Fwiw, I have a need to implement similar zip extraction, here's the code I ended up with: https://gist.github.com/jcorbin/fd0716bd0bc4c49cd22d73e4aad23b38

• extracts in GOMAXPROCS-parallel
• slightly better reliability / error flow in the file extraction path
• slightly better file permission restoration
• more efficient for large archives, since it uses io.CopyBuffer internally with a re-used buffer, rather than io.Copy dynamically allocated a 32KB buffer for every extracted file (causing GC pressure scaling with size of archive)
• oh and it supports context cancellation, in case that matters (e.g. timing out a large archive extraction, or otherwise)...
• ...around this code, I'm running the extraction into a sibling temporary directory, and then either os.Rename-ing into place when done, or os.RemoveAll-ing it in case of (e.g. cancellation) error (similar to github.com/google/reanmio.PendingFile, but for directories