`i` argument could get with=FALSE

Open
#4,485 3 comments 3 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

  • #4488 by @jangorecki — closed without merging

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Stale
Tech stack
r
Domain
data

Research direction

Start by reading the [.data.table implementation and the linked context in issue #3735. Review how the existing with argument is interpreted for i and j, then determine the tests needed for the scalar, named, and length-two forms described here. Done means the proposed interface works while preserving existing scalar behavior.

Written by the indexing model from the issue text.

Description

feature request performance

Argument i is getting very good massaging before it gets evaluated, and even then it is evaluated with try catches, typo checks, etc. Every piece of that process covers some feature, which is nice. This FR asks for an interface to skip all those features, and the same time reduce the overhead that the process imposes, by fall back to a basic integer/logical subset of rows. An example use case can be found in #3735.
That FR could be redundant if we would manage to rewrite [.data.table to escape early what could be escaped, but to be fair, the quantity of features we have in i makes that probably impossible. So giving a control over that to user seems to be a way to go.
The proposal (credits to @chnynf and @franknarf1) is to extend existing with argument to accept length 2 logical vector, possibly named. Example use cases.

DT[1L, with=c(i=FALSE)]
DT[1L, col1+col2, with=c(i=FALSE, j=TRUE)]
DT[1L, 3L, with=c(FALSE, FALSE)]

scalar with would of course refer to j argument for backward compatibility, either length 2, or named would be interpretted for i/j accordingly.

Dominant language
R
Stars
3.9k
Forks
1.1k
Avg merge
14h 4m
Merged PRs (30d)
4

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Rdatatable/data.table

All issues in Rdatatable/data.table

Similar issues

More R issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.