bbustin / dmarc_aggregate_parser Goto Github PK

Hi @bbustin,
I have taken a longer look at parse_dir() and I think there are a few issues with the function. As you mentioned you put this together while learning Rust. I myself am still learning Rust :). I would personally use it for my own small projects. But it has some issues when being considered for use in the larger more complex system.

• Unable to mark files as processed, or rename, remove files that have successfully been processed. This means that files need be manually updated to avoid duplication or reprocessing.
• I also suspect that if there is a file that causes an error, any following files would not be processed until the offending file is fixed/removed.

The general approach I have used and seen used is something along the lines of the following dodgy pseudo code

for file in dir {
  file = parse()
  if file is ok {
    Do any work that needs to be done. 
    If there is a problem log and report issue and move to the next file without making any system changes.
    If the work completes successfully. Mark the file as processed or remove it.
  } else {
    log and report file as a problem
  }
}

This basically is the long way of saying the parse_dir function is great for testing :) but not so much for production environment as I see things.

Hi,
I am interested in contributing to this repo. I was thinking of writing my own. But this seems to be a great place to start and contribute.

Some of the issues that I see in the code that I think I can help with.

• Code has a panic hard coded. I would like to remove this and make the function return a result
• There are a number of unwraps that might fail on a malformed xml file.
• Have the program handleparsing a directory as a Path in a nice way.

I have already raised two PRs to address some very simple things.

• Prevent the tests for failing with test_parse_dir by adding the #[should_panic] See pr #3
• Made parse_reader private as this does not need to be public in the current code.