In ARROW-12644, we added support for decoding the file paths when reading datasets. So a valid follow-up question: should we also support encoding when writing datasets?

(see also #11027)

Rereading ARROW-12644, there wasn't yet much discussion on that aspect.

cc @westonpace @lidavidm

Reporter: Joris Van den Bossche / @jorisvandenbossche

Related issues:

• [C++][Dataset] Support reading date/time-partitioned datasets accounting for URL encoding (Spark) (is related to)
• [C++] Add URL Encoding kernel (depends upon)

_{Note: This issue was originally created as ARROW-13813. Please see the migration documentation for further details.}