Unicode Character Inspector
Inspect each Unicode character, encoding, category, script and normalization form.
Mojibake Repair
Detect and repair text that was decoded with the wrong character encoding, then compare the most plausible results.
Repair processing occurs on the UnicodeNow server only after you press Find repair candidates. Submitted text is used only to generate repair candidates and is not stored in the database, Redis, analytics, URLs or application logs.
Select a candidate to populate this output. The original input is never overwritten.
Repair candidates
Paste garbled text to look for mojibake repair candidates.
Changed sequences
| Original sequence | Repaired sequence | Original code points | Repaired code points | Recovered bytes | Explanation | Pass |
|---|
No candidate selected.
What is mojibake?
Mojibake occurs when bytes written using one character encoding are decoded using another. For example, café stored as UTF-8 can appear as café when the UTF-8 bytes are interpreted through Windows-1252 or ISO-8859-1.
What this tool does
The tool detects common mojibake patterns, tries a bounded set of justified repair paths, scores and deduplicates candidates, explains each transformation, and lets you choose the result. Repair is heuristic and cannot be guaranteed.
Why several candidates may appear
Encoding repair can be ambiguous. Windows-1252 and ISO-8859-1 can produce the same result for some text, while punctuation and emoji often require Windows-1252. Showing candidates helps avoid silently choosing the wrong repair.
Common repair paths
Common paths include Windows-1252 → UTF-8, ISO-8859-1 → UTF-8, bounded double-encoded UTF-8 repair and selected legacy encodings from an allowlist. Arbitrary codec names are not accepted.
Double encoding
café
→ café
→ caféDouble encoding needs more than one bounded pass. The tool stops early when output stops improving, repeats, or reaches the configured pass limit.
Replacement character warning
� is U+FFFD REPLACEMENT CHARACTER. It usually means a decoder already lost byte information, so exact restoration may not be possible.
Can repair damage correct text?
Yes. Already-correct text such as café and it’s ready should not be repaired blindly. When no strong mojibake indicators are present, this tool warns and avoids automatic selection.
Mojibake versus Unicode normalization
NFC, NFD, NFKC and NFKD do not repair wrongly decoded bytes. Optional NFC normalization here is separate from encoding repair and is reported separately.
Common use cases
- Legacy database imports.
- CSV exports and CMS migrations.
- Broken email content and old application exports.
- Damaged curly punctuation, accented names and emoji.
- Encoding migration diagnostics.
Technical details
Processing is server-side and bounded. The service reconstructs bytes from visible mojibake using an explicit encoding allowlist, decodes those bytes as UTF-8, scores candidates with documented heuristics, deduplicates identical outputs, caps candidate count and never invokes arbitrary codecs or shell commands.
Privacy
Repair processing occurs on the UnicodeNow server. Submitted text is used only to generate repair candidates and is not stored in the database, Redis, analytics, URLs or application logs. Responses are returned with Cache-Control: no-store.
FAQ
What is mojibake?
Mojibake is garbled text caused when bytes written in one encoding are decoded using a different encoding.
Why does café appear instead of café?
The UTF-8 bytes for café were likely interpreted as Windows-1252 or ISO-8859-1 text, producing café.
What does Windows-1252 to UTF-8 repair mean?
The visible mojibake is encoded back as Windows-1252 bytes, then those bytes are decoded as UTF-8.
What is double-encoded UTF-8?
It is text that went through a wrong decode more than once, such as café becoming café and then café after two passes.
Can this tool fix double-encoded text?
It can try bounded one-pass and two-pass repairs and records the pass history when a candidate improves.
Can repair damage already-correct text?
Yes. If no clear mojibake indicators are present, the tool warns and does not automatically select a repair.
Why are several repair candidates shown?
Encoding repair can be ambiguous. The tool shows plausible bounded paths instead of assuming the first result is correct.
How is confidence determined?
Confidence is a heuristic based on removed mojibake markers, remaining replacement characters, control characters, pass count and whether the result improves.
What does the replacement character � mean?
U+FFFD usually means a decoder already lost some byte information, so exact recovery may not be possible.
Can every mojibake string be repaired?
No. Repair requires enough reversible information to reconstruct the original bytes.
Does Unicode normalization fix mojibake?
No. Normalization compares equivalent code-point sequences; mojibake is an encoding mismatch.
Which encodings are supported?
The bounded allowlist supports Windows-1252, ISO-8859-1 and Windows-1251 repair paths, plus bounded double-encoded UTF-8 repair.
Is processing local or server-side?
Repair candidate generation runs on the UnicodeNow server after you press the button. It is not submitted while typing.
Is submitted text stored?
No. Submitted text is processed in memory only and is not stored in the database, Redis, analytics, URLs or application logs.
Related tools
UTF-8 Encoder and Decoder
Convert text to UTF-8 bytes and validate byte sequences.
UTF-8 Validator
Validate hexadecimal byte sequences as UTF-8.
Character Encoding Detector
Compare likely text encodings from raw bytes or byte-like input.
Unicode Text Cleaner
Normalize, trim and clean problematic Unicode text safely.
Text to Hex
Convert UTF-8 text bytes into hexadecimal values.
Hex to Text
Decode hexadecimal byte values into UTF-8 text.
Related guides
Unicode vs UTF-8
Unicode defines characters and code points. UTF-8 encodes those code points as bytes for files, databases, web pages, APIs and network messages.
What Is Mojibake?
Learn why text becomes garbled, how UTF-8 bytes turn into mojibake such as café and It’s, and how to repair it safely.
How to Fix Broken UTF-8 Text
A safe workflow for diagnosing broken UTF-8, validating bytes, reversing mojibake, handling double encoding and auditing repairs.