Commit b7323da7 authored by Jan Reimes's avatar Jan Reimes
Browse files

fix(database): correct sorting key in TDocDatabase 

* Update sorting key to use 'date_created' instead of 'date_retrieved'.
* Enhance date parsing in DocumentList to support additional formats.
parent 34e0cc28

src/tdoc_crawler/database/tdocs.py

+1 −1
Original line number Diff line number Diff line
@@ -181,7 +181,7 @@ class TDocDatabase(MeetingDatabase): 
        descending = config.order.value.lower() == "desc"

        records.sort(

            key=lambda record: (

                record.date_retrieved or datetime.min.replace(tzinfo=UTC),

                record.date_created or datetime.min.replace(tzinfo=UTC),

                record.tdoc_id,

            ),

            reverse=descending,

src/tdoc_crawler/tdocs/sources/doclist.py

+3 −2
Original line number Diff line number Diff line
@@ -348,12 +348,13 @@ def _parse_date(date_value: str | None) -> datetime | None: 
        return None



    try:

        # Handle different date formats

        if isinstance(date_value, datetime):

            return date_value

        if isinstance(date_value, str):

            # Try common date formats

            # Try common date formats (with and without time)

            formats = [

                "%d/%m/%Y %H:%M:%S",

                "%Y-%m-%d %H:%M:%S",

                "%Y-%m-%d",

                "%d/%m/%Y",

                "%m/%d/%Y",