Initial commit for PE parser

[Schamper]

While reviewing #10 it quite quickly became apparent that it was more beneficial to create a new parsing base to work from. This PR implements that base. This means that almost all code here is made with the intent to parse, not to write. The idea is that instead of trying to wreck our brain on how to create a nice interface to both read and write, we first focus on making a nice parsing/reading interface, and to take the lessons learned from #10 regarding writing and at some later point apply them to this new base.

Notably this base focuses on:

• Lazy parsing wherever possible
• More natural Pythonic interfaces for interacting with different parts of a PE file
• Same other nice QOL parsing features over Initial commit for dissect.executable.pe #10

No hard feelings to anyone involved with #10. Love you @sud0woodo.

Todo for this PR:

• Unit tests
• Implement bound import directory

Depends on #29 for some minor project cleanup.

[codecov]

Codecov Report

❌ Patch coverage is 79.45708% with 280 lines in your changes missing coverage. Please review.
✅ Project coverage is 81.90%. Comparing base (e312272) to head (b5e33d5).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
dissect/executable/pe/directory/resource.py	75.52%	81 Missing ⚠️
dissect/executable/pe/pe.py	74.33%	68 Missing ⚠️
dissect/executable/pe/directory/delay_import.py	76.19%	25 Missing ⚠️
dissect/executable/pe/directory/debug.py	82.85%	24 Missing ⚠️
dissect/executable/pe/directory/imports.py	76.92%	21 Missing ⚠️
dissect/executable/pe/directory/export.py	80.68%	17 Missing ⚠️
dissect/executable/pe/directory/bound_import.py	84.05%	11 Missing ⚠️
dissect/executable/pe/directory/exception.py	80.00%	6 Missing ⚠️
dissect/executable/pe/directory/load_config.py	83.33%	6 Missing ⚠️
dissect/executable/pe/directory/com_descriptor.py	89.36%	5 Missing ⚠️
... and 5 more

Additional details and impacted files

@@             Coverage Diff             @@
##             main      #28       +/-   ##
===========================================
- Coverage   92.33%   81.90%   -10.44%     
===========================================
  Files           5       24       +19     
  Lines         300     1663     +1363     
===========================================
+ Hits          277     1362     +1085     
- Misses         23      301      +278     

Flag	Coverage Δ
unittests	81.90% <79.45%> (-10.44%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Miauwkeru]

I am a bit torn on passing the PE object to every directory and section etc, while in most cases passing just the file handle will do (with some additional flags). This makes it a bit harder in the future to test specific parts without creating a PE object first as it is strongly coupled to the PE object.

But then again, how often would we encounter a file or a set of bytes where we know it is a BaseReloactionDirectory for example, and just want to parse that part without the PE file.

So we could leave it as it is and only change it when it is actually needed. aka, if we want to parse directories or sections without a PE file.

[Schamper]

I am a bit torn on passing the PE object to every directory and section etc, while in most cases passing just the file handle will do (with some additional flags). This makes it a bit harder in the future to test specific parts without creating a PE object first as it is strongly coupled to the PE object.

But then again, how often would we encounter a file or a set of bytes where we know it is a BaseReloactionDirectory for example, and just want to parse that part without the PE file.

So we could leave it as it is and only change it when it is actually needed. aka, if we want to parse directories or sections without a PE file.

Cross that bridge when we get there.

[Miauwkeru]

Why is this TODO here?

[Schamper]

Because there are a ton of structures and enums related to symbols and dynamic relocations that aren't in this file yet but don't need to be in there for now. Can be a future improvement.

[Miauwkeru]

From what I have read OS2 is a separate executable format so it doesn't fit inside the PE class. So while good enough for now, if we ever want to fully implement the OS2 format we should move it to a different class. If that case ever happens

[Schamper]

From what I have read OS2 is a separate executable format so it doesn't fit inside the PE class. So while good enough for now, if we ever want to fully implement the OS2 format we should move it to a different class. If that case ever happens

I know, that's why I also didn't build it out any further.