scorm | Parse and create SCORM packages

My scorm file is in a different domain then the LMS.

This is your problem.

The only link the SCORM package has to your LMS (and therefore to your reporting) is the SCORM API. Modern browsers do not allow CORS (cross-origin resource sharing) by default because it opens you to XSS (cross-site scripting) attacks.

The package may be able to find the API if one exists on the external domain, but it will not be able to communicate with the LMS directly.

The way I worked around this was to add a wrapper .js file on the external domain that hosted the SCORM API and used Window.postMessage to pass calls back to the LMS.

You then use window.open or your iframe to open a page on the external domain which has access to scorm_wrapper.js and its own iframe that hosts your package.

Because the Window.opener method is not reliable and not fully-featured when the current window was opened from a different origin, we will use the domain directly.

Since we control the LMS, we know the domain it is running on and since we just opened the package, we know it should be active. We can postMessage it to ask for a reply.

Django searches for a templates folder within each app for the relevant template to render. Looks like the first it found was index.html from within the symposium app.

The convention of directory structure for templates is to put your templates within another folder with the name of the app in order to avoid this exact situation with templates of the same name.

Move your index.html file from

/elearning/templates/index.html

to

/elearning/templates/elearning/index.html

and then change your home view function to

This might help to solve & analyze your problem locally.

I have created the index:

PUT /resourcelibrary

Here are a couple options:

• Select the "single activity" course format: This prevents the /course/view.php page from appearing. Users still have to click a button to enter the SCORM activity, but takes clicks and page views out of the UX. This option can be selected in the course settings.

• Customize: Create a version of the SCORM activity that launches without requiring the user to click the entry button on the SCORM activity page. This requires replicating what occurs when the the user clicks the entry button on the SCORM page, but is possible.

Hope one of these is helpful - good luck w/ it!

For comparable use with xAPI as the communication protocol I'd recommend considering the cmi5 profile. It is intended specifically for the packaging and LMS launch scenario while leveraging xAPI as its data storage and communication mechanism.

cmi5 provides for a mechanism to launch into remotely hosted content (AU) and provides via the launch URL enough information then to communicate back to the LMS/LRS in a cross origin way. xAPI is itself compatible with this approach assuming the LRS can provide the correct CORS headers (most do).

SCORM does not generally work (natively) in this type of setup because it is based on a JavaScript API that must be made accessible to the content itself which generally can't be accessed across different origins. There are ways around this, but they are supplements to the core specification.

First your current code is vulnerable to sql injection, you really should use parameters Moodle Documentation. I'd also break modules out into another query.

The issue has been resolved, and it is specific to the SCORM protocol.
When you open a piece of SCORM content, which is a bunch of html,flash,js and other media files, the first thing it does is trying to find the SCORM api.
The SCORM api is JS that has to be in the window that loads the SCORM content.
In some SCORM packages, the function that looks for the API, looks recursively for the TOP window, i.e. the initial window that started the entire process.
In my case it was a window on my site, that poped the SCORM launcher on my CDN, which has a different domain name.

All I had to do to solve this issue, is cheat the browser to think the SCORM launcher (the window I poped on the CDN) is the TOP, by adding the following two js lines in it:

actor is not one of the available query parameters. You would need to use agent, and to filter statements by an Agent you have to provide the IFI, in your case the account value, name is not an acceptable identifier in xAPI so it can't be used as a filter value. The list of available properties matches the set of query string parameters listed in the GET /statements resource here: https://github.com/adlnet/xAPI-Spec/blob/master/xAPI-Communication.md#213-get-statements

inferSchema for CSV file expects data in the specific format: yyyy-MM-dd'T'HH:mm:ss.SSSXXX, and the XXX pattern is matching to following specification of the timezone: Z; -08; -0830; -08:30; -083015; -08:30:15;, that doesn't match your data.

To fix this you may revert to the creating a table via notebook, and add following option to specify the timestamp format: .option("timestampFormat", "yyyy-MM-dd'T'HH:mm:ss.SSSZ"), the whole read command would look as following: