Skip to main content

​
Description

With this manipulator you import tabular CSV data into a notebook.

​
Application

To do any meaningful in a notebook you have to start with a data import. You are likely to start your notebook with this manipulator to import tabular data and start working with it.

​
How to use

The importer dialog consists of three sections as shown in the image below.
  1. Import type option.
  2. Selected files are displayed here
  3. Click this button to open a file/folder selection dialog
  4. Assign a name to the dataset which is created from the imported file(s). All other manipulators are going to refer to the imported dataset via this name.

​
Import Type

The import type in the top left of the Tabular importer drives the way the File Selection Dialog is configured which opens if you click the button Add/Edit datasets.

Files

Select data on a file-by-file basis. This is the default mode. Read here how to select files.

Folders

Enables to load files according to a file name pattern from a set of folders. This option even lets you import files recursively from a deeper folder structure. How to use the Folders option is explained in depth here.

​
Data file requirements

The data files you select will need to follow certain requirements:
  • File format: CSV
  • Decimal sign: dot (.)
  • File encoding: utf-8
  • Multiple files loaded together must all contain the same columns
Importing multiple files. When you select several files, their rows are stacked into a single dataset — the rows of one file follow the next — and columns that share the same name are combined into one column. Rows are not aligned on any index, so files that share values (for example the same timestamps) produce one row per file rather than a single merged row. All selected files must contain the same columns: unlike the MF4 Loader (which combines files with differing channels and leaves missing values blank), a Tabular import whose files have different columns will fail. If you need to combine files with different columns, import them separately and join them downstream.

​
Advanced Options

To show the advanced options click on Show/Hide Advanced Options in the bottom right part of the edit dialog. With these options you can control the following parameters:

Column Separator

  • Comma (default)

  • Semi-Colon

  • Tabs

  • Spaces

For Spaces you can use any number of spaces between two columns.

Header row

Here you can specify which line in your file is the header line with the column names.

This settings becomes important if you have a longer header which includes additional meta-data like test-ID, testbench-ID, test operator, …

The index is zero-based, meaning the first line in your file is row 0, the second line row 1, and so on.
Every physical line in the file counts toward this index — including blank rows, metadata rows, or partially empty rows.
Select the header row based on the exact line number in the raw file (if you are familiar with looking at your data on a spreadsheet, you should think spreadsheet row number − 1).

Data start row

This option lets you specify in which line your actual data body starts.

This option is relevant if you have a longer header and your data line doesn’t start in the second line.

Your data section doesn’t have to start in the line after the header; a gap between header and data section can be handled.

The data start row also uses zero-based indexing and refers to the literal row number in the file.
If this row is blank, that blank row will be imported as the first data row.
The importer does not skip or ignore blank lines, so ensure you account for them when selecting where data begins.

​
How Blank Rows Are Treated

Blank rows are treated as normal rows and counted in the row index.
They are not skipped, collapsed, or automatically removed.
This ensures consistent and reproducible imports, but means you must account for any blank or spacer rows when selecting the header row and data start row.

​
Column Formatting Limitations

  • Column names in notebooks will only contain:
    • Letters: a-z
    • Underscore: _
    • Numbers: 0-9 (except as first character)
  • Other characters (whitespaces, parenthesis, quotation marks, semi-colons, commas, curly brackets, equal signs, etc.) are replaced by an underscore when loaded (sanitisation)
  • Column names must be unique (case insensitive)
  • All letters are lowercased during data processing
  • Column names can’t be more than 255 characters
Before you import, please confirm your files follow these rules. Otherwise, the Monolith platform may change column names or misrepresent your data during import.
Last modified on July 23, 2026