Bino Reports VS Code Extension

Enhanced YAML editing for Bino Reports report manifests with intelligent autocompletion, navigation, and project overview.

Features

• Language Server: Completion, hover, diagnostics, go-to-definition, find-references, rename, and quick-fixes are all served by the real bino lsp Language Server (see the LSP docs)
• Bino Explorer Tree View: Browse all Bino documents grouped by kind (DataSource, DataSet, ReportArtefact, etc.) in the Explorer sidebar
• Go to Definition / Find References / Rename: Ctrl+Click (Cmd+Click on macOS) on references like dataset:, source:, secret:, signingProfile:, selectedStyle:, or items in dependencies: to jump to their definitions; Shift+F12/F2 work project-wide
• Dataset Autocompletion: When typing dataset:, suggests all DataSet names and $-prefixed DataSource names from your project
• Column Introspection: When editing scenarios:/variances: arrays or a query:/prql: block, suggests real column names from the referenced dataset via DuckDB
• Reference Completions: Smart completions for signingProfile, selectedStyle, and other cross-document references
• Kind Completions: Suggests all valid document kinds when typing kind:
• Designer: A schema-driven property-form editor for Table/Chart/Text/Tree/Grid manifests (Bino: Open Designer)
• DataSource Wizard: Scaffolds a DataSource + DataSet pair from a file or database connection (Bino: New DataSource from File/Database…)
• Tree/Table Editor & Rows Preview: A grid editor for Tree-kind manifests, plus a live 100-row data sample for any DataSource/DataSet
• PRQL Integration: When editing spec.prql blocks in DataSet manifests, right-click to open a dedicated PRQL editor or SQL preview (requires the PRQL extension)

The extension talks to bino lsp for all live editor intelligence, and to an optional persistent bino daemon (see bino.daemon.enabled) shared by the LSP, preview, build, and data-introspection features so they stay warm across requests. With no daemon running, the extension falls back to discrete bino lsp-helper CLI calls.

Requirements

• RedHat YAML Extension (automatically installed as dependency; provides generic YAML folding/formatting)
• bino CLI must be installed and available in PATH (or configured via bino.binPath)
• PRQL Extension (optional, for enhanced PRQL editing)

Extension Settings

• bino.binPath: Path to the bino CLI executable. If not set, uses 'bino' from PATH.
• bino.columnCacheTTL: Time in milliseconds to cache column introspection results (default: 60 seconds).
• bino.previewPort: Port number for the bino preview server (default: 3000).
• bino.daemon.enabled: Enable the persistent background daemon for faster indexing, validation, and data introspection (default: enabled).
• bino.wizard.sampleRowLimit: Max sample rows the DataSource wizard fetches when introspecting a source (default: 100).

Commands

• Bino: Refresh Index - Manually refresh the workspace index (also available via refresh button in Bino Explorer)
• Bino: Open PRQL Editor - Extract spec.prql from the current DataSet and open it in a PRQL editor with syntax highlighting
• Bino: Open PRQL SQL Preview - Open the PRQL SQL Preview panel to see the compiled SQL (requires PRQL extension)

PRQL Support

Bino supports PRQL (Pipelined Relational Query Language) as an alternative to SQL for DataSet queries. When you have a DataSet with spec.prql:

apiVersion: bino.bi/v1alpha1
kind: DataSet
metadata:
  name: sales_summary
spec:
  prql: |
    from sales_csv
    filter amount > 0
    group {region} (
      aggregate {total = sum amount}
    )
    sort {-total}
  dependencies:
    - sales_csv

You can:

1. Edit with PRQL tooling: Place your cursor inside the prql: block and run "Bino: Open PRQL Editor" (or right-click → "Open PRQL Editor"). This opens the PRQL in a dedicated editor with syntax highlighting and diagnostics.

2. Preview compiled SQL: Run "Bino: Open PRQL SQL Preview" to see the SQL that DuckDB will execute. This uses the official PRQL VS Code extension's SQL Preview feature.

3. Execute via bino build: When you run bino build, PRQL queries are compiled and executed directly by DuckDB using the prql community extension.

For the best PRQL editing experience, install the PRQL extension. Bino will prompt you to install it when it detects PRQL usage in your workspace.

How It Works

1. On activation, the extension spawns bino lsp (the Language Server) and indexes all YAML files in the workspace that contain apiVersion: bino.bi
2. The Bino Explorer in the sidebar shows all indexed documents grouped by kind
3. When you request completions, the language server provides context-aware suggestions:
   • dataset:/source: → Lists matching DataSet/DataSource names
   • scenarios:/variances: → Executes the referenced dataset query via DuckDB and returns column names
   • signingProfile:/selectedStyle: → Lists all matching SigningProfile/ComponentStyle names
4. Go to Definition / Find All References / Rename work on reference fields - Ctrl+Click, Shift+F12, or F2 to navigate/refactor
5. File changes trigger cache invalidation, re-indexing, and a fresh project-wide validation pass (lint findings, missing ${VAR}, engine-compat warnings)