Carson Sievert

Results 165 issues of Carson Sievert

This changes the default behaivor of ggplotly() to render `geom_boxplot()` as low-level polygons rather than plotly.js box traces. As result, we'll be able to exactly replicate the ggplot2 result in...

TODO: how hard would it be to get this respecting aesthetic mappings? For example, `plot_ly(iris) %>% add_parcoords(color = ~Species)`

See #1401 for more context ## Review questions - [ ] Is it safe to remember event data by attaching an attribute to `graphDiv` in this way? Should try this...

TODO: - [ ] Converters for **ggforce**'s `GeomMarkCircle`/`GeomMarkEllipsis`/`GeomMarkHull` - [ ] Converter for **ggraph**'s `GeomAxisHive` (path + text?). For example, see https://github.com/thomasp85/ggraph/blob/8d23845/R/geom_axis_hive.R#L123-L142 - [ ] Add relevant `key` attributes. -...

TODO: - [ ] tests (seems as though opacity dimming isn't working as of 951727c) - [ ] benchmarks

- [ ] find a way to query default system font, perhaps similar to this -- https://github.com/r-lib/svglite/blob/dcda7d885b108ef0d8cf1dfbc16167e0914ce827/R/fonts.R#L5-L19 also https://cran.r-project.org/web/packages/svglite/vignettes/fonts.html

This change is kinda scary, but the speed improvement seems to be worth it. See, for example, https://github.com/ropensci/plotly/tree/master/inst/examples/shiny/Diamonds UPDATE: this appears to break some crosstalk+shiny use cases, such as https://github.com/cpsievert/housing-data-challenge-plotly

Note to self: this is really the only relevant part -- https://github.com/ropensci/plotly/pull/938/files#diff-bfe74656b9837557ae3edcc5e1daa831R22 May want to cherry-pick that code as merging might be a disaster