The grid package provides low-level functions to create graphical objects (grobs), and position them on a page in specific viewports. The gtable package introduced a higher-level layout scheme, arguably more amenable to user-level interaction. With the arrangeGrob/grid.arrange() pair of functions, gridExtra builds upon gtable to arrange multiple grobs on a page.
In this example we mix a few grobs and plots,
library(gridExtra) library(grid) library(ggplot2) library(lattice) p <- qplot(1,1) p2 <- xyplot(1~1) r <- rectGrob(gp=gpar(fill="grey90")) t <- textGrob("text") grid.arrange(t, p, p2, r, ncol=2)gs <- lapply(1:9, function(ii) grobTree(rectGrob(gp=gpar(fill=ii, alpha=0.5)), textGrob(ii))) grid.arrange(grobs=gs, ncol=4, top="top label", bottom="bottom\nlabel", left="left label", right="right label") grid.rect(gp=gpar(fill=NA))We can provide a matrix defining the layout,
lay <- rbind(c(1,1,1,2,3), c(1,1,1,4,5), c(6,7,8,9,9)) grid.arrange(grobs = gs, layout_matrix = lay)The layout itself may contain holes, but note that for any given grob index the region must be simply connected (no hole),
hlay <- rbind(c(1,1,NA,2,3), c(1,1,NA,4,NA), c(NA,7,8,9,NA)) select_grobs <- function(lay) { id <- unique(c(t(lay))) id[!is.na(id)] } grid.arrange(grobs=gs[select_grobs(hlay)], layout_matrix=hlay)All cells are of equal size by default, but users may pass explicity widths and/or heights in any valid grid units, or as relative numbers (interpreted as null),
grid.arrange(grobs=gs[1:3], ncol=2, widths = 1:2, heights=unit(c(1,10), c("in", "mm")))arrangeGrobThe grid.arrange() function draws on the device; for more complex layouts, we may want to store the gtable and combine it with other objects, e.g. forming nested layouts. To this end, use arrangeGrob(),
g1 <- arrangeGrob(grobs = gs, layout_matrix = t(lay)) g2 <- arrangeGrob(grobs = gs, layout_matrix = lay) grid.arrange(g1, g2, ncol=2)Finally, we may want to place grobs on multiple pages; the marrangeGrob() function provides a convenient interface for this, also compatible with ggsave().
set.seed(123) pl <- lapply(1:11, function(.x) qplot(1:10, rnorm(10), main=paste("plot", .x))) ml <- marrangeGrob(pl, nrow=2, ncol=2) ## non-interactive use, multipage pdf ## ggsave("multipage.pdf", ml) ## interactive use; calling `dev.new` multiple times ml