bino add

bino add provides interactive wizards for creating new manifest documents. It supports three interaction modes: fully interactive, semi-interactive (with partial flags), and non-interactive for CI/scripting.

Subcommands

• bino add dataset – create a DataSet manifest
• bino add datasource – create a DataSource manifest

Usage

bino add dataset [name] [flags]
bino add datasource [name] [flags]

Interaction Modes

Fully Interactive

Run without arguments to use the step-by-step wizard:

bino add dataset

The wizard guides you through:

1. Name and description
2. Query type selection (SQL, PRQL, or pass-through)
3. Dependencies (DataSets and DataSources)
4. Constraints
5. Output file location
6. YAML preview and confirmation

Semi-Interactive

Provide some flags and the wizard prompts only for missing values:

bino add dataset sales_summary --sql-file queries/sales.sql

Non-Interactive

Use --no-prompt for CI pipelines or scripting. All required values must be provided via flags:

bino add dataset sales_summary \
  --sql "SELECT * FROM sales" \
  --output datasets/sales.yaml \
  --no-prompt

---

bino add dataset

Create a new DataSet manifest with SQL or PRQL queries.

Flags

Query source (one required in non-interactive mode):

• --sql – inline SQL query
• --sql-file – path to external SQL file (uses $file() syntax)
• --prql – inline PRQL query
• --prql-file – path to external PRQL file (uses $file() syntax)
• --source – pass-through from another DataSet

Configuration:

• --deps – dependencies (DataSet or DataSource names), repeatable
• --constraint – constraints in field:operator:value format, repeatable
• --description – optional description

Output:

• --output, -o – output file path (creates new file)
• --append-to – append to existing multi-document YAML file

Behavior:

• --no-prompt – disable interactive prompts (fail if required values missing)
• --open-editor – open $EDITOR for query input

Examples

Create a dataset interactively:

bino add dataset

Create with inline SQL:

bino add dataset monthly_sales \
  --sql "SELECT month, SUM(amount) FROM orders GROUP BY month" \
  --deps orders_source \
  --output datasets/monthly_sales.yaml \
  --no-prompt

Create with external SQL file:

bino add dataset quarterly_report \
  --sql-file queries/quarterly.sql \
  --deps sales_data \
  --constraint "quarter:in:Q1,Q2,Q3,Q4" \
  --output datasets/quarterly.yaml \
  --no-prompt

Create a pass-through dataset:

bino add dataset filtered_orders \
  --source orders \
  --constraint "status:eq:completed" \
  --output datasets/filtered.yaml \
  --no-prompt

---

bino add datasource

Create a new DataSource manifest for databases or files.

Flags

Type (required in non-interactive mode):

• --type – source type: postgres, mysql, csv, parquet, excel, json

Database connection (for postgres and mysql types):

• --db-host – database host
• --db-port – database port
• --db-database – database name
• --db-schema – database schema (optional, PostgreSQL only)
• --db-user – database username
• --db-secret – ConnectionSecret name for credentials
• --db-query – SQL query to execute

File source (for file types):

• --file – path to source file

CSV options:

• --csv-delimiter – field delimiter (default: ,)
• --csv-header – file has header row (default: true)
• --csv-skip-rows – rows to skip before data

Configuration:

• --constraint – constraints in field:operator:value format, repeatable
• --description – optional description

Output:

• --output, -o – output file path (creates new file)
• --append-to – append to existing multi-document YAML file

Behavior:

• --no-prompt – disable interactive prompts

Examples

Create a PostgreSQL datasource with structured connection:

bino add datasource sales_db \
  --type postgres \
  --db-host localhost \
  --db-port 5432 \
  --db-database analytics \
  --db-user reporting \
  --db-secret postgresCredentials \
  --db-query "SELECT * FROM sales" \
  --output datasources/sales_db.yaml \
  --no-prompt

Create a CSV datasource:

bino add datasource sales_csv \
  --type csv \
  --file data/sales.csv \
  --csv-delimiter ";" \
  --output datasources/sales.yaml \
  --no-prompt

Create a Parquet datasource:

bino add datasource events \
  --type parquet \
  --file data/events.parquet \
  --output datasources/events.yaml \
  --no-prompt

---

Name Validation

DataSet and DataSource names must follow these rules:

• Start with a lowercase letter
• Contain only lowercase letters, numbers, and underscores
• Maximum 64 characters
• Must not start with _inline_ (reserved prefix)
• Must be unique within the project for that kind

Valid examples: sales_data, orders2024, monthly_report

Invalid examples: SalesData (uppercase), 2024_orders (starts with number), sales-data (hyphen)

The wizard suggests corrections for invalid names.

File Organization

The wizard detects your project’s file organization pattern:

• Separate files – each manifest in its own file (e.g., datasets/sales.yaml)
• Multi-document – multiple manifests in one file separated by ---

Your preference is saved to .bino/config.json for future use.

Editor Integration

When using --open-editor or when the wizard needs query input, bino opens your preferred editor:

1. $VISUAL environment variable
2. $EDITOR environment variable
3. vim (fallback)
4. nano (final fallback)

For GUI editors like VS Code, bino automatically adds the --wait flag.

See also

• DataSet reference
• DataSource reference
• bino init
• bino lint