How does it read a tar without unpacking it?

A tar archive is a sequence of 512-byte header blocks, each followed by the file's data padded to a 512-byte boundary. The tool reads each header to list the entry, so it can show the full contents without writing any file to disk.

Does it handle gzipped tarballs (.tar.gz / .tgz)?

Yes. When the file starts with the gzip magic bytes 0x1f 0x8b it is decompressed first using the browser's native gzip DecompressionStream, then the resulting tar stream is parsed. Everything happens locally.

Is my archive uploaded?

No. The .tar or .tar.gz is read with FileReader, decompressed in your browser if needed, and parsed with JavaScript. Nothing is transmitted to any server.

What do the mode and owner columns mean?

The mode is the Unix permission bits in octal (for example 0755), stored in the header. Owner shows the uid/gid or the user/group names recorded in the USTAR header, which matter when the archive is restored on a Unix system.

What file types appear besides regular files?

Tar marks each entry with a type flag: regular file, directory, symbolic link, hard link, character or block device, FIFO, or a long-name/PAX extension. Only regular files can be extracted to download.

What is the TAR / TAR.GZ Inspector?

Free TAR and TAR.GZ inspector. Drop a .tar or .tar.gz file to list every entry with its size, file mode, owner, type, and modified time — parsed from the USTAR headers. Gzip is decompressed natively in-browser. Extract individual files locally. No upload. It runs free in your browser on Gera Tools, with nothing uploaded.

TAR / TAR.GZ Inspector

Name: TAR / TAR.GZ Inspector
Creator: Gera Tools
License: https://creativecommons.org/licenses/by/4.0/

Get one useful tool a week

Like this tool? Enter your email and we'll send you one genuinely useful Gera tool a week — plus a link to come back to this one. No spam, one-click unsubscribe any time.

A tar archive (“tape archive”) is the standard way to bundle many files on Unix systems, often gzip-compressed as .tar.gz or .tgz. Unlike a ZIP file, a tar archive has no central index at the end — it is simply a flat stream of file headers followed by data blocks. This inspector reads those headers in your browser so you can see everything inside, including Unix-specific metadata like file mode and owner, and extract individual files without unpacking the whole archive.

How it works

Tar stores each entry as a 512-byte USTAR header block immediately followed by the file’s data, padded up to the next 512-byte boundary. The tool reads the stream block by block, interpreting fixed-offset fields within each header:

Header offset	Field	Format
0–99	File name	NUL-terminated ASCII
100–107	File mode	Octal string
108–115	Owner UID	Octal string
116–123	Owner GID	Octal string
124–135	File size	Octal string
136–147	Modification time	Octal Unix timestamp
156	Type flag	Single ASCII character
265–499	USTAR name prefix	NUL-terminated ASCII

The USTAR prefix field (bytes 265–499) is concatenated with the name field to reconstruct paths longer than 100 characters. Sizes and timestamps are stored as octal text — a detail that trips up naive byte parsers and causes silent misreads when not handled.

If the file begins with the gzip magic bytes 0x1f 0x8b, the tool first decompresses it using the browser’s native DecompressionStream('gzip') and then parses the resulting tar stream. To extract an individual file, it slices the data bytes that immediately follow that entry’s header without re-reading any other part of the archive.

Type flags and what they mean

The type flag byte distinguishes entry types. Common values:

Flag	Meaning
`0` or `\0`	Regular file
`1`	Hard link
`2`	Symbolic link
`5`	Directory
`3` / `4`	Character / block device
`6`	Named pipe (FIFO)
`L`	GNU long name extension
`x` / `g`	PAX extended headers

Only regular files can be extracted to download. Directories, symlinks, and device entries are listed for inspection but contain no extractable data bytes.

Notes and practical tips

Two consecutive all-zero 512-byte blocks mark the end of the archive; the tool stops there and ignores padding beyond that point.
GNU long-name (L) and PAX extended headers are recognised, so paths longer than 100 characters display correctly.
The file mode shows Unix permission bits in octal (for example 0755 = rwxr-xr-x). These matter if you restore the archive on a Unix system — the tool shows them so you can verify that scripts and executables have the right permissions before extraction.
Extraction is entirely local; downloaded file bytes never touch a server.