DataFusion weekly project plan (Andrew Lamb) - April 29, 2024

[alamb]

Follow on to #10172

It would be great for other contributors to DataFusion who plan non trivial work could try to make them visible somehow as well 🙏 (feel free to copy / modify the format)

My Highlights from Last Week

• We have a new website https://datafusion.apache.org/ (with redirects from old urls) 🎉
• We have worked through [EPIC] Tasks for a new Top Level Apache Project #9691 -- thanks to @phillipleblanc, @comphead, @andygrove and @tisonkun and @lewiszlw)
• We merged quite a few PRs for [EPIC] Stop copying LogicalPlan during OptimizerPasses #9637 (e.g. 10184 and 10212. etc. Thanks to @peter-toth, @Lordworms, @kevinmingtarja)
• @jayzhan and @liukun cleaned up how zero argument handling works ScalarUDF: Remove supports_zero_argument and avoid creating null array for empty args #10193
• @Jefffrey Added MEDIAN(DISTINCT ..) support: The last aggregate without distinct support
• @simonvandel made date_bin (2x faster) (https://github.com/apache/datafusion/pull/10215[)](https://github.com/apache/datafusion/commit/169701e0128911f16ed07e1ea714a4fdc1e90ee0) 🚀

My (personal) goals

1000 projects are built using DataFusion 📈

My (personal) plans for this week

• Support [EPIC] Tasks for a new Top Level Apache Project #9691 (complete Blog Post about graduating to a new top level project #10135)
• Keep removing copies to make planning faster: [EPIC] Stop copying LogicalPlan during OptimizerPasses #9637
• (Stretch goal) Document ParquetExec more (so it is clear / easier to understand what is already supported and what else could be useful API in ParquetExec to pass in RowSelections to ParquetExec (enable custom indexes, finer grained pushdown) #9929)

Project Queue (list of future projects)

• API in ParquetExec to pass in RowSelections to ParquetExec (enable custom indexes, finer grained pushdown) #9929 (likely not this week)

Projects I plan to help actively help review / plan

• [EPIC] Implement StringViewArray and BinaryViewArray arrow-rs#5374 (I am struggling to find time, but we are making progress, slowly)
• [Epic] A Collection of Sort Based Optimizations #10313
• [EPIC] Support converting Exprs and LogicalPlans --> SQL Strings #9494
• Supporting using parallel parquet writer outside of Datafusion query execution #9493 with @wiedld

Algorithm for (my) prioritizing PR reviews

Note there are many committers who can and do review and merge PRs, so this is not the priorities of the project as a whole, just the approximate algorithm I use to prioritize my own time.

Priority:

1. Bug fixes (where something is just incorrect), especially regressions (where it used to work and now does not)
2. Improvements directly related to features needed for InfluxDB (my employer)
3. Documentation and test improvements (I view these as very strategically important)
4. PRs that I think are strategically important
5. Other new features / additions to functionality (note this is the lowest)

The top strategically important projects in my head are:

• Anything that makes it easier to use DataFusion as a user (docs, examples, better APIs, etc), including Getting started guide for new users (who want to use DataFusion in their project)
• Anything that improves DataFusion's quality such as bug fixes, and improved / expanded tests, etc
• Making function packages modular and easier to mix/match: [EPIC] Unify Function Interface (remove BuiltInScalarFunction) #8045

Thus, if you are interested in contributing to DataFusion and are interested in a fast turn around time I would recommend looking into bug fixes / test improvements / documentation or the projects named above.

If you propose adding new functionality, especially if the PR is large/complex and not connected to a wider need, the review cycle will likely be longer. You can make it a shorter cycle by looking at the comments on other recent PRs and following the same model (e.g. ensure there are tests in sqllogictest for example, the CI passes, includes documentation, etc)

Background

The idea of this ticket is make my plans for DataFusion visible, largely for my own personal organizational needs, but also to:

1. Communicate / coordinate in the community
2. Help provide an interesting summary of what is happening in DataFusion this week

[alamb]

Review Queue

• Fix Coalesce casting logic to follows what Postgres and DuckDB do. Introduce signature that do non-comparison coercion #10268
• fix: preserve more dictionaries when coercing types #10221
• feat: Determine ordering of file groups #9593
• fix: schema error when parsing order-by expressions #10234

SQL Parser:

• Improve parsing of JSON accesses on Postgres and Snowflake sqlparser-rs/sqlparser-rs#1215

[alamb]

Inspired by #9593 from @suremarc , and #10257 from @nga, I started organizing sort based optimizations in a new epic: #10313

[alamb]

Review Queue

• feat: run expression simplifier in a loop until a fixedpoint or 3 cycles #10358 (review)
• Add simplify method to aggregate function #10354
• Array agg groups accumulator #10149
• feat: support grouping aggregate function #10208

Arrow

• docs: add sizing explanation to bloom filter docs in parquet arrow-rs#5705 (review)
• Add the ability for Maps to cast to another case where the field names are different arrow-rs#5703
• Support casting StringView/BinaryView --> StringArray/BinaryArray. arrow-rs#5704
• Implement GenericByteViewArray::gc for compacting StringViewArray and ByteViewArray arrow-rs#5707

[liukun]

• @jayzhan and @liukun cleaned up how zero argument handling works ScalarUDF: Remove supports_zero_argument and avoid creating null array for empty args #10193

Should be @jayzhan211 and @liukun4515 .
Not me ;p

[alamb]

Sorry @liukun -- my bad

[alamb]

Next week: #10395