仅使用基数R将`=`转换为`<-`

时间:2018-10-28 19:06:37

标签: r parsing text-styling

问题

  1. 将分配等号变成分配箭头。
  2. 仅使用基数R(不能使用stylerformatR)。

上下文

https://github.com/ropensci/drake/issues/562

示例

输入:

f = function(x = 1){}

所需的输出:

f <- function(x = 1){}

1 个答案:

答案 0 :(得分:2)

已发布,但不妨尝试一些技巧:

library(magrittr)

raw_src <- "z = {f('#') # comment

x <- 5
y = 'test'
    }"

# so we can have some tasty parse data
first <- parse(text = raw_src, keep.source = TRUE)

# this makes a nice data frame of the tokenized R source including line and column positions of the source bits
src_info <- getParseData(first, TRUE)

# only care about those blasphemous = assignments
elements_with_equals_assignment <- subset(src_info, token == "EQ_ASSIGN")

# take the source and split it into lines
raw_src_lines <- strsplit(raw_src, "\n")[[1]]

# for as many instances in the data frame replace the = with <-
for (idx in 1:nrow(elements_with_equals_assignment)) {
  stringi::stri_sub(
    raw_src_lines[elements_with_equals_assignment[idx, "line1"]],
    elements_with_equals_assignment[idx, "col1"],
    elements_with_equals_assignment[idx, "col2"]
  ) <- "<-"
}

# put the lines back together and do the thing
parse(
  text = paste0(raw_src_lines, collapse="\n"),
  keep.source = FALSE
)[[1]] %>%
  deparse() %>%
  cat(sep = "\n")
## z <- {
##     f("#")
##     x <- 5
##     y <- "test"
## }