Anonymise PDF¶

The Anonymise PDF tool lets you produce a redacted copy of a bank statement PDF suitable for sharing — for example, when attaching a failing statement to a GitHub issue. All letters are scrambled by default; you control exactly which numeric strings (sort codes, account numbers) are also scrambled, and which structural words must remain intact so that the parser can still process the anonymised file.

Opening the tool¶

The Anonymise tool can be opened from two places:

• Admin dialog → Anonymise PDF section → Open Anonymise Tool button.
• Debug Info dialog → Anonymise button on any REVIEW or FAILURE row — the tool opens pre-loaded with that statement's PDF path.

Selecting a source PDF¶

Use the Browse… button to open a file picker and select the PDF you want to anonymise. The full path is shown in the field to the left of the button.

If the tool was opened from the Debug Info dialog, the path is pre-filled with the failing statement's path.

Editing anonymise.toml¶

The central panel shows the contents of config/user/anonymise.toml inside the active project folder. This file is created automatically when the project is initialised, pre-populated with sensible defaults including common UK bank transaction type codes and month names.

Edit the file directly in the text editor. Three sections control the anonymisation:

[numbers_to_scramble]¶

A list of substrings. Any PDF text fragment containing one of these strings has its digit characters replaced with random digits. Separators (hyphens, spaces) are preserved.

Use this to scramble sort codes and account numbers that would otherwise be left as-is:

[numbers_to_scramble]
values = [
    "11-22-33",   # sort code
    "12345678",   # account number
]

[words_to_not_scramble]¶

A list of words and phrases that must appear unchanged in the output. Matching ignores case and all whitespace, so multi-word phrases are matched even when the PDF renders them across multiple text fragments.

Pre-populate this with any structural text the parser needs to read — column headers, balance labels, transaction type codes:

[words_to_not_scramble]
exclude = [
    "Balance Brought Forward",
    "Date",
    "Money In",
    "Money Out",
    # ... bank-specific phrases
]

[filename_replacements]¶

Key/value pairs applied to the output filename stem before the anonymised_ prefix is prepended. Use this to remove real names or account references from the filename:

[filename_replacements]
"RealSurname" = "Testname"

Saving the config¶

Click Save Config to write your edits back to disk. The editor validates that the file is valid TOML before saving; an error dialog is shown if the syntax is invalid.

Running the anonymisation¶

Click Run Anonymisation. The tool calls bsp.anonymise_pdf in a background thread so the UI remains responsive. The status line updates when the run completes, showing the full path of the output file.

The anonymised PDF is written alongside the source file with anonymised_ prepended to the filename (after any filename replacements are applied).

Viewing the results¶

Once a run completes, Open Original PDF and Open Anonymised PDF both become active. Click either button to open the file in your system's default PDF viewer. Open both to compare them side-by-side.

anonymise.toml location¶

The config file lives at:

<project folder>/config/user/anonymise.toml

It is created automatically when a project is first initialised or connected. You can also edit it directly outside of openstan with any text editor.