Server-side data saving

Overview

Exercises: 20 min

Objectives

• Handle participant data sent to the server, and save it to the OSF.

Create a new file in the jsPsych quickstart project called save_data.php.

Paste the code below into the file:

save_data.php

<?php
/*
* secrets.php should contain the definition of the variable $OSF_PAT
* It should look like:
* <?php
* $OSF_PAT="osf_personal_access_token_with_osf.nodes.data_write_scope";
*/
include('secrets.php');

// Your OSF data component id (https://osf.io/osf_repository_id/)
$WHERE_TO_SAVE="26j4q";

$content = $_POST['public_data']; // Data sent from the participant's machine
$participant_id = $_POST['participant_id']; // Participant id sent from participant's machine

// Because the above came from the participant's machine, we sanitize the input
if(!preg_match("/^[0-9a-zA-Z]+$/", $participant_id)) {
    http_response_code(400);
    die("{error: 'Invalid participant Id'}");
}

// This may need changing if you don't use Frankfurt (DE) for your data storage
$url = "https://files.de-1.test.osf.io/v1/resources/$WHERE_TO_SAVE/providers/osfstorage/?kind=file&name=$participant_id.csv";

/*
 * The cURL request below is adapted from https://stackoverflow.com/a/5676572
 * We create a request to the $url above, with the following key properties:
 * - PUT method (because we're creating a file)
 * - Authorization header with the token
 * - Body content of the file we want to create
 */

// open connection
$ch = curl_init();

// set the url, number of POST vars, POST data
curl_setopt($ch,CURLOPT_URL, $url);
curl_setopt($ch,CURLOPT_POST, true);
curl_setopt($ch,CURLOPT_POSTFIELDS, $content);
curl_setopt($ch, CURLOPT_CUSTOMREQUEST, "PUT");

// set the request headers, used to authenticate using the PAT
curl_setopt($ch, CURLOPT_HTTPHEADER, array(
    "Authorization: Bearer $OSF_PAT"
));

// have curl_exec return the contents of the cURL; rather than echoing it
curl_setopt($ch,CURLOPT_RETURNTRANSFER, true);

// execute post
$result = curl_exec($ch);

// Inform the participant that we saved the data successfully
echo $result;

Notice at the top of the file we include('secrets.php'). secrets.php doesn’t exist at the moment, so let’s create that.

Its contents are explained in the comment in save_data.php: we need to declare a variable $OSF_PAT that holds the personal access token we created earlier. Once you’ve done that, make sure secrets.php is listed in your .gitignore file (if you’re using git) so that it doesn’t get made public when you commit your changes.

We also need to edit $WHERE_TO_SAVE to point to the OSF component identifier we made a note of earlier.

With those changes made, we should be ready to go! Do your experiment one more time, and you should see the success message that we wrote in index.html. When you navigate to your OSF test component, you should see it has created a .csv file with your experimental data.

From now on, when participants complete your experiment, their data will be saved directly to the OSF. When you access the data, you’ll be accessing it in the same way that others do. All that remains is to write a data dictionary so that others can understand the variables (don’t worry too much if you don’t understand all of the variables, just note down the ones you do understand).

For this example, you can use the minimal data dictionary here:

You can copy this table and paste it into a spreadsheet program such as Microsoft Excel or Google Sheets, and then save in .CSV format as something like data_dictionary.csv. Upload the saved file to your data component so that people accessing that component can use the dictionary.

Now you’re done! You now have

• participant data save automatically to the OSF
• data automatically made public and indexable by search engines
• a license that tells people what they can do with your data
• a data dictionary that tells people (including you in 6 months) what your variables mean and where they came from

While we’re here…

You don’t actually need to make the data public to upload them to the OSF. We could have kept the component private, and perhaps made it public later. If the component is private it doesn’t need a license, but you should still include a metadata file, and you’ll still need a PAT to upload data.

Key Points

• Never trust data sent from the participant’s computer.

• Don’t put your PAT on GitHub.