Merge branch 'main' into tile_panel_resolution

Author: teunbrand

.github/workflows/R-CMD-check.yaml

@@ -25,8 +25,6 @@ jobs:
           - {os: macos-latest,   r: 'release'}
 
           - {os: windows-latest, r: 'release'}
-          # Use 3.6 to trigger usage of RTools35
-          - {os: windows-latest, r: '3.6'}
           # use 4.1 to check with rtools40's older compiler
           - {os: windows-latest, r: '4.1'}
 
@@ -64,7 +62,7 @@ jobs:
           extra-packages: >
             any::rcmdcheck,
             Hmisc=?ignore-before-r=4.1.0,
-            quantreg=?ignore-before-r=3.6.0,
+            quantreg=?ignore-before-r=4.3.0,
           needs: check
 
       - uses: r-lib/actions/check-r-package@v2

DESCRIPTION

@@ -1,5 +1,5 @@
 Package: ggplot2
-Version: 3.5.0.9000
+Version: 3.5.1.9000
 Title: Create Elegant Data Visualisations Using the Grammar of Graphics
 Authors@R: c(
     person("Hadley", "Wickham", , "[email protected]", role = "aut",
@@ -45,7 +45,7 @@ Imports:
     scales (>= 1.3.0),
     stats,
     tibble,
-    vctrs (>= 0.5.0),
+    vctrs (>= 0.6.0),
     withr (>= 2.5.0)
 Suggests:
     covr,

NAMESPACE

@@ -422,6 +422,7 @@ export(get_alt_text)
 export(get_element_tree)
 export(get_guide_data)
 export(gg_dep)
+export(ggpar)
 export(ggplot)
 export(ggplotGrob)
 export(ggplot_add)

NEWS

@@ -28,7 +28,7 @@ NEW FEATURES
   `example(position_jitterdodge)` for a potential usage. (@kevinushey, #932)
 
 * Allow specifying only one of the limits in a scale and use the automatic
-  calculation of the other limit by passing NA to to the limit function,
+  calculation of the other limit by passing NA to the limit function,
   `xlim()` or `ylim()` (@jimhester, #557).
 
 * Allow to use brewer palettes for continuous scales, through the new

NEWS.md

@@ -1,17 +1,92 @@
-
 # ggplot2 (development version)
 
-* Fixed performance loss when the `.data` pronoun is used in `aes()` (#5730).
+* The bounded density option in `stat_density()` uses a wider range to
+  prevent discontinuities (#5641).
+* `geom_raster()` now falls back to rendering as `geom_rect()` when coordinates
+  are not Cartesian (#5503).
+* `stat_ecdf()` now has an optional `weight` aesthetic (@teunbrand, #5058).
+* Position scales combined with `coord_sf()` can now use functions in the 
+ `breaks` argument. In addition, `n.breaks` works as intended and 
+ `breaks = NULL` removes grid lines and axes (@teunbrand, #4622).
+* (Internal) Applying defaults in `geom_sf()` has moved from the internal 
+  `sf_grob()` to `GeomSf$use_defaults()` (@teunbrand).
+* `facet_wrap()` has new options for the `dir` argument to more precisely
+  control panel directions (@teunbrand, #5212)
+* Prevented `facet_wrap(..., drop = FALSE)` from throwing spurious errors when
+  a character facetting variable contained `NA`s (@teunbrand, #5485).
+* When facets coerce the faceting variables to factors, the 'ordered' class
+  is dropped (@teunbrand, #5666).
+* `geom_curve()` now appropriately removes missing data instead of throwing
+  errors (@teunbrand, #5831).
+* `update_geom_defaults()` and `update_stat_defaults()` have a reset mechanism
+  when using `new = NULL` and invisible return the previous defaults (#4993).
+* Fixed regression in axes where `breaks = NULL` caused the axes to disappear
+  instead of just rendering the axis line (@teunbrand, #5816).
+* `geom_point()` can be dodged vertically by using 
+  `position_dodge(..., orientation = "y")` (@teunbrand, #5809).
+* Fixed bug where `na.value` was incorrectly mapped to non-`NA` values 
+  (@teunbrand, #5756).
+* Fixed bug in `guide_custom()` that would throw error with `theme_void()` 
+  (@teunbrand, #5856).
+* New helper function `ggpar()` to translate ggplot2's interpretation of 
+  graphical parameters to {grid}'s interpretation (@teunbrand, #5866).
+* `scale_{x/y}_discrete()` can now accept a `sec.axis`. It is recommended to
+  only use `dup_axis()` to set custom breaks or labels, as discrete variables 
+  cannot be transformed (@teunbrand, #3171).
+* `stat_density()` has the new computed variable: `wdensity`, which is
+  calculated as the density times the sum of weights (@teunbrand, #4176).
+* `theme()` gets new `spacing` and `margins` arguments that all other spacings
+  and (non-text) margins inherit from (@teunbrand, #5622).
+* `geom_ribbon()` can have varying `fill` or `alpha` in linear coordinate
+  systems (@teunbrand, #4690).
+* `geom_tile()` computes default widths and heights per panel instead of
+  per layer (@teunbrand, #5740).
+
+# ggplot2 3.5.1
+
+This is a small release focusing on fixing regressions from 3.5.0 and 
+documentation updates.
+
+## Bug fixes
+
 * Fixed bug where discrete scales could not map aesthetics only consisting of
   `NA`s (#5623)
-* Facet evaluation is better at dealing with inherited errors 
-  (@teunbrand, #5670).
 * Fixed spurious warnings from `sec_axis()` with `breaks = NULL` (#5713).
 * Patterns and gradients are now also enabled in `geom_sf()` 
   (@teunbrand, #5716).
+* The default behaviour of `resolution()` has been reverted to pre-3.5.0 
+  behaviour. Whether mapped discrete vectors should be treated as having 
+  resolution of 1 is controlled by the new `discrete` argument.
+* Fixed bug in `guide_bins()` and `guide_coloursteps()` where discrete breaks,
+  such as the levels produced by `cut()`, were ordered incorrectly 
+  (@teunbrand, #5757).
+  
+## Improvements
+
+* When facets coerce the faceting variables to factors, the 'ordered' class
+  is dropped (@teunbrand, #5666).
+* `coord_map()` and `coord_polar()` throw informative warnings when used
+  with the guide system (#5707).
+* When passing a function to `stat_contour(breaks)`, that function is used to
+  calculate the breaks even if `bins` and `binwidth` are missing 
+  (@teunbrand, #5686).
+* `geom_step()` now supports `lineend`, `linejoin` and `linemitre` parameters 
+  (@teunbrand, #5705).
+* Fixed performance loss when the `.data` pronoun is used in `aes()` (#5730).
+* Facet evaluation is better at dealing with inherited errors 
+  (@teunbrand, #5670).
 * `stat_bin()` deals with non-finite breaks better (@teunbrand, #5665).
-* `geom_tile()` computes default widths and heights per panel instead of
-  per layer (@teunbrand, #5740).
+* While axes in `coord_radial()` don't neatly fit the top/right/bottom/left
+  organisation, specifying `position = "top"` or `position = "right"` 
+  in the scale will flip the placement of the radial axis (#5735)
+* Theme elements that do not exist now throw warnings instead of errors (#5719).
+* Fixed bug in `coord_radial()` where full circles were not treated as such 
+  (@teunbrand, #5750).
+* When legends detect the presence of values in a layer, `NA` is now detected
+  if the data contains values outside the given breaks (@teunbrand, #5749).
+* `annotate()` now warns about `stat` or `position` arguments (@teunbrand, #5151)
+* `guide_coloursteps(even.steps = FALSE)` now works with discrete data that has 
+  been formatted by `cut()` (@teunbrand, #3877).
 
 # ggplot2 3.5.0
 
@@ -1444,7 +1519,7 @@ accompanying issue #2890.
 
     We recognise that this is a big change and if you're not already familiar
     with rlang, there's a lot to learn. If you are stuck, or need any help,
-    please reach out on <https://community.rstudio.com>.
+    please reach out on <https://forum.posit.co/>.
 
 *   Error: Column `y` must be a 1d atomic vector or a list
 

R/annotation-logticks.R

@@ -175,14 +175,14 @@ GeomLogticks <- ggproto("GeomLogticks", Geom,
         ticks$x_b <- with(data, segmentsGrob(
           x0 = unit(xticks$x, "native"), x1 = unit(xticks$x, "native"),
           y0 = unit(xticks$start, "cm"), y1 = unit(xticks$end, "cm"),
-          gp = gpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth * .pt)
+          gp = ggpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth)
         ))
       }
       if (grepl("t", sides) && nrow(xticks) > 0) {
         ticks$x_t <- with(data, segmentsGrob(
           x0 = unit(xticks$x, "native"), x1 = unit(xticks$x, "native"),
           y0 = unit(1, "npc") - unit(xticks$start, "cm"), y1 = unit(1, "npc") - unit(xticks$end, "cm"),
-          gp = gpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth * .pt)
+          gp = ggpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth)
         ))
       }
     }
@@ -213,14 +213,14 @@ GeomLogticks <- ggproto("GeomLogticks", Geom,
         ticks$y_l <- with(data, segmentsGrob(
           y0 = unit(yticks$y, "native"), y1 = unit(yticks$y, "native"),
           x0 = unit(yticks$start, "cm"), x1 = unit(yticks$end, "cm"),
-          gp = gpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth * .pt)
+          gp = ggpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth)
         ))
       }
       if (grepl("r", sides) && nrow(yticks) > 0) {
         ticks$y_r <- with(data, segmentsGrob(
           y0 = unit(yticks$y, "native"), y1 = unit(yticks$y, "native"),
           x0 = unit(1, "npc") - unit(yticks$start, "cm"), x1 = unit(1, "npc") - unit(yticks$end, "cm"),
-          gp = gpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth * .pt)
+          gp = ggpar(col = alpha(colour, alpha), lty = linetype, lwd = linewidth)
         ))
       }
     }

R/annotation-map.R

@@ -95,9 +95,9 @@ GeomAnnotationMap <- ggproto("GeomAnnotationMap", GeomMap,
 
     polygonGrob(coords$x, coords$y, default.units = "native",
       id = grob_id,
-      gp = gpar(
+      gp = ggpar(
         col = data$colour, fill = alpha(data$fill, data$alpha),
-        lwd = data$linewidth * .pt)
+        lwd = data$linewidth)
       )
   },
 

R/annotation.R

@@ -74,12 +74,19 @@ annotate <- function(geom, x = NULL, y = NULL, xmin = NULL, xmax = NULL,
   }
 
   data <- data_frame0(!!!position, .size = n)
+
+  params <- list2(na.rm = na.rm, ...)
+  reject <- intersect(names(params), c("position", "stat"))
+  if (length(reject) > 0) {
+    cli::cli_warn(
+      "{.fn annotate} can't accept {.or {.arg {reject}}} argument{?s}."
+    )
+    params <- params[setdiff(names(params), reject)]
+  }
+
   layer(
     geom = geom,
-    params = list(
-      na.rm = na.rm,
-      ...
-    ),
+    params = params,
     stat = StatIdentity,
     position = PositionIdentity,
     data = data,

R/autolayer.R

@@ -8,7 +8,7 @@
 #' @param ... other arguments passed to specific methods
 #' @return a ggplot layer
 #' @export
-#' @seealso [autoplot()], [ggplot()] and [fortify()]
+#' @family plotting automation topics
 autolayer <- function(object, ...) {
   UseMethod("autolayer")
 }

R/autoplot.R

@@ -1,3 +1,115 @@
+#' Tailoring plots to particular data types
+#'
+#' @description
+#' There are three functions to make plotting particular data types easier:
+#' `autoplot()`, `autolayer()` and `fortify()`. These are S3 generics for which
+#' other packages can write methods to display classes of data. The three
+#' functions are complementary and allow different levels of customisation.
+#' Below we'll explore implementing this series of methods to automate plotting
+#' of some class.
+#'
+#' Let's suppose we are writing a packages that has a class called 'my_heatmap',
+#' that wraps a matrix and we'd like users to easily plot this heatmap.
+#'
+#' ```r
+#' my_heatmap <- function(...) {
+#'   m <- matrix(...)
+#'   class(m) <- c("my_heatmap", class(m))
+#'   m
+#' }
+#'
+#' my_data <- my_heatmap(volcano)
+#' ```
+#'
+#' # Automatic data shaping
+#'
+#' One of the things we have to do is ensure that the data is shaped in the long
+#' format so that it is compatible with ggplot2. This is the job of the
+#' `fortify()` function. Because 'my_heatmap' wraps a matrix, we can let the
+#' fortify method 'melt' the matrix to a long format. If your data is already
+#' based on a long-format `<data.frame>`, you can skip implementing a
+#' `fortify()` method.
+#'
+#' ```r
+#' fortify.my_heatmap <- function(model, ...) {
+#'   data.frame(
+#'     row = as.vector(row(model)),
+#'     col = as.vector(col(model)),
+#'     value = as.vector(model)
+#'   )
+#' }
+#'
+#' fortify(my_data)
+#' ```
+#'
+#' When you have implemented the `fortify()` method, it should be easier to
+#' construct a plot with the data than with the matrix.
+#'
+#' ```r
+#' ggplot(my_data, aes(x = col, y = row, fill = value)) +
+#'   geom_raster()
+#' ```
+#'
+#' # Automatic layers
+#'
+#' A next step in automating plotting of your data type is to write an
+#' `autolayer()` method. These are typically wrappers around geoms or stats
+#' that automatically set aesthetics or other parameters. If you haven't
+#' implemented a `fortify()` method for your data type, you might have to
+#' reshape the data in `autolayer()`.
+#'
+#' If you require multiple layers to display your data type, you can use an
+#' `autolayer()` method that constructs a list of layers, which can be added
+#' to a plot.
+#'
+#' ```r
+#' autolayer.my_heatmap <- function(object, ...) {
+#'   geom_raster(
+#'     mapping = aes(x = col, y = row, fill = value),
+#'     data = object,
+#'     ...,
+#'     inherit.aes = FALSE
+#'   )
+#' }
+#'
+#' ggplot() + autolayer(my_data)
+#' ```
+#'
+#' As a quick tip: if you define a mapping in `autolayer()`, you might want
+#' to set `inherit.aes = FALSE` to not have aesthetics set in other layers
+#' interfere with your layer.
+#'
+#' # Automatic plots
+#'
+#' The last step in automating plotting is to write an `autoplot()` method
+#' for your data type. The expectation is that these return a complete plot.
+#' In the example below, we're exploiting the `autolayer()` method that we
+#' have already written to make a complete plot.
+#'
+#' ```r
+#' autoplot.my_heatmap <- function(object, ..., option = "magma") {
+#'   ggplot() +
+#'     autolayer(my_data) +
+#'     scale_fill_viridis_c(option = option) +
+#'     theme_void()
+#' }
+#'
+#' autoplot(my_data)
+#' ```
+#'
+#' If you don't have a wish to implement a base R plotting method, you
+#' can set the plot method for your class to the autoplot method.
+#'
+#' ```r
+#' plot.my_heatmap <- autoplot.my_heatmap
+#' plot(my_data)
+#' ```
+#'
+#' @family plotting automation topics
+#'
+#' @name automatic_plotting
+NULL
+
 #' Create a complete ggplot appropriate to a particular data type
 #'
 #' `autoplot()` uses ggplot2 to draw a particular plot for an object of a
@@ -8,7 +120,7 @@
 #' @param ... other arguments passed to specific methods
 #' @return a ggplot object
 #' @export
-#' @seealso [autolayer()], [ggplot()] and [fortify()]
+#' @family plotting automation topics
 autoplot <- function(object, ...) {
   UseMethod("autoplot")
 }