Search code examples
node.jsamazon-s3php-ziparchive

Upload zip archive files to S3 with node


I download a ziparchive from an api, which contains gzipped files, and i need to take the gz files and save to s3. Don't want to uncompress or anything. Just move to S3.

When i open the archive, it has a folder w/ random numbers, /12345/file1.gz, and many files, /12345/file2.gz, etc.

I've tried yauzl and adm-zip, but don't understand how to take each entry in the archive and just send to s3. I have s3-stream-upload package, which i can use to send. Just can't get it right. Thanks for any help

yauzl.open("output.zip", {lazyEntries: true}, function(err, zipfile) {
  if (err) console.error('zip err: ', err);
  console.log(zipfile);
  //upload.write(zipfile);
  zipfile.readEntry();
  zipfile.on("entry", function(entry) {

      // file entry
      zipfile.openReadStream(entry, function(err, readStream) {
        if (err) console.error('readstream err: ', err);
        readStream.on("end", function() {
          zipfile.readEntry();
        });
        console.log(entry);
        readStream.pipe(upload) //upload is an s3-upload-stream
        .on('finish', function() { console.log('finished'); })
        .on('error', function(err) { console.error('stream err: ',err); });
      });

  });
});

This gives me write after end error, i think bcz the readstream is the actual data of the entries/files. been at this a while could use some help. thanks


Solution

  • Answer was doing a straight s3 put with readStream as the body of the object...

    yauzl.open("output.zip", {lazyEntries: true}, function(err, zipfile) {
      if (err) console.error('zip err: ', err);
      zipfile.readEntry();
      zipfile.on("entry", function(entry) {
    
          // file entry
          zipfile.openReadStream(entry, function(err, readStream) {
            if (err) console.error('readstream err: ', err);
            readStream.on("end", function() {
              zipfile.readEntry();
            });
            readStream.length = entry.uncompressedSize;
    
                s3.putObject({
                    Bucket: "bark-data-team",
                    Key: "amplitude-data/raw/" + startDate + "/" + entry.fileName,
                    Body: readStream
                }, function(err, data) {
                    console.log(err);
                    console.log(data);
                });
          });
    
      });
    });