GH-32320: [R] Docs unclear about potential disadvantages of dictionary encoding

[thisisnic]

Rationale for this change

Unclear docs - dictionary encoding can slow stuff down but it's not apparent

What changes are included in this PR?

Talk about it

Are these changes tested?

Nah

Are there any user-facing changes?

Nope

• GitHub Issue: [R] Docs unclear about potential disadvantages of dictionary encoding #32320

[github-actions]

⚠️ GitHub issue #32320 has been automatically assigned in GitHub to PR creator.

[Copilot]

Pull request overview

This PR clarifies Arrow’s Parquet writer documentation (R and Python) to note that dictionary encoding—while often beneficial—can be counterproductive for high-cardinality columns by increasing file size and reducing compression effectiveness.

Changes:

• Expanded the R write_parquet() roxygen docs for use_dictionary with guidance about high-cardinality columns.
• Expanded the Python pyarrow.parquet writer argument docs for use_dictionary with the same cautionary note.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
r/R/parquet.R	Adds an explicit warning in write_parquet() docs about dictionary encoding potentially worsening size/compression for many-unique-value columns.
python/pyarrow/parquet/core.py	Adds the same cautionary note to the Parquet writer argument documentation string under use_dictionary.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[thisisnic]

Closing this as we have a fallback and on reflection the original use case may have been a little niche and adding this to the docs may confuse things further