[C++][Python] Binary search for sorted tables.
- Dominant language
- C++
- Stars
- 17.1k
- Forks
- 4.3k
- Avg merge
- 3d 18h
- Merged PRs (30d)
- 91
Description
### Describe the enhancement requested
While I support the decision for tables to not have index columns, I have still found it useful to enable fast binary search when it's known that a table is sorted. Currently I'm using `schema.pandas_metadata['index_columns']` to detect sorting. It would be nice if arrow had it own metadata standard, instead of a pandas dependency.
The feature could go through stages:
1. define and document metadata standard
2. `Table.sort_by` adds the metadata
3. a binary search compute function is added
### Component(s)
C++, Python
Contributor guide
Research direction
Start by reviewing the existing Table.sort_by behavior and compute-function entry points, then define the proposed metadata standard independently of pandas. Document the standard, make Table.sort_by add it, and add a binary-search compute function; done means all three stages are specified and implemented for the C++ and Python components.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, python
- Domain
- data
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 30/100