Releases: me9hanics/wikidata-SparQL-data-collection
Releases · me9hanics/wikidata-SparQL-data-collection
Version 2.0.0
This is the first version that is released on GitHub, so it might be odd to call it 2.0.0, however this version is fundamentally different than the codebase before (even a week before, or especially 3 months ago).
Updates:
- make the primary functions
get_all_person_info_strict
andget_multiple_people_all_info_fast_retry_missing
as reliable and fast as possible (select only human instances, faster querying through other languages, etc.) - "Most common" and "Above threshold" label values selection: for queries with multiple responses (e.g. multiple profiles with the same name), select the most common result / all results above some threshold
- many now redundant functions and code removed (~300 lines, e.g. retry-after methods); more to come + renaming
- more accurate Wikidata ID selection (most common label value)
- more use of the
sparql_query
method in other methods, removing the need to handle retry delay logic in other methods, and redundant endpoint_url arguments - refactoring (rename, reorder), adding clean abstractions in sparql_query_by_dict, construct_person_query, etc. Also more input handling, and offering more abstract outputs in multiple methods (e.g. get_places_from_response offering list and different string outputs)
- many new functions: key-value counts methods, get_id_from_results
- no nonstandard library dependencies, except Pandas for those who use the
results_dataframe
function.