Module dsl 

Canonical pipeline parsing and execution. Canonical document-first pipeline DSL.

A pipeline is a command followed by zero or more transformation stages separated by |. The stages transform the rows returned by the command:

"orch task list | F status=running | S created | L 10"
 ─────────────────  ─────────────────────────────────
    command              pipeline stages

Data flow:

raw line
  │
  ▼  parse_pipeline(line)
  Pipeline { command: "orch task list",
             stages:  ["F status=running", "S created", "L 10"] }
  │
  │  caller dispatches the command and gets rows back
  │
  ▼  apply_pipeline(rows, &stages)
  OutputResult  ← filtered · sorted · limited rows

The main public surface:

• crate::dsl::parse_pipeline turns a raw line into a crate::dsl::Pipeline value
• crate::dsl::parse_stage parses one stage when the command has already been split away
• crate::dsl::apply_pipeline / crate::dsl::apply_output_pipeline apply stages to existing rows
• crate::dsl::registered_verbs returns metadata for all supported stage verbs
• crate::dsl::verb_info looks up a single verb by name

Choose the smallest entrypoint that matches your starting point:

• if you only need to split "command | stages" into structured pieces, use crate::dsl::parse_pipeline
• if you already have one raw stage like "F uid=alice" and need to inspect its verb/spec shape, use crate::dsl::parse_stage
• if your command already produced Vec<Row>, use crate::dsl::apply_pipeline
• if you already have an crate::core::output_model::OutputResult and want to preserve its semantic document or render metadata, use crate::dsl::apply_output_pipeline
• if your rows come from an iterator and you want streamable stages to stay streamable for as long as possible, use crate::dsl::execute_pipeline_streaming

Common verbs: F (filter), P (project), S (sort), G (group), A (aggregate), L (limit), V/K (quick search), U (unroll), JQ (jq expression). See crate::dsl::registered_verbs for the full list with streaming notes in crate::dsl::VerbStreaming.

Structs

Pipeline

Parsed command line split into a command segment and trailing DSL stages.

VerbInfo

Static metadata for one registered DSL verb.

Enums

VerbStreaming

Streaming behavior for a DSL verb.

Functions

apply_output_pipeline

Apply a pipeline to existing output without flattening grouped data first.

apply_pipeline

Apply a pipeline to plain row output.

execute_pipeline

Execute a pipeline starting from plain rows.

execute_pipeline_streaming

Execute a pipeline from any row iterator.

is_registered_explicit_verb

Returns whether verb is a registered non-meta verb.

parse_pipeline

Split a full command line into its command portion and raw pipe stages.

parse_stage

Parse a raw stage string into the structured form the evaluator consumes.

registered_verbs

Returns metadata for all registered DSL verbs, including meta-only verbs.

render_streaming_badge

Returns the display badge for a verb’s streaming behavior, if any.

verb_info

Returns verb metadata for verb, matched case-insensitively.