n8n Extract from File nodeConfigure Extract from File in n8n.

Transforms a CSV file into output items. Comma separated values is the format every tool exports to, and this operation is the one you reach for most often.

• Input Binary Field: required, the name of the input field containing the file data to be processed. Point it at the field the previous node produced, often {{ $json.data }}.
• Delimiter: sets the field delimiter, usually a comma. Worth setting when the export uses semicolons.
• Header Row: tells the node whether the first row of the file contains the header names. Leave it off and your columns come back numbered.

Takes a table inside an HTML file and turns it into output items. Useful when a system publishes its report as a web page instead of a data file.

• Input Binary Field: required, the field holding the HTML file to be processed.
• Header Row: whether the first row of the file contains the header names, which decides whether your keys read as labels or positions.
• Include Empty Cells: switch it on to keep empty cells, filled with an empty string, so every row keeps the same shape.

Reads an iCalendar file and turns it into output items. Calendar invitations travel as .ics attachments, and this is how a workflow reads what is inside one.

• Destination Output Field: required here, the name of the output field that will contain the extracted data, placeholder e.g data.
• Keep Source: choose json, binary or both to carry the input item forward alongside the extraction.

Pulls JSON data out of a binary file. The content is already structured, but as long as it sits in a file the workflow treats it as bytes, so this operation unwraps it.

• Destination Output Field: required, and it names the output field that will contain the extracted data.
• File Encoding: specify the encoding of the file, defaults to UTF-8. The list runs from ascii to windows1258.

Transforms an ODS file into output items. Open document spreadsheets come out of office suites that are not Excel, and the node reads them with the same set of options.

• Sheet Name: name of the sheet to read from in the spreadsheet, placeholder e.g. mySheet. Without it you are at the mercy of whichever tab comes first.
• Range: the range to read from the table. A number is read as the starting row, a string as an A1 style notation range.

Extracts the content and metadata from a PDF file. Contracts, invoices and reports arrive in this format constantly, and the text inside them is unreachable until this operation runs.

• Max Pages: maximum number of pages to include. Handy on long documents where only the first pages carry the data.
• Join Pages: on, the text of all pages is joined; off, you get an array of text per page.
• Password: fill it in if the PDF is encrypted, otherwise the file stays closed.

Turns a table held in a rich text format file into output items. Older document workflows still produce RTF, and the node treats it like any other tabular source.

• Read As String: in some cases and file formats it is necessary to read as string so special characters are interpreted correctly. Accented text is the usual reason to enable it.
• RAW Data: returns raw data instead of parsing it, which is how you inspect what the file really contains when the result looks wrong.

Extracts the content of a text file. No structure, no columns, just the characters, dropped into the field you name.

• Destination Output Field: required, the name of the output field that will contain the extracted data, placeholder e.g data.
• Strip BOM: strips the byte order mark from the file, which helps in an environment where the presence of the BOM is causing issues or inconsistencies.

Plenty of accounting and logistics systems still export nothing but XML. This operation extracts the content of that file and makes it usable in a workflow.

• Input Binary Field: required, the field carrying the XML file, placeholder e.g data.
• Destination Output Field: required too, and it decides under which key the extracted tree appears in the output item.

Transforms an Excel file into output items, in the older XLS format. Files that have been passed around for years are often still saved this way.

• Sheet Name: name of the sheet to read from in the spreadsheet. Set it whenever the file carries more than one tab.
• Max Number of Rows to Load: stops handling records after the requested number of rows. Use -1 to load them all, placeholder e.g. 10 for a quick test.

Same treatment for the current Excel format: the XLSX file becomes output items. This is the operation that replaces the weekly export someone opens by hand.

• Header Row: whether the first row of the file contains the header names, so your items come back keyed by column title.
• Starting Line: start handling records from the requested line number, counting from 0, placeholder e.g. 0. Set it when the file opens with a logo and three blank rows.

Converts a file into a base64 encoded string, a text friendly way of carrying bytes. Nothing is parsed here: the file becomes a value you can put in a field.

• Input Binary Field: required, the field that currently holds the file.
• Destination Output Field: required, the output field that will contain the extracted data, which is the string itself.
• Keep Source: json, binary or both, depending on what the next node still needs from the input item.