Added changelog generator script and configuration.

[jorgecarleitao]

Closes #274

[jorgecarleitao]

The changelog up to today looks as below. I find placing the PRs nice because they give explicit credit to the contributors, but we can also trim them.

@ianmcook , do you think the content below would be a sufficient form of documentation about what changed?

---

For older versions, see apache/arrow/CHANGELOG.md

Changelog

4.0.1 (2021-05-14)

Full Changelog

Implemented enhancements:

• Improve performance of bound checking option #280 [arrow]
• CSV Reader String Dictionary Support #228 [arrow]
• Add Builder interface for adding Arrays to record batches #210 [arrow]
• Support auto-vectorization for min/max #209 [arrow]
• Read list field correctly in <struct<list>> #167 [parquet]
• Support LargeUtf8 in sort kernel #25 [arrow]

Fixed bugs:

• Rust 1.52 Clippy error #266
• NaNs can break parquet statistics #255 [parquet]
• u64::MAX does not roundtrip through parquet #254 [parquet]
• Integration tests failing to compile (flatbuffer) #249 [arrow]
• Investigate compatibility quirks between arrow and parquet structs #245 [parquet]
• Unable to write non-null Arrow structs to Parquet #244 [parquet]
• Dev PR / Process (pull_request) Failing on PRs #242
• schema: missing field metadata when deserialize #241 [arrow]
• Arrow does not compile #238 [arrow]
• Sort with limit panics for the limit includes some but not all nulls, for large arrays #235 [arrow]
• Fix SEGFAULT/ SIGILL in child-data ffi #206 [arrow]

Documentation updates:

• Comment out the instructions in the PR template #277
• Update links to datafusion and ballista in README.md #19
• Update "repository" in Cargo.toml #12
• Improve docs for NullArray, new_null_array and new_empty_array #240 [arrow] (alamb)

Merged pull requests:

• Fix subtraction underflow when sorting string arrays with many nulls #285 (medwards)
• Speed up bound checking in take #281 (Dandandan)
• Update PR template by commenting out instructions #278 (nevi-me)
• Fix null struct and list roundtrip #270 (nevi-me)
• 1.52 clippy fixes #267 (nevi-me)
• Fix typo in csv/reader.rs #265 (domoritz)
• Fix empty Schema::metadata deserialization error #260 (hulunbier)
• update datafusion and ballista doc links #259 (Jimexist)
• support full u32 and u64 roundtrip through parquet #258 [parquet] (crepererum)
• fix NaN handling in parquet statistics #256 (crepererum)
• [MINOR] Added env to run rust in integration. #253 (jorgecarleitao)
• [Minor] Made integration tests always run. #248 (jorgecarleitao)
• fix parquet max_definition for non-null structs #246 (nevi-me)
• Disabled rebase needed until demonstrate working. #243 (jorgecarleitao)
• pin flatbuffers to 0.8.4 #239 (ritchie46)
• sort_primitive result is capped to the min of limit or values.len #236 (medwards)
• Read list field correctly #234 [parquet] (nevi-me)
• Fix code examples for RecordBatch::try_from_iter #231 (alamb)
• Support string dictionaries in csv reader (CSV Reader String Dictionary Support #228) #229 (tustvold)
• support LargeUtf8 in sort kernel #26 (ritchie46)
• Removed unused files #22 (jorgecarleitao)
• ARROW-12504: Buffer::from_slice_ref set correct capacity #18 [arrow] (tustvold)
• Add GitHub templates #17 (andygrove)
• ARROW-12493: Add support for writing dictionary arrays to CSV and JSON #16 [arrow] (tustvold)
• ARROW-12426: [Rust] Fix concatentation of arrow dictionaries #15 [arrow] (tustvold)
• Update repository and homepage urls #14 [arrow] [arrow-flight] [parquet] (Dandandan)
• Added rebase-needed bot #13 (jorgecarleitao)
• Added Integration tests against arrow #10 (jorgecarleitao)
• ARROW-12343: [Rust] Support auto-vectorization for min/max #9 [arrow] (Dandandan)
• ARROW-12411: [Rust] Create RecordBatches from Iterators #7 [arrow] (alamb)

* This Changelog was automatically generated by github_changelog_generator

[codecov-commenter]

Codecov Report

Merging #289 (7d232ed) into master (ce8e67c) will increase coverage by 0.00%.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #289   +/-   ##
=======================================
  Coverage   82.49%   82.49%           
=======================================
  Files         162      162           
  Lines       43980    43980           
=======================================
+ Hits        36282    36283    +1     
+ Misses       7698     7697    -1     

Impacted Files	Coverage Δ
arrow/src/util/display.rs	29.72% <0.00%> (-0.83%)	⬇️
...-testing/src/bin/flight-test-integration-client.rs	0.00% <0.00%> (ø)
...-testing/src/bin/flight-test-integration-server.rs	0.00% <0.00%> (ø)
parquet/src/encodings/encoding.rs	95.04% <0.00%> (+0.19%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ce8e67c...7d232ed. Read the comment docs.

[alamb]

Looks very cool @jorgecarleitao

I ran CHANGELOG_GITHUB_TOKEN=$ARROW_GITHUB_API_TOKEN ./change_log.sh locally

It all seemed to go well until the end:

Generated log placed in /usr/local/src/your-app/CHANGELOG.md
sed: 1: "CHANGELOG.md": invalid command code C

I think as we gear up for our first release (🎉 ) of code from arrow-rs, we should start documenting some processes. For example, maybe we could write up a few words about how to make release notes in https://github.com/apache/arrow-rs/blob/master/dev/README.md

[alamb]

I wonder if this should be something like

Suggested change

	sed -i "s/\\\n/\n\n/" CHANGELOG.md
	sed -i -e "s/\\\n/\n\n/" CHANGELOG.md

Though I am not a sed 🧙

[jorgecarleitao]

This was due to a bug on the generator, github-changelog-generator/github-changelog-generator#962, that it was fixed and released in less than 24h (big kudos to @olleolleolle for this 💯 ). I need to check that the new docker image works as intended, so that we can remove this hack. 👍

[alamb]

I plan to merge this in once I can get a clean PR run - and then I will use it to create part of the initial proposed 4.0.1 release

[jorgecarleitao]

sounds good. I agree that a dry-run over a concrete release helps flushing out any issues around this 👍