Downloading from Zenodo (2025)

download.R

# this has worked as of 2024-03-01
library(jsonlite)
library(dplyr)
library(tidyr)
library(stringr)

# Specific DOI - resolves to a fixed version
zenodo.id <- "6568295"
# We will recover the rest from Zenodo API
zenodo.api = "https://zenodo.org/api/records/"
# where to save
dataloc <- here::here("data","downloaded")
if ( ! file.exists(dataloc)) {
  dir.create(dataloc,recursive = TRUE)
}


download.file(paste0(zenodo.api,zenodo.id),destfile=file.path(dataloc,"metadata.json"))
latest <- fromJSON(file.path(dataloc,"metadata.json"))

print(paste0("DOI: ",latest$links$doi))
print(paste0("Current: ",latest$links$html))
print(paste0("Latest: ",latest$links$latest_html))


# we download all the files! (Implement some selection if only interested in specific files)
file.list <- as.data.frame(latest$files$links) %>%  gather() 

for ( value in file.list$value ) {
  print(value)
  file.name <- basename(value %>% str_remove("/content"))
    message(paste0("Downloading... ",file.name))
    #download.file(value,destfile=file.path(dataloc,file.name))
}

download_with_python.sh

# this achieves the same thing as the various R and Stata code above
pip install zenodo_get
dataloc="data/downloaded"
[[ -d $dataloc ]] || mkdir -p "$dataloc"
zenodo_get 6568295 -o "$dataloc"

download_zenodo.do

// Download files from Zenodo in Stata
global zenodoid "6568295"
global zenodoapi "https://zenodo.org/api/records/"
global zenodosfx "/content"

foreach year of numlist 2018/2020 {
  foreach dmon of numlist 1/12 {
    if `dmon' < 10 {
       local month "0`dmon'"
       }
    else {
       local month "`dmon'"
    foreach dval of numlist 1/31 {
       if `dval' < 10 {
         local day "0`dval'"
         }
       else {
         local day "`dval'"
         }
         // now download
         
         local filename "sucursales_`year'-`month'-`day'.csv.gz"
         local url "${zenodoprefix}${zenodoid}/`filename'${zenodosfx}"
         copy `url' `filename'
         
       } // end day
    } // end month
  } // end year  
     
         




// Alternatively, just hard-code the paths once!

copy https://zenodo.org/api/records/6568295/files/sucursales_2018-02-27.csv.gz/content sucursales_2018-02-27.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-02-28.csv.gz/content sucursales_2018-02-28.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-01.csv.gz/content sucursales_2018-03-01.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-02.csv.gz/content sucursales_2018-03-02.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-03.csv.gz/content sucursales_2018-03-03.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-04.csv.gz/content sucursales_2018-03-04.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-05.csv.gz/content sucursales_2018-03-05.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-06.csv.gz/content sucursales_2018-03-06.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-07.csv.gz/content sucursales_2018-03-07.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-08.csv.gz/content sucursales_2018-03-08.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-09.csv.gz/content sucursales_2018-03-09.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-10.csv.gz/content sucursales_2018-03-10.csv.gz
copy https://zenodo.org/api/records/6568295/files/sucursales_2018-03-11.csv.gz/content sucursales_2018-03-11.csv.gz