Skip to content


The Firefly III data importer (FIDI) generates an import file each time you import data. You can download the configuration file during or after the import.

You can use it to quickly configure your import the way you want to.

Pre-made import files

There's a repository on GitHub with import configurations for common banks and financial institutions .

Example file

This is a typical configuration file with all options:

  "date": "Ymd",
  "default_account": 49,
  "delimiter": "comma",
  "headers": true,
  "rules": true,
  "skip_form": false,
  "add_import_tag": true,
  "specifics": [
  "roles": [
  "do_mapping": {
    "3": true,
    "4": true,
    "0": false,
    "1": false,
    "2": false,
    "5": false
  "mapping": {
    "3": {
      "Savings Account": 3
    "4": {
      "Savings Account": 3
  "duplicate_detection_method": "classic",
  "ignore_duplicate_lines": false,
  "ignore_duplicate_transactions": true,
  "unique_column_index": 1,
  "unique_column_type": "external-id",
  "version": 3,
  "flow": "csv",
  "identifier": "123456",
  "connection": "789543",
  "ignore_spectre_categories": false,
  "map_all_data": false,
  "accounts": [

  "date_range": "all",
  "date_range_number": 30,
  "date_range_unit": "d",
  "date_not_before": "2021-01-01",
  "date_not_after": "2021-06-01",
  "nordigen_country": "NL",
  "nordigen_bank": "ABN_AMRO_XY",
  "nordigen_requisitions": [],
  "conversion": true

Fields and values


Sets the date format of the date entries in the CSV file. If your file contains internationalized dates, you can prefix the date format with your country code, like it: or nl:. Then, enter your date format.

Read more about the format in the PHP documentation.

Here are some examples:

  • Ymd. Will convert 20210318
  • F/j/Y. Will convert January/17/2021
  • nl:d F Y. Will convert Dutch date 5 mei 2021


The ID of the default asset accounts into which transactions will be imported if there's no source account. This is some time useful when files just list the transactions and the destination, nothing more.


The delimiter of the entries in the CSV files. Most files use "comma", but you can also use "semicolon" or "tab".


True or false if the CSV file has headers, a top row that explains what each column is for.


Whether or not to apply your rules to the import. It is useful to run your rules after the import although it will significantly slow down the import. The execution of rules can be slow when you have a lot of them.


If checked, next time you use the importer it will skip the configuration form. You can always go back to the form when using the user interface.


"Specifics" are the bank specific scripts that can be applied to your import. They are decrepated and may be removed in a future version of the CSV importer. You can read more about them on the page about import configuration.


Each column in your CSV file gets a single role. This array in the JSON file indicates the role for each column. The user interface shows you all the possible roles, but you can also read the code to learn which roles are available.


Whether or not these columns should be 'mapped' to data in your Firefly III installation. Mapping is a process where values in your CSV file will always point in a specific value in your Firefly III installation. For each column, it will tell the CSV importer if it should use the mapping in the next array. In the user interface, this is indicated by little check boxes after each column.

Here are some examples:

  "do_mapping": [

Column 0 and column 1 should be mapped, and column 2 should not. Keep in mind that the array counts from 0, so in your CSV file this would indicate: map the first two columns but don't map the third column.

In some JSON files, the order of true and false gets mixed up, which is why you'll see something like this:

  "do_mapping": {
    "0": true,
    "2": false,
    "1": true

What you see here is the same thing as the previous example. Map the first two columns, don't map the last one.


The mapping. The value in quotes is what's found in the CSV file. The numeric ID links to the object in your Firefly III installation. Here are some examples:

  "0": {
    "Groceries": 3,
    "Grozeries": 3,
    "Bills": 2,
    "Going out": 21

The 0 in this example refers to the first column, because we count from zero. In this example, if column 0 contains either Groceries or the misspelled variant the CSV importer should link to a category with ID #3. From the mapping array alone you can't tell that I'm talking about columns. This is something you can find or set in the roles array.

Another example:

  "4": {
    "NL21INGB9861487085": 15,
    "NL35ABNA6289099205": 23,
    "289099205": 23

This example maps the account numbers in column #4 to account ID's in Firefly III.

You can configure a mapping for any column, but not all roles allow themselves to be mapped. For example, you can't map the "transaction amount". If you try, it will be ignored.


The CSV importer has three methods to detect duplicate transactions. The import configuration page has all the details. The JSON file reflects your choices.

  • none. Do no duplicate detection.
  • classic. Do content-based duplicate detection.
  • cell. Do identifier-based duplicate detection.

Again, to learn more about these options, read about import configuration.


Regardless of the choice of duplicate detection, you should leave this on true, so duplicated lines in your CSV files will be removed.


This variable is not really used anymore, but it's an indicator of which duplicate_detection_method you're using. For classic, it will always be true. Otherwise it's false.


This contains the column number that contains the unique value for the duplicate detection. It's only ever used when you use the cell-method, which stands for identifier-based duplicate detection. Read more about import configuration.


This contains the field type of the unique identifier. It's only ever used when you use the cell-method, which stands for identifier-based duplicate detection. Read more about import configuration.


The version should always be "3" so the importer knows how to parse your file.


If set to true, the import will the same tag to all transactions in the import.


Determines what is in your import. csv, spectre or nordigen


Spectre only, and contains the bank identifier.


Spectre only, and contains the bank connection.


Spectre only. Set to true, this will make the importer ignore Spectre's suggested categories.


Spectre and Nordigen only. Is an indication if you wish to map data.


Spectre and Nordigen only. A list of accounts from your provider that you wish to import, and into which Firefly III account you want to import.


Spectre and Nordigen only. What date range to import from your provider.

  • all: Import everything
  • partial: Go back date_range_number units of date_range_unit (see below)
  • range: Import a specific range


Spectre and Nordigen only. d for days, w for weeks, m for months, y for years. Used only if date_range is partial.


Spectre and Nordigen only. Number of days/weeks/months/years to go back. Used only if date_range is partial.


Spectre and Nordigen only. Used only if date_range is range. Will not import from dates before the set date.


Spectre and Nordigen only. Used only if date_range is range. Will not import from dates after the set date.


Nordigen only. The country of the bank you selected to import from.


Nordigen only. The bank you selected to import from.


Nordigen only. A list of valid "requisitions" (permissions to import) from your bank.


CSV only. If the importer should attempt to convert the file to UTF-8.

Last update: 2022-01-14