Will Glynn

@m4b I'm entirely on board :+1: `pdb` depends on `uuid`, and matching those PE UUIDs with corresponding PDB UUIDs is the point of that field. `uuid` everywhere makes that use...

@m4b Oh yeah, I know `gnu ld` emits UUIDs in ELF note sections when `--build-id=uuid`. No idea how common those are in practice, but I once needed them in the...

@luser: Good point. `--build-id=uuid` is a 128-bit UUID, while `--build-id=md5` is a 128-bit MD5, and I don't think those are distinguishable after the fact. Sounds like we can skip ELF...

Maybe we should ask if `uuid` would take a PR adding an optional dependency on `scroll`? That'd be consistent with how `uuid` presently interacts with `serde`, and it seems like...

One thought [in reply to this](https://github.com/m4b/goblin/issues/58#issuecomment-343612776): > if we peek, the if condition guarantees us that the Unknown variant cannot occur, hence its unreachable. But this is runtime information (hence...

@m4b Sure. I just wanted to tamp down expectations. A `peek()` that finds a valid magic makes it highly likely that `parse()` will see the same thing, it just doesn't...

+1. I needed to construct a command like: ``` convert -size 100x100 xc:none \ image_a.png -geometry +0+0 -composite \ image_b.png -geometry +10+20 -composite \ output.png ``` This was possible through...

> Is it possible to modify parts of a PDB or rewrite it entirely with this? Browsing through the docs, it looks like it only reads PDBs. Correct, this library...

The simplifying conceit used by `pdb` is that MSF streams can be represented as contiguous byte slices. This means that there's no complex I/O layer scattered around – all the...

Yep, you've got it. `Stream` represents a possibly-expensive copy of a PDB stream. Think of it as a `Vec` (and note that it is a `Vec` unless you wrote your...