Tod Kurt comments

Results 368 comments of


Tod Kurt

Strings like the product string should be available on an opened device

The deviceInfos can't get cached efficiently, at least with the current underlying hid-api library. The `HID.devices()` call causes an explicit USB bus enumeration, which can be costly on some OSes....

Strings like the product string should be available on an opened device

Not a problem. This aspect of hid-api has been bugging me in my C projects for a few years now. :-)

Android Cannot Open device (hid_open)

Does your app have the correct permission filter in its AndroidManifest.xml and your code have the correct PendingIntent sent to `UsbManager.requestPermission(device, permissionIntent)`? Here is an example USB HID device library...

Powermate LED not changing

Hi Ziriax, Have you tried `node-powermate`? https://github.com/sandeepmistry/node-powermate

feat: run async read on std::thread instead of uv threadpool

This looks really good, thanks! What OSes have you tested this on?

feat: run async read on std::thread instead of uv threadpool

Apologies I've been dealing with family issues. I've not had access to my hid testing hardware. Could I ask you to update the README and changing the various examples in...

feat: run async read on std::thread instead of uv threadpool

This is great, thank you. I'll test this out this week on some devices and OSes.

feat: run async read on std::thread instead of uv threadpool

Which changes would you like to see? Let's try to get them in too. One small thing I'm doing is fixing `hidapi` version to latest 0.12 in the git submodule...

feat: run async read on std::thread instead of uv threadpool

The underlying `hidapi` library is explicitly not thread-safe, how are these plans impacted by that?

feat: run async read on std::thread instead of uv threadpool

To answer your previous questions: I agree that a second class for async sounds like a good idea, for the reasons you mention. I don't mind if it's a total...