How to normalise messy whitespace and control characters

1. Paste the text

   Drop the source into the "Text" field. Typical inputs: copy-pasted PDF text with weird line wraps, web-scraped strings with non-breaking spaces, or messages full of zero-width characters used to hide trackers.

2. Press Run

   Result returns a single `cleaned` field. Multiple consecutive spaces collapse to one; runs of newlines collapse to one blank line max; tabs become single spaces; zero-width and BOM characters disappear.

3. What gets touched and what doesn't

   Letters, digits, regular punctuation, and Unicode beyond ASCII (like emoji and CJK characters) are left alone. Only whitespace, control codes, and invisible-format characters are normalised.

4. When NOT to run it

   Source code with significant whitespace (YAML, Python, Markdown indentation) would be corrupted — tab → space changes meaning. CSV with empty fields could lose them if leading/trailing spaces matter. Diff or sample first.