Cast Data Types

Cast Data Types changes the data type of one or more columns in your dataset. Every column is shown at once with its current source type and a dropdown to select the target type. Change only the columns you need. Unchanged columns are left as-is. No new column is created.

The target types available in the dropdown are determined by your connected destination database (BigQuery, Snowflake, PostgreSQL, etc.). The dropdown shows only types that are valid for that destination.

When to Use Cast Data Types

• Fixing schema mismatches after ingestion. A numeric column ingested as string cannot be summed or averaged. Cast it to integer or float before aggregating.
• Preparing for joins. Join keys must match in type. If one table has order_id as string and another has it as integer, cast before joining.
• Converting date strings to datetime. Dates stored as "2024-01-15" strings need to be cast to Date or Datetime before using in date range filters or datetime aggregations.
• Standardizing boolean columns. Values like "true", "yes", "1" stored as strings can be cast to Boolean for downstream logic.
• Matching destination schema. Your target table expects specific types. Cast here to avoid load errors.

Sample Dataset

The examples in this doc use the Edilitics sample orders dataset. Download it to follow along in your own workspace.

Relevant columns for Cast Data Types examples:

How Casting Works by Target Type

How to Apply Cast Data Types

Before and After

Four columns cast in one operation:

Before:

After:

"N/A" cannot be cast to float so it becomes null. "no" is not in the true-list so it becomes false.

Code Equivalent

SELECT
  CAST(order_id   AS INTEGER)                     AS order_id,
  CAST(order_date AS DATE)                        AS order_date,
  TRY_CAST(revenue AS FLOAT)                      AS revenue,
  CASE
    WHEN LOWER(is_returned) IN ('true','1','y','t','yes') THEN TRUE
    ELSE FALSE
  END                                             AS is_returned
FROM orders;
-- TRY_CAST returns null on failure (PostgreSQL: CAST with error catch)

import polars as pl

df = df.with_columns([
    pl.col("order_id").cast(pl.Int64, strict=False),
    pl.col("order_date").cast(pl.Utf8).str.to_datetime(strict=False).dt.date(),
    pl.col("revenue").cast(pl.Float64, strict=False),
    pl.col("is_returned").map_elements(
        lambda x: str(x).lower() in ["true", "1", "y", "t", "yes"],
        return_dtype=pl.Boolean
    ),
])

Operation Reference

Frequently Asked Questions

Next Steps

Need help? Email support@edilitics.com with your workspace, job ID, and context. We reply within one business day.