**Source URL:** https://general.veevavault.dev/sitevault/direct-data-api/getting-started/download-direct-data-file.md # Download a Direct Data File As an alternative to using the [Vault Postman Collection](/library/references/developer-resources#Postman_Resources), you can use the example Shell script below to download Direct Data files from your Vault. The script uses your provided credentials and the filters specified to download all file parts of the latest available Direct Data file. Your security profile must have the [appropriate permissions](/direct-data-api/references/direct-data-permissions) to use Direct Data API endpoints. The script requires the following variables: * **vault_dns**: The DNS of your Vault. * **session_id**: An active session ID for the Vault. * **extract_type**: The type of extract you wish to download: `full_directdata`, `incremental_directdata`, or `log_directdata` * **start_time**: The start time for which to capture data that will be in the Direct Data file, in the format `YYYY-MM-DDTHH:MMZ`. Always use `2000-01-01T00:00Z` if `extract_type=full_directdata`. * **stop_time**: The stop time for which to capture data that will be in the Direct Data file, in the format `YYYY-MM-DDTHH:MMZ`. This script runs natively in OSX/UNIX systems. On Windows operating systems, the script requires Bash. If you have Git installed, you can use Git Bash. Run this script from the directory where you would like to download the Direct Data file. If there are multiple file parts, the script combines them into a single .tar.gz file. ``` # Add the vault_dns of your Vault vault_dns="your-vault.veevavault.com" # Add in your session ID session_id="YOUR_SESSION_ID" # Add "full_directdata", "incremental_directdata", or "log_directdata" extract_type="full_directdata" # For "full_directdata" always use 2000-01-01T00:00Z as the start_time start_time="2000-01-01T00:00Z" # Add the stop_time stop_time="2024-06-01T15:15Z" # Will place the files in the current folder where the script runs target_directory="$(pwd)" # Perform the API call to retrieve the list of Direct Data files direct_data_file_list_response=$(curl -s -X GET -H "Authorization: $session_id" \ -H "Accept: application/json" \ "https://$vault_dns/api/v24.1/services/directdata/files?extract_type=$extract_type&start_time=$start_time&stop_time=$stop_time") # Extract the response status from the API response response_status=$(echo "$direct_data_file_list_response" | grep -o '"responseStatus":"[^"]*' | sed 's/"responseStatus":"//') # Check if the API call was successful if [ "$response_status" != "SUCCESS" ]; then error_message=$(echo "$direct_data_file_list_response" | grep -o '"message":"[^"]*' | sed 's/"message":"//' | tr -d '"') if [ -z "$error_message" ]; then printf "Retrieve Available Direct Data Files call failed. Exiting script.\n" else printf "Retrieve Available Direct Data Files call failed with error: %s\n" "$error_message" fi exit 1 else printf "Retrieve Available Direct Data Files call succeeded.\n" # Extract data array data=$(echo "$direct_data_file_list_response" | grep -o '"data":\[[^]]*\]' | sed 's/"data":\[//' | tr -d ']') # Count file parts fileparts=$(echo "$data" | grep -o '"fileparts":[0-9]*' | sed 's/"fileparts"://') # Check if fileparts is null or empty if [ -z "$fileparts" ]; then printf "No Direct Data Extract Files found for '$extract_type' with start_time = '$start_time' and stop_time = '$stop_time'.\n" exit 0 fi if [ "$fileparts" -gt 1 ]; then printf "Multiple file parts.\n" # Handling multiple file parts filepart_details=$(echo "$data" | grep -o '"filepart_details":\[{"[^]]*' | sed 's/"filepart_details":\[//' | tr -d ']') filepart_details=$(echo "$filepart_details" | sed 's/},{/}\n{/g') filename=$(echo "$data" | grep -o '"filename":"[^"]*' | sed 's/"filename":"//' | tr -d '"' | head -n 1) while IFS= read -r filepart_detail; do filepart_url=$(echo "$filepart_detail" | grep -o '"url":"[^"]*' | sed 's/"url":"//' | tr -d '"') output_filepart_name=$(echo "$filepart_detail" | grep -o '"filename":"[^"]*' | sed 's/"filename":"//' | tr -d '"') curl -o "$output_filepart_name" -X GET -H "Authorization: $session_id" \ -H "Accept: application/json" \ "$filepart_url" done <<< "$filepart_details" # Combine file parts name=$(echo "$data" | grep -o '"name":"[^"]*' | sed 's/"name":"//' | tr -d '"' | head -n 1) cat "$name."* > "$filename" full_path="$target_directory/$name" if [ ! -d "$full_path" ]; then # Directory does not exist, create it mkdir -p "$full_path" printf "Directory '%s' created.\n" "$full_path" else printf "Directory '%s' already exists.\n" "$full_path" fi tar -xzvf "$filename" -C "$full_path" else printf "Only one file part.\n" # Handling single file part filepart_detail=$(echo "$data" | grep -o '"filepart_details":\[{"[^]]*' | sed 's/"filepart_details":\[//' | tr -d '{}') filepart_url=$(echo "$filepart_detail" | grep -o '"url":"[^"]*' | sed 's/"url":"//' | tr -d '"') filename=$(echo "$data" | grep -o '"filename":"[^"]*' | sed 's/"filename":"//' | tr -d '"' | head -n 1) curl -o "$filename" -X GET -H "Authorization: $session_id" \ -H "Accept: application/json" "$filepart_url" name=$(echo "$data" | grep -o '"name":"[^"]*' | sed 's/"name":"//' | tr -d '"' | head -n 1) full_path="$target_directory/$name" if [ ! -d "$full_path" ]; then # Directory does not exist, create it mkdir -p "$full_path" printf "Directory '%s' created.\n" "$full_path" else printf "Directory '%s' already exists.\n" "$full_path" fi tar -xzvf "$filename" -C "$full_path" fi fi ``` --- **Previous:** [Getting Started](/sitevault/direct-data-api/getting-started) **Next:** [Guides](/sitevault/direct-data-api/guides)