Ted Mielczarek

Trying to sanely handle the complexity of HID gamepads/joysticks across platform is really hard. I've been trying to get something useful spec'ed in the W3C Gamepad API spec for years...

This got a fully-fleshed-out implementation in the `fix-stacks` binary: https://github.com/mozilla/fix-stacks/blob/c30daaef43bd15486fba189b06487aa24dad3492/src/main.rs#L433-L502

FWIW, since this was filed I have fleshed out enough of moria that it can locate dSYMs on mac. [Here's an example of using it](https://github.com/luser/rust-disasm/blob/b5be8d39d7fa9aae8704d9473952ddb65c039955/src/lib.rs#L297) in my rust-disasm crate.

> replace lots of pub fields with accessors Is there a reason for this aside from hiding the implementation details? > clean up type names (should we really prefix so...

Mentioning this here as well: https://github.com/luser/rust-minidump/pull/312#issuecomment-976752323

Is this a stepping-stone towards dealing with crashes involving non-canonical pointers? (Because it sure feels like it!)

> I did this mostly because I needed some trait anyway to make it easy to unit test without synthesizing my own fake memory list. I originally had a GetMemoryAtAddress...

I think with a little bit of up-front work you can write a function that generates suitable synthetic minidumps for your tests so that the individual tests can be very...

My original thoughts with `SynthMinidump` were to make it easy to set up various minidump configurations including common streams, so having helpers for modules/threads/memory/exceptions made sense. For other streams, I...

> I’m not sure if we need the whole cpu context for that, probably not, so maybe a `instruction: u64` is sufficient. This is most often used in the stack...