highperfocused/danswer
1
0
Fork 0
mirror of https://github.com/danswer-ai/danswer.git synced 2025-06-17 11:30:58 +02:00
Code Issues Packages Projects Releases Wiki Activity

Notion Recurse Empty Final Field (#2819)

Browse Source
This commit is contained in:
Yuhong Sun 2024-10-15 23:03:37 -07:00 committed by GitHub
parent 11372aac8f
commit c148fa5bfa
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 25 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
backend/danswer/connectors/notion/connector.py
View File

@ -216,17 +216,28 @@ class NotionConnector(LoadConnector, PollConnector):
def _properties_to_str(properties: dict[str, Any]) -> str: def _properties_to_str(properties: dict[str, Any]) -> str:
"""Converts Notion properties to a string""" """Converts Notion properties to a string"""
def _recurse_properties(inner_dict: dict[str, Any]) -> str: def _recurse_properties(inner_dict: dict[str, Any]) -> str | None:
if not inner_dict:
# Edge case handling, should not happen
return "N/A"
while "type" in inner_dict: while "type" in inner_dict:
type_name = inner_dict["type"] type_name = inner_dict["type"]
inner_dict = inner_dict[type_name] inner_dict = inner_dict[type_name]
# If the innermost layer is None, the value is not set
if not inner_dict:
return None
if isinstance(inner_dict, list): if isinstance(inner_dict, list):
return ", ".join( list_properties = [
[_recurse_properties(item) for item in inner_dict if item] _recurse_properties(item) for item in inner_dict if item
]
return (
", ".join(
[
list_property
for list_property in list_properties
if list_property
]
)
or None
) )
# TODO there may be more types to handle here # TODO there may be more types to handle here
@ -244,11 +255,13 @@ class NotionConnector(LoadConnector, PollConnector):
return f"Until {end}" return f"Until {end}"
if "id" in inner_dict: if "id" in inner_dict:
logger.debug("Skipping Notion Id field") # This is not useful to index, it's a reference to another Notion object
return "Unreadable Property" # and this ID value in plaintext is useless outside of the Notion context
logger.debug("Skipping Notion object id field property")
return None
logger.debug(f"Unreadable property from innermost prop: {inner_dict}") logger.debug(f"Unreadable property from innermost prop: {inner_dict}")
return "Unreadable Property" return None
result = "" result = ""
for prop_name, prop in properties.items(): for prop_name, prop in properties.items():
@ -258,7 +271,8 @@ class NotionConnector(LoadConnector, PollConnector):
inner_value = _recurse_properties(prop) inner_value = _recurse_properties(prop)
# Not a perfect way to format Notion database tables but there's no perfect representation # Not a perfect way to format Notion database tables but there's no perfect representation
# since this must be represented as plaintext # since this must be represented as plaintext
result += f"{prop_name}: {inner_value}\t" if inner_value:
result += f"{prop_name}: {inner_value}\t"
return result return result