將水平圖例包裹在多行中

Question

假設我有以下數據：

    lab <- "A really really long string!"
    dat <- data.frame(grp = paste(1:6,lab),x=1:6,y=runif(6))

當用字符串繪制這個漫長的傳奇時，有時候讓傳說很合適可能是一個挑戰。 如果必須的話，我總是可以縮短字符串以縮短它們，但我想知道是否有可能（最有可能使用一些grid魔法）將圖例“包裹”在多個行或列中。 例如，假設我將圖例水平放置在底部：

    ggplot(dat,aes(x=x,y=y,colour=grp)) + geom_point() + 
        opts(legend.position="bottom",legend.direction="horizontal")

是否有可能將此圖例顯示為兩行三行，而不是一行六行？

Answer 1

要包裝長字符串，請使用strwrap 。

lipsum <- "Lorem ipsum dolor sit amet, consectetur adipiscing elit. Curabitur ullamcorper tellus vitae libero placerat aliquet egestas tortor semper. Maecenas pellentesque euismod tristique. Donec semper interdum magna, commodo vehicula ante hendrerit vitae. Maecenas at diam sollicitudin magna mollis lobortis. In nibh elit, tincidunt eu lobortis ac, molestie a felis. Proin turpis leo, iaculis non commodo quis, venenatis at justo. Duis in magna vel erat fringilla gravida quis non nisl. Nunc lacus magna, varius eu luctus vel, luctus tristique sapien. Suspendisse mi dolor, vestibulum at facilisis elementum, lacinia vitae metus. Etiam ut nisl urna, vel tempus mi. In hac habitasse platea dictumst. Quisque pretium volutpat felis, nec tempor diam faucibus at. Praesent volutpat posuere sapien, eu vulputate risus molestie vitae. Proin iaculis quam non leo porttitor hendrerit."

strwrap(lipsum)
cat(strwrap(lipsum), sep = "\n")
# Lorem ipsum dolor sit amet, consectetur adipiscing elit. Curabitur ullamcorper tellus
# vitae libero placerat aliquet egestas tortor semper. Maecenas pellentesque euismod
# tristique. Donec semper interdum magna, commodo vehicula ante hendrerit vitae. Maecenas
# at diam sollicitudin magna mollis lobortis. In nibh elit, tincidunt eu lobortis ac,
# molestie a felis. Proin turpis leo, iaculis non commodo quis, venenatis at justo. Duis
# in magna vel erat fringilla gravida quis non nisl. Nunc lacus magna, varius eu luctus
# vel, luctus tristique sapien. Suspendisse mi dolor, vestibulum at facilisis elementum,
# lacinia vitae metus. Etiam ut nisl urna, vel tempus mi. In hac habitasse platea
# dictumst. Quisque pretium volutpat felis, nec tempor diam faucibus at. Praesent
# volutpat posuere sapien, eu vulputate risus molestie vitae. Proin iaculis quam non leo
# porttitor hendrerit.

Answer 2

試試這個。 我為很長的標題寫了這個，但它適用於任何長字符串。 你仍然需要弄清楚你的實例的線長。

# splits title of plot if to long
splittitle=function(title,linelength=40)
{
    spltitle<-strsplit(title,' ')
    splt<-as.data.frame(spltitle)
    title2<-NULL
    title3<-NULL
    titlelength<-round(nchar(title)/round(nchar(title)/linelength))
    dimsplt<-dim(splt)
    n=1
    doonce2=0
    for(m in 1:round(nchar(title)/linelength)){
  doonce=0
    doonce2=0
    for(l in n:dimsplt[1]){
        if(doonce==0){title2<-title3}
        title2=paste(title2,splt[l,],sep=' ')
        if(doonce2==0){if(nchar(title2)>=(titlelength*m)){title3=paste(title2,'\n',sep='')
        n<-(l+1)
        doonce2=1}
        }
        doonce=1
    }
    }
    title2
}

lab <- "A really really long string!A really really long string!A really really long string!A really really long string!A really really long string!A really really long string!A really really long string!A really really long string!"
lab2<-splittitle(lab)
cat(lab)
cat(lab2)


library('ggplot2')

1個原件

dat <- data.frame(grp = paste(1:6,lab2),x=1:6,y=runif(6))

ggplot(dat,aes(x=x,y=y,colour=grp)) + geom_point() + 
    opts(legend.position="bottom",legend.direction="horizontal")

2使用splittitle

dat <- data.frame(grp = paste(1:6,lab2),x=1:6,y=runif(6))
ggplot(dat,aes(x=x,y=y,colour=grp)) + geom_point() + 
    opts(legend.position="bottom",legend.direction="horizontal")

Answer 3

前面提到的splittitle 幾乎可以工作，但是例如

> splittitle("abc defg hi jkl m", 6)
[1] " abc defg\\n hi\\n jkl m"

並沒有真正給你你想要的東西......

一個技巧是使用RGraphics::splitString

“將一個字符串拆分成多行（通過插入換行符），以便輸出適合當前視口。”

然后，您只需暫時更改視口。 下面的函數為我做了訣竅，但仍然只是一個快速和骯臟的解決方案。 我用它來包裝一個傳奇標題。

library(RGraphics)
multiLines <- function(text, maxWidth=11) {
  textLen = nchar(text)
  maxHeight = ceiling(textLen/maxWidth)*1.5
  vp=viewport(width=maxWidth,height=maxHeight, default.units="char")
  pushViewport(vp) #activate the viewport
  text2 = splitString(text) #given vp, split the text
  popViewport() #get rid of it
  return(text2)
}