data are now retrieved from the versioned childes-db dataset on Redivis (https://redivis.com/datapages/datasets/childes_db) instead of a MySQL server; the get_ function interface and outputs are unchanged
filtering of the large content tables (tokens, utterances, types) happens server-side on Redivis, so only matching rows are transferred
get_sql_query() now runs BigQuery Standard SQL (Redivis’s query engine) instead of MySQL SQL
get_tokens(replace = TRUE) now correctly drops the replacement column; previously a bug kept it
dropped dependencies on DBI, dbplyr, and RMySQL; added a Suggests dependency on the redivis client (installable from https://langcog.r-universe.dev)
the current database version is now childes-db 2026.1 (Redivis v1.4). It adds a token_morpheme table (accessible via get_sql_query(); dedicated getters may follow) and reworks morphological parsing, so morpheme-based measures (e.g. mlu_m, num_morphemes) are not comparable with earlier versions. As always, numeric ids are internal to a release – pin db_version for reproducible analyses and use pid to link transcripts across versions
childesr 0.2.3
CRAN release: 2022-01-26
update url for database connection
childesr 0.2.2
CRAN release: 2021-10-04
add checks for database connection errors
childesr 0.2.1
CRAN release: 2020-11-13
fix age filtering in get_participants()
add support for SQL Query strings
update for compatibility with dplyr 2.0.0
childesr 0.1.2
CRAN release: 2019-10-17
enable connections to previous versions of childes-db
add function for clearing all database connections
don’t run code in vignette if unable to connect to childes-db