This is research code in its ugly and glorious form, used just for testing ideas. The complaint dept is located in /dev/null; otherwise, you are welcome to fix anything you have an issue with.

Jace is Just Another Coaching Engine. HMC's Just Another Coaching Engine project serving NSF, NIH, and other funded projects

Jace accepts incoming posts via its API and then disseminates that information into a user_model. The user_model consists of a User, Question, Answer, and Survey. The user_model also contains a rules function, this allows updates to the User_Model's models, or updates caused by.

Normaly when a rule is triggered it tells the communication manager to send a message to the application server. This message is to turn on or off a question set. There is also the ability to send a fully loaded question if needed.

1. Make a container for alpine-linux
2. Make a container for redis and postgres A. Make sure postgres is setup with matching user name and passwords in settings.py
3. Redis should be set to port 36379, postgres to 32768
4. Set docker port in alpine-linux to 8000, and the published port to 8888
5. In alpine-linux general make a new environmental setting called POSTGRES and set it to 32768
6. Also, in general, make a new environmental setting called REDIS and set it to 36379
7. $ apk update
8. $ apk add python3, gcc, git
9. $ git clone http://github.com/parc/jace.git
10. sh setup.sh

First create your content, in the example json format. Make sure that there is some variable in the attributes section. This value controls whether or not a content item will be seen. Note that for some cases it is important to have multiple permutations

Note the formula is ((2^n)-n+1) where n is the number of conditions.

The other basic logic formula for determining if a stimulus should fire is

if (Item.1 or Item.2 or Item.X), then Target

After content creation, go to usermodels/rules.py and write a function, you may find that you need to add in slots in the user model, if so edit accordingly, and then run. $ sh update_all_models.sh

sh run_jace.sh

./run_local
./manage.py run_huey

Just Another Coaching Engine project serving NSF, NIH, and other funded projects

django settings

scheduler settings

main subfields for API links

contains Infromation on the user,questions, answers

rules for updatings users

locattion of django models

This is where messages are processed at the moment of posting. It works by looking for a signal that a item was posted to communications, and then sorts it into the user model's models. It also has the feature of updating the last heard from date for a particular user.

Serializer for the user models

MUST be names tasks.py for HUEY to find it, contains the scheduled item(s), and the schudler.

placeholder for future urls, don't delete

contains all programs invovled in the API

creates, and serializes objects when the api is called.

serializor for the reports objects

Contains a catch all store for all incoming messages to jace, stores information about the data and source as reports

This is where the api is generated. It uses regex to build, note however all of these are in the subdirectory communications. Example

   localhost:8888/communications/reports/

checks to see if there are any missing datas

python manage.py runscript updates

$ ./run_local

$ sh setup.sh

try flushing DB

$ python manage.py flush

    try 
    $ ./run_local

Source Code, Non-Commercial Use Permitted under the following conditions

This software is licensed under the Aladdin Free Public License. A quick summary of this license is that you CAN Modify and Distribute, you CANNOT Hold Liable, Sublicense, Place Warranty, or Use Comercially, and you MUST Include License, Include Original Code, and State Your Changes.

Content, not code, Non-Commercial Use Permitted under the following conditions

The non-code content in this repository is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0). A quick summary of this license is that you ARE FREE to Share and Adapt PROVIDED that you Provide Attribution, Use Non-Commercially, and Share Alike.

For those interested in Commercial Use, please contact PARC, Inc. by email at [email protected]