Wei Yang Tham

PhD candidate in Economics at The Ohio State University

Downloading zip files in R

1 minute read

In theory this should be a straightforward task with the download.file function, but as I recently found, it’s a little trickier than that (but really not that much trickier).1 The main things you have to do are

  1. Specify mode = "wb" in download.file
  2. Use unz to unzip

Example code below. I used a real url in the example that downloads data on NIH grants from this page.

# download zipped folder
download.file(url = "https://exporter.nih.gov/CSVs/final/RePORTER_PRJ_C_FY2013.zip", 
              destfile = "destination/folder/RePORTER_PRJ_C_FY2013.zip", 
              mode = "wb")

# extract file from zipped folder
x = readr::read_csv(unz(description = "/destination/folder/RePORTER_PRJ_C_FY2013.zip", 
                        filename = "RePORTER_PRJ_C_FY2013.csv"))

This is what you should get at the end:

## # A tibble: 100 × 45
##    APPLICATION_ID ACTIVITY ADMINISTERING_IC APPLICATION_TYPE ARRA_FUNDED
##             <int>    <chr>            <chr>            <int>       <chr>
## 1         8504974      R01               CA                5           N
## 2         8710587      P30               CA                3           N
## 3         8578352      R01               CA                2           N
## 4         8519269      U51               IP                5           N
## 5         8476221      T34               GM                5           N
## 6         8514650      R01               GM                5           N
## 7         8415817      R01               NS                5           N
## 8         8519478      R01               GM                5           N
## 9         8516121      R01               NS                5           N
## 10        8633290      R43               DK                1           N
## # ... with 90 more rows, and 40 more variables: AWARD_NOTICE_DATE <chr>,
## #   BUDGET_START <chr>, BUDGET_END <chr>, CFDA_CODE <int>,
## #   CORE_PROJECT_NUM <chr>, ED_INST_TYPE <chr>, FOA_NUMBER <chr>,
## #   FULL_PROJECT_NUM <chr>, FUNDING_ICs <chr>, FUNDING_MECHANISM <chr>,
## #   FY <int>, IC_NAME <chr>, NIH_SPENDING_CATS <chr>, ORG_CITY <chr>,
## #   ORG_COUNTRY <chr>, ORG_DEPT <chr>, ORG_DISTRICT <int>, ORG_DUNS <int>,
## #   ORG_FIPS <chr>, ORG_NAME <chr>, ORG_STATE <chr>, ORG_ZIPCODE <chr>,
## #   PHR <chr>, PI_IDS <chr>, PI_NAMEs <chr>, PROGRAM_OFFICER_NAME <chr>,
## #   PROJECT_START <chr>, PROJECT_END <chr>, PROJECT_TERMS <chr>,
## #   PROJECT_TITLE <chr>, SERIAL_NUMBER <int>, STUDY_SECTION <chr>,
## #   STUDY_SECTION_NAME <chr>, SUBPROJECT_ID <int>, SUFFIX <chr>,
## #   SUPPORT_YEAR <int>, DIRECT_COST_AMT <int>, INDIRECT_COST_AMT <int>,
## #   TOTAL_COST <int>, TOTAL_COST_SUB_PROJECT <int>

  1. I relied on this Stack Overflow post to figure this out, and it has a few variants on the solution that I’m using here. ↩

Tags:

Updated:

Leave a Comment

You May Also Enjoy