Compare commits

..

1 Commits

Author SHA1 Message Date
pablonyx
575cfa5224 fix 2025-03-27 11:16:58 -07:00
20 changed files with 52 additions and 266 deletions

View File

@@ -9,10 +9,6 @@ on:
- cron: "0 16 * * *"
env:
# AWS
AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
# Confluence
CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
CONFLUENCE_TEST_SPACE: ${{ secrets.CONFLUENCE_TEST_SPACE }}

View File

@@ -25,10 +25,6 @@ SAML_CONF_DIR = os.environ.get("SAML_CONF_DIR") or "/app/ee/onyx/configs/saml_co
#####
# Auto Permission Sync
#####
DEFAULT_PERMISSION_DOC_SYNC_FREQUENCY = int(
os.environ.get("DEFAULT_PERMISSION_DOC_SYNC_FREQUENCY") or 5 * 60
)
# In seconds, default is 5 minutes
CONFLUENCE_PERMISSION_GROUP_SYNC_FREQUENCY = int(
os.environ.get("CONFLUENCE_PERMISSION_GROUP_SYNC_FREQUENCY") or 5 * 60
@@ -43,7 +39,6 @@ CONFLUENCE_ANONYMOUS_ACCESS_IS_PUBLIC = (
CONFLUENCE_PERMISSION_DOC_SYNC_FREQUENCY = int(
os.environ.get("CONFLUENCE_PERMISSION_DOC_SYNC_FREQUENCY") or 5 * 60
)
NUM_PERMISSION_WORKERS = int(os.environ.get("NUM_PERMISSION_WORKERS") or 2)
@@ -77,13 +72,6 @@ OAUTH_GOOGLE_DRIVE_CLIENT_SECRET = os.environ.get(
"OAUTH_GOOGLE_DRIVE_CLIENT_SECRET", ""
)
GOOGLE_DRIVE_PERMISSION_GROUP_SYNC_FREQUENCY = int(
os.environ.get("GOOGLE_DRIVE_PERMISSION_GROUP_SYNC_FREQUENCY") or 5 * 60
)
SLACK_PERMISSION_DOC_SYNC_FREQUENCY = int(
os.environ.get("SLACK_PERMISSION_DOC_SYNC_FREQUENCY") or 5 * 60
)
# The posthog client does not accept empty API keys or hosts however it fails silently
# when the capture is called. These defaults prevent Posthog issues from breaking the Onyx app

View File

@@ -3,8 +3,6 @@ from collections.abc import Generator
from ee.onyx.configs.app_configs import CONFLUENCE_PERMISSION_DOC_SYNC_FREQUENCY
from ee.onyx.configs.app_configs import CONFLUENCE_PERMISSION_GROUP_SYNC_FREQUENCY
from ee.onyx.configs.app_configs import GOOGLE_DRIVE_PERMISSION_GROUP_SYNC_FREQUENCY
from ee.onyx.configs.app_configs import SLACK_PERMISSION_DOC_SYNC_FREQUENCY
from ee.onyx.db.external_perm import ExternalUserGroup
from ee.onyx.external_permissions.confluence.doc_sync import confluence_doc_sync
from ee.onyx.external_permissions.confluence.group_sync import confluence_group_sync
@@ -68,13 +66,13 @@ GROUP_PERMISSIONS_IS_CC_PAIR_AGNOSTIC: set[DocumentSource] = {
DOC_PERMISSION_SYNC_PERIODS: dict[DocumentSource, int] = {
# Polling is not supported so we fetch all doc permissions every 5 minutes
DocumentSource.CONFLUENCE: CONFLUENCE_PERMISSION_DOC_SYNC_FREQUENCY,
DocumentSource.SLACK: SLACK_PERMISSION_DOC_SYNC_FREQUENCY,
DocumentSource.SLACK: 5 * 60,
}
# If nothing is specified here, we run the doc_sync every time the celery beat runs
EXTERNAL_GROUP_SYNC_PERIODS: dict[DocumentSource, int] = {
# Polling is not supported so we fetch all group permissions every 30 minutes
DocumentSource.GOOGLE_DRIVE: GOOGLE_DRIVE_PERMISSION_GROUP_SYNC_FREQUENCY,
DocumentSource.GOOGLE_DRIVE: 5 * 60,
DocumentSource.CONFLUENCE: CONFLUENCE_PERMISSION_GROUP_SYNC_FREQUENCY,
}

View File

@@ -505,8 +505,11 @@ async def setup_tenant(tenant_id: str) -> None:
try:
token = CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
# Run Alembic migrations
await asyncio.to_thread(run_alembic_migrations, tenant_id)
# Run Alembic migrations in a way that isolates it from the current event loop
# Create a new event loop for this synchronous operation
loop = asyncio.get_event_loop()
# Use run_in_executor which properly isolates the thread execution
await loop.run_in_executor(None, lambda: run_alembic_migrations(tenant_id))
# Configure the tenant with default settings
with get_session_with_tenant(tenant_id=tenant_id) as db_session:

View File

@@ -1,5 +1,6 @@
from datetime import timedelta
from typing import Any
from typing import cast
from celery import Celery
from celery import signals
@@ -9,10 +10,12 @@ from celery.utils.log import get_task_logger
import onyx.background.celery.apps.app_base as app_base
from onyx.background.celery.tasks.beat_schedule import CLOUD_BEAT_MULTIPLIER_DEFAULT
from onyx.configs.constants import ONYX_CLOUD_REDIS_RUNTIME
from onyx.configs.constants import ONYX_CLOUD_TENANT_ID
from onyx.configs.constants import POSTGRES_CELERY_BEAT_APP_NAME
from onyx.db.engine import get_all_tenant_ids
from onyx.db.engine import SqlEngine
from onyx.server.runtime.onyx_runtime import OnyxRuntime
from onyx.redis.redis_pool import get_redis_replica_client
from onyx.utils.variable_functionality import fetch_versioned_implementation
from shared_configs.configs import IGNORED_SYNCING_TENANT_LIST
from shared_configs.configs import MULTI_TENANT
@@ -138,6 +141,8 @@ class DynamicTenantScheduler(PersistentScheduler):
"""Only updates the actual beat schedule on the celery app when it changes"""
do_update = False
r = get_redis_replica_client(tenant_id=ONYX_CLOUD_TENANT_ID)
task_logger.debug("_try_updating_schedule starting")
tenant_ids = get_all_tenant_ids()
@@ -147,7 +152,16 @@ class DynamicTenantScheduler(PersistentScheduler):
current_schedule = self.schedule.items()
# get potential new state
beat_multiplier = OnyxRuntime.get_beat_multiplier()
beat_multiplier = CLOUD_BEAT_MULTIPLIER_DEFAULT
beat_multiplier_raw = r.get(f"{ONYX_CLOUD_REDIS_RUNTIME}:beat_multiplier")
if beat_multiplier_raw is not None:
try:
beat_multiplier_bytes = cast(bytes, beat_multiplier_raw)
beat_multiplier = float(beat_multiplier_bytes.decode())
except ValueError:
task_logger.error(
f"Invalid beat_multiplier value: {beat_multiplier_raw}"
)
new_schedule = self._generate_schedule(tenant_ids, beat_multiplier)

View File

@@ -21,7 +21,6 @@ BEAT_EXPIRES_DEFAULT = 15 * 60 # 15 minutes (in seconds)
# we have a better implementation (backpressure, etc)
# Note that DynamicTenantScheduler can adjust the runtime value for this via Redis
CLOUD_BEAT_MULTIPLIER_DEFAULT = 8.0
CLOUD_DOC_PERMISSION_SYNC_MULTIPLIER_DEFAULT = 1.0
# tasks that run in either self-hosted on cloud
beat_task_templates: list[dict] = []

View File

@@ -17,7 +17,6 @@ from redis.exceptions import LockError
from redis.lock import Lock as RedisLock
from sqlalchemy.orm import Session
from ee.onyx.configs.app_configs import DEFAULT_PERMISSION_DOC_SYNC_FREQUENCY
from ee.onyx.db.connector_credential_pair import get_all_auto_sync_cc_pairs
from ee.onyx.db.document import upsert_document_external_perms
from ee.onyx.external_permissions.sync_params import DOC_PERMISSION_SYNC_PERIODS
@@ -64,7 +63,6 @@ from onyx.redis.redis_connector_doc_perm_sync import RedisConnectorPermissionSyn
from onyx.redis.redis_pool import get_redis_client
from onyx.redis.redis_pool import get_redis_replica_client
from onyx.redis.redis_pool import redis_lock_dump
from onyx.server.runtime.onyx_runtime import OnyxRuntime
from onyx.server.utils import make_short_id
from onyx.utils.logger import doc_permission_sync_ctx
from onyx.utils.logger import format_error_for_logging
@@ -108,10 +106,9 @@ def _is_external_doc_permissions_sync_due(cc_pair: ConnectorCredentialPair) -> b
source_sync_period = DOC_PERMISSION_SYNC_PERIODS.get(cc_pair.connector.source)
# If RESTRICTED_FETCH_PERIOD[source] is None, we always run the sync.
if not source_sync_period:
source_sync_period = DEFAULT_PERMISSION_DOC_SYNC_FREQUENCY
source_sync_period *= int(OnyxRuntime.get_doc_permission_sync_multiplier())
return True
# If the last sync is greater than the full fetch period, we run the sync
next_sync = last_perm_sync + timedelta(seconds=source_sync_period)

View File

@@ -72,7 +72,6 @@ from onyx.redis.redis_pool import get_redis_replica_client
from onyx.redis.redis_pool import redis_lock_dump
from onyx.redis.redis_pool import SCAN_ITER_COUNT_DEFAULT
from onyx.redis.redis_utils import is_fence
from onyx.server.runtime.onyx_runtime import OnyxRuntime
from onyx.utils.logger import setup_logger
from onyx.utils.variable_functionality import global_version
from shared_configs.configs import INDEXING_MODEL_SERVER_HOST
@@ -402,11 +401,7 @@ def check_for_indexing(self: Task, *, tenant_id: str) -> int | None:
logger.warning(f"Adding {key_bytes} to the lookup table.")
redis_client.sadd(OnyxRedisConstants.ACTIVE_FENCES, key_bytes)
redis_client.set(
OnyxRedisSignals.BLOCK_BUILD_FENCE_LOOKUP_TABLE,
1,
ex=OnyxRuntime.get_build_fence_lookup_table_interval(),
)
redis_client.set(OnyxRedisSignals.BLOCK_BUILD_FENCE_LOOKUP_TABLE, 1, ex=300)
# 1/3: KICKOFF

View File

@@ -382,7 +382,6 @@ ONYX_CLOUD_TENANT_ID = "cloud"
# the redis namespace for runtime variables
ONYX_CLOUD_REDIS_RUNTIME = "runtime"
CLOUD_BUILD_FENCE_LOOKUP_TABLE_INTERVAL_DEFAULT = 600
class OnyxCeleryTask:

View File

@@ -87,7 +87,7 @@ class BlobStorageConnector(LoadConnector, PollConnector):
credentials.get(key)
for key in ["aws_access_key_id", "aws_secret_access_key"]
):
raise ConnectorMissingCredentialError("Amazon S3")
raise ConnectorMissingCredentialError("Google Cloud Storage")
session = boto3.Session(
aws_access_key_id=credentials["aws_access_key_id"],

View File

@@ -28,9 +28,8 @@ from onyx.connectors.models import TextSection
from onyx.file_processing.extract_file_text import detect_encoding
from onyx.file_processing.extract_file_text import extract_file_text
from onyx.file_processing.extract_file_text import get_file_ext
from onyx.file_processing.extract_file_text import is_accepted_file_ext
from onyx.file_processing.extract_file_text import is_text_file_extension
from onyx.file_processing.extract_file_text import OnyxExtensionType
from onyx.file_processing.extract_file_text import is_valid_file_ext
from onyx.file_processing.extract_file_text import read_text_file
from onyx.utils.logger import setup_logger
from onyx.utils.retry_wrapper import request_with_retries
@@ -70,9 +69,7 @@ def _process_egnyte_file(
file_name = file_metadata["name"]
extension = get_file_ext(file_name)
if not is_accepted_file_ext(
extension, OnyxExtensionType.Plain | OnyxExtensionType.Document
):
if not is_valid_file_ext(extension):
logger.warning(f"Skipping file '{file_name}' with extension '{extension}'")
return None

View File

@@ -22,9 +22,8 @@ from onyx.db.engine import get_session_with_current_tenant
from onyx.db.pg_file_store import get_pgfilestore_by_file_name
from onyx.file_processing.extract_file_text import extract_text_and_images
from onyx.file_processing.extract_file_text import get_file_ext
from onyx.file_processing.extract_file_text import is_accepted_file_ext
from onyx.file_processing.extract_file_text import is_valid_file_ext
from onyx.file_processing.extract_file_text import load_files_from_zip
from onyx.file_processing.extract_file_text import OnyxExtensionType
from onyx.file_processing.image_utils import store_image_and_create_section
from onyx.file_store.file_store import get_default_file_store
from onyx.utils.logger import setup_logger
@@ -52,7 +51,7 @@ def _read_files_and_metadata(
file_content, ignore_dirs=True
):
yield os.path.join(directory_path, file_info.filename), subfile, metadata
elif is_accepted_file_ext(extension, OnyxExtensionType.All):
elif is_valid_file_ext(extension):
yield file_name, file_content, metadata
else:
logger.warning(f"Skipping file '{file_name}' with extension '{extension}'")
@@ -123,7 +122,7 @@ def _process_file(
logger.warning(f"No file record found for '{file_name}' in PG; skipping.")
return []
if not is_accepted_file_ext(extension, OnyxExtensionType.All):
if not is_valid_file_ext(extension):
logger.warning(
f"Skipping file '{file_name}' with unrecognized extension '{extension}'"
)

View File

@@ -20,8 +20,8 @@ from onyx.connectors.models import ConnectorMissingCredentialError
from onyx.connectors.models import Document
from onyx.connectors.models import SlimDocument
from onyx.connectors.models import TextSection
from onyx.file_processing.extract_file_text import ALL_ACCEPTED_FILE_EXTENSIONS
from onyx.file_processing.extract_file_text import extract_file_text
from onyx.file_processing.extract_file_text import VALID_FILE_EXTENSIONS
from onyx.indexing.indexing_heartbeat import IndexingHeartbeatInterface
from onyx.utils.logger import setup_logger
@@ -298,7 +298,7 @@ class HighspotConnector(LoadConnector, PollConnector, SlimConnector):
elif (
is_valid_format
and file_extension in ALL_ACCEPTED_FILE_EXTENSIONS
and file_extension in VALID_FILE_EXTENSIONS
and can_download
):
# For documents, try to get the text content

View File

@@ -821,30 +821,26 @@ class VespaIndex(DocumentIndex):
num_to_retrieve: int = NUM_RETURNED_HITS,
offset: int = 0,
) -> list[InferenceChunkUncleaned]:
vespa_where_clauses = build_vespa_filters(
filters, include_hidden=True, remove_trailing_and=True
vespa_where_clauses = build_vespa_filters(filters, include_hidden=True)
yql = (
YQL_BASE.format(index_name=self.index_name)
+ vespa_where_clauses
+ '({grammar: "weakAnd"}userInput(@query) '
# `({defaultIndex: "content_summary"}userInput(@query))` section is
# needed for highlighting while the N-gram highlighting is broken /
# not working as desired
+ f'or ({{defaultIndex: "{CONTENT_SUMMARY}"}}userInput(@query)))'
)
yql = YQL_BASE.format(index_name=self.index_name) + vespa_where_clauses
params: dict[str, str | int] = {
"yql": yql,
"query": query,
"hits": num_to_retrieve,
"offset": 0,
"ranking.profile": "admin_search",
"timeout": VESPA_TIMEOUT,
}
if len(query.strip()) > 0:
yql += (
' and ({grammar: "weakAnd"}userInput(@query) '
# `({defaultIndex: "content_summary"}userInput(@query))` section is
# needed for highlighting while the N-gram highlighting is broken /
# not working as desired
+ f'or ({{defaultIndex: "{CONTENT_SUMMARY}"}}userInput(@query)))'
)
params["yql"] = yql
params["query"] = query
return query_vespa(params)
# Retrieves chunk information for a document:

View File

@@ -7,8 +7,6 @@ from collections.abc import Callable
from collections.abc import Iterator
from collections.abc import Sequence
from email.parser import Parser as EmailParser
from enum import auto
from enum import IntFlag
from io import BytesIO
from pathlib import Path
from typing import Any
@@ -37,7 +35,7 @@ logger = setup_logger()
TEXT_SECTION_SEPARATOR = "\n\n"
ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS = [
PLAIN_TEXT_FILE_EXTENSIONS = [
".txt",
".md",
".mdx",
@@ -51,7 +49,7 @@ ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS = [
".yaml",
]
ACCEPTED_DOCUMENT_FILE_EXTENSIONS = [
VALID_FILE_EXTENSIONS = PLAIN_TEXT_FILE_EXTENSIONS + [
".pdf",
".docx",
".pptx",
@@ -59,21 +57,12 @@ ACCEPTED_DOCUMENT_FILE_EXTENSIONS = [
".eml",
".epub",
".html",
]
ACCEPTED_IMAGE_FILE_EXTENSIONS = [
".png",
".jpg",
".jpeg",
".webp",
]
ALL_ACCEPTED_FILE_EXTENSIONS = (
ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS
+ ACCEPTED_DOCUMENT_FILE_EXTENSIONS
+ ACCEPTED_IMAGE_FILE_EXTENSIONS
)
IMAGE_MEDIA_TYPES = [
"image/png",
"image/jpeg",
@@ -81,15 +70,8 @@ IMAGE_MEDIA_TYPES = [
]
class OnyxExtensionType(IntFlag):
Plain = auto()
Document = auto()
Multimedia = auto()
All = Plain | Document | Multimedia
def is_text_file_extension(file_name: str) -> bool:
return any(file_name.endswith(ext) for ext in ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS)
return any(file_name.endswith(ext) for ext in PLAIN_TEXT_FILE_EXTENSIONS)
def get_file_ext(file_path_or_name: str | Path) -> str:
@@ -101,20 +83,8 @@ def is_valid_media_type(media_type: str) -> bool:
return media_type in IMAGE_MEDIA_TYPES
def is_accepted_file_ext(ext: str, ext_type: OnyxExtensionType) -> bool:
if ext_type & OnyxExtensionType.Plain:
if ext in ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS:
return True
if ext_type & OnyxExtensionType.Document:
if ext in ACCEPTED_DOCUMENT_FILE_EXTENSIONS:
return True
if ext_type & OnyxExtensionType.Multimedia:
if ext in ACCEPTED_IMAGE_FILE_EXTENSIONS:
return True
return False
def is_valid_file_ext(ext: str) -> bool:
return ext in VALID_FILE_EXTENSIONS
def is_text_file(file: IO[bytes]) -> bool:
@@ -412,9 +382,6 @@ def extract_file_text(
"""
Legacy function that returns *only text*, ignoring embedded images.
For backward-compatibility in code that only wants text.
NOTE: Ignoring seems to be defined as returning an empty string for files it can't
handle (such as images).
"""
extension_to_function: dict[str, Callable[[IO[Any]], str]] = {
".pdf": pdf_to_text,
@@ -438,9 +405,7 @@ def extract_file_text(
if extension is None:
extension = get_file_ext(file_name)
if is_accepted_file_ext(
extension, OnyxExtensionType.Plain | OnyxExtensionType.Document
):
if is_valid_file_ext(extension):
func = extension_to_function.get(extension, file_io_to_text)
file.seek(0)
return func(file)

View File

@@ -1,19 +1,10 @@
import io
from typing import cast
from PIL import Image
from onyx.background.celery.tasks.beat_schedule import CLOUD_BEAT_MULTIPLIER_DEFAULT
from onyx.background.celery.tasks.beat_schedule import (
CLOUD_DOC_PERMISSION_SYNC_MULTIPLIER_DEFAULT,
)
from onyx.configs.constants import CLOUD_BUILD_FENCE_LOOKUP_TABLE_INTERVAL_DEFAULT
from onyx.configs.constants import ONYX_CLOUD_REDIS_RUNTIME
from onyx.configs.constants import ONYX_CLOUD_TENANT_ID
from onyx.configs.constants import ONYX_EMAILABLE_LOGO_MAX_DIM
from onyx.db.engine import get_session_with_shared_schema
from onyx.file_store.file_store import PostgresBackedFileStore
from onyx.redis.redis_pool import get_redis_replica_client
from onyx.utils.file import FileWithMimeType
from onyx.utils.file import OnyxStaticFileManager
from onyx.utils.variable_functionality import (
@@ -96,72 +87,3 @@ class OnyxRuntime:
)
return OnyxRuntime._get_with_static_fallback(db_filename, STATIC_FILENAME)
@staticmethod
def get_beat_multiplier() -> float:
"""the beat multiplier is used to scale up or down the frequency of certain beat
tasks in the cloud. It has a significant effect on load and is useful to adjust
in real time."""
beat_multiplier: float = CLOUD_BEAT_MULTIPLIER_DEFAULT
r = get_redis_replica_client(tenant_id=ONYX_CLOUD_TENANT_ID)
beat_multiplier_raw = r.get(f"{ONYX_CLOUD_REDIS_RUNTIME}:beat_multiplier")
if beat_multiplier_raw is not None:
try:
beat_multiplier_bytes = cast(bytes, beat_multiplier_raw)
beat_multiplier = float(beat_multiplier_bytes.decode())
except ValueError:
pass
if beat_multiplier <= 0.0:
return 1.0
return beat_multiplier
@staticmethod
def get_doc_permission_sync_multiplier() -> float:
"""Permission syncs are a significant source of load / queueing in the cloud."""
value: float = CLOUD_DOC_PERMISSION_SYNC_MULTIPLIER_DEFAULT
r = get_redis_replica_client(tenant_id=ONYX_CLOUD_TENANT_ID)
value_raw = r.get(f"{ONYX_CLOUD_REDIS_RUNTIME}:doc_permission_sync_multiplier")
if value_raw is not None:
try:
value_bytes = cast(bytes, value_raw)
value = float(value_bytes.decode())
except ValueError:
pass
if value <= 0.0:
return 1.0
return value
@staticmethod
def get_build_fence_lookup_table_interval() -> int:
"""We maintain an active fence table to make lookups of existing fences efficient.
However, reconstructing the table is expensive, so adjusting it in realtime is useful.
"""
interval: int = CLOUD_BUILD_FENCE_LOOKUP_TABLE_INTERVAL_DEFAULT
r = get_redis_replica_client(tenant_id=ONYX_CLOUD_TENANT_ID)
interval_raw = r.get(
f"{ONYX_CLOUD_REDIS_RUNTIME}:build_fence_lookup_table_interval"
)
if interval_raw is not None:
try:
interval_bytes = cast(bytes, interval_raw)
interval = int(interval_bytes.decode())
except ValueError:
pass
if interval <= 0.0:
return CLOUD_BUILD_FENCE_LOOKUP_TABLE_INTERVAL_DEFAULT
return interval

View File

@@ -1,77 +0,0 @@
import os
from unittest.mock import MagicMock
from unittest.mock import patch
import pytest
from onyx.configs.constants import BlobType
from onyx.connectors.blob.connector import BlobStorageConnector
from onyx.connectors.models import Document
from onyx.connectors.models import TextSection
from onyx.file_processing.extract_file_text import ACCEPTED_DOCUMENT_FILE_EXTENSIONS
from onyx.file_processing.extract_file_text import ACCEPTED_IMAGE_FILE_EXTENSIONS
from onyx.file_processing.extract_file_text import ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS
from onyx.file_processing.extract_file_text import get_file_ext
@pytest.fixture
def blob_connector(request: pytest.FixtureRequest) -> BlobStorageConnector:
connector = BlobStorageConnector(
bucket_type=BlobType.S3, bucket_name="onyx-connector-tests"
)
connector.load_credentials(
{
"aws_access_key_id": os.environ["AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS"],
"aws_secret_access_key": os.environ[
"AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS"
],
}
)
return connector
@patch(
"onyx.file_processing.extract_file_text.get_unstructured_api_key",
return_value=None,
)
def test_blob_s3_connector(
mock_get_api_key: MagicMock, blob_connector: BlobStorageConnector
) -> None:
"""
Plain and document file types should be fully indexed.
Multimedia and unknown file types will be indexed by title only with one empty section.
This is intentional in order to allow searching by just the title even if we can't
index the file content.
"""
all_docs: list[Document] = []
document_batches = blob_connector.load_from_state()
for doc_batch in document_batches:
for doc in doc_batch:
all_docs.append(doc)
#
assert len(all_docs) == 19
for doc in all_docs:
section = doc.sections[0]
assert isinstance(section, TextSection)
file_extension = get_file_ext(doc.semantic_identifier)
if file_extension in ACCEPTED_PLAIN_TEXT_FILE_EXTENSIONS:
assert len(section.text) > 0
continue
if file_extension in ACCEPTED_DOCUMENT_FILE_EXTENSIONS:
assert len(section.text) > 0
continue
if file_extension in ACCEPTED_IMAGE_FILE_EXTENSIONS:
assert len(section.text) == 0
continue
# unknown extension
assert len(section.text) == 0

View File

@@ -148,8 +148,7 @@ export function Explorer({
clearTimeout(timeoutId);
}
let doSearch = true;
if (doSearch) {
if (query && query.trim() !== "") {
router.replace(
`/admin/documents/explorer?query=${encodeURIComponent(query)}`
);

View File

@@ -1384,7 +1384,6 @@ export function ChatPage({
if (!packet) {
continue;
}
console.log("Packet:", JSON.stringify(packet));
if (!initialFetchDetails) {
if (!Object.hasOwn(packet, "user_message_id")) {
@@ -1730,7 +1729,6 @@ export function ChatPage({
}
}
} catch (e: any) {
console.log("Error:", e);
const errorMsg = e.message;
upsertToCompleteMessageMap({
messages: [
@@ -1758,13 +1756,11 @@ export function ChatPage({
completeMessageMapOverride: currentMessageMap(completeMessageDetail),
});
}
console.log("Finished streaming");
setAgenticGenerating(false);
resetRegenerationState(currentSessionId());
updateChatState("input");
if (isNewSession) {
console.log("Setting up new session");
if (finalMessage) {
setSelectedMessageForDocDisplay(finalMessage.message_id);
}

View File

@@ -102,7 +102,7 @@ export function UserProvider({
};
// Use the custom token refresh hook
// useTokenRefresh(upToDateUser, fetchUser);
useTokenRefresh(upToDateUser, fetchUser);
const updateUserTemperatureOverrideEnabled = async (enabled: boolean) => {
try {