# EPUBCheck PKG-027: file names not in UTF-8, how to fix it

Source: https://epubcc.com/epubcheck/pkg-027/ (updated 2026-11-04)

EPUBCheck PKG-027 is a fatal error message: “Could not extract EPUB ZIP content, probably due to file names not encoded in UTF-8.” EPUBCheck could not extract the archive, most likely because the file names inside the ZIP are not encoded in UTF-8. It is a fatal error: the content cannot be checked until the archive is rebuilt.

- Official message (EPUBCheck 5.4.0): Could not extract EPUB ZIP content, probably due to file names not encoded in UTF-8.
- Default severity: Fatal error

## Typical causes

- A ZIP created by an old tool or on a system that stores file names in a local code page, often with accented characters in the names.
- Files named in another script (Cyrillic, Japanese) zipped by a tool without UTF-8 support.

## How to fix it

1. Rename files to ASCII names, or rebuild the archive with a tool that stores names in UTF-8.
2. Rebuild the archive with the mimetype file first and stored uncompressed, for example from the folder that contains the book: zip -X0 book.epub mimetype and then zip -rX9 book.epub META-INF OEBPS (use your own folder names).
3. Run EPUBCheck again on the rebuilt file and fix any message that appears next: a damaged archive can hide other problems until it can be read.

## When it's not this

Non-ASCII names that do extract correctly only give the PKG-012 usage message.

## Sources

- EPUB 3.3, §4.3.2 ZIP file requirements: https://www.w3.org/TR/epub-33/#sec-zip-container-zipreqs (read 2026-11-04)
- EPUB 3.3, §4.2.3 File paths and file names: https://www.w3.org/TR/epub-33/#sec-container-filenames (read 2026-11-04)

Message text and severity: EPUBCheck (W3C, BSD-3-Clause). Explanation: EpubCC.