Z01 File Documentation

Summary

A WinZip First Split Zip File is the first segment of a single ZIP archive that was divided into parts (.z01, .z02, …, and a final .zip). It is not a standalone archive: it holds only raw compressed data, while the archive’s index lives in the last part. To extract, put every part in one folder and open the final .zip (not the .z01) with 7-Zip or WinRAR, which reassemble the volumes automatically.

Technical details

FeatureValue
Full nameSplit ZIP Archive — First Part
File extension.z01
MIME typeapplication/zip
Format typeFirst volume of a multi-part (split) ZIP archive, binary
RoleData segment only — not independently openable
Naming conventionname.z01, name.z02…, last part name.zip
Central directoryNot present — stored in the final .zip part
File to openThe final .zip, with all parts in one folder
DeveloperPKWARE (split/spanned convention); popularised by WinZip, WinRAR, 7-Zip
Byte orderLittle-endian (ZIP structures)
Typical start bytes50 4B 03 04 (local file header), or 50 4B 07 08 spanning marker
Split vs spannedSplit parts are named per segment; spanned parts share one name. Internally identical
Open standardYes — defined in PKWARE APPNOTE (ZIP 6.3.x)
Related extensions.zip, .z02, .z03, .zipx, .001
Specificationpkware.com — APPNOTE.TXT
File signature (magic bytes)
50 4B 03 04

Offset 0. A .z01 holds raw ZIP data and usually begins with a normal local file header, ASCII PK\x03\x04, because the first file’s compressed data starts here and continues across the volumes. Some tools instead write a 4-byte spanning marker 50 4B 07 08 (PK\x07\x08) at the very start of the first segment. Either way the .z01 is not a complete archive: the central directory that indexes the files lives at the end of the last part (the .zip), so a .z01 opened on its own has no table of contents.

What is a Z01 file?

A .z01 is the first part of a split ZIP archive, not a format in its own right. When a ZIP archive is divided into several smaller files, to fit size-limited media, slip under an email or upload limit, or make a large download more manageable, tools like WinZip, WinRAR, 7-Zip and PeaZip name the segments in sequence: name.z01, name.z02, name.z03, and so on, with the last part keeping the plain .zip extension. A three-part archive is therefore data.z01, data.z02, data.zip. The convention comes from PKWARE’s ZIP specification (APPNOTE.TXT) and was popularised by WinZip.

The one fact that governs everything about a .z01 is where the archive’s index lives. A ZIP records its table of contents at the end of the archive, so in a split set that index sits in the final .zip part, never in the .z01. The .z01 contains only raw compressed bytes that continue across the volumes. That is why double-clicking a .z01 produces an error and why the correct file to open is always the last one. The sections below explain the ZIP structures involved and the honest limits of what a lone .z01 can do.

Why the index lives in the last part

A ZIP archive is read back-to-front. Each stored file is written as a local file header followed by its compressed data, but the authoritative list of what the archive contains, the central directory, is appended after all the file data, and a small end of central directory (EOCD) record sits at the very end pointing to it.

[local header + data]  file 1
[local header + data]  file 2
...
[central directory]    one entry per file: name, sizes, offset, CRC-32
[end of central dir]   count of entries and offset of the central directory

An extractor opens a ZIP by seeking to the end, reading the EOCD, then reading the central directory it points to, so it learns every file’s name and where its data begins before touching any of that data. When the archive is split, this whole index structure falls in the final segment, which is exactly why that segment keeps the .zip extension: the extension signals “this is the part with the directory, start here.” The .z01 holds the beginning of the file data (its first local file header at offset 0, ASCII PK\x03\x04), but nothing that tells a reader how many files there are or where they start.

Split versus spanned, and the PK0708 marker

Two related schemes exist. In a split archive the segments have different names by number (.z01, .z02…, .zip), which is the common on-disk case. In a spanned archive, the original floppy-disk scheme, every volume has the same name and the tool prompts for each disk in turn. Internally the two are identical byte-for-byte; you can convert between them just by renaming the segments.

Some tools mark a spanned or split set by writing a 4-byte spanning signature 50 4B 07 08 (PK\x07\x08) at the very start of the first volume, before the first file’s data. That signature tells an extractor “this is a multi-volume archive.” Whether or not the marker is present, the defining property is the same: the data stream runs continuously across the volumes, and the index is only in the last one.

Extracting: open the last part with every segment present

To extract a split ZIP, gather all the parts into one folder with their original names intact, then open the final .zip, not the .z01. 7-Zip (right-click the .zip › 7-Zip › Extract), WinRAR, PeaZip and WinZip detect the sibling .z01, .z02… volumes automatically and stream through them in order as if they were one file. On the command line, 7z x archive.zip does the same. On macOS the built-in Archive Utility often cannot handle split volumes, so The Unarchiver or Keka is the reliable choice.

There is no separate “merge” step and no button that rebuilds a single .zip from the parts: the extractor reads the segments as a continuous stream and unpacks the contents directly. If you specifically want one normal ZIP again afterwards, you extract the files and then re-compress them into a fresh non-split archive.

The incomplete-set problem

The failure mode unique to split archives is not corruption but a missing part. Because the data stream is continuous and the index is only in the final .zip, the set only works if it is complete and unbroken. If you have the .z01 but not the closing .zip, you have data with no index, and nothing can extract it. If a middle volume that the data passes through is missing, the stream has a hole and files that span the gap cannot be reconstructed. The bytes that are absent simply are not present anywhere, so no tool can recover them; the only fix is to obtain the missing segments.

Before relying on a split set, confirm you have a continuous run of numbered parts (.z01, .z02, …) plus the final .zip, with no gap in the numbering. The most common real situation behind a “my .z01 won’t open” question is exactly this: someone downloaded or was sent only some of the parts.

Frequently asked questions

Which file do I open in a split ZIP set?

Always the one ending in .zip, the last part, with every other segment in the same folder. That final part holds the central directory that ties the archive together. Opening the .z01 or .z02 directly fails because those parts have no index.

I only have the .z01. Can I extract it?

No. The .z01 holds part of the compressed data and none of the archive’s index, which is in the final .zip. Without every part including that closing .zip, the archive cannot be opened, and the missing bytes cannot be recovered. You need to obtain the rest of the set.

Why is the last part a .zip instead of .z03?

By the split-ZIP convention the final volume keeps the .zip extension because it contains the central directory and end-of-central-directory record. Giving it the standard extension lets an extractor find the index quickly and start there; the earlier data-only volumes are numbered .z01, .z02, and so on.

References