Skip to content

docs: adopt issue #4419 terminology in Understanding Comet Plans guide#4650

Open
andygrove wants to merge 1 commit into
apache:mainfrom
andygrove:docs-terminology-4419
Open

docs: adopt issue #4419 terminology in Understanding Comet Plans guide#4650
andygrove wants to merge 1 commit into
apache:mainfrom
andygrove:docs-terminology-4419

Conversation

@andygrove

Copy link
Copy Markdown
Member

Which issue does this PR close?

Part of #4419.

Rationale for this change

Issue #4419 points out that the word "native" is overloaded across Comet's docs, carrying three different meanings (Rust implementation, Comet pipeline membership, and scan type). The issue proposes a terminology framework that keeps a few unambiguous compound forms (such as "native Rust", "native shuffle", and operator names like CometNativeScan) but otherwise replaces bare "native" with precise terms.

The "Understanding Comet Plans" page is the central user-facing doc that explains these concepts, so it is the highest-impact place to start. This PR keeps the change focused on that one page rather than attempting to clean up every doc at once.

What changes are included in this PR?

In docs/source/user-guide/latest/understanding-comet-plans.md:

  • Replace bare "native" with the implementation-language term where it means Rust ("Rust-implemented", "implemented in Rust", "runs in Rust via DataFusion").
  • Use "Comet pipeline" / "runs in Comet" / "falls back to Spark" for pipeline-membership and fallback wording.
  • Rename the "Native Execution Operators" section to "Native Rust Operators" to pair symmetrically with "JVM-Side Operators".
  • Keep the compound forms the issue allows: "native Rust", "native shuffle" (paired with the JVM columnar shuffle), and operator/config names.

How are these changes tested?

Documentation-only change. Verified with prettier --check.

…s guide

Replace bare uses of "native" with precise terms from the terminology
framework: "Rust-implemented" / "implemented in Rust" for the
implementation language, "Comet pipeline" for pipeline membership, and
"runs in Comet" / "falls back to Spark" for fallback. Keep the compound
forms the issue allows (native Rust, native shuffle, and operator names
such as CometNativeScan).
@andygrove andygrove added this to the 0.17.0 milestone Jun 13, 2026
A "fallback" happens when Comet cannot translate part of a plan into native
execution. Fallback can be partial (a subtree falls back while the rest stays
native) or full (no Comet nodes appear).
A "fallback" happens when Comet cannot run part of a plan in the Comet

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we also mention fallback to columnar?

@comphead comphead left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks @andygrove

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants