Remove extra spaces and exact copy/paste artifacts, normalize selected punctuation, and keep your Unicode text private in your browser.
Cleaning stays in this browser. Press Escape, then Tab, to move focus out of the editor.
0 / 100,000 characters
Named presets have fixed definitions. Editing any checkbox creates a temporary Custom selection.
Each tab becomes this fixed number of spaces. The editor’s Tab key inserts five spaces independently.
Review exactly what changed before copying or applying the complete cleaned text.
Your result
Your cleaned text will appear here
Enter or paste text, then choose a preset or individual cleanup operations.
Text Cleaner applies only the operations you select, in a fixed deterministic order. It is designed for copied text from PDFs, rendered webpages, spreadsheets, email, documents, content systems, logs, captions, and generated drafts.
The source remains editable and undoable. The complete cleaned result stays in browser memory for review, copying, or confirmed application back to the source.
Whitespace options are deliberately narrow and do not remove line breaks or infer document layout.
Remove the exact Text Cleaner horizontal-whitespace set only at the source boundaries or physical line endings.
Replace each tab with one through eight fixed spaces and convert only U+00A0 NBSP to a normal space.
Collapse runs of ordinary U+0020 spaces after selected tab and NBSP conversion.
Text Cleaner horizontal whitespace is TAB, U+0020, U+00A0, U+1680, U+2000–U+200A, U+202F, U+205F, and U+3000. The product does not use ECMAScript \s or a broad Unicode whitespace category.
Vertical Tab and Form Feed are handled exclusively by Remove Control Characters. CR, LF, NEXT LINE, Unicode line and paragraph separators, and U+FEFF are also excluded from horizontal-whitespace cleanup, which prevents overlapping summary attribution.
Remove only zero width space, word joiner, soft hyphen, and U+FEFF while retaining language and emoji shaping characters.
Optionally convert the documented curly quotes, typographic dashes, and ellipsis toward ASCII punctuation.
Deliberately remove the documented C0, DEL, and C1 ranges while always preserving TAB, LF, and CR.
Default cleanup preserves Hindi, Arabic, Persian, Hebrew, CJK, accented text, combining marks, mathematical Unicode, emoji sequences, ZWJ, ZWNJ, variation selectors, modifiers, and bidirectional controls.
No custom selection is universally safe for every document. Review destructive artifact, control, and typography choices when exact source semantics matter.
Text Cleaner handles character-level whitespace, selected hidden artifacts, exact controls, and optional punctuation normalization. It preserves physical line structure.
Remove Line Breaks owns structural work: joining lines, removing or collapsing blank lines, trimming every line, and normalizing line endings. Keeping these responsibilities separate avoids hidden transformations.
The editable source accepts up to 100,000 Unicode code points. The cleaning engine has no matching output ceiling, so replacing many tabs can legitimately produce a much larger complete result.
An oversized result still receives a grapheme-safe preview and can be copied completely. Apply Result is disabled instead of truncating or partially replacing the source.
Text Cleaner does not repair grammar, spelling, OCR, encoding, markup, or document layout. It does not strip HTML, decode entities, or remove content by script or character class.
Literal HTML, Markdown, BBCode, URLs, email addresses, numbers, symbols, and other punctuation remain unchanged unless one exact selected operation documents otherwise.
Last reviewed: July 2026
Technical review: Lalit Giani
Horizontal-whitespace membership, artifact removal, control removal, typography mappings, and preset memberships are explicit TextCraftHub product definitions. They are intentionally narrower than broad Unicode or ECMAScript character categories.
No. Text Cleaner processes the source and result locally in your browser. It does not upload your text, put it in a URL, or save persistent history.
Basic Cleanup trims the documented horizontal whitespace at the outer boundaries, collapses repeated normal spaces, replaces tabs, replaces U+00A0 non-breaking spaces, and removes trailing horizontal whitespace from physical lines.
Text Cleaner uses an exact product-defined set: TAB, normal space, NBSP, Ogham Space Mark, U+2000 through U+200A spacing characters, narrow NBSP, medium mathematical space, and ideographic space. It does not use a broad whitespace expression.
They are control characters in Text Cleaner’s explicit model and are handled only by Remove Control Characters. Keeping the operations separate makes behavior and summary attribution predictable.
No. It preserves CRLF, LF, and lone CR line structure. Use Remove Line Breaks to join lines, remove or collapse blank lines, or normalize line endings.
The option removes exactly U+200B ZERO WIDTH SPACE, U+2060 WORD JOINER, U+00AD SOFT HYPHEN, and U+FEFF. It is not an all-invisible-character filter.
Yes by default. ZWJ, ZWNJ, variation selectors, emoji modifiers, combining marks, bidirectional marks, and multilingual text are preserved. Exact destructive options still change only the documented characters.
It removes C0 controls except TAB, LF, and CR, plus DELETE and the C1 control range. It is an advanced destructive option and no preset enables it.
Named preset definitions never change. Editing a checkbox switches the displayed preset to Custom. Selecting any named preset again discards temporary Custom selections and restores that preset’s exact canonical operations.
No. It works one way toward ASCII: four curly quote characters become straight quotes, U+2010 through U+2014 become hyphen-minus, and the ellipsis character becomes three periods.
No. Literal markup and entities remain unchanged. Text Cleaner does not use innerHTML, DOMParser, or regex-based tag stripping.
Apply waits until the current source and options have finished cleaning. After confirmation, the result is recomputed synchronously from the latest source and options before it can replace the editor text.
The editable source accepts at most 100,000 Unicode code points, but cleanup can expand output—for example when tabs become multiple spaces. Oversized results remain complete and copyable but cannot be applied to the limited editor.
No. Large results show 500 complete grapheme clusters by default, while Copy Result always uses the complete in-memory output.
TextCraftHub uses cookies and similar technologies for analytics and advertising, including Google AdSense, according to your choices. You can accept all, reject all, or manage your preferences. Privacy Policy
