在数据帧的循环中使用 curl 库下载图像答案

【问题标题】：downloading images using curl library in a loop over data frame在数据帧的循环中使用 curl 库下载图像
【发布时间】：2019-06-13 05:29:30
【问题描述】：

我正在尝试下载我从 Google 表单（实际上是该表单的回复表）研究的鸟类的图像。在这个简短的数据集中，我有每只鸟的 ID、捕获日期，以及从不同角度（尾巴、frontPop 等）拍摄的多张鸟的图像。

我想遍历这个数据集，下载每种类型的图像，并通过鸟的 ID、捕获日期和图像类型（对应于列名）来命名：

189307140_2019-01-14_tail.jpg

我可以使用“curl”包从电子表格中的每个 Google Drive 链接下载图像，如下所示：

library(curl)
curl_download(url = 'https://drive.google.com/uc?export=download&id=1kra8bSf4WMpoK8BTyFip2OxmUuz30Thl', 
              destfile = 'bird.jpg')

但我被困在一个更优雅的循环中（在我的数据集中的每个鸟和照片类型上）......

以下是数据示例：

dat <- structure(list(markerID = c(189307136L, 189307145L, 183337360L
), date = structure(c(17907, 17910, 17910), class = "Date"), 
    tail = c("https://drive.google.com/uc?export=download&id=13S9s_j6acfndEz4HbpG-v2ZRyT8LLgji", 
    "https://drive.google.com/uc?export=download&id=19XpBTLws94wtCtgPF6oXKM_GZ_cV4oMf", 
    "https://drive.google.com/uc?export=download&id=1I5zA8tJdEv26EzI9rwBSN5tVsASyT4Cl"
    ), frontPop = c("https://drive.google.com/uc?export=download&id=1lJgl3hin9sWQcV40aJgdYPQM6jzch2Lb", 
    "https://drive.google.com/uc?export=download&id=1QdB1KmyHrlKkTlux0fkyI1Aw-Pe15sYF", 
    "https://drive.google.com/uc?export=download&id=1xqNT9CPVfMj2ksxqMONGbWXyJKaKtLuR"
    ), backPop = c("https://drive.google.com/uc?export=download&id=1YAe4S7_LIrLsbOW2qCBuWjyOr_SgT54T", 
    "https://drive.google.com/uc?export=download&id=1QOC8rPDjWfy6PVSaXFycf4jDyUnV-Vbv", 
    "https://drive.google.com/uc?export=download&id=1Gyo4lXgp0nXbsdd0jA_kk1m2jW8RGGIY"
    )), class = c("tbl_df", "tbl", "data.frame"), row.names = c(NA, 
-3L))

    # A tibble: 3 x 5
   markerID date       tail                                         frontPop                                     backPop                                     
      <int> <date>     <chr>                                        <chr>                                        <chr>                                       
1 189307136 2019-01-11 https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download…
2 189307145 2019-01-14 https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download…
3 183337360 2019-01-14 https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download… https://drive.google.com/uc?export=download…

【问题讨论】：

标签： r for-loop curl tidyverse

【解决方案1】：

这是使用library(purrr) 解决问题的一种优雅方法：

library(tidyverse)
library(curl)

dat <- structure(
  list(markerID = c(189307136L, 189307145L, 183337360L), 
       date = structure(c(17907, 17910, 17910), class = "Date"), 
       tail = c("https://drive.google.com/uc?export=download&id=13S9s_j6acfndEz4HbpG-v2ZRyT8LLgji", 
                "https://drive.google.com/uc?export=download&id=19XpBTLws94wtCtgPF6oXKM_GZ_cV4oMf", 
                "https://drive.google.com/uc?export=download&id=1I5zA8tJdEv26EzI9rwBSN5tVsASyT4Cl"), 
       frontPop = c("https://drive.google.com/uc?export=download&id=1lJgl3hin9sWQcV40aJgdYPQM6jzch2Lb", 
                    "https://drive.google.com/uc?export=download&id=1QdB1KmyHrlKkTlux0fkyI1Aw-Pe15sYF", 
                    "https://drive.google.com/uc?export=download&id=1xqNT9CPVfMj2ksxqMONGbWXyJKaKtLuR"), 
       backPop = c("https://drive.google.com/uc?export=download&id=1YAe4S7_LIrLsbOW2qCBuWjyOr_SgT54T", 
                   "https://drive.google.com/uc?export=download&id=1QOC8rPDjWfy6PVSaXFycf4jDyUnV-Vbv", 
                   "https://drive.google.com/uc?export=download&id=1Gyo4lXgp0nXbsdd0jA_kk1m2jW8RGGIY")), 
  class = c("tbl_df", "tbl", "data.frame"), 
  row.names = c(NA, -3L))

read_and_title <- as_mapper(~curl_download(url = ..4, 
                                           destfile = paste0(..1,"_",..2,"_",..3,".png")))

dat %>% 
  gather(key = "photo_type", value = "url", 3:5) %>% 
  pmap_chr(read_and_title)
#> [1] "189307136_17907_tail.png"     "189307145_17910_tail.png"    
#> [3] "183337360_17910_tail.png"     "189307136_17907_frontPop.png"
#> [5] "189307145_17910_frontPop.png" "183337360_17910_frontPop.png"
#> [7] "189307136_17907_backPop.png"  "189307145_17910_backPop.png" 
#> [9] "183337360_17910_backPop.png"

^{由reprex package (v0.2.1) 于 2019-01-18 创建}

【讨论】：