fix: regression to artifact handling in util/taskcluster.py

Author: ahal

src/taskgraph/util/taskcluster.py

@@ -6,6 +6,7 @@
 import copy
 import datetime
 import functools
+import io
 import logging
 import os
 from typing import Dict, List, Union
@@ -70,12 +71,30 @@ def get_taskcluster_client(service: str):
 
 
 def _handle_artifact(path, response):
+    # When taskcluster client returns non-JSON responses, it wraps them in {"response": <Response>}
+    # When it returns JSON, it's already parsed
+    if isinstance(response, dict) and "response" in response:
+        response = response["response"]
+
     if path.endswith(".json"):
+        # If response is already a dict (parsed JSON), return it directly
+        if isinstance(response, dict):
+            return response
         return response.json()
 
     if path.endswith(".yml"):
         return yaml.load_stream(response.content)
 
+    # For non-JSON/YAML content, return raw response content
+    if getattr(response, "raw", None):
+        # Try to read from raw, but if it's empty, fall back to content
+        content = response.raw.read()
+        if not content and hasattr(response, "content"):
+            content = response.content
+        return io.BytesIO(content)
+    elif hasattr(response, "content"):
+        return io.BytesIO(response.content)
+
     response.raw.read = functools.partial(response.raw.read, decode_content=True)
     return response.raw
 

test/test_util_taskcluster.py

@@ -84,7 +84,6 @@ def test_get_artifact_url(monkeypatch):
     assert tc.get_artifact_url(task_id, path) == expected_proxy
 
 
-@pytest.mark.xfail
 def test_get_artifact(responses, root_url):
     tid = "abc123"
     tc.get_taskcluster_client.cache_clear()
@@ -194,7 +193,6 @@ def test_find_task_id_batched(responses, root_url):
     assert result == {"index.abc": "abc", "index.def": "def"}
 
 
-@pytest.mark.xfail
 def test_get_artifact_from_index(responses, root_url):
     index = "foo"
     path = "file.txt"