Contributing to this Repo: remove panic and unwrap about dmarc_aggregate_parser

[bbustin]

Hi @Bas-Man. I agree, there are a lot of areas this could be improved. I really hacked it together quickly while initially learning Rust. Your help is much appreciated. Removing the panics and unwraps would be great.

I incorrectly closed the pull request about making parse_reader private. You are correct. It should not be public. I mis-read it as parse and closed the pull request. I will try to re-open and merge it. Sorry about that...

The second pull request is also a good one. Ideally parse_dir would not try to parse files that are not .xml, .gz, or .zip. But for now making this change is a good idea.

[Bas-Man]

Yes. Ideally parse_dir would only parse valid files. The [should_panic] is a temp fix to get all tests passing including the panic.

I am already thinking about how to resolve the parse_dir so that it does not actually panic.

But moving step by step to avoid conflicts.

I also would like to make most of the struct data private with accessors. But that might be over engineering.

[Bas-Man]

@bbustin could you point to a good source of demarc samples.

[bbustin]

I do not know a good source of samples, unfortunately. I was just using ones I was receiving on my mail server from Google, Yahoo, and other places.

[Bas-Man]

No worries. I will work with what we have.

I would like to remove the guard in parse_dir()

The path.is_dir() {}
This appears to be preventing any IO errors from being returned. This is because it evaluates before any IO operations. At least from what I have tried. When it is removed. IO errors are returned for unreadable files and directories.

[bbustin]

👍