Evan Pezent

Results 68 comments of Evan Pezent

It's not currently possible. matplotlib also suffers from [this issue](https://stackoverflow.com/questions/26759813/bars-width-are-wrong-using-log-scale-of-x-axis), and the recommended solution there is to pass an array of variable bar widths. We don't support that yet. Your...

Thanks Omar! Very thoughtful of you to share these. Will defintely address these.

You can insert a NaN for dates without values, and ImPlot will treat them as missing data (as demoed in the `NaN Values` section of the demo): ![image](https://user-images.githubusercontent.com/29577475/174466059-ad5545e2-9b26-4176-b539-08ea29684f7c.png)

It's not currently doable with our Time scale ticker. We would need to adopt an ordinal scale or have some complicated transformation logic to handle discontinuities in lines that cross...

We can look into adding an API to detect state changes to plots/axes. A function that returns fllags with the current state changes (a la `ImGuiTableColumnFlags`) is probably the right...

We don't currently provide an API to reorder either. You could fake it though, by plotting the digital items with "##" prefix in the name to hide their legend labels,...

Thanks for the feedback. I just made a few changes to the algorithm in https://github.com/epezent/implot/commit/2dc2a4cfd5e3f0419b1f864cdaedba94ec8ef3a5. Can you give the latest master a try and let me know if this improves...

Good to know. I think we can solve this case with a second stage of pruning. Shouldn't be to hard to implement.

For one, you are misusing PlotText -- you are passing pixel coordinates (i.e. return value of `GetPlotSize()`), not plot coordinates. `ImPlot::PlotText("Image Not Found.", 960, 540);` would be more appropriate. But...

First, this is amazing work! Thank you for opening this PR > I know you are not convinced by the need of implicit lifetime management of the color pointer but...