Question

我获得了一些使用情况统计数据。其数月的数据分成多个缺少标题的TSV文件;

07-01-2017_01.tsv
07-01-2017_02.tsv
07-02-2017_01.tsv
07-02-2017_02.tsv
07-03-2017_01.tsv
07-03-2017_02.tsv
07-04-2017_01.tsv
07-04-2017_02.tsv
07-04-2017_03.tsv

我想每天合并数据，添加标题并将其导出为CSV格式。我设法用以下代码做了1天，但我想知道是否有任何方法可以自动化它，这样我就不需要为每个月的每一天运行代码。

  data_part1 <-  read.delim("~/07-01-2017_01.tsv", header = FALSE, sep = "\t", quote = "", stringsAsFactors=FALSE)
  data_part2 <-  read.delim("~/07-01-2017_02.tsv", header = FALSE, sep = "\t", quote = "", stringsAsFactors=FALSE)
  data_merged <- rbind(data_part1, data_part2)
  names(data_merged) <-
    c(
      "post_visid_high",
      "post_visid_low",
      "quarterly_visitor",
      "visid_timestamp",
      "visid_type",
      "visit_keywords",
      "visit_num",
      "visit_page_num",
      "visit_ref_domain",
      "visit_ref_type",
      "visit_referrer",
      "visit_search_engine",
      "visit_start_page_url",
      "visit_start_pagename",
      "visit_start_time_gmt",
    )
  write.csv(data_merged, "~/07-01-2017_02.csv")

预期产出

07-01-2017_merged.csv
07-02-2017_merged.csv
07-03-2017_merged.csv
07-04-2017_merged.csv

Answer 1

您可以这样做：

setwd("path")

temp = list.files(pattern = "*.tsv")

allCsv <- data.frame()

for(file in 1:length(temp)){

  currFile <- temp[file]
  k <- read.tsv(currFile)
  allCsv <- rbind(allCsv,k)
  }

colnames(allCsv) =c(
                       "post_visid_high",
                       "post_visid_low",
                       "quarterly_visitor",
                       "visid_timestamp",
                       "visid_type",
                       "visit_keywords",
                       "visit_num",
                       "visit_page_num",
                       "visit_ref_domain",
                       "visit_ref_type",
                       "visit_referrer",
                       "visit_search_engine",
                       "visit_start_page_url",
                       "visit_start_pagename",
                       "visit_start_time_gmt",
                     )

write.csv(allCsv, "outputPath")

Answer 2

您只需根据具体情况调整列表解决方案即可。 for循环正常工作：

all_files = list.files(pattern = "\\.tsv$")
output_path = "/path/to/output_dir/"
cnames =  c(
      "post_visid_high",
      "post_visid_low",
      "quarterly_visitor",
      "visid_timestamp",
      "visid_type",
      "visit_keywords",
      "visit_num",
      "visit_page_num",
      "visit_ref_domain",
      "visit_ref_type",
      "visit_referrer",
      "visit_search_engine",
      "visit_start_page_url",
      "visit_start_pagename",
      "visit_start_time_gmt"
    )

all_files = data.frame(fn = all_files, date = sub(pattern = "_[0-9]+\\.tsv", replacement = "",  x = all_files))

for (d in unique(all_files$date)) {
  data_list = lapply(all_files$fn[all_files$date == d],
    read.delim, header = FALSE, sep = "\t",
    quote = "", stringsAsFactors = FALSE
  )
  merged_data = do.call(rbind, data_list)
  names(merged_data) = cnames
  write.csv(merged_data, paste0(output_path, d, "_merged.csv"))
}

如果您想提高速度，请使用library(data.table)并将read.delim替换为fread，将do.call(rbind...)替换为rbindlist(data_list)，将write.csv替换为fwrite protected void btnsubmit_Click(object sender, EventArgs e) { i = Int32.Parse(lblStart.Text); i++; lblStart.Text = i.ToString(); }。

自动合并TSV中的数据，导出CSV

2 个答案: