[nealrichardson]

Rationale for this change

See #35710

What changes are included in this PR?

• Get LIB_DIR from pkg-config where possible, to handle the possibility that it is lib64 and not just lib on some platforms.
• Allow x.y.z.1 to use x.y.z C++ library, so that apt/yum official release packages can be used with patched versions submitted to CRAN.

Are these changes tested?

The version check change has a unit test. The LIB_DIR change hopefully is well enough covered by our existing CI.

• Closes: [R] Issues with configure script #35710

[github-actions]

⚠️ GitHub issue #35710 has been automatically assigned in GitHub to PR creator.

[kou]

+1

Can we use more descriptive PR title?

How about writing commit message contents to the PR description too because our merge script only uses the PR description for merged commit. (Commit messages are discarded.)

[nealrichardson]

@kou done, that look better?

[kou]

Yes! Thanks!

[conbench-apache-arrow]

Conbench analyzed the 6 benchmark runs on commit 3f6dfb38.

There were 12 benchmark results indicating a performance regression:

• Commit Run on arm64-t4g-linux-compute at 2023-07-03 23:09:19Z
  • params=simple_expression/batch_size:100000/null_prob:0/bool_true_prob:50/real_time, source=cpp-micro, suite=arrow-acero-filter-benchmark
  • params=complex_expression/batch_size:10000/null_prob:0/bool_true_prob:50/real_time, source=cpp-micro, suite=arrow-acero-filter-benchmark
• and 10 more (see the report linked below)

The full Conbench report has more details.