[refactor] - Create separate handler for non-archive data

[ahrav]

Description:

This PR splits the original default handler into two separate handlers: default and archive. The default handler now processes non-archive files, and the archive handler manages all archive file types supported by the archiver library.

Additionally, this PR restores the git cat file functionality for binary diffs to a streaming method. This change is due to HandleFile now properly consuming the entire reader before beginning file processing.

Note: Most of the important changes are in handlers.go.

Checklist:

• Tests passing (make test-community)?
• Lint passing (make lint this requires golangci-lint)?

[ahrav]

Most of this code was moved from the old default handler.

[rosecodym]

I have a non-blocking thought about a possible different way to structure the code (that you may have already considered) and a question about a suspicious comment.

[rosecodym]

It looks like three of these fields are only used to later choose a handler type. Did you consider just creating a field for a handler here and setting it in newFileReader, thus eliminating the need for selectHandler and therefore the fields needed for it to work? (It looks like you'd need to add some additional functionality to individual archive implementations, e.g. to support archive skipping, so I don't think this is a slam dunk, but I thought I'd ask.)

[ahrav]

It should also be used in archive.go L88-91 here, sorry I should've explicitly called it out since the rest of the code was a lift and shift.

[ahrav]

I have a non-blocking thought about a possible different way to structure the code (that you may have already considered) and a question about a suspicious comment.

Replied inline.

[rosecodym]

🎉