The details of which files are "model files" and which are "data files" are
detailed in the bbr::check_up_to_date() docs.
# S3 method for bbi_stan_model
check_up_to_date(.bbi_object, .build_data = TRUE, ...)
# S3 method for bbi_stan_gq_model
check_up_to_date(.bbi_object, .build_data = TRUE, ...)
# S3 method for bbi_stan_summary
check_up_to_date(.bbi_object, .build_data = TRUE, ...)
# S3 method for bbi_stan_gq_summary
check_up_to_date(.bbi_object, .build_data = TRUE, ...)the object to check. Could be
a bbi_{.model_type}_model object,
a bbi_{.model_type}_summary object,
or a bbi_log_df tibble.
If TRUE, the default, run -standata.R and save the
output to a temp file and check the hash of the temp file against the
bbi_config.json hash. This option actually runs the code and,
importantly, verifies that the input data to -standata.R has not changed
either. If FALSE, check the hashes of the -standata.R and
-standata.json against the hashes in bbi_config.json but do not run
the -standata.R script. This option is less secure and primarily exists
for quicker checking if building the data is time consuming for certain
models.
Arguments passed through (currently none).
The model files:
<run>.stan
<run>-stanargs.R
<run>-init.R, if the model is not a standalone generated quantities
model
<run>-fitted-params.R, if the model is a standalone generated
quantities model (i.e. a model that inherits from bbi_stan_gq_model).
The data files (see .build_data argument):
<run>-standata.R
<run>-standata.json
For standalone generated quantities, the bbi_config.json files in the
output directory of the models listed in the gq_parent fields are also
considered "data" files. If a gq_parent is re-submitted and anything
recorded in bbi_config.json changes, the linked "stan_gq" model is
considered out of date.