Working with workspaces¶

Create and work with files and folders¶

In a workspace you can use a familiar IDE and source control conventions to author, organize, and run code.

1. Sign in to Snowsight.
2. In the navigation menu, select Projects » Workspaces.
3. Select the + next to the appropriate folder. If you’re using Workspaces for the first time, select + Add New.
4. Select from the following options to create a new file or folder or to upload an existing file or folder:
   • SQL File: Creates a new, blank SQL file in the editor as a tab in the editor. By default, .sql is appended to unnamed files. The editor recognizes it as a SQL file and enables syntax highlighting and autocomplete.
   • File: Creates a new file. Name the file and its extension. If the extension is recognized by the editor (for example, Java, JavaScript, or Scala), code highlighting and autocomplete are enabled.
   • Folder: Creates a new, empty folder inside the workspace.
   • Upload Files: Upload one or more files to any location in your workspace. The editor uses the file extension and applies the appropriate icon, behavior, and syntax highlighting to the file when it’s opened. For example, .sql files show SQL-specific features.
   • Upload Folder: Select one or more files or folders to add to the selected workspace.

Manage files¶

You can rename, delete, move, and organize your workspaces, files, and folders.

To rename or delete a workspace, file, or folder, follow these steps:

1. Hover over the target and select the vertical ellipsis (more actions).
2. Select Rename or Delete. If you choose to delete, you are prompted to confirm.

• To create a folder in a workspace, select the + next to the workspace or an existing folder.
• To organize files and folders, drag any file or folder into a different location in the same workspace. You can also drag a worksheet into a workspace.

Format SQL code¶

Workspaces include a built-in functionality to format and standardize SQL code for improved readability and maintenance.

1. In the Workspaces editor, select the horizontal ellipsis.

   

2. Select Format SQL or use the keyboard shortcut command + shift + O (Windows: CTRL + Alt + O).

Organize sections of code¶

Use code folding to collapse and expand large blocks of code, allowing you to focus on specific sections and improve overall code navigation.

1. In the Workspaces editor, locate the code section to collapse.
2. Hover the mouse to the right of the line numbers. A code folding icon () appears at the fold line.
3. Toggle the icon to fold or unfold the section of code.

View multiple files or results in one layout¶

Managing multiple files with tabs and split panes offers several advantages:

• Compare code or results side by side: Quickly reference one worksheet query while working on another.
• Multitask more efficiently: View different cells, outputs, or files at once with less switching.

To adjust your Workspaces layout, select the vertical ellipsis () in the Workspaces pane and choose the appropriate option:

1. Split right
2. Split down
3. Close others

Exploring query results¶

When you run a query in Workspaces, you can use interactive features to filter, analyze, and explore your results without writing additional SQL. These features help you quickly understand your data and identify patterns.

Use interactive column statistics¶

Each column in your query results includes interactive visual statistics (mini graphs or histograms) that help you understand data distribution and quality. You can click these statistics to open a detailed panel and create filters.

To view column statistics:

1. Run a query in a Workspaces SQL file.

2. In the results table, show the column statistics by doing one of the following:

   • Click Show column stats in the top-left corner of the table (next to the column headers).
   • Click the ellipsis button in any column header and select Show column stats.

   Mini graphs (histograms or distribution charts) appear in each column header showing the data distribution.

3. Click a histogram to view sum and average values for that entire column in the bottom-right of the table.

   Alternatively, you can select a range of cells in the results table to view statistics in the bottom-right of the table. For numeric columns, sum and average values are displayed. For non-numeric columns, the count is displayed.

To filter using column statistics:

1. Click Show column stats in the top-left corner of the results table, or click the ellipsis button in any column header and select Show column stats.

2. Click the histogram for the column you want to filter by. A popover displays detailed statistics for that column, including:

   • Sum and average values for numeric columns
   • Distribution charts showing value frequency
   • Data quality metrics such as null and filled percentages

3. In the popover, select the values or ranges you want to filter by.

4. Select Apply to apply the filter to your results.

This interactive filtering helps you explore your data visually without writing WHERE clauses or other SQL filter logic.

Inspect cell values¶

The cell inspector provides detailed information about individual cells or selections in your query results.

To inspect a single cell:

1. In the results table, double-click any cell to open the Inspector Panel.
2. Review the detailed value, including formatting and data type information.

To view aggregate statistics for multiple cells:

• In the results table, select multiple cells by clicking and dragging across rows and columns.

  A statistics bar appears at the bottom showing:

  • Sum of numeric values
  • Average of numeric values
  • Count of selected cells
  • Min and max values in the selection

This feature is useful for quick calculations and data exploration without creating new queries.

Keyboard shortcuts¶

Worksheets provide keyboard shortcuts to help you quickly navigate, customize your view, and edit queries. The following table identifies commonly used keyboard shortcuts:

Recover a workspace from a dropped user¶

When a user is dropped, their personal database (PDB) and all files within their workspaces are retained. The PDB is renamed to DROPPED_USER$<dropped_user_name>_<timestamp> and is owned by the primary role that was active when the DROP USER statement ran (referred to here as <owning_role>).

Recovery usually involves two people:

• an administrator — someone with the privileges to drop users, with access to <owning_role>
• the recovering user — the person inheriting the work from the dropped user, who wants the files in their own private workspace.

In order for the recovering user to recover deleted workspace into their own private workspaces they must run the final CREATE WORKSPACE step themselves targetting their own USER$ PDB.

To locate the dropped user’s PDB the administrator lists them and identifies the one to recover, those that start with the reserved prefix DROPPED_USER$, then they locate the specific workspaces inside to recover:

SHOW DATABASES STARTS WITH 'DROPPED_USER$';
-- surfaces database "DROPPED_USER$user_123"
SHOW WORKSPACES IN DATABASE "DROPPED_USER$user_123";
-- surfaces workspace "DROPPED_USER$user_123".PUBLIC."deleted_ws"

Option 1: Use the owning role. Use this when the recovering user already has access to <owning_role>, or when <owning_role> can be granted to them.

-- Administrator (skip if the recovering user already has the role):
GRANT ROLE <owning_role> TO USER <recovering_user>;

-- Recovering user:
USE ROLE <owning_role>;

The administrator can revoke the role after recovery with REVOKE ROLE <owning_role> FROM USER <recovering_user>.

Option 2: Create a temporary recovery role. Use this when you don’t want to grant <owning_role>. An administrator creates a role scoped to just the dropped PDB and grants it to the recovering user.

-- Administrator:
CREATE ROLE recovery_temp;
GRANT ROLE recovery_temp TO USER <recovering_user>;
GRANT USAGE ON DATABASE "DROPPED_USER$user_123" TO ROLE recovery_temp;
GRANT USAGE ON SCHEMA "DROPPED_USER$user_123".PUBLIC TO ROLE recovery_temp;
GRANT READ ON WORKSPACE "DROPPED_USER$user_123".PUBLIC."deleted_ws" TO ROLE recovery_temp;

-- Recovering user:
USE ROLE recovery_temp;

The administrator can drop the role after recovery with DROP ROLE recovery_temp; the recovered workspace is unaffected.

Recover the workspace. With a role that has access active, the recovering user runs the following steps.

1. List the origin workspace files to confirm access:

   Copy code

   LS 'snow://workspace/"DROPPED_USER$user_123".PUBLIC."deleted_ws"/versions/head/';
   

2. Create a copy of the workspace into your own PDB, then add a live version so it’s ready to edit:

   Copy code

   CREATE WORKSPACE USER$.PUBLIC."recovered_ws"
     FROM 'snow://workspace/"DROPPED_USER$user_123".PUBLIC."deleted_ws"/versions/head';
   ALTER WORKSPACE USER$.PUBLIC."recovered_ws" ADD LIVE VERSION FROM LAST;
   

Limitations¶

• Column statistics may take longer to generate as the number of columns increases.
• Snowflake Copilot is not available in Workspaces.
• Query filters are not supported. Any queries containing filters will fail.
• Workspaces files are not included in Universal Search results.
• Opening and editing the same worksheet in the new Workspaces UI and old Worksheets UI simultaneously can result in lost changes.
• For worksheets, execution context settings (role, warehouse, and namespace) are not synchronized across the new Workspaces UI and the old Worksheets UI.