尽管deparse（替代），函数内部的ggplot不起作用答案

【问题标题】：ggplot inside function not working despite deparse(substitute尽管deparse（替代），函数内部的ggplot不起作用
【发布时间】：2014-05-01 01:15:40
【问题描述】：

我以http://www.ling.upenn.edu/~joseff/rstudy/plots/graphics/phila_bar.R 的示例和http://www.ling.upenn.edu/~joseff/rstudy/summer2010_ggplot2_intro.html 的命令为例（在此页面上搜索 Gender:Name）。我试图将命令放在一个函数中，并尝试在变量名周围不带引号的情况下调用该函数，但尽管使用了替换（deparse...：

phil <- structure(list(Name = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 
3L, 3L, 3L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 
4L), .Label = c("Bucks County", "Delaware County", "Montgomery County", 
"Philadelphia County"), class = "factor"), Gender = structure(c(1L, 
1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 
2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 
1L, 1L, 2L, 2L, 2L, 2L, 2L), .Label = c("Female", "Male"), class = "factor"), 
    value = c(15288L, 43054L, 25788L, 62853L, 33622L, 50792L, 
    27184L, 71321L, 43593L, 94877L, 32442L, 54872L, 43751L, 18346L, 
    25545L, 87732L, 46656L, 63640L, 39675L, 25468L, 43636L, 34558L, 
    59923L, 26979L, 17550L, 27492L, 71404L, 39946L, 50107L, 96580L, 
    24957L, 17433L, 31468L, 40585L, 53239L, 21899L, 62542L, 38352L, 
    47008L, 31485L), Level = structure(c(1L, 4L, 2L, 5L, 3L, 
    3L, 1L, 4L, 2L, 5L, 3L, 5L, 4L, 1L, 2L, 5L, 3L, 4L, 2L, 1L, 
    4L, 3L, 5L, 2L, 1L, 1L, 4L, 2L, 3L, 5L, 2L, 1L, 3L, 4L, 5L, 
    1L, 5L, 3L, 4L, 2L), .Label = c("LessHigh", "High", "SomeAssoc", 
    "Bachelors", "GradProf"), class = "factor")), .Names = c("Name", 
"Gender", "value", "Level"), class = "data.frame", row.names = c(3463L, 
3465L, 3466L, 3467L, 3468L, 3471L, 3473L, 3475L, 3478L, 3479L, 
8741L, 8742L, 8743L, 8746L, 8750L, 8751L, 8752L, 8754L, 8756L, 
8757L, 22925L, 22926L, 22927L, 22928L, 22929L, 22933L, 22937L, 
22938L, 22939L, 22940L, 25844L, 25845L, 25846L, 25847L, 25849L, 
25854L, 25855L, 25856L, 25858L, 25860L))


testfn = function(gdf, Level, value, Gender, Name){
    print(
    ggplot(gdf, aes(deparse(substitute(Level)), deparse(substitute(value)),
         color = deparse(substitute(Name)),
         linetype = deparse(substitute(Gender)),
         group = deparse(substitute(Gender)):deparse(substitute(Name))))+
       geom_point()+
       geom_line() 
    )
}

testfn(phil, Level, value, Gender, Name)
Error in deparse(substitute(Gender)):deparse(substitute(Name)) : 
  NA/NaN argument
In addition: Warning messages:
1: In deparse(substitute(Gender)):deparse(substitute(Name)) :
  numerical expression has 4 elements: only the first used
2: In deparse(substitute(Gender)):deparse(substitute(Name)) :
  numerical expression has 5 elements: only the first used
3: In eval(expr, envir, enclos) : NAs introduced by coercion
4: In eval(expr, envir, enclos) : NAs introduced by coercion
5: In deparse(substitute(Gender)):deparse(substitute(Name)) :
  numerical expression has 4 elements: only the first used
6: In deparse(substitute(Gender)):deparse(substitute(Name)) :
  numerical expression has 5 elements: only the first used
7: In eval(expr, envir, enclos) : NAs introduced by coercion
8: In eval(expr, envir, enclos) : NAs introduced by coercion
>

问题可能出在哪里。感谢您的帮助。

【问题讨论】：

在testfn 中，您有gdf 作为参数但没有使用它（请用它替换phil）。您是否尝试删除 deparse 电话？此外，即使使用正确的语法，group= 行也不会工作，ggplot 也不会像 linetype 那样切换。
感谢您指出这一点。我已经在上面的函数中更正了它。但是错误仍然完全相同。它必须在访问 gdf 之前发生。我实际上已经按照此页面上的建议放入了 deparse 部分：stackoverflow.com/questions/23382822/…
你没有使用aes_string有什么特别的原因吗？
只是因为不带引号更容易打字。

标签： r plot ggplot2

【解决方案1】：

你确定吗？（看起来删除deparse 使一些我认为可能不起作用的东西，实际上也起作用了:-)

testfn = function(gdf, Level, value, Gender, Name) {

  gg <- ggplot(gdf, aes(x = substitute(Level),
                        y = substitute(value),
                        color = substitute(Name),
                        linetype = substitute(Gender),
                        group = substitute(Gender):substitute(Name)
  ))

  gg <- gg + geom_point()
  gg <- gg + geom_line() 

  print(gg)

}

testfn(phil, Level, value, Gender, Name)

【讨论】：