Skip to main content

User Guide

Welcome to the Sigilweaver Loom User Guide. This documentation covers everything you need to build data transformation workflows.

Getting Started​

Sigilweaver Loom is a visual data pipeline tool that lets you build complex data transformations through an intuitive drag-and-drop interface. Under the hood, it uses Polars for fast, efficient data processing with lazy evaluation.

Documentation Sections​

Interface​

Learn how to navigate Studio:

Expressions​

The expression language used in the Custom/Advanced mode of Filter and Formula tools. Most users won't need this - the visual builders handle the vast majority of use cases. But when you need more power, the expression editor has full autocomplete:

  • Basics - Autocomplete, column references, and operators
  • Syntax Rules - Parentheses, multi-line expressions, and gotchas
  • Common Operations - String manipulation, dates, casting, and null handling

Data Types​

Understand the data types Sigilweaver Loom supports and when to use each.

Tools​

Reference documentation for every tool:

CategoryTools
Input/OutputInput, Output
PreparationFilter, Select, Sort, Formula, Sample, Unique
JoinUnion, Join
AggregateSummarize, Pivot, Unpivot

Core Concepts​

Lazy Evaluation​

Sigilweaver Loom uses Polars' lazy evaluation model. This means:

  • Operations are not executed immediately when you build the workflow
  • The entire pipeline is optimized before execution
  • Data is processed efficiently in a streaming fashion when possible

This allows you to work with datasets much larger than your available memory.

Workflows​

A workflow is a directed acyclic graph (DAG) of tools connected by wires:

  • Tools perform operations on data (filtering, joining, aggregating, etc.)
  • Wires connect tool outputs to tool inputs, defining data flow
  • Sockets are the connection points on tools (inputs on the left, outputs on the right)

Expressions​

Many tools use Polars expressions for configuration. The Filter and Formula tools provide visual builders (Basic mode) that handle most use cases without any coding. For advanced logic, Custom mode lets you write expressions directly:

  • Reference columns: pl.col("column_name")
  • Apply transformations: pl.col("price") * 1.1
  • Filter rows: pl.col("status") == "active"
  • Combine conditions: (pl.col("a") > 10) & (pl.col("b") < 20)

See the Expressions section for a complete guide.

External Resources​

For Administrators​

If you're managing a Sigilweaver Loom Hub deployment for your team, see the Administrator Guide for: