Compare commits

...

7 Commits

Author SHA1 Message Date
Yuhong Sun
5dfc226fe7 Backport Test 7 (#2971) 2024-10-28 05:55:57 +00:00
Yuhong Sun
94b0973ad1 Backport Test 5 (#2969) 2024-10-28 03:07:42 +00:00
Yuhong Sun
813ff48217 Backport Test 4 (#2968) 2024-10-28 02:41:19 +00:00
Yuhong Sun
ef92ece8bd Backport Test (#2958) 2024-10-27 23:06:51 +00:00
Yuhong Sun
8f2a754e4f Backport Test (#2957) 2024-10-27 22:55:21 +00:00
Yuhong Sun
17fc1ad58f Harmless Backport Test (#2950) 2024-10-27 21:49:26 +00:00
Yuhong Sun
fc2bbd1bbb Backport Test (#2940) 2024-10-26 21:27:01 +00:00
6 changed files with 19 additions and 2 deletions

View File

@@ -80,6 +80,7 @@ from danswer.utils.telemetry import RecordType
from danswer.utils.variable_functionality import fetch_versioned_implementation
from shared_configs.configs import current_tenant_id
logger = setup_logger()

View File

@@ -24,6 +24,10 @@ from danswer.connectors.models import Document
from danswer.connectors.models import Section
from danswer.utils.logger import setup_logger
logger = setup_logger()
# List of directories/Files to exclude
exclude_patterns = [
"logs",
@@ -31,7 +35,6 @@ exclude_patterns = [
".gitlab/",
".pre-commit-config.yaml",
]
logger = setup_logger()
def _batch_gitlab_objects(

View File

@@ -22,6 +22,7 @@ from danswer.connectors.models import Document
from danswer.connectors.models import Section
from danswer.utils.logger import setup_logger
logger = setup_logger()
@@ -230,5 +231,7 @@ if __name__ == "__main__":
print("All docs", all_docs)
current = datetime.datetime.now().timestamp()
one_day_ago = current - 30 * 24 * 60 * 60 # 30 days
latest_docs = list(test_connector.poll_source(one_day_ago, current))
print("Latest docs", latest_docs)

View File

@@ -20,10 +20,13 @@ from danswer.connectors.models import Document
from danswer.connectors.models import Section
from danswer.utils.logger import setup_logger
logger = setup_logger()
# Fairly generous retry because it's not understood why occasionally GraphQL requests fail even with timeout > 1 min
SLAB_GRAPHQL_MAX_TRIES = 10
SLAB_API_URL = "https://api.slab.com/v1/graphql"
logger = setup_logger()
def run_graphql_request(

View File

@@ -429,6 +429,7 @@ if __name__ == "__main__":
current = time.time()
one_day_ago = current - 24 * 60 * 60 # 1 day
document_batches = connector.poll_source(one_day_ago, current)
print(next(document_batches))

View File

@@ -9,12 +9,15 @@ from pytest_mock import MockFixture
from danswer.connectors.mediawiki import wiki
# These tests are disabled for now
@pytest.fixture
def site() -> pywikibot.Site:
return pywikibot.Site("en", "wikipedia")
@pytest.mark.skip(reason="Test disabled")
def test_pywikibot_timestamp_to_utc_datetime() -> None:
timestamp_without_tzinfo = pywikibot.Timestamp(2023, 12, 27, 15, 38, 49)
timestamp_min_timezone = timestamp_without_tzinfo.astimezone(datetime.timezone.min)
@@ -80,6 +83,7 @@ class MockPage(pywikibot.Page):
)
@pytest.mark.skip(reason="Test disabled")
def test_get_doc_from_page(site: pywikibot.Site) -> None:
test_page = MockPage(site, "Test Page", _has_categories=True)
doc = wiki.get_doc_from_page(test_page, site, wiki.DocumentSource.MEDIAWIKI)
@@ -103,6 +107,7 @@ def test_get_doc_from_page(site: pywikibot.Site) -> None:
assert doc.id == f"MEDIAWIKI_{test_page.pageid}_{test_page.full_url()}"
@pytest.mark.skip(reason="Test disabled")
def test_mediawiki_connector_recurse_depth() -> None:
"""Test that the recurse_depth parameter is parsed correctly.
@@ -132,6 +137,7 @@ def test_mediawiki_connector_recurse_depth() -> None:
assert connector.recurse_depth == recurse_depth
@pytest.mark.skip(reason="Test disabled")
def test_load_from_state_calls_poll_source_with_nones(mocker: MockFixture) -> None:
connector = wiki.MediaWikiConnector("wikipedia.org", [], [], 0, "test")
poll_source = mocker.patch.object(connector, "poll_source")