Philippe Rivière comments

Results 1049 comments of


                                            Philippe Rivière

A transform to consolidate ordinal values outside the top n into an “other” category, perhaps in conjunction with the group transform.

We now have sort:{ fx: { value: …, **limit** } } in #442 ; the only thing missing is "others".

A transform to consolidate ordinal values outside the top n into an “other” category, perhaps in conjunction with the group transform.

A more advanced experiment here https://observablehq.com/@observablehq/plot-stacking-others-144

A transform to consolidate ordinal values outside the top n into an “other” category, perhaps in conjunction with the group transform.

Yet another take https://observablehq.com/@observablehq/plot-group-top-n-with-others

SSR Hydration attribute mismatch on `transform` attribute

I believe this is due to this _offset_ https://github.com/observablehq/plot/blob/7d18179ef5c46db1dc0f463b9e4d0888766cc80d/src/style.js#L9 We should probably align its value to the most common use case (which is possibly the high rather than low DPI...

SSR Hydration attribute mismatch on `transform` attribute

The offset makes lines sharp, avoiding antialiasing. Without it, a black line of "1px" against a white background is rendered as two vertical lines of grey pixels, instead of 1...

implicit zero

I don't mind it being applied to color as well, in the linear case. But it's probably less important. Adding a default rule seems a tad too opinionated. The new...

treemap

Love it! As a random remark, if we let users specify their own tiling method, can we anticipate that some tiles might be generic polygons rather that rectangles (and maybe...

treemap

I'm pushing a few experiments in #1732. We'll also need a strategy for faceting; in particular, the surface area between facets should be comparable—a proportion reducer would equalize the total...

The default tickFormat with a collapsed domain lacks sufficient precision

Thanks. Did you test [#287](https://github.com/d3/d3-scale/pull/287)?