Rep-specific seeds in tar_rep(), tar_map_rep(), etc.

[wlandau]

Prework

• I understand and agree to the code of conduct and contributing guidelines.
• If there is already a relevant issue, whether open or closed, comment on the existing thread instead of posting a new issue.
• New features take time and effort to create, and they take even more effort to maintain. So if the purpose of the feature is to resolve a struggle you are encountering personally, please consider first posting a "trouble" or "other" issue so we can discuss your use case and search for existing solutions first.

Proposal

I originally thought this might obfuscate targets' seed-setting behavior, but currently there is a lot of frustration when the batching scheme changes. Ideally the seeds should stay the same when batches * reps stays the same. A tarchetypes at-runtime function could set its own seed that depends on the pattern target name, batch index, and rep index. These special seeds would be saved with target output.

[wlandau]

Hash of the pattern target name and ‘rep_index + n_reps_per_batch * (batch_index - 1)’

[wlandau]

Should affect most non-deprecated functions with batching:

• tar_rep_raw() (see tar_rep_run_map_rep())
• tar_map_rep_raw()
• tar_rep2_raw() (see tar_rep2_run_rep())
• tar_map2_raw() (see tar_map2_run_rep()) (automatically gets us tar_map2_size() and tar_map2_count() for free)
• tar_render_rep_raw() (see tar_render_rep_rep())
• tar_quarto_rep_raw() (see tar_quarto_rep_raw()) (even though the Quarto CLI runs in a separate process and seems to set a random seed. This could change at some point.)

Also:

• Document new output column "tar_seed".

[wlandau]

Fixed in #113

[wlandau]

I will need to reimplement this along the lines of ropensci/targets#1139.

[wlandau]

From ropensci/targets#1139 (comment), I may not have plans in tarchetypes to rework seeds.