Anton Firszov

Windows-only coverage is already much better, than no coverage :)

I'm not big fan of lazy solutions. Makes the lifecycle and program behaviour unpredictable. Eg: reading a pixel could trigger an expensive, multi-thread jpeg decoder execution ... brrr! I think...

@brianpopow nice! Only two more left.

Nice catch! Looks like a bug sneaked into our `PngEncoderTests` code and we are comparing the image to itself instead of the reference file :) https://github.com/SixLabors/ImageSharp/blob/75dc96a27f5120451a5a8c00c75d0eaeff68437c/tests/ImageSharp.Tests/Formats/Png/PngEncoderTests.cs#L624-L628 Other encoder tests seem...

> I was writing tests for jpeg encoder PR and looked at the tests which compare saved images to reference images Note that the mentioned PNG tests are validating generated...

Hey, @IldarKhayrutdinov thanks for your interest! Feel free to pick up, and feel free to discuss the topic and ask any question either here or on [gitter](https://gitter.im/ImageSharp/General).

Говорю, ну я с детства в Венгрии живу, поэтому девелоперский диалект совсем пропустил. Последнее времья стараюсь его в [Ютубе изучать](https://www.youtube.com/watch?v=eWrRzOUaSBQ). Заговорить на нём мне пока не просто )) Ну я...

А можем в [гиттере](https://gitter.im/ImageSharp/General)? Там просто более просше обсуждать код, чем в майле, и не будем спамить других сдесь. ПМ меня - есле ОК. Есле нет, напишу майл.

Personally, I would be fine exposing the raw XML in some read-only form (maybe as `string`). We don't want to make it editable if we cannot validate the contents. BTW...

This is important in long term, but improving Jpeg decoder performance has higher priority at the moment IMO, so if @br3aker is more interested in Jpeg he's doing the right...