我有以下包含文件路径的示例数据。
在我的数据文件夹中，有带有站点标识符的子文件夹。每个子文件夹包含特定于该站点的数据。例如，文件夹`data/A101/`包含5个数据文件`bill1-separated.xlsx`，`bill1.xlsx`，`bill2.xls`，`bill2-separated.xlsx`，`bill3.xlsx`。
我只想保留文件路径中名称中包含“separated”的文件，前提是存在一个分隔版本和原始副本。例如，对于`bill1-separated.xlsx`和`bill1.xlsx`，我只想保留“separated”版本。
最终我想要这样的结果：
```R
desired <- c(NA_character_, "data/A101/bill1-separated.xlsx", NA_character_, "data/B215/usage2-separated.xlsx",
             NA_character_, "data/A101/bill2-separated.xlsx", NA_character_, "data/C145/account1-separated.xlsx",
             "data/A101/bill3.xlsx", "data/B215/usage1.csv", "data/B215/usage3.xls", "data/C145/account2.xlsx") %>% sort()

library(stringr)
library(dplyr)
# 尝试
temp <- data.frame(path = paths)
temp %>% 
  mutate(
    file = str_extract(paths, "\\/.*\\/(.+)", group=1),
    attempt = ifelse(!grepl("separated", file),
                     NA, # 是否嵌套另一个ifelse？但不确定要使用什么条件/正则表达式
                     path
                     )
  )

<details>
<summary>英文:</summary>
I have the following example data that contain file paths.

In my data folder, there are subfolders with identifiers for sites. Each subfolder contains data specific to the site. For example, folder `data/A101/` contains 5 data files `bill1-separated.xlsx`, `bill1.xlsx`, `bill2.xls`, `bill2-separated.xlsx`, `bill3.xlsx`.
I want to keep only the file paths with "separated" in the name only if there are a separated version and the original copy. For example, for `bill1-separated.xlsx` and `bill1.xlsx`, I want to keep only the "separated" version.
I want something like this in the end:

My attempt was to use `ifelse` and `grepl` to capture the correct file paths, but I couldn't really get it to work. Any help will be much appreciated.

</details>
# 答案1
**得分**: 2
由于您忽略了文件扩展名，我认为我们可以使用这个方案。（我从您的代码中推测出了 [tag:dplyr]。）
```r
library(dplyr)
df %>%
  mutate(
    dir = dirname(path),
    pathnoext = tools::file_path_sans_ext(path),
    notsep = sub("-separated", "", pathnoext)
  ) %>%
  mutate(
    res = if_else(any(grepl("-separated", path)) & pathnoext == notsep, 
                  path[NA], path),
    .by = notsep
  ) %>%
  select(path, res, desire, everything())
#                                 path                               res                            desire       dir                    pathnoext             notsep
# 1               data/A101/bill1.xlsx                              <NA>                              <NA> data/A101              data/A101/bill1    data/A101/bill1
# 2     data/A101/bill1-separated.xlsx    data/A101/bill1-separated.xlsx    data/A101/bill1-separated.xlsx data/A101    data/A101/bill1-separated    data/A101/bill1
# 3               data/B215/usage2.csv                              <NA>                              <NA> data/B215             data/B215/usage2   data/B215/usage2
# 4    data/B215/usage2-separated.xlsx   data/B215/usage2-separated.xlsx   data/B215/usage2-separated.xlsx data/B215   data/B215/usage2-separated   data/B215/usage2
# 5                data/A101/bill2.xls                              <NA>                              <NA> data/A101              data/A101/bill2    data/A101/bill2
# 6     data/A101/bill2-separated.xlsx    data/A101/bill2-separated.xlsx    data/A101/bill2-separated.xlsx data/A101    data/A101/bill2-separated    data/A101/bill2
# 7            data/C145/account1.xlsx                              <NA>                              <NA> data/C145           data/C145/account1 data/C145/account1
# 8  data/C145/account1-separated.xlsx data/C145/account1-separated.xlsx data/C145/account1-separated.xlsx data/C145 data/C145/account1-separated data/C145/account1
# 9               data/A101/bill3.xlsx              data/A101/bill3.xlsx              data/A101/bill3.xlsx data/A101              data/A101/bill3    data/A101/bill3
# 10              data/B215/usage1.csv              data/B215/usage1.csv              data/B215/usage1.csv data/B215             data/B215/usage1   data/B215/usage1
# 11              data/B215/usage3.xls              data/B215/usage3.xls              data/B215/usage3.xls data/B215             data/B215/usage3   data/B215/usage3
# 12           data/C145/account2.xlsx           data/C145/account2.xlsx           data/C145/account2.xlsx data/C145           data/C145/account2 data/C145/account2

df <- structure(list(path = c("data/A101/bill1.xlsx", "data/A101/bill1-separated.xlsx", "data/B215/usage2.csv", "data/B215/usage2-separated.xlsx", "data/A101/bill2.xls", "data/A101/bill2-separated.xlsx", "data/C145/account1.xlsx", "data/C145/account1-separated.xlsx",  "data/A101/bill3.xlsx", "data/B215/usage1.csv", "data/B215/usage3.xls", "data/C145/account2.xlsx"), desire = c(NA, "data/A101/bill1-separated.xlsx", NA, "data/B215/usage2-separated.xlsx", NA, "data/A101/bill2-separated.xlsx", NA, "data/C145/account1-separated.xlsx",  "data/A101/bill3.xlsx", "data/B215/usage1.csv", "data/B215/usage3.xls", "data/C145/account2.xlsx")), class = "data.frame", row.names = c(NA, -12L))