Skip to content

Commit 60d37b6

Browse files
alexander-alderman-webbalexander-alderman-webb
committed
test(huggingface_hub): Only check relevant data with ApproxDict
1 parent f4d4af7 commit 60d37b6

1 file changed

Lines changed: 15 additions & 100 deletions

File tree

‎tests/integrations/huggingface_hub/test_huggingface_hub.py‎

Lines changed: 15 additions & 100 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,5 @@
11
import re
22
from typing import TYPE_CHECKING
3-
from unittest import mock
43

54
import pytest
65
import responses
@@ -10,6 +9,7 @@
109
from sentry_sdk.consts import SPANDATA
1110
from sentry_sdk.integrations.huggingface_hub import HuggingfaceHubIntegration
1211
from sentry_sdk.utils import package_version, safe_serialize
12+
from tests.conftest import ApproxDict
1313

1414
try:
1515
from huggingface_hub.utils._errors import HfHubHTTPError
@@ -522,19 +522,8 @@ def test_text_generation(
522522
"gen_ai.response.finish_reasons": "length",
523523
"gen_ai.response.streaming": False,
524524
"gen_ai.usage.total_tokens": 10,
525-
"process.runtime.name": mock.ANY,
526-
"process.runtime.version": mock.ANY,
527-
"sentry.environment": "production",
528525
"sentry.op": "gen_ai.text_completion",
529526
"sentry.origin": "auto.ai.huggingface_hub",
530-
"sentry.release": mock.ANY,
531-
"sentry.sdk.name": "sentry.python",
532-
"sentry.sdk.version": mock.ANY,
533-
"sentry.segment.id": mock.ANY,
534-
"sentry.segment.name": "test",
535-
"server.address": mock.ANY,
536-
"thread.id": mock.ANY,
537-
"thread.name": mock.ANY,
538527
}
539528

540529
if send_default_pii and include_prompts:
@@ -547,7 +536,7 @@ def test_text_generation(
547536
assert "gen_ai.request.messages" not in expected_data
548537
assert "gen_ai.response.text" not in expected_data
549538

550-
assert span["attributes"] == expected_data
539+
assert span["attributes"] == ApproxDict(expected_data)
551540

552541
# text generation does not set the response model
553542
assert "gen_ai.response.model" not in span["attributes"]
@@ -585,8 +574,6 @@ def test_text_generation(
585574
"gen_ai.response.finish_reasons": "length",
586575
"gen_ai.response.streaming": False,
587576
"gen_ai.usage.total_tokens": 10,
588-
"thread.id": mock.ANY,
589-
"thread.name": mock.ANY,
590577
}
591578

592579
if send_default_pii and include_prompts:
@@ -599,7 +586,7 @@ def test_text_generation(
599586
assert "gen_ai.request.messages" not in expected_data
600587
assert "gen_ai.response.text" not in expected_data
601588

602-
assert span["data"] == expected_data
589+
assert span["data"] == ApproxDict(expected_data)
603590

604591
# text generation does not set the response model
605592
assert "gen_ai.response.model" not in span["data"]
@@ -661,19 +648,9 @@ def test_text_generation_streaming(
661648
"gen_ai.response.finish_reasons": "length",
662649
"gen_ai.response.streaming": True,
663650
"gen_ai.usage.total_tokens": 10,
664-
"process.runtime.name": mock.ANY,
665-
"process.runtime.version": mock.ANY,
666651
"sentry.environment": "production",
667652
"sentry.op": "gen_ai.text_completion",
668653
"sentry.origin": "auto.ai.huggingface_hub",
669-
"sentry.release": mock.ANY,
670-
"sentry.sdk.name": "sentry.python",
671-
"sentry.sdk.version": mock.ANY,
672-
"sentry.segment.id": mock.ANY,
673-
"sentry.segment.name": "test",
674-
"server.address": mock.ANY,
675-
"thread.id": mock.ANY,
676-
"thread.name": mock.ANY,
677654
}
678655

679656
if send_default_pii and include_prompts:
@@ -684,7 +661,7 @@ def test_text_generation_streaming(
684661
assert "gen_ai.request.messages" not in expected_data
685662
assert "gen_ai.response.text" not in expected_data
686663

687-
assert span["attributes"] == expected_data
664+
assert span["attributes"] == ApproxDict(expected_data)
688665

689666
# text generation does not set the response model
690667
assert "gen_ai.response.model" not in span["attributes"]
@@ -723,8 +700,6 @@ def test_text_generation_streaming(
723700
"gen_ai.response.finish_reasons": "length",
724701
"gen_ai.response.streaming": True,
725702
"gen_ai.usage.total_tokens": 10,
726-
"thread.id": mock.ANY,
727-
"thread.name": mock.ANY,
728703
}
729704

730705
if send_default_pii and include_prompts:
@@ -735,7 +710,7 @@ def test_text_generation_streaming(
735710
assert "gen_ai.request.messages" not in expected_data
736711
assert "gen_ai.response.text" not in expected_data
737712

738-
assert span["data"] == expected_data
713+
assert span["data"] == ApproxDict(expected_data)
739714

740715
# text generation does not set the response model
741716
assert "gen_ai.response.model" not in span["data"]
@@ -804,19 +779,9 @@ def test_chat_completion(
804779
"gen_ai.usage.input_tokens": 10,
805780
"gen_ai.usage.output_tokens": 8,
806781
"gen_ai.usage.total_tokens": 18,
807-
"process.runtime.name": mock.ANY,
808-
"process.runtime.version": mock.ANY,
809782
"sentry.environment": "production",
810783
"sentry.op": "gen_ai.chat",
811784
"sentry.origin": "auto.ai.huggingface_hub",
812-
"sentry.release": mock.ANY,
813-
"sentry.sdk.name": "sentry.python",
814-
"sentry.sdk.version": mock.ANY,
815-
"sentry.segment.id": mock.ANY,
816-
"sentry.segment.name": "test",
817-
"server.address": mock.ANY,
818-
"thread.id": mock.ANY,
819-
"thread.name": mock.ANY,
820785
}
821786

822787
if send_default_pii and include_prompts:
@@ -837,7 +802,7 @@ def test_chat_completion(
837802
assert "gen_ai.request.messages" not in expected_data
838803
assert "gen_ai.response.text" not in expected_data
839804

840-
assert span["attributes"] == expected_data
805+
assert span["attributes"] == ApproxDict(expected_data)
841806
else:
842807
events = capture_events()
843808

@@ -874,8 +839,6 @@ def test_chat_completion(
874839
"gen_ai.usage.input_tokens": 10,
875840
"gen_ai.usage.output_tokens": 8,
876841
"gen_ai.usage.total_tokens": 18,
877-
"thread.id": mock.ANY,
878-
"thread.name": mock.ANY,
879842
}
880843

881844
if send_default_pii and include_prompts:
@@ -890,7 +853,7 @@ def test_chat_completion(
890853
assert "gen_ai.request.messages" not in expected_data
891854
assert "gen_ai.response.text" not in expected_data
892855

893-
assert span["data"] == expected_data
856+
assert span["data"] == ApproxDict(expected_data)
894857

895858

896859
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
@@ -955,19 +918,9 @@ def test_chat_completion_streaming(
955918
"gen_ai.response.finish_reasons": "stop",
956919
"gen_ai.response.model": "test-model-123",
957920
"gen_ai.response.streaming": True,
958-
"process.runtime.name": mock.ANY,
959-
"process.runtime.version": mock.ANY,
960921
"sentry.environment": "production",
961922
"sentry.op": "gen_ai.chat",
962923
"sentry.origin": "auto.ai.huggingface_hub",
963-
"sentry.release": mock.ANY,
964-
"sentry.sdk.name": "sentry.python",
965-
"sentry.sdk.version": mock.ANY,
966-
"sentry.segment.id": mock.ANY,
967-
"sentry.segment.name": "test",
968-
"server.address": mock.ANY,
969-
"thread.id": mock.ANY,
970-
"thread.name": mock.ANY,
971924
}
972925
# usage is not available in older versions of the library
973926
if HF_VERSION and HF_VERSION >= (0, 26, 0):
@@ -991,7 +944,7 @@ def test_chat_completion_streaming(
991944
assert "gen_ai.request.messages" not in expected_data
992945
assert "gen_ai.response.text" not in expected_data
993946

994-
assert span["attributes"] == expected_data
947+
assert span["attributes"] == ApproxDict(expected_data)
995948
else:
996949
events = capture_events()
997950

@@ -1027,8 +980,6 @@ def test_chat_completion_streaming(
1027980
"gen_ai.response.finish_reasons": "stop",
1028981
"gen_ai.response.model": "test-model-123",
1029982
"gen_ai.response.streaming": True,
1030-
"thread.id": mock.ANY,
1031-
"thread.name": mock.ANY,
1032983
}
1033984
# usage is not available in older versions of the library
1034985
if HF_VERSION and HF_VERSION >= (0, 26, 0):
@@ -1046,7 +997,7 @@ def test_chat_completion_streaming(
1046997
assert "gen_ai.request.messages" not in expected_data
1047998
assert "gen_ai.response.text" not in expected_data
1048999

1049-
assert span["data"] == expected_data
1000+
assert span["data"] == ApproxDict(expected_data)
10501001

10511002

10521003
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
@@ -1104,21 +1055,11 @@ def test_chat_completion_api_error(
11041055
expected_data = {
11051056
"gen_ai.operation.name": "chat",
11061057
"gen_ai.request.model": "test-model",
1107-
"process.runtime.name": mock.ANY,
1108-
"process.runtime.version": mock.ANY,
11091058
"sentry.environment": "production",
11101059
"sentry.op": "gen_ai.chat",
11111060
"sentry.origin": "auto.ai.huggingface_hub",
1112-
"sentry.release": mock.ANY,
1113-
"sentry.sdk.name": "sentry.python",
1114-
"sentry.sdk.version": mock.ANY,
1115-
"sentry.segment.id": mock.ANY,
1116-
"sentry.segment.name": "test",
1117-
"server.address": mock.ANY,
1118-
"thread.id": mock.ANY,
1119-
"thread.name": mock.ANY,
11201061
}
1121-
assert span["attributes"] == expected_data
1062+
assert span["attributes"] == ApproxDict(expected_data)
11221063
else:
11231064
events = capture_events()
11241065

@@ -1160,10 +1101,8 @@ def test_chat_completion_api_error(
11601101
expected_data = {
11611102
"gen_ai.operation.name": "chat",
11621103
"gen_ai.request.model": "test-model",
1163-
"thread.id": mock.ANY,
1164-
"thread.name": mock.ANY,
11651104
}
1166-
assert span["data"] == expected_data
1105+
assert span["data"] == ApproxDict(expected_data)
11671106

11681107

11691108
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
@@ -1305,19 +1244,9 @@ def test_chat_completion_with_tools(
13051244
"gen_ai.usage.input_tokens": 10,
13061245
"gen_ai.usage.output_tokens": 8,
13071246
"gen_ai.usage.total_tokens": 18,
1308-
"process.runtime.name": mock.ANY,
1309-
"process.runtime.version": mock.ANY,
13101247
"sentry.environment": "production",
13111248
"sentry.op": "gen_ai.chat",
13121249
"sentry.origin": "auto.ai.huggingface_hub",
1313-
"sentry.release": mock.ANY,
1314-
"sentry.sdk.name": "sentry.python",
1315-
"sentry.sdk.version": mock.ANY,
1316-
"sentry.segment.id": mock.ANY,
1317-
"sentry.segment.name": "test",
1318-
"server.address": mock.ANY,
1319-
"thread.id": mock.ANY,
1320-
"thread.name": mock.ANY,
13211250
}
13221251

13231252
if send_default_pii and include_prompts:
@@ -1333,7 +1262,7 @@ def test_chat_completion_with_tools(
13331262
assert "gen_ai.response.text" not in expected_data
13341263
assert "gen_ai.response.tool_calls" not in expected_data
13351264

1336-
assert span["attributes"] == expected_data
1265+
assert span["attributes"] == ApproxDict(expected_data)
13371266
else:
13381267
events = capture_events()
13391268

@@ -1371,8 +1300,6 @@ def test_chat_completion_with_tools(
13711300
"gen_ai.usage.input_tokens": 10,
13721301
"gen_ai.usage.output_tokens": 8,
13731302
"gen_ai.usage.total_tokens": 18,
1374-
"thread.id": mock.ANY,
1375-
"thread.name": mock.ANY,
13761303
}
13771304

13781305
if send_default_pii and include_prompts:
@@ -1388,7 +1315,7 @@ def test_chat_completion_with_tools(
13881315
assert "gen_ai.response.text" not in expected_data
13891316
assert "gen_ai.response.tool_calls" not in expected_data
13901317

1391-
assert span["data"] == expected_data
1318+
assert span["data"] == ApproxDict(expected_data)
13921319

13931320

13941321
@pytest.mark.parametrize("stream_gen_ai_spans", [True, False])
@@ -1467,19 +1394,9 @@ def test_chat_completion_streaming_with_tools(
14671394
"gen_ai.response.finish_reasons": "tool_calls",
14681395
"gen_ai.response.model": "test-model-123",
14691396
"gen_ai.response.streaming": True,
1470-
"process.runtime.name": mock.ANY,
1471-
"process.runtime.version": mock.ANY,
14721397
"sentry.environment": "production",
14731398
"sentry.op": "gen_ai.chat",
14741399
"sentry.origin": "auto.ai.huggingface_hub",
1475-
"sentry.release": mock.ANY,
1476-
"sentry.sdk.name": "sentry.python",
1477-
"sentry.sdk.version": mock.ANY,
1478-
"sentry.segment.id": mock.ANY,
1479-
"sentry.segment.name": "test",
1480-
"server.address": mock.ANY,
1481-
"thread.id": mock.ANY,
1482-
"thread.name": mock.ANY,
14831400
}
14841401

14851402
if HF_VERSION and HF_VERSION >= (0, 26, 0):
@@ -1501,7 +1418,7 @@ def test_chat_completion_streaming_with_tools(
15011418
assert "gen_ai.response.text" not in expected_data
15021419
assert "gen_ai.response.tool_calls" not in expected_data
15031420

1504-
assert span["attributes"] == expected_data
1421+
assert span["attributes"] == ApproxDict(expected_data)
15051422
else:
15061423
events = capture_events()
15071424

@@ -1542,8 +1459,6 @@ def test_chat_completion_streaming_with_tools(
15421459
"gen_ai.response.finish_reasons": "tool_calls",
15431460
"gen_ai.response.model": "test-model-123",
15441461
"gen_ai.response.streaming": True,
1545-
"thread.id": mock.ANY,
1546-
"thread.name": mock.ANY,
15471462
}
15481463

15491464
if HF_VERSION and HF_VERSION >= (0, 26, 0):
@@ -1565,7 +1480,7 @@ def test_chat_completion_streaming_with_tools(
15651480
assert "gen_ai.response.text" not in expected_data
15661481
assert "gen_ai.response.tool_calls" not in expected_data
15671482

1568-
assert span["data"] == expected_data
1483+
assert span["data"] == ApproxDict(expected_data)
15691484

15701485

15711486
DATA_COLLECTION_TOOLS = [

0 commit comments

Comments
 (0)