Unpacking TAR, .tar.gz, .tgz, and converting to ZIP
Using a browser-based TAR & .tar.gz Extractor lets engineers unpack .tar, .tar.gz, .tgz, and .gz bundles locally without opening a terminal or uploading source code. Everything from inspection to .zip conversion runs inside Web Worker memory on your machine.
Inspecting entries, octal modes, and text previews
Dropping an archive parses entries inside everyday.worker.ts via archiveInspect, or you can generate the four-file sample bundle from makeSampleTarGz to test offline immediately. Tables paginate at 200 rows per page, showing relative paths, byte sizes, and octal file permissions.
Clicking supported text files opens a UTF-8 preview pane capped at 256 KB across 27 recognized formats such as .json, .ts, .md, .yaml, .csv, and .log. Following the Library of Congress Tape Archive format specification, directory entries (typeflag 5) organize hierarchies visually while downloadable files save cleanly one by one.
Converting between .tar.gz, .tar, .gz, and .zip
Teams collaborating across Linux build servers and Windows desktops frequently need format translation without losing directory depth. Running archiveConvert inside the worker repacks validated entries cleanly between .tar.gz, .tar, single-file .gz, and .zip containers.
If you only need to inspect or pull files out of plain .zip folders from colleagues, you can also unpack standalone ZIP archives directly without extra GZIP overhead. Every converted archive downloads straight from memory blobs with zero server round-trips.
How do USTAR, PAX, and RFC 1952 headers preserve filenames?
Classical TAR headers reserve only 100 bytes for a filename plus 155 bytes for a path prefix before overflowing on deep repositories or non-ASCII text. Modern extractors stay compatible by reading PAX key-value overrides alongside RFC 1952 GZIP wrapper metadata so long nested paths unpack without truncation.
Decoding PAX type x records and GNU LongLink paths
Each TAR member begins with a 512-byte header carrying magic bytes ustar at offset 257 and an octal checksum covering all 512 header bytes. The browser tarball extractor checks both unsigned and signed byte sums so archives from legacy BSD tools open cleanly alongside modern Linux builds.
When paths exceed 100 bytes or contain UTF-8 characters, POSIX.1-2001 PAX headers (typeflag x per-file or g global) and GNU LongLink (typeflag L) supply full path strings ahead of the file block. If your workflow exports standard desktop ZIPs instead, use our browser ZIP folder packer to build lightweight bundles.
Validating RFC 1952 GZIP headers and CRC-32 trailers
Before any .tar.gz stream enters the TAR parser, decompression follows the IETF RFC 1952 GZIP File Format Specification starting from magic bytes 0x1f 0x8b. Single .gz files also expose optional FNAME metadata so uncompressed output retains its original stem.
At stream completion, the worker checks the uncompressed 32-bit CRC32 digest and ISIZE byte counter modulo two raised to thirty-two against decoded payload bytes. Corrupt transfers trigger immediate integrity errors instead of silently writing truncated source files or broken datasets.
Archive format limits, DEFLATE compression, and safety checks
Bounded worker quotas protect desktop and mobile browsers against decompression bombs while blocking unsafe archive entry types before rendering downloadable paths. Decompression chunks stream at 16 KB per iteration using the IETF RFC 1951 DEFLATE Compressed Data Format engine inside worker memory.
| Safety guard or parameter | Enforced limit or value | Scope inside archive worker | Security or stability purpose | Target format affected |
|---|---|---|---|---|
MAX_ARCHIVE_ENTRIES |
4,000 files maximum |
Per archive inspection or conversion | Stops inode exhaustion from tiny files | .tar, .tar.gz, .tgz, .zip |
MAX_ARCHIVE_SINGLE_BYTES |
200 MB per file |
Single extracted entry payload | Prevents tab crash from huge binaries | All unpacked files and .gz |
MAX_ARCHIVE_TOTAL_BYTES |
350 MB total bytes |
Cumulative unpacked archive size | Defends against GZIP decompression bombs | .tar.gz, .tgz, .zip, .tar |
MAX_TAR_METADATA_BYTES |
4 MB extended buffer |
PAX (x/g) and GNU LongLink (L) blocks |
Caps memory allocated to malformed headers | .tar, .tar.gz, .tgz |
| Link and node typefilter | Skips 1, 2, 3, 4, 6 |
Symlinks, hardlinks, device, FIFO entries | Blocks traversal attacks and OS device nodes | .tar, .tar.gz, .tgz |
Path sanitization strips leading / roots and .. parent traversal segments so malicious entries cannot escape target folder structures. Teams unpacking exported frame sheets for inspection can jump straight to Video Frame Extractor after extracting media archives.
Tip: Open unfamiliar CI or vendor tarballs here before running command-line extraction on your workstation. Browser inspection isolates
PAXheaders, filters symlinks and device nodes automatically, and lets you preview.sh,.json, or configuration scripts safely up to256 KB.
Creating deterministic .tar.gz bundles in the browser
Packaging release assets, static site exports, or build inputs inside browser memory removes dependence on local tar flags that differ between macOS BSD and GNU Linux. Archive creation follows the POSIX.1-2017 pax and ustar format standard with deterministic field formatting across platforms.
Editing queued relative paths and PAX UTF-8 wrapping
Adding local files enters a staging queue where you can rewrite every entry’s relative destination path before packing. When a customized folder path exceeds classical USTAR limits or includes international UTF-8 characters, createTarGzBytes automatically injects a PAX x path= header record.
Parent directories normalize cleanly with forward slashes regardless of whether files originated on Windows or macOS. After packaging raw audio recordings into clean archives, editors needing chapter breaks can use our sibling tool to keep batch podcast chapters organized offline locally.
Reproducible zero-mtime tarballs for media and build assets
Clicking build runs archiveCreateTar or archiveCreateTarGz inside the worker, fixing modification timestamps at mtime: 0 (1970-01-01) and locking GZIP DEFLATE compression at level: 6. Removing wall-clock variance ensures identical file trees always yield matching cryptographic hashes.
For single large SQL dumps or logs that need standalone .gz compression without a TAR wrapper, archiveGzipSingle writes a compliant RFC 1952 FNAME header directly from memory. Closing your browser tab drops every cached array buffer cleanly.