Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
In the Fabric portal, you can use Load to tables from the Lakehouse home page to turn CSV or Parquet files into Delta tables.
Note
In the current Load to tables flow, you can't define a custom column schema (for example, explicit column names and data types) from the Lakehouse home page. Use a notebook when you need explicit column schema control.
What you can do in the Fabric portal
In the Fabric portal, the Lakehouse home page supports the following actions when you use Load to tables:
- Start from either a single file or a folder.
- Load into either a new table or an existing table.
- Load CSV or Parquet data and keep output in Delta format with V-Order optimization.
For field-by-field behavior (including when options like File type, Append/Overwrite, Column header, and Separator appear), see Fields by load path.
Fields by load path
When you use Load to tables, the dialog title and fields change based on your path:
- Source: file or folder.
- Target: new table or existing table.
- Format: CSV or Parquet.
Dialog titles follow those choices (for example, Load file to new table, Load folder to new table, Load file to existing table, or Load folder to existing table).
Fields by scenario
Use the following table to quickly identify which fields appear for each load scenario, what values to provide, and how validation behaves.
| Field | When you see it | What to enter or select | Validation and behavior |
|---|---|---|---|
| Schema (dropdown) | New table and existing table paths | Destination Lakehouse schema namespace | Select the namespace where the table is created or updated. |
| New table name | New table paths only | Name for the destination Delta table | Use alphanumeric characters and underscores (_) only, up to 256 characters. Dashes (-) and spaces aren't allowed. |
| Load mode (Append or Overwrite) | Existing table paths only | How to load into the selected existing table | Choose Append to add rows, or Overwrite to replace existing data. |
| File type | Folder paths only | Folder load format (CSV or Parquet) | All files in one folder load action must match the selected file type. |
| Column header (checkbox) | CSV paths only | Whether to use the first row as column names | If checked, Fabric uses first-row values as column names. If not checked (or headers don't exist), Fabric assigns defaults such as _c0, _c1, _c2. |
| Separator (text box) | CSV paths only | CSV delimiter value | Can't be empty, can't be longer than 8 characters, and can't include (, ), [, ], {, }, ', ", or whitespace. |
| Column names (resulting names) | CSV paths after header handling | Resulting column names from file headers or defaults | Names can include English letters (upper or lower case), underscores (_), and UTF characters (for example, Chinese), up to 128 characters. Invalid characters are replaced with underscores. If a valid name can't be produced, the load fails. |
After required fields are set for your path, select Load.