zipread is a ZIP file reader designed for speed and memory efficiency. Unlike other modules, it doesn't read the whole file into memory and uses native ZLIB functions for decompression.
var zipread = require("zipread");
var zip = zipread("file.zip");
var contents = zip.readFileSync("some_file.txt").toString();
// Or for async version:
zip.readFile("some_file.txt", function (err, buf) {
contents = buf.toString();
});
zipread also provides a set of fs hooks to require node modules directly from ZIP.
Create a modules archive:
$ (cd node_modules && zip -r ../mods.zip *)
$ rm -r node_modules
Require a module explicitly:
require("zipread/hooks"); // Install hooks
require("./mods.zip/request"); // Load 'request' from mods.zip
Or implicitly by setting NODE_PATH:
$ NODE_PATH=./mods.zip node ./yourapp.js
In this case, you still have to require the hook but the rest can be untouched:
require("zipread/hooks");
require("request");
A good use case for this would be shipping desktop applications. For example, Windows doesn't like long paths and putting your huge node_modules directory into a single archive should solve this problem.