Filename Unicode Normalizer

Sanitize Windows/POSIX invalid characters with explicit policy, normalize Unicode NFC/NFD via platform Unicode APIs, prevent traversal/reserved names/case-fold collisions, never overwrite.

Filename Unicode Normalizer workspace

Original files remain untouched. Renaming exports a ZIP of actual file bytes; it does not rename files on disk.

Limits: 12 files, 20 MiB each, 64 MiB input; 2,000 archive entries, 32 MiB per decoded file, 128 MiB total, maximum 100× expansion. Choose files or a folder; empty folders may be unavailable. Folder selection varies by browser; the file picker is the fallback.

Or drop selected files here.

No files selected.

    Candidate workspace — select local files to begin.

    Progress counts files read; native codec completion has no reliable time estimate. Cancel terminates this job's worker.

    How to use Filename Unicode Normalizer

    Mode and preview, before/after, collision report, ZIP and CSV mapping. Select local files, inspect the preview, adjust settings, then process and review the actual result before exporting. Reset clears the job and restores defaults.

    Compatibility and safe processing

    NFC and NFD names converge in preview; Windows CON, trailing dot, slash and emoji are handled explicitly. All processing happens locally in a disposable worker. ZIP supports Store and Deflate, UTF-8 paths, CRC checks and safe regular files. ZIP64, legacy non-UTF8 paths, alternate Unicode path fields, encrypted entries and unsupported methods fail closed. TAR supports USTAR and bounded local PAX path records; special entries are rejected. GZIP is one stream, not a multi-file archive; concatenated members are unsupported. Decompression limits apply to actual output. Unicode path comparisons use NFC and case folding. CSV is UTF-8 RFC4180 with optional BOM. Selected folder trees omit empty folders unavailable to browser APIs. No tool can inspect unselected directories.

    Related tools in this family