Flatten a /results/metadata answer into a tibble

Turns the nested JSON of POST /results/metadata (studies, their study results, their component results) into a flat tibble with one row per component result. Study-level and study-result-level values repeat down their component results; every column name says which level it belongs to, so length(unique(meta$study_result_id)) counts participants while nrow(meta) counts component results. Column names are snake case, times are POSIXct in UTC, sizes are bytes.

Usage

jatos_flatten_metadata(metadata)

Arguments

metadata

The metadata as the server sent it, not yet a tibble: a parsed answer (the whole envelope or its data element) as returned by jsonlite::read_json(simplifyVector = FALSE), or the path of a metadata.json file.

Value

A tibble with, per study: study_id, study_uuid, study_title; per study result: study_result_id, study_result_uuid, study_code, worker_id, worker_type, batch_id, batch_uuid, batch_title, group_id, study_state, study_start_time, study_end_time, study_duration (minutes), last_seen, comment, study_message (the text shown when a run was aborted or failed), confirmation_code (MTurk), quota_reached, url_query (list of the URL query parameters the worker arrived with); per component result: component_result_id, component_id, component_uuid, component_state, component_start_time, component_end_time, component_duration (minutes), path (location inside a result zip), data_size (bytes on the server; 0 for reloads and unfinished runs), data_file, n_files and files (list of attached-file entries).

Examples

meta <- jatos_flatten_metadata(
  system.file("extdata", "metadata.json", package = "jatosr")
)
meta[, c("study_result_id", "component_result_id", "component_state", "data_size")]
#> # A tibble: 6 × 4
#>   study_result_id component_result_id component_state data_size
#>             <int>               <int> <chr>               <dbl>
#> 1            9001                7001 FINISHED             2048
#> 2            9002                7002 RELOADED                0
#> 3            9002                7003 FINISHED             1536
#> 4            9003                7004 STARTED                 0
#> 5            9004                7005 FINISHED              300
#> 6            9004                7006 FINISHED              700