Corrupt a Text File Online
Corrupt a .txt file online for free. Injects invalid UTF-8 byte sequences so editors show mojibake or refuse to decode the file entirely.
- Aucun envoi
- Aucune limite de taille
- Aucune inscription
Corrompez votre fichier text
Réglé pour .txt, .log, .ini, .conf, et fonctionne tout aussi bien avec n'importe quel autre format.
Pourquoi les fichiers text sont difficiles à corrompre correctement
Plain text is the one case where scrambling bytes genuinely is enough, but scrambling them into other printable characters just produces different text, not a broken file. The trick is injecting byte sequences that are not valid UTF-8 at all, so the file stops being decodable rather than merely becoming nonsense.
Ce que cet outil détruit
- Invalid UTF-8 sequences. Bytes in the
0x80–0xFFrange are injected without valid lead bytes, which is a decoding error rather than a strange character. - The opening bytes. The start of the file is overwritten, so any byte-order mark or format sniffing is defeated immediately.
- Line structure. Newline boundaries are broken up, so tools that read the file line by line lose their framing.
- The tail. The final bytes are replaced, which breaks any trailing marker or terminating newline a parser expects.
Ce que vous verrez ensuite
| Application | Comportement |
|---|---|
| Notepad / TextEdit | Opens showing replacement characters and mojibake instead of readable text. |
| VS Code | Warns that the file contains invalid characters for the detected encoding. |
Python open() | Raises UnicodeDecodeError in strict mode. |
| grep / diff | Reports the file as binary and refuses to process it as text. |
Raisons courantes de le faire
- Testing that a log ingestion pipeline survives a partially written file.
- Verifying an importer handles encoding errors rather than crashing.
- Producing a broken fixture for a text-parsing test suite.
- Demonstrating what encoding corruption looks like in a training session.
Comment corrompre un fichier text
- Déposez votre fichier .txt dans la zone ci-dessus, ou cliquez pour parcourir.
- Laissez le niveau sur Standard, sauf si vous voulez des dommages partiels (Light) ou devez mettre en échec un logiciel de récupération (Heavy).
- Cliquez sur Corrompre le fichier. Le traitement a lieu sur votre appareil, il se termine donc presque instantanément.
- Téléchargez le résultat. Par défaut, il conserve le nom de fichier et la longueur en octets d'origine.
Questions fréquentes sur Text
Why not just shuffle the letters?
Shuffled letters are still valid text, the file opens perfectly and simply reads as nonsense. Injecting invalid UTF-8 makes the file fail to decode, which is what real corruption looks like.
Will the file still open in Notepad?
Notepad is very permissive and will usually display something, but it will be mojibake rather than your content. Stricter tools refuse it outright.
Does this work for source code files?
Yes. Any text-shaped extension: .py, .js, .java, .php and so on, uses the same profile.
Is the original file changed?
No. Corruption is applied to a copy in memory and offered as a download. The file you selected is only read.