fix: follow redirects for redacted audio downloads

assemblyai/__version__.py

@@ -1 +1 @@
-__version__ = "1.9.0"
+__version__ = "1.9.1"

assemblyai/prerecorded/v2/async_transcript.py

@@ -262,7 +262,9 @@ async def save_redacted_audio(self, filepath: str) -> None:
         # The redacted audio lives behind a pre-signed URL, so it is fetched
         # with a bare client rather than the API-authenticated one.
         async with httpx.AsyncClient() as client:
-            async with client.stream(method="GET", url=url) as response:
+            async with client.stream(
+                method="GET", url=url, follow_redirects=True
+            ) as response:
                 if response.status_code not in (
                     httpx.codes.OK,
                     httpx.codes.NOT_MODIFIED,

assemblyai/prerecorded/v2/transcript.py

@@ -206,7 +206,9 @@ def save_redacted_audio(self, filepath: str):
         Args:
             filepath: The path to save the redacted audio file to.
         """
-        with httpx.stream(method="GET", url=self.get_redacted_audio_url()) as response:
+        with httpx.stream(
+            method="GET", url=self.get_redacted_audio_url(), follow_redirects=True
+        ) as response:
             if response.status_code not in (httpx.codes.OK, httpx.codes.NOT_MODIFIED):
                 raise types.RedactedAudioUnavailableError(
                     f"Fetching redacted audio failed with status code {response.status_code}",

tests/unit/test_redacted_audio_redirects.py

@@ -0,0 +1,63 @@
+"""Signed redacted-audio downloads may redirect to a different host."""
+
+import pytest
+
+import assemblyai as aai
+from tests.unit import factories
+
+
+def prepare(httpx_mock, status):
+    data = factories.generate_dict_factory(
+        factories.TranscriptCompletedResponseFactory
+    )()
+    data.update(
+        redact_pii=True, redact_pii_audio=True, redact_pii_policies=["person_name"]
+    )
+    response = aai.types.TranscriptResponse(**data)
+    settings = aai.Settings(api_key="test")
+    httpx_mock.add_response(
+        url=f"{settings.base_url}/v2/transcript/{response.id}/redacted-audio",
+        json={
+            "status": "redacted_audio_ready",
+            "redacted_audio_url": "https://audio.example/redacted.wav",
+        },
+    )
+    httpx_mock.add_response(
+        url="https://audio.example/redacted.wav",
+        status_code=status,
+        headers={"location": "https://cdn.example/redacted.wav"},
+    )
+    httpx_mock.add_response(
+        url="https://cdn.example/redacted.wav", content=b"redacted audio"
+    )
+    return response, settings
+
+
+@pytest.mark.parametrize("status", [301, 302, 307, 308])
+def test_save_redacted_audio_follows_redirect(httpx_mock, tmp_path, status):
+    response, settings = prepare(httpx_mock, status)
+    transcript = aai.Transcript.from_response(
+        client=aai.Client(settings=settings), response=response
+    )
+    target = tmp_path / "redacted.wav"
+    transcript.save_redacted_audio(str(target))
+    assert target.read_bytes() == b"redacted audio"
+    assert all(
+        "authorization" not in request.headers
+        for request in httpx_mock.get_requests()[1:]
+    )
+
+
+@pytest.mark.asyncio
+@pytest.mark.parametrize("status", [301, 302, 307, 308])
+async def test_async_save_redacted_audio_follows_redirect(httpx_mock, tmp_path, status):
+    response, settings = prepare(httpx_mock, status)
+    async with aai.AsyncClient(settings=settings) as client:
+        transcript = aai.AsyncTranscript.from_response(client=client, response=response)
+        target = tmp_path / "redacted.wav"
+        await transcript.save_redacted_audio(str(target))
+    assert target.read_bytes() == b"redacted audio"
+    assert all(
+        "authorization" not in request.headers
+        for request in httpx_mock.get_requests()[1:]
+    )