Philippe Rivière

TODO: - [ ] add a textAnchor option - [ ] add a dx, dy option

TODO: - [ ] should filter be separated into filterTick and filterLabel? Often times you want the tick but not the label? - [ ] I suppose reach===0 should create...

sometimes you want more grid lines than ticks

Confirmed in d3v6. Do you have an idea to solve this? I think the issue is in d3.pointer, not specifically d3.brush. It's fun to see that it doesn't happen with...

Is it something similar to https://blockbuilder.org/Fil/076415313a08e76ae60d468e2cb25909 ? (related: https://github.com/d3/d3-zoom/issues/222)

when you are brushing the selection *s* can be [387, 387] (it will be nulled when you release the mouse). At this point the code uses `x.domain(s.map(x2.invert, x2));` and `d3.zoomIdentity.scale(width...

Maybe a case of touch-action: none or touchStart.preventDefault ; see https://observablehq.com/d/8c6133113d33b60c

tested on an Android phone [this version](https://observablehq.com/d/660b25f59013163c) seems good (fixing the issue both on Chrome and FF): the difference is `svg.on("touchstart", event => event.preventDefault())` I'm not sure if it should...

I think this is similar to what @nbremer did in the “Creating a better chord ordering” section of [Data-based orientations for gradients in a d3.js chord diagram](https://www.visualcinnamon.com/2016/06/orientation-gradient-d3-chord-diagram). (Except the position...

We could keep the API mostly unchanged by passing the group's id *i* as a third argument to sortSubgroup?