App Engine application for the Udacity training course.
- Update the value of
applicationinapp.yamlto the app ID you have registered in the App Engine admin console and would like to use to host your instance of this sample. - Update the values at the top of
settings.pyto reflect the respective client IDs you have registered in the Developer Console. - Update the value of CLIENT_ID in
static/js/app.jsto the Web client ID - (Optional) Mark the configuration files as unchanged as follows:
$ git update-index --assume-unchanged app.yaml settings.py static/js/app.js - Run the app with the devserver using
dev_appserver.py DIR, and ensure it's running by visiting your local server's address (by default localhost:8080.) - (Optional) Generate your client library(ies) with the endpoints tool.
- Deploy your application.
To implement sessions to the app, I modeled the session classes and methods after the extant Conference classes and methods. The some of the support methods already in place were useful in this regard but others had to be redesigned to fit my needs, in general sessions are much more simple than confrences so it was straight forward to proccess their contents.
For the session classes I chose data types that, for the most part, should be self evident. The required fields are only the Name, the Speaker websafeConferenceKey, this is for simpicities sake, From the user perspective however this allows them to make sessions and advertise what will be talked about without while details are not yet concreate or if they just do not want to commit to our session paradigm they do not have to. For the session type, I made it a repeated string so that if a session could fit into multiple catagories, for example a lecture followed by a workshop. Finally for the start time and duration fields I chose integers because of the requirement that they be in 24 hour notation. startTime should be a number between 0 and 2359 excluding numbers ending with 60-99, these number corrospond with 0:00 to 23:59 and are checked when making a Session. Durration is just the amount of time in minutes.
I decided to keep the speaker entity simple, becasue linking it to several factors would have been convoluted and pointless. All it really needs is a list of sessions that it will be speaking at. I decided not to link it to a profile because there is no garentee that a speaker would have a profile and it wouldn't add any functionality other than blocking them from adding sessions that they were speaking at to their wish list.
The Speaker has three fields a Name, Bio and speakerSessionKeys. Name holds the speakers name so I chose a String type to hold it. Name is a required field because it is how the speaker is brought up. For Bio I wanted a short description of the speaker so I also chose string although this field is not required. The speakerSessionKeys is a repeated string because it is ment to hold the string type session keys of all seasons that the speaker is speaking at.
Speaker are either made manually with the createSpeaker() endpoint or by createSession() endpoint if there is not already a speaker with the name that is given to the endpoint(). The other endpoint that uses speaker objects getSessionsBySpeacker(), however a I use a session query to get the sessions rather than using the speakers list of session keys because it is easier to just filter by name than get each session from the keys in its list.
Adding wish lists was relatively simple. All that needed to be done was to add a repeated StringField to the Profile Class the required methods
For the two addittional querys I included getSessionAttendees(websafeSessionKey) and getConferenceAttendees(websafeConferenceKey). I chose these because for the three classes Conference, Profile and Session there was a way to get Sessions of Conferences from a Profile, and a way to get Sessions from a Conference, but no way to get Profiles from a Conference or Session.
Question: Let’s say that you don't like workshops and you don't like sessions after 7 pm. How would you handle a query for all non-workshop sessions before 7 pm? What is the problem for implementing this query? What ways to solve it did you think of?
The problem is that Google datastore API doesn't support multiple inequality filters(!= and >=) on different properties for a single query. The way I would solve it is to make multiple queries for each inequality filter and compare the results.
I implemented this ability by checking queries for multiple inequality filters if they were there. If they where it would be routed to a new function that seperates inequality filters from the other filters. Then for each inequality filter a query is made and filtered by that inequality filter and the other normal filters. Finally each of these queries are compared against eachother and only conferences that are in each of the queries are returned back from the function. I did not need to make a new endpoint for this solution. The functions I wrote for this are: _checkMultiInequality() _formatMultiFilters() _getMultiQuery()
Implemented featured speaker functionality using task queue and memcache. Logic for functionality is in main.py while taskqueue call is made in conference.py. The task queries session and filters for the given conference key. It then finds which speaker is speaking at the most sessions and sets that speaker with all of their session names as the memcache entry.