Summarise a local cache batch by batch

Reads every metadata.json of a cache, as jatos_read_metadata() does, and reports per batch how many component results it lists, how many are on disk, how many a run of jatos_download_results() would fetch, and how many data.txt files sit in the batch directory without a metadata row (orphans, counted inside that batch directory only; a fresh download of the batch's metadata covers them). No request is made.

Usage

jatos_cache_status(path, batch_id = NULL)

Arguments

path

The cache directory, one batch_<id> directory inside it, or the path of a single metadata.json file.

batch_id

Optional ids to keep only some batches.

Value

A tibble with one row per metadata.json: batch_id (NA when the file holds several batches), path (its directory), n_results (component results listed), n_downloaded (with a local file), n_pending (server size above 0 and no local file, or a local file smaller than the server's), n_shrunk (local file larger than the server's), n_orphans (data files without a metadata row) and bytes (size of the local data files). With batch_id, the counts cover the rows of those batches and a metadata.json without any of them is left out; n_orphans still compares the directory with every row of its metadata.json, so the files of another batch listed there are not orphans.

Examples

cache <- system.file("extdata", "JATOS_data", package = "jatosr")
jatos_cache_status(cache)
#> # A tibble: 1 × 8
#>   batch_id path        n_results n_downloaded n_pending n_shrunk n_orphans bytes
#>      <int> <chr>           <int>        <int>     <int>    <int>     <int> <dbl>
#> 1       34 /home/runn…         4            2         0        0         0  3584
jatos_cache_status(cache, batch_id = 34)
#> # A tibble: 1 × 8
#>   batch_id path        n_results n_downloaded n_pending n_shrunk n_orphans bytes
#>      <int> <chr>           <int>        <int>     <int>    <int>     <int> <dbl>
#> 1       34 /home/runn…         4            2         0        0         0  3584