Most simply, you can use flangio strictly as a web-based manager for MongoDB.
You can also use flangio as a starting point for your own application or own
customized API.  Many of the features are optional and can be turned of
and on to suit your specific needs by adjusting the `settings.py` file..


You can use flangio as a "backend database" that is accessed primarily by
external systems, via RESTful API, and you can build your own Django application
right inside the flangio project, (therefore accessing MongoDB natively as opposed
through the RESTful API). You simply build out your own Django app (and related urls)
inside `flangio/apps`. If you create an application on flangio you can access
MongoDB natively, via pymongo or another 3rd party MongoDB client.

Sample Project - Quantified-Clan
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

There is a sample application bundled with flangio called "qualified-clan". This
is designed to function as a "family-based" PHR where everyone with an account
can see everyone else's data.  More about that here.


This is a web application that plots weight, blood pressure measurements, over
time for a group. It also has a Twitter integration option. (Read more about that
here <insert link>.)

Division of Labor
~~~~~~~~~~~~~~~~~

There are a handful of core flangio applications.  These are `accounts`,
`socialgraph`, `mongodj`, `dataimport`, and `search`. Most fucntions are
accessiable via web site or a RESTFul API. There project's applications' key
functions are as follows:


* `accounts` - Manages users based on Django's traditional `auth.user` with some
enhancements. OAuth urls for Facebook, Twitter, etc also live under `accounts`.

* `socialgraph` - The social graph assocates users bettween one another via 

* `mongodj` - Create, Upddate, Delete and Index operations for MongoDB Database
and Collectionmanagement.

* `search` - Read and seeach operations on MongoDBs.  Included in this application
is the ability to create stored searches and convert default JSON output to other 
formats including `.csv`, `.xls`, and `.xml`.  

* `dataimport` - Import data form other sources such as CSV into MongoDB.


Following the Design Pattern
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The `socialgraph` features in flangio require that you provide some basic
fields to your dataset schema.  This list of schema reccomendations define some
metadata about your data including:

* Infomation for timezone-aware, accurate, time series information.
* Identity information. who sends(sender) what(the data) whom(receiver) and
who was data about (subject).  These may all be the same in some circumstances,
but when people are acting on other's behalf, they are not the same.  This is
application specific and this is further explained in the next section
"Configuration Options".


Configuration Options
~~~~~~~~~~~~~~~~~~~~~

flangio extends Django's `settings.py` file for its custom settings parameters.


    # Django default settings for Flangio------------------------------------
    # Anything in this file can overridden by creating a
    # settings_local.py file and placing additional settngs there.
    .
    .
    # flangio Default Settings -------------------------------------------------
    
    # The next 2 settings ensure that the transaction time is no more off +/-
    # MAX_TIME_SKEW_MIN
    
    
    ENFORCE_TIME_SANITY  = False
    
    # Maximum time in minutes the transaction time submitted by the client can be
    # off from UTC.
    
    MAX_TIME_SKEW_MIN    = 5   
    
    #Deprecated
    OTHER_LABELS         = False # If True, use label value from DataLabelMeta instead of default.



    # If following MUESLI format, then users related must exist before DB IO can
    # This goes for all three actors; `sender`, `receiver` and `subject`. 
    USERS_MUST_EXIST     = True
    
    
    # If set to False it creates an "admin party" arrangement amongst all
    # registered users.
    RESPECT_SOCIAL_GRAPH = False
    
    
    #When an account is created, then create a self follow social graph.
    AUTO_SELF_FOLLOW     = True
    
    # Deprecated
    ALLOW_UPDATE_TX      = True
    
    
    # Deprecated
    ALLOW_DELETE_TX      = True
    
    
    
    # Change this to change how files uploaded are stored. It can be stored on
    the same server  LOCAL means save it to this server, AWSS3 means use S3.
    # You must have your AWS keys defined in settings (or local_settings) for this to work.
    # Allowable values are 'LOCAL' (default), 'AWSS3'.
    
    BINARY_STORAGE = "LOCAL"
    

    # On the master database/collection sort the resulting columns by these names.
    # If a blank tuple (default), no sorting will occur.
    
    SORTCOLUMNS= ()
    
    
    
    #Mongo DB settings
    
    MONGO_HOST = "127.0.0.1"
    MONGO_PORT = 27017
    MONGO_DB_NAME = "flangio"
    MONGO_MASTER_COLLECTION = "quantified-clan"
    #MONGO_HISTORYDB_NAME = "history"
    
    # Default limit for search queries. This can be overridden by user.
    MONGO_LIMIT = 200
    
    
    # This file creates an ever incrementing integer ID for every MUESLI-compliant
    # transaction. This is basicly identical to Twitter's sinceid field.
    # In production on high-volume systems place this on its own filesystem or
    replace this with a database.
    
    SINCE_ID_FILE = os.path.join(BASE_DIR, 'db/since.id')
    
    
    # Defaulted to MUESLI See http://code.google.com/p/muesli for more information.
    # MUESLI = Mobile Universal Encapsulated Serialized Longitudinal Information
    
    
    ALLOWABLE_TRANSACTION_TYPES = ("text","tweet", "omhe")
    
    
    ALLOWABLE_HEALTH_TRANSACTION_TYPES = "text","omhe", "indivo", "hdata", "pdf",
                                  "png", "jpg", "bmp", "tiff", "jpeg200",
                                   "ms_word_doc", "ms_word_docx","ms_excel_xls",
                                   "ms_excel_xlsx", "tweet", "snomedct", "hl7v2",
                                   "cda-c32", "ccda", "ccr", "ccd", "unk", "icd9",
                                   "icd10", "cpt", "script", "hicpcs", "loinc",
                                   "lab", "rx", "cd-iso", "dvd-iso", "zip",
                                   "tar.gz", "rxnorm", "dicom_image",
                                   "dicom_structured_report", "x10", "idc",
                                   "edi", )


Installation on Ubuntu 14.04:
-----------------------------


flangio is built atop Django and MongoDB so these need to be setup first.

These instructions are for Ubuntu 13.04, but can be modified to support other
non-debian Linux sources. Install some prerequisites.

    sudo apt-get update
    sudo apt-get upgrade
    sudo-apt-get install python-pip build-essential python-virtualenv apache2 libapache2-mod-wsgi python-pymongo python-bson python-dev nmap git

    sudo apt-get install python-setuptools
    sudo pip install virtualenvwrapper

Now install MongoDB. See
(http://docs.mongodb.org/manual/tutorial/install-mongodb-on-ubuntu/) for more
information.

    sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 7F0CEB10
    echo 'deb http://downloads-distro.mongodb.org/repo/ubuntu-upstart dist 10gen' | sudo tee /etc/apt/sources.list.d/mongodb.list
    sudo apt-get update
    sudo apt-get install mongodb-10gen
    sudo apt-get update
    
    
Create a working directory to hold our Django projects, then change into that
directory.
    
    mkidir django-projects
    cd django-projects

Download most recent flangio using Git:
    
    git clone https://github.com/videntity/flangio.git
    cd flangio
    
Install other python requirements from a requirements file (Django, etc.)

     sudo pip install -r flangio/config/requirements.txt
     

Using Django's `manage.py` command, create the project's relational database
(i.e. not MongoDB).  

    python manage.py syncdb


These tables manage users, accounts, and other information. Some tables are
created as part of Django and 3rd party packaes.
By Default flangio uses SQLite. This is SQLite by default and is stored
in a directory called `db` in the project's root directory.

Create a file to hold our sinceid index counter.

    echo "0" > db/since.id
   
Change the permissions on the entire folder so Apache can use it. (SQLite specific)

    chmod -R 777 db   
    
Flangio also needs to have permission to use the uploads folder. 

    chmod -R 777 uploads
    
Add a couple lines to the file `/etc/apache2/sites-available/000-default.conf` inside the
`<VirtualHosts></VirtualHosts>` tags.

    WSGIScriptAlias / /home/ubuntu/django-projects/flangio/flangio/config/apache/django.wsgi
    WSGIPassAuthorization off

Add the following lines to the same file, inside the same <VirtualHosts></VirtualHosts> section. 
This is needed to allow Apache to access the WSGI file used by Flangio

    <Directory /home/ubuntu/django-projects/flangio/flangio/config/apache>
      <IfVersion < 2.3 >
       Order allow,deny
       Allow from all
      </IfVersion>
      <IfVersion >= 2.3>
       Require all granted
      </IfVersion>
    </Directory>



Restart Apache

    sudo service apache2 restart
    
    
If you haven't started MongoDB do that now:

    sudo /etc/init.d/mongod start
    
or using the command:

    sudo start mongod
    
For more information on configuring Django with apache or other Web servers
such as NGINX, see https://docs.djangoproject.com/en/1.3/howto/deployment/modwsgi/.
For performance, static content should not be served via Django.

Need help? flangio is free open source software, but it just so happens we
offer commercial support for flangio. This can include managed hosting within
EC2 and other options. Contact us at sales@videntity.com