ARROW-15040: [R] Enable write_csv_arrow to take a Dataset or arrow_dplyr_query as input

[thisisnic]

No description provided.

[github-actions]

https://issues.apache.org/jira/browse/ARROW-15040

[github-actions]

⚠️ Ticket has not been started in JIRA, please click 'Start Progress'.

[jonkeane]

Oh, this is exciting! One comment about types. It also looks like there are some(?) related segfaults in the CI

[paleolimbot]

It may be worth rebasing this and seeing if it still segfaults! I know there were some updates to the Scanner that fixed some segfaults for me when I was writing tests for the Python bridge (from the Python side). Another way to write this test that would be independent of that is to Scan a Table or InMemoryDataset rather than a FileSystemDataset.

I'm also a fan of the "allow a general record batch reader" approach, which might also make this easier to test.

[paleolimbot]

Also linking ARROW-15128 since it seems like a blocker for this PR.

[thisisnic]

Thanks for the suggestion @paleolimbot but the segfault is still happening after a rebase.

[thisisnic]

Have now updated this and removed the blocker ticket which isn't a blocker!

[nealrichardson]

Should also write a test for accepting RecordBatchReader, and need to update the docs too

[nealrichardson]

A couple of final suggestions

[ursabot]

Benchmark runs are scheduled for baseline = 676b49f and contender = 13045f4. 13045f4 is a master commit associated with this PR. Results will be available as each benchmark for each run completes.
Conbench compare runs links:
[Finished ⬇️0.0% ⬆️0.0%] ec2-t3-xlarge-us-east-2
[Finished ⬇️0.63% ⬆️0.67%] test-mac-arm
[Finished ⬇️0.0% ⬆️3.21%] ursa-i9-9960x
[Finished ⬇️0.3% ⬆️0.68%] ursa-thinkcentre-m75q
Supported benchmarks:
ec2-t3-xlarge-us-east-2: Supported benchmark langs: Python. Runs only benchmarks with cloud = True
test-mac-arm: Supported benchmark langs: C++, Python, R
ursa-i9-9960x: Supported benchmark langs: Python, R, JavaScript
ursa-thinkcentre-m75q: Supported benchmark langs: C++, Java