_error == "Oh No a NUL byte: !"

From e89663d1b72072f0dd2d7e1436fe1d921f6091ee Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tom=C3=A1s=20Far=C3=ADas=20Santana?= <tomas@tomasfarias.dev>
Date: Tue, 31 Dec 2024 11:39:27 +0100
Subject: [PATCH 3/3] chore(typing): Remove unused comment

---
 posthog/temporal/tests/batch_exports/test_run_updates.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/posthog/temporal/tests/batch_exports/test_run_updates.py b/posthog/temporal/tests/batch_exports/test_run_updates.py
index b719ca455f7db..904c79d6dd5cd 100644
--- a/posthog/temporal/tests/batch_exports/test_run_updates.py
+++ b/posthog/temporal/tests/batch_exports/test_run_updates.py
@@ -245,7 +245,7 @@ async def test_finish_batch_export_run_handles_nul_bytes(activity_environment, t
     await activity_environment.run(finish_batch_export_run, finish_inputs)
 
     runs = BatchExportRun.objects.filter(id=run_id)
-    run = await sync_to_async(runs.first)()  # type:ignore
+    run = await sync_to_async(runs.first)()
     assert run is not None
     assert run.status == "Failed"
     assert run.latest_error == "Oh No a NUL byte: !"