---
title: Support to iterate datasource schema from the UI
excerpt: 'Change the schema of your Data Source from the UI. Also, select or de-select all columns at once when new columns are detected'
date: '2024-07-25'
---

# Change Data Source schema from the UI

It's now easier to update your Data Source schema from the UI:

{% image src="/assets/changelog/column-options.png" alt="Column options" /%}

Now, you can drop an existing column from the Data Source schema directly from the UI:

{% image src="/assets/changelog/drop-column.gif" alt="Drop column" /%}

You can also make a column nullable.

{% image src="/assets/changelog/make-nullable.gif" alt="Make nullable" /%}

And you can set a default value for a column.

{% image src="/assets/changelog/change-default.gif" alt="Set default" /%}

## De-select/select all columns at once when new columns are detected

In addition to these changes, you can now also select or de-select all columns at once when there are more than 10 columns detected. Who has time to select columns individually when SPEED WINS?!

{% image src="/assets/changelog/column-select-all.gif" alt="Column select all" /%}

## Small improvements and bug fixes

- Added support to `json` and `jsonl` formats when creating S3 Data Sources. You can read more at [S3 docs](/forward/ingest-data/connectors/s3).
- You can now create an API Endpoint from the UI, even if there is a SQL error in the Node. This feature-matches what was already possible when using the CLI, and is helpful for users validating parameters in an Endpoint.
- The "Processed data" metric in any Endpoint Chart shows now the **total** processed, instead of the average.
- Any `String` columns sent to quarantine are no longer encased in double quotes (`""`), making it easier to recover data from quarantine.
