orangutan | Simulate native events on Android-like devices

In order to accomplish this, you should have the lists for the y values along with a list where the percentages will go outside of the go.bar like this:

You could use the selector argument:

By default, the entire page is captured. If you'd like to capture a specific part of the screen, you can use the selector parameter to specify a CSS selector. For example, if you have a plot with ID myplot then you can use screenshot(selector="#myplot")

In this case :

It's a bit of a hack, but if you know there will always only be one pageid (or if you'll always want the first one), you can use Object.values:

If I understand the problem correctly, you should use Promise when fetching data, after that, you need a utility to format the response something like the following example:

You were close, you have to build the sum inside the innerLoop and then print out the sum in the outer loop. Also note that you have to reset it to zero again in the outerloop

There is a more efficient way to this. Your solution will have runtime. Thats not very good if you need to join more than a few items.

The easiest way to improve on your solution is to generate a lookup table from the original data beforehand by creating an associative array which has title/name as it's key. When doing the merge afterwards a lookup will be fast (and does not get slower with more entries).

You only loop twice instead of n*m times. Which would lead to a linear runtime

One solution is to use a Gantt Chart which may handle dates better (especially defining where it starts and ends).

Here's a working example:

As @Maximilian Peters pointed out, there isn't a good way of referencing a point an arbitrary amount between categorical coordinates.

The best you can do is set the xref to paper meaning that the x-axis is scaled to be between 0 and 1. Then you have to guess and check where the x-coordinates should be to line up where you want them. Since your y-axis is numerical, you can reference the y-values you like directly. After some trial and error, drawing lines between (0.10, 20), (0,10, 25), (0.44, 25), (0.44, 14) gets you pretty close to your desired annotation.

This is definitely hacky, and not generalizable if you add more bars, or change the width of the bars in any way.

Judging from:

I was looking into the fig.update_xaxes function but was only able to change the colors of the lines which are white currently on the grid.

It sounds like you're actually asking how to change the background color for particular segments of the background along the y-axis. Particularly since you're also stating that:

[...] so that the bars are more visible.

If that is in fact the case, then I would shapes for sensible intervals with alternating background colors, and set the shapes to appear "below" (behind) the traces of the figure to get this:

And if you'd like to hide the gridlines you can just throw xaxis=dict(showgrid=False) into the mix: