Example files/directory

让我们创建一个包含一些示例文件的示例目录.这将让我们证明该解决方案是有效的，并且是可重复的解决方案的关键.

library(dplyr)
library(writexl)
library(readxl)

# Example directory with example files
dir.create(path = "KOMP")
write.csv(data.frame(x = 5), file = "KOMP/foo.csv")
write.csv(data.frame(x = 20), file = "KOMP/foo.nrrd.csv")
write.csv(data.frame(x = 1), file = "KOMP/foo2.nrrd.csv")
write.csv(data.frame(z = 2), file = "KOMP/bar.csv")
write.csv(data.frame(z = 5), file = "KOMP/bar.rrdr.csv")

# Example Excel file
write_xlsx(data.frame(name = c("foo", "hotdog")),
           path = "KOMP/all_data.xlsx")

Solution

现在，我们可以使用我们的示例文件和目录来显示问题的解决方案.

# Get file paths in a data.frame for those that contain ".nrrd"
# Use data.frame() to avoid row names instead of as.data.frame()
# Need to use \\ to escape the period in the regular expression
file_names <- list.files(
  path = "KOMP/",
  pattern = "\\.nrrd",
  all.files = TRUE,
  full.names = TRUE,
  recursive = TRUE
) %>%
  data.frame(paths = .)

# Extract part of file name (i.e. removing directory substrings) that
# comes before .nrrd and add a column. Can get file name with basename()
# and use regular expressions for the other part.
file_names$match_string <- file_names %>%
  pull(paths) %>%
  basename() %>%
  gsub(pattern = "\\.nrrd.*", replacement = "")

file_names$match_string
#> [1] "foo"  "foo2"

# Read in excel file with file names to match (if possible)
name_data <- read_excel("KOMP/all_data.xlsx")

name_data$name
#> [1] "foo"    "hotdog"

# Create match indicator and row number
name_data <- name_data %>%
  mutate(
    matched = case_when(name %in% file_names$match_string ~ 1,
                        TRUE ~ 0),
    rowID = row_number()
  )

# Create excel spreadsheet of files already downloaded
name_data %>%
  filter(matched == 1) %>%
  write_xlsx(path = "KOMP/already_downloaded.xlsx")