Text / OCR Cleaner

• Strip invisible charactersRemove BOM, zero-width characters, and stray control characters (keeps newlines and tabs).
• Normalize Unicode spacesReplace no-break, narrow no-break, figure, thin, hair, en/em, and similar spaces with a normal space (common in PDF text).
• Normalize whitespaceCollapse runs of spaces and tabs on each line into a single space.
• Trim trailing spacesRemove spaces and other trailing whitespace at the end of each line.
• Trim leading whitespace per lineRemove indentation at the start of each line. Turn off when you need preserved indents (lists, code, poetry).
• Tidy punctuation spacingRemove spaces before . , ! ? ; : and tighten spaces around ( ) [ ] { }. English-oriented; turn off for French-style punctuation.
• Join hyphenated line breaksMerge lines when a word breaks at end-of-line hyphen and the next line starts with a lowercase letter (e.g. book OCR).
• Reflow paragraphs (unwrap soft breaks)Join consecutive non-empty lines into one line per paragraph; blank lines still separate paragraphs. Heuristic: turn off for tables or verse.
• Remove duplicate linesDrop consecutive identical lines (common when OCR repeats a row).
• Collapse blank linesReplace multiple empty lines in a row with a single blank line.
• Trim leading & trailing blank linesAfter all other steps (including line regex and find/replace), remove empty or whitespace-only lines at the start and end.

Runs after the options above. Each line is tested against the pattern; matching lines are removed. Invalid patterns are ignored until fixed.

Literal text only (not regex). Multiline strings are allowed. Rules run top to bottom; each step sees the result of the previous one.

• Rule 1Remove
  Find

  Replace with