Runs on your device · Your file never leaves your device.
Drop a .zip and see every entry with its size and its packed size, in the folder tree the archive was built with. Download a single file, download them all one by one, or preview an image or a text file first without extracting anything.
It works on the phone as well as the laptop — someone sends you an archive on the go, you open it here and save just the PDF you need. The archive is read on your device; nothing is sent anywhere. File names that older archivers wrote in a local code page — Cyrillic, Arabic, Greek, Hebrew, Turkish, Japanese, Chinese, Korean — are read in that code page.
Password-protected files are the one thing this cannot take out: the format marks them, so they are shown as protected and everything else in the archive comes out. Damaged or truncated files are reported plainly, and so is an archive too large for the device’s memory.
Not for: Taking out password-protected files, or opening RAR and 7z archives — only standard ZIP entries without a password come out.
The tool checks the first two bytes of the file, the letters PK that begin every ZIP, not its extension. Several common formats are ZIP archives inside: Word, Excel and PowerPoint files (.docx, .xlsx, .pptx), EPUB e-books and Android .apk packages. Drag one onto the drop area, since the choose button lists only .zip files, and its parts are listed. In a .docx the pictures are in the word/media folder, and each can be downloaded on its own.
A file without those two bytes is refused with "This file is not a ZIP archive." That is what a RAR or 7z archive gets, and also a "zip" that is really a web page saved by a failed download.
A ZIP keeps its table of contents at the very end. When that end is missing, as after an interrupted download, the tool reports that the ZIP may be incomplete or damaged. The same message appears for one part of an archive split into pieces, such as .z01 or .zip.001; those parts cannot be joined here, so you need the archive as a single file. An archive too large for this device's memory is reported as exactly that.
A file with a password is marked as password-protected in the list, a message says how many there are, and every other file still comes out. Only when every file has a password is the whole archive refused. A file packed with a method other than Store or Deflate, which few programs use, cannot be extracted and is reported by name.
Names are read as UTF-8 when the archive marks them that way, carries a Unicode copy of the name, or simply is valid UTF-8. Archives made by older programs stored names in a local code page; those names are read in the code page that makes them words, such as cp866 for Russian, cp720 for Arabic, cp737 for Greek, cp857 for Turkish, cp862 for Hebrew, or Shift_JIS, GBK, Big5 and EUC-KR for Japanese, Chinese and Korean.
Size is the unpacked size and Packed is the size inside the archive; the Packed column is hidden on narrow screens. When the two are almost equal, the file was already compressed, as photos and videos usually are. Folders are listed first, and numbers in names sort by value, so page2 comes before page10. The text preview shows the first 4,000 characters, read as UTF-8, and besides TXT, CSV, JSON, XML, HTML, Markdown and log files it also opens TSV, CSS, JavaScript, YAML, INI, SRT and VTT files.
This page in other languages: العربية · Deutsch · Français · Español · Русский · Português · 中文 · हिन्दी · Türkçe · עברית · فارسی · اردو · Bahasa Indonesia · বাংলা