We have a pipeline that produces millions of .tar.gz files, each with one text file in them. We need to deliver the text files (gzipped, but not tarred).

We have a pipeline that produces millions of .tar.gz files, each with one text file in them. We need to deliver the text files (gzipped, but not tarred).

We have a pipeline that produces millions of .tar.gz files, each with one text file in them. We need to deliver the text files (gzipped, but not tarred).

We have a pipeline that produces millions of .tar.gz files, each with one text file in them. We need to deliver the text files (gzipped, but not tarred).

We have a pipeline that produces millions of .tar.gz files, each with one text file in them. We need to deliver the text files (gzipped, but not tarred).

I have a number of compressed files (*.tar.gz) in a directory. The file names have the created date in them. Each contains multiple files. I need to know the number of line that contain a specific word in a file with specific naming pattern. Is there a way to do this without uncompressing and storing every file?