p3-insta485-clientside

EECS485 P3: Client-side Dynamic Pages

Due: 8pm Thursday, February 20, 2020. This is a group project to be completed in groups of two to three.

Change log

Initial Release: 6.0-w20

Version 6.1: Change the sample Likes component in the React tutorial section to pass Eslint. 02/17/2020 Version 6.2: Changed test_index.py and utils.py to better search for api calls in flask log. Nothing should change if you were passing the test. Fixes issues for students having trouble mainly with test_index.py::test_refresh.

Introduction

An Instagram clone implemented with client-side dynamic pages. This is the third of an EECS 485 three project sequence: a static site generator from templates, server-side dynamic pages, and client-side dynamic pages.

Build an application using client-side dynamic pages and a REST API. Reuse server-side code from project 2, refactoring portions of it into a REST API. Write a client application in JavaScript that runs in the browser and makes AJAX calls to the REST API.

The learning goals of this project are client-side dynamic pages, JavaScript programming, asynchronous programming (AJAX), and REST APIs. You’ll also gain more practice with the command line.

This spec will walk you through several parts:

  1. Setup
  2. REST API Specification
  3. Client-side Insta485 specification
  4. Testing
  5. Deploy to AWS
  6. Submitting and grading
  7. FAQ

Setup

Group registration

Register your group on the Autograder.

AWS account and instance

You will use Amazon Web Services (AWS) to deploy your project. AWS account setup may take up to 24 hours, so get started now. Create an account, launch and configure the instance. Don’t deploy yet. AWS Tutorial.

Project folder

Create a folder for this project (instructions). Your folder location might be different.

WARNING: avoid file paths containing spaces in the project. Spaces can cause problems with the local tool installations.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside

Version control

Set up version control using the Version control tutorial.

Be sure to check out the Version control for a team tutorial.

After you’re done, you should have a local repository with a “clean” status and your local repository should be connected to a remote GitLab repository.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ git status
On branch master
Your branch is up-to-date with 'origin/master'.

nothing to commit, working tree clean
$ git remote -v
origin	https://gitlab.eecs.umich.edu/awdeorio/p3-insta485-clientside.git (fetch)
origin	https://gitlab.eecs.umich.edu/awdeorio/p3-insta485-clientside.git (push)

You should have a .gitignore file (instructions).

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ head .gitignore
This is a sample .gitignore file that's useful for EECS 485 projects.
...

Python virtual environment

Create a Python virtual environment using the Project 1 Python Virtual Environment Tutorial.

Check that you have a Python virtual environment, and that it’s activated (remember source env/bin/activate).

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ ls -d env
env
$ echo $VIRTUAL_ENV
/Users/awdeorio/src/eecs485/p3-insta485-clientside/env

Install utilities

Linux and Windows 10 Subsystem for Linux

$ sudo apt-get install sqlite3 curl httpie

MacOS

$ brew install sqlite3 curl httpie coreutils

Starter files

Download and unpack the starter files.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ wget https://eecs485staff.github.io/p3-insta485-clientside/starter_files.tar.gz
$ tar -xvzf starter_files.tar.gz

Move the starter files to your project directory and remove the original starter_files/ directory and tarball.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ mv starter_files/* .
$ mv starter_files/.eslintrc.js .
$ rm -rf starter_files starter_files.tar.gz

You should see these files.

$ tree --matchdirs -I 'env|__pycache__'
.
├── VERSION
├── package-lock.json
├── package.json
├── setup.py
├── sql
│   └── uploads
        ...
│       └── e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg
├── tests
    ...
│   └── util.py
└── webpack.config.js

Here’s a brief description of each of the starter files.

VERSION Version of the starter files
package-lock.json JavaScript packages with dependencies
package.json JavaScript packages
setup.py Insta485 python package configuration
sql/uploads/ Sample image uploads
tests/ Public unit tests
webpack.config.js JavaScript bundler config

Before making any changes to the clean starter files, it’s a good idea to make a commit to your Git repository.

Copy project 2 code

You’ll reuse much of your code from project 2. Copy these files and directories from project 2 to project 3:

Do not copy:

Your directory should now look like this:

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ tree --matchdirs -I 'env|__pycache__'
.
├── bin
│   ├── insta485db
│   ├── insta485run
│   └── insta485test
├── insta485
│   ├── __init__.py
│   ├── api
│   │   ├── __init__.py
        ...
│   ├── config.py
│   ├── model.py
│   ├── static
│   │   ├── css
│   │   │   └── style.css
│   │   ├── images
│   │   │   └── logo.png
│   │   └── js
│   │       └── bundle.js
│   ├── templates
        ...
│   │   └── index.html
│   └── views
│       ├── __init__.py
├── package-lock.json
├── package.json
├── setup.py
├── sql
│   ├── data.sql
│   ├── schema.sql
│   └── uploads
        ...
│       └── e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg
├── tests
    ...
│   └── util.py
└── webpack.config.js

Use pip to install the insta485 package.

$ pip install -e .

Run your project 2 code and make sure it still works by navigating to http://localhost:8000/.

$ ./bin/insta485db reset
$ ./bin/insta485run

Commit these changes and push to your Git repository.

REST API

The Flask REST API Tutorial will show you how to create a small REST API with Python/Flask.

Run the Flask development server.

$ ./bin/insta485run

Navigate to http://localhost:8000/api/v1/p/1/likes/. You should see this JSON response:

{
  "logname_likes_this": 1,
  "likes_count": 3,
  "postid": 1,
  "url": "/api/v1/p/1/likes/"
}

Commit these changes and push to your Git repository.

REST API tools

The REST API Tools Tutorial will show you how to use curl and HTTPie (the http command) to test a REST API from the command line.

You should now be able to log in and make a REST API call via the command line. Login using HTTPie to fill out the login form. This will save a file called session.json containing a cookie set by the server.

$ http \
  --session=./session.json \
  --form POST \
  "http://localhost:8000/accounts/login/" \
  username=awdeorio \
  password=password \
  submit=login
HTTP/1.0 302 FOUND
...

$ http \
  --session=./session.json \
  "http://localhost:8000/api/v1/p/1/likes/"
HTTP/1.0 200 OK
...
{
    "likes_count": 3,
    "logname_likes_this": 1,
    "postid": 1,
    "url": "/api/v1/p/1/likes/"
}

React/JS

The React/JS Tutorial will get you starter with a “hello world” React app and development toolchain.

After completing the tutorial, you have local JavaScript libraries and tools installed. Your versions may be different.

$ ls -d node_modules
node_modules
$ echo $VIRTUAL_ENV
/Users/awdeorio/src/eecs485/p3-insta485-clientside/env
$ which npm
/Users/awdeorio/src/eecs485/p3-insta485-clientside/env/bin/npm
$ npm --version
6.5.0
$ which node
/Users/awdeorio/src/eecs485/p3-insta485-clientside/env/bin/node
$ node --version
v11.9.0

More tools written in JavaScript were installed via npm.

$ npx webpack --version
4.41.5
$ npx eslint --version
v6.8.0

You can check the style of your code using eslint.

$ npx eslint --ext jsx insta485/js/

Build the front end using webpack and then start a Flask development server.

$ npx webpack
$ ./bin/insta485run

Browse to http://localhost:8000/ where you should see the test “Likes” React Component.

Commit these changes and push to your Git repository.

End-to-end testing

The End-to-end Testing Tutorial describes how to test a website implemented with client-side dynamic pages.

After completing the tutorial, you should have Google Chrome and Chrome Driver installed. The first part of version should match (77 in this example). While your versions should match, they might be different than this example.

$ google-chrome --version
Google Chrome 79.0.3945.130
$ chromedriver --version
ChromeDriver 79.0.3945.36 (3582db32b33893869b8c1339e8f4d9ed1816f143-refs/branch-heads/3945@{#614})

Run an end-to-end test provided with the starter files.

$ pytest -v --log-cli-level=INFO tests/test_index.py::test_anything
...
INFO     autograder:conftest.py:77 Setup test fixture 'app'
INFO     autograder:conftest.py:130 Setup test fixture 'base_driver'
INFO     autograder:conftest.py:160 IMPLICIT_WAIT_TIME=10
INFO     autograder:conftest.py:192 Setup test fixture 'driver'
INFO     werkzeug:_internal.py:122  * Running on http://localhost:50504/ (Press CTRL+C to quit)
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET / HTTP/1.1" 302 -
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET /accounts/login/ HTTP/1.1" 200 -
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET / HTTP/1.1" 302 -
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET /accounts/login/ HTTP/1.1" 200 -
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET /static/css/style.css HTTP/1.1" 200 -
INFO     werkzeug:_internal.py:122 127.0.0.1 - - [22/Jan/2020 08:25:52] "GET /static/images/logo.png HTTP/1.1" 200 -
PASSED                                                                          ...

Install script

Installing the tool chain requires a lot of steps! Write a bash script bin/insta485install to install your app. Don’t forget to check for shell script pitfalls.

Remember to add bin/insta485install to your Git repo and push.

WARNING Do not commit automatically generated or binary files to your Git repo! They can cause problems when running the code base on other computers, e.g., on AWS or a group member’s machine. These should all be in your .gitignore.

Fresh install

These instructions are useful for a group member installing the toolchain after checking out a fresh copy of the code.

Check out a fresh copy of the code and change directory.

$ git clone <your git URL here>
$ cd p3-insta485-clientside/

If you run into trouble with packages or dependencies, you can delete these automatically generated files.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ rm -rf env/ node_modules/ insta485.egg-info/ insta485/static/js/bundle.js

Run the installer created during the setup tutorial.

$ ./bin/insta485install

Activate the newly created virtual environment.

$ source env/bin/activate

That’s it!

Database

Use the same database schema and starter data as in the Project 2 Database instructions.

After copying data.sql and schema.sql from project 2, your sql/ directory should look like this.

$ tree sql
sql
├── data.sql
├── schema.sql
└── uploads
    ...
    └── e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg

Database management shell script

Reuse your same database management shell script (insta485db) from project 2. Your script should already support these subcommands:

$ insta485db create
$ insta485db destroy
$ insta485db reset
$ insta485db dump

Add the insta485db random subcommand, which will generate 100 posts in the database each with owner awdeorio and a random photo (selected from the starter photos).

Here is a bash snippet that adds 100 posts to the database each with owner awdeorio and a random photo. Note: you will not need to modify this bash snippet, but you will need to add the random subcommand to your bash script.

SHUF=shuf
# If shuf is not on this machine, try to use gshuf instead
if ! type shuf 2> /dev/null; then
  SHUF=gshuf
fi
DB_FILENAME=var/insta485.sqlite3
FILENAMES="122a7d27ca1d7420a1072f695d9290fad4501a41.jpg
           ad7790405c539894d25ab8dcf0b79eed3341e109.jpg
           9887e06812ef434d291e4936417d125cd594b38a.jpg
           2ec7cf8ae158b3b1f40065abfb33e81143707842.jpg"
for i in `seq 1 100`; do
  # echo $FILENAMES      print string
  # shuf -n1             select one random line from multiline input
  # awk '{$1=$1;print}'  trim leading and trailing whitespace

  # Use '${SHUF}' instead of 'shuf'
  FILENAME=`echo "$FILENAMES" | ${SHUF} -n1 | awk '{$1=$1;print}'`
  OWNER="awdeorio"
  sqlite3 -echo -batch ${DB_FILENAME} "INSERT INTO posts(filename, owner) VALUES('${FILENAME}','${OWNER}');"
done

MacOS: the insta485db random code above using the shuf (or gshuf) command-line utility, which not installed by default. Install the coreutils package, which includes gshuf.

$ brew install coreutils

REST API Specification

This section describes the REST API implemented by the server. It implements the functionality needed to implement the main insta485 page. You might find this tutorial on REST APIs using Python/Flask helpful.

The following table describes each REST API method.

HTTP Method Example URL Action
GET /api/v1/ Return API resource URLs
GET /api/v1/p/ Return 10 newest posts
GET /api/v1/p/?size=N Return N newest posts
GET /api/v1/p/?page=N Return N’th page of posts
GET /api/v1/p/<postid>/ Return post metadata: URL, username, etc.
GET /api/v1/p/<postid>/comments/ Return comments for one post
POST /api/v1/p/<postid>/comments/ Create comment
GET /api/v1/p/<postid>/likes/ Return number of likes
POST /api/v1/p/<postid>/likes/ Create like
DELETE /api/v1/p/<postid>/likes/ Delete like

Reminder: all the following examples require a logged in user. Here’s how to log in and save session cookies using curl:

$ curl \
  --request POST \
  --cookie-jar cookies.txt \
  --form 'username=awdeorio' \
  --form 'password=password' \
  --form 'submit=login' \
  "http://localhost:8000/accounts/login/"

GET /api/v1/

Return a list of services available. The output should look exactly like this example. Note that curl -b is the same as curl --cookie.

$ curl -b cookies.txt "http://localhost:8000/api/v1/"
{
  "posts": "/api/v1/p/",
  "url": "/api/v1/"
}

GET /api/v1/p/

Return the 10 newest posts. The posts should meet the following criteria: each post is made by a user which the logged in user follows or the post is made by the logged in user. The URL of the next page of posts is returned in next. Note that postid is an int, not a string.

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/"
{
  "next": "",
  "results": [
    {
      "postid": 3,
      "url": "/api/v1/p/3/"
    },
    {
      "postid": 2,
      "url": "/api/v1/p/2/"
    },
    {
      "postid": 1,
      "url": "/api/v1/p/1/"
    }
  ],
  "url": "/api/v1/p/"
}

Request a specific number of results with ?size=N.

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/?size=1"
{
  "next": "/api/v1/p/?size=1&page=1",
  "results": [
    {
      "postid": 3,
      "url": "/api/v1/p/3/"
    }
  ],
  "url": "/api/v1/p/"
}

Request a specific page of results with ?page=N.

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/?page=1"
{
  "next": "",
  "results": [],
  "url": "/api/v1/p/"
}

Put size and page together.

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/?size=1&page=1"
{
  "next": "/api/v1/p/?size=1&page=2",
  "results": [
    {
      "postid": 2,
      "url": "/api/v1/p/2/"
    }
  ],
  "url": "/api/v1/p/"
}

Both size and page must be non-negative integers. Hint: let Flask coerce to the integer type in a query string like this: flask.request.args.get("size", default=<some number>, type=int).

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/?page=-1"
{
  "message": "Bad Request",
  "status_code": 400
}
$ curl -b cookies.txt "http://localhost:8000/api/v1/p/?size=-1"
{
  "message": "Bad Request",
  "status_code": 400
}

HINT: Use an SQL query with LIMIT and OFFSET, which you can compute from the page and size parameters. NOTE: “age” should not be returned as humanreadble from API.

Pro-tip: Returning the newest posts can be tricky due to the fact that all the posts are generated at nearly the same instant. If you tried to order by timestamp, this could potentially cause ‘ties’. Instead of using timestamp, use the fact that post ID is auto incremented in the order of creation to get the correct order.

GET /api/v1/p/<postid>/

Return the details for one post. Example:

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/3/"
{
  "age": "2017-09-28 04:33:28",
  "img_url": "/uploads/9887e06812ef434d291e4936417d125cd594b38a.jpg",
  "owner": "awdeorio",
  "owner_img_url": "/uploads/e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg",
  "owner_show_url": "/u/awdeorio/",
  "post_show_url": "/p/3/",
  "url": "/api/v1/p/3/"
}

HINT: <postid> must be an integer. Let Flask enforce the integer type in a URL like this:

@insta485.app.route('/api/v1/p/<int:postid_url_slug>/', methods=["GET"])
def get_post(postid_url_slug):

GET /api/v1/p/<postid>/comments/

Return a list of comments for one post. Example:

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/3/comments/"
{
  "comments": [
    {
      "commentid": 1,
      "owner": "awdeorio",
      "owner_show_url": "/u/awdeorio/",
      "postid": 3,
      "text": "#chickensofinstagram"
    },
    {
      "commentid": 2,
      "owner": "jflinn",
      "owner_show_url": "/u/jflinn/",
      "postid": 3,
      "text": "I <3 chickens"
    },
    {
      "commentid": 3,
      "owner": "michjc",
      "owner_show_url": "/u/michjc/",
      "postid": 3,
      "text": "Cute overload!"
    }
  ],
  "url": "/api/v1/p/3/comments/"
}

POST /api/v1/p/<postid>/comments/

Add one comment to a post. Include the ID of the new comment in the return data. Return 201 on success.

HINT: sqlite3 provides a special function to retrieve the ID of the most recently inserted item: SELECT last_insert_rowid().

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request POST \
  --data '{"text":"Comment sent from curl"}'\
  http://localhost:8000/api/v1/p/3/comments/
HTTP/1.0 201 CREATED
Content-Type: application/json
Content-Length: 135
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:38:06 GMT

{
  "commentid": 8,
  "owner": "awdeorio",
  "owner_show_url": "/u/awdeorio/",
  "postid": 3,
  "text": "Comment sent from curl"
}

The new comment appears in the list now.

$ curl -ib cookies.txt 'http://localhost:8000/api/v1/p/3/comments/'
HTTP/1.0 200 OK
Content-Type: application/json
Content-Length: 655
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:38:19 GMT

{
  "comments": [
    {
      "commentid": 1,
      "owner": "awdeorio",
      "owner_show_url": "/u/awdeorio/",
      "postid": 3,
      "text": "#chickensofinstagram"
    },
    {
      "commentid": 2,
      "owner": "jflinn",
      "owner_show_url": "/u/jflinn/",
      "postid": 3,
      "text": "I <3 chickens"
    },
    {
      "commentid": 3,
      "owner": "michjc",
      "owner_show_url": "/u/michjc/",
      "postid": 3,
      "text": "Cute overload!"
    },
    {
      "commentid": 8,
      "owner": "awdeorio",
      "owner_show_url": "/u/awdeorio/",
      "postid": 3,
      "text": "Comment sent from curl"
    }
  ],
  "url": "/api/v1/p/3/comments/"
}

GET /api/v1/p/<postid>/likes/

Return the number of likes for one post. Also include whether the logged in user like this (1) or not (0). Example:

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/3/likes/"
{
  "logname_likes_this": 1,
  "likes_count": 1,
  "postid": 3,
  "url": "/api/v1/p/3/likes/"
}

DELETE /api/v1/p/<postid>/likes/

Delete one “like”. Return 204 on sucess.

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request DELETE \
  --data '{}' \
  http://localhost:8000/api/v1/p/3/likes/
HTTP/1.0 204 NO CONTENT
Content-Type: application/json
Content-Length: 0
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:48:03 GMT

It’s OK to delete a “like” twice.

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request DELETE \
  --data '{}' \
  http://localhost:8000/api/v1/p/3/likes/
HTTP/1.0 204 NO CONTENT
Content-Type: application/json
Content-Length: 0
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:50:28 GMT

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request DELETE \
  --data '{}' \
  http://localhost:8000/api/v1/p/3/likes/
HTTP/1.0 204 NO CONTENT
Content-Type: application/json
Content-Length: 0
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:50:28 GMT

POST /api/v1/p/<postid>/likes/

Create one “like”. Return 201 on success. Example:

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request POST \
  --data '{}' \
  http://localhost:8000/api/v1/p/3/likes/
HTTP/1.0 201 CREATED
Content-Type: application/json
Content-Length: 44
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:51:30 GMT

{
  "logname": "awdeorio",
  "postid": 3
}

If the “like” already exists, return the same data with a 409 error and JSON description.

$ curl -ib cookies.txt \
  --header 'Content-Type: application/json' \
  --request POST \
  --data '{}' \
  http://localhost:8000/api/v1/p/3/likes/
HTTP/1.0 409 CONFLICT
Content-Type: application/json
Content-Length: 44
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 17:57:59 GMT

{
  "logname": "awdeorio",
  "message": "Conflict",
  "postid": 3,
  "status_code": 409
}

HTTP Response codes

The Flask documentation has a helpful section on implementing API exceptions. Errors returned by the REST API should take the form:

{
  "message": "<describe the problem here>",
  "status_code": <int goes here>
  ...
}

All routes require a login, except /api/v1/. Return 403 if user is not logged in.

$ curl -i 'http://localhost:8000/api/v1/p/'  # didn't send cookies
HTTP/1.0 403 FORBIDDEN
Content-Type: application/json
Content-Length: 52
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 20:12:26 GMT

{
  "message": "Forbidden",
  "status_code": 403
}
$ curl -i 'http://localhost:8000/api/v1/'  # didn't send cookies
HTTP/1.0 200 OK
Content-Type: application/json
Content-Length: 50
Server: Werkzeug/1.0.0 Python/3.7.6
Date: Wed, 19 Feb 2020 13:43:43 GMT

{
  "posts": "/api/v1/p/",
  "url": "/api/v1/"
}

Note that requests to user-facing pages should still return HTML. For example, if the user isn’t logged in, the / redirects to /accounts/login/.

$ curl -i 'http://localhost:8000/'
HTTP/1.0 302 FOUND
Content-Type: text/html; charset=utf-8
Content-Length: 239
Location: http://localhost:8000/accounts/login/
Server: Werkzeug/0.12.2 Python/3.6.1
Date: Wed, 28 Jun 2017 20:13:04 GMT

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<title>Redirecting...</title>
<h1>Redirecting...</h1>
<p>You should be redirected automatically to target URL: <a href="/accounts/login/">/accounts/login/</a>.  If not click the link.

Post IDs that are out of range should return a 404 error.

$ curl -b cookies.txt "http://localhost:8000/api/v1/p/1000/"
{
  "message": "Not Found",
  "status_code": 404
}
$ curl -b cookies.txt "http://localhost:8000/api/v1/p/1000/comments/"
{
  "message": "Not Found",
  "status_code": 404
}
$ curl -b cookies.txt "http://localhost:8000/api/v1/p/1000/likes/"
{
  "message": "Not Found",
  "status_code": 404
}

Checking output style

All returned JSON should conform to standard formatting. Whitespace doesn’t matter. You can check it using jsonlint. Example:

$ curl -b cookies.txt 'http://localhost:8000/api/v1/p/' | jsonlint
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100   222  100   222    0     0  24960      0 --:--:-- --:--:-- --:--:-- 27750
{
  "next": "",
  "results": [
    {
      "postid": 3,
      "url": "/api/v1/p/3/"
    },
    {
      "postid": 2,
      "url": "/api/v1/p/2/"
    },
    {
      "postid": 1,
      "url": "/api/v1/p/1/"
    }
  ],
  "url": "/api/v1/p/"
}

At this point, the REST API autograder tests should pass.

$ pytest -v --log-cli-level=INFO tests/test_rest_api.py

Client-side Insta485 Specification

This project includes the same pages as project 2. The only modified page is the index /. All other user-facing pages are identical to project 2. This section applies only to the main page.

The main page displays the same feed of posts as in project 2. In this project, posts will be rendered client-side by JavaScript code. All posts, including comments, likes, photo, data about the user who posted, etc. must be generated from JavaScript. Continue to use server-side rendering (AKA HTML templates) for the navigation bar at the top of the page.

Pro-tip: start with a React/JS mock-up and hard coded data. Gradually add features, like retrieving data from the REST API, one at a time. See the Thinking in React docs for a good example.

Pro-tip: Commit and push features to your git repo, one at a time. Follow the Version control for a team work flow.

HTML Modifications from Project 2

In order for the autograder to correctly navigate the insta485 site, you’ll need to add a few HTML tags to your HTML forms. This section only applies to the main page; all other pages and their HTML elements may be left as they were in project 2.

The comment form must contain the class name attribute “comment-form”. You may use this HTML form code. Feel free to style it and add other HTML attributes.

<form class="comment-form">
  <input type="text" value=""/>
</form>

The like button must contain the class name attribute “like-unlike-button”. You may use this HTML code. Feel free to style it and add other HTML attributes.

<button className="like-unlike-button">
  FIXME-button-text-here
</button>

Main page HTML

The navigation bar should be rendered server-side, just like project 2. Include a link to / in the upper left hand corner. If not logged in, redirect to /accounts/login/. If logged in, include a link to /explore/ and /u/<user_url_slug>/ in the upper right hand corner.

Here’s an outline of the rendered HTML for the main page. Notice that there is no feed content. Rather, there is an entry point for JavaScript to add the feed content.


...
<body>
  <!-- Plain old HTML and jinja2 nav bar goes here -->


  <div id="reactEntry">
    Loading ...
  </div>
  <!-- Load JavaScript -->
  <script type="text/javascript" src="{{ url_for('static', filename='js/bundle.js') }}"></script>
</body>
...

Note: Rendered html text should be in the appropriate tags.

// good
render(){
  <p>{this.state.some_bool ? 'Hello' : 'Goodbye'}</p>
}

// bad
render(){
  {this.state.some_bool ? 'Hello' : 'Goodbye'}
}

Human readable timestamps

The API call GET /api/v1/p/<postid> returns the age of a post in the format Year-Month-Day Hour:Minutes:Seconds.

Your React code should convert this timestamp into human readable form e.g a few seconds ago. You should only use the moment.js library, which is already included for you in package.json, to achieve this. Using any other libraries, including react-moment, will cause you to fail tests the Autograder.

Response time

The main page should load without errors (exceptions), even when the REST API takes a long time to respond. Keep in mind that the render() method may be called asynchronously, and may be called multiple times.

Put another way, render() will likely be called by the React framework before any AJAX data arrives. The page should still render without errors.

Likes and comments update

Likes and comments added by the logged in user should appear immediately on the user interface without a page reload.

Comments are added by pressing the enter (return) key. The user interface shall not contain any comment submit button. The React docs on forms are very helpful for this feature.

Likes demo video.

Comments update demo video.

Double Click to Like

Double clicking on an unliked image should like the image. Likes added by double clicking on the image should appear immediately on the user interface without a page reload. The like count and text on the like button should also update immediately.

Double clicking on a liked image should do nothing. Do not unlike an image when it is double clicked.

The video below shows a demo. The heart animation is optional.

Double click demo video.

Infinite scroll

Scrolling to the bottom of the page causes additional posts to be loaded and displayed. Load and display the next 10 posts as specified by the next parameter of the most recent API call to /api/v1/p/. Do not reload the page.

Infinite scroll demo video.

We recommend the React Infinite Scroll Component, which is already included in package.json.

Note: if infinite scroll has been triggered and more than 10 posts are present on the main page, reloading the page should only display the 10 most recent posts (including any new posts made before the reload).

Pro-tip to test this feature, you can use insta485db random.

(Note that in some visual demos, we use the same pictures multiple times, which might make you think that infinite scroll should at some point “cycle back to the start.” That’s NOT how it should work. Infinite scroll should keep scrolling until there are no more pictures available.)

Browser history

Don’t break the back button. Here’s an example:

  1. awdeorio loads /, which displays 10 posts.
  2. awdeorio scrolls, triggering the infinite scroll mechanism. Now the page contains 20 posts.
  3. jflinn adds a new post, which updates the database.
  4. awdeorio clicks on a post to view the post details at /p/<postid>/.
  5. awdeorio clicks the back button on his browser, returning to /.
  6. The exact same 20 posts from step 2 are loaded. jflinn’s new post is not included.
  7. awdeorio refreshes the page. the 10 most recent posts are shown including jflinn’s new post.

As you might notice in the previous example, we do not specify whether the comments and / or likes on the 20 posts (step 6) are updated after returning to the index page using the back button. This is intentional. It is the student’s choice whether to update the comments and likes of each post after returning to the main page using the back button.

Here’s an example of two accceptable scenarios depending on the student’s choice:

Scenario 1:

  1. awdeorio loads /, which displays 10 posts (post ids 1-10).
  2. awdeorio scrolls, triggering the infinite scroll mechanism. Now the page contains 20 posts (post ids 1-20).
  3. jflinn adds a new post, which updates the database.
  4. jflinn likes post id 20 (displayed to awdeorio in step 2), which updates the database.
  5. awdeorio clicks on a post to view the post details at /p/5/.
  6. awdeorio clicks the back button on his browser, returning to /.
  7. The exact same 20 posts from step 2 are loaded. jflinn’s new post is not included but his like on post id 20 is shown.
  8. awdeorio refreshes the page. the 10 most recent posts are shown including jflinn’s new post.

Scenario 2:

  1. awdeorio loads /, which displays 10 posts (post ids 1-10).
  2. awdeorio scrolls, triggering the infinite scroll mechanism. Now the page contains 20 posts (post ids 1-20).
  3. jflinn adds a new post, which updates the database.
  4. jflinn likes post id 20 (displayed to awdeorio in step 2), which updates the database.
  5. awdeorio clicks on a post to view the post details at /p/5/.
  6. awdeorio clicks the back button on his browser, returning to /.
  7. The exact same 20 posts from step 2 are loaded. jflinn’s new post is not included and his like on post id 20 is not shown.
  8. awdeorio refreshes the page. the 10 most recent posts are shown including jflinn’s new post.

The same example could be given to illustrate the student’s freedom by taking scenario 1 and 2 and replacing jflinn’s “like” in step 4 with a comment instead. In this case, the student could choose whether or not to display the comment in step 7.

The Mozilla documentation on the history API will be helpful.

Hint: this requires very few code modifications! Use the History API to manipulate browser history and use the PerformanceNavigationTiming API to check how the user is navigating to and from a page. Don’t use other libraries for this feature (they only make it harder). Do not modify the URL.

REST API calls and logging

We’re going to grade your REST API by inspecting the server logs. We’ll also be checking that your client-side javascript is making the correct API calls by inspecting the server logs. Loading the main page with the default database configuration, while logged in as awdeorio should yield the following logs. Note that the order of likes and comments doesn’t matter.

127.0.0.1 - - [06/Jul/2017 12:02:09] "GET / HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /static/js/bundle.js HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/3/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/3/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/3/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/2/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/2/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/2/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/1/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/1/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /api/v1/p/1/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /uploads/e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /uploads/9887e06812ef434d291e4936417d125cd594b38a.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /uploads/505083b8b56c97429a728b68f31b0b2a089e5113.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /uploads/ad7790405c539894d25ab8dcf0b79eed3341e109.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:02:09] "GET /uploads/122a7d27ca1d7420a1072f695d9290fad4501a41.jpg HTTP/1.1" 200 -

Press the like button a couple of times:

127.0.0.1 - - [06/Jul/2017 12:03:08] "DELETE /api/v1/p/3/likes/ HTTP/1.1" 204 -
127.0.0.1 - - [06/Jul/2017 12:03:09] "POST /api/v1/p/3/likes/ HTTP/1.1" 201 -

Add a comment:

127.0.0.1 - - [06/Jul/2017 12:03:27] "POST /api/v1/p/3/comments/ HTTP/1.1" 201 -

An example of infinite scroll. First, we load the main page from a database populated with 100 random posts.

127.0.0.1 - - [06/Jul/2017 12:04:59] "GET / HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /static/js/bundle.js HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/104/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/104/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/104/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/103/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/103/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/103/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/101/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/101/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/101/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/100/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/100/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/100/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/99/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/99/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/99/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/98/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/98/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/98/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/96/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/96/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/96/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/95/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/95/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/95/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/94/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/94/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/94/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/93/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/93/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /api/v1/p/93/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/505083b8b56c97429a728b68f31b0b2a089e5113.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/e1a7c5c32973862ee15173b0259e3efdb6a391af.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/ad7790405c539894d25ab8dcf0b79eed3341e109.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/5ecde7677b83304132cb2871516ea50032ff7a4f.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/122a7d27ca1d7420a1072f695d9290fad4501a41.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/2ec7cf8ae158b3b1f40065abfb33e81143707842.jpg HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:04:59] "GET /uploads/9887e06812ef434d291e4936417d125cd594b38a.jpg HTTP/1.1" 200 -

Scroll to the bottom and infinite scroll is triggered.

127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/?size=10&page=1 HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/91/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/91/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/91/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/90/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/90/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/90/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/89/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/89/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/89/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/87/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/87/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/87/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/85/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/85/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/85/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/84/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/84/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/84/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/83/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/83/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/83/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/81/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/81/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/81/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/80/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/80/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/80/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/79/likes/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/79/comments/ HTTP/1.1" 200 -
127.0.0.1 - - [06/Jul/2017 12:05:18] "GET /api/v1/p/79/ HTTP/1.1" 200 -

Code style

As in project 2, all HTML should be W3C compliant, as reported by html5validator. Python code should be contain no errors or warnings from pycodestyle, pydocstyle, and pylint. Use pylint --disable=cyclic-import.

All JSON returned by the REST API should be jsonlint clean.

All JavaScript source code should conform to the AirBnB javascript coding standard. Use eslint to test it. Refer to the setup / eslint tutorial.

You may only use JavaScript libraries that are contained in package.json from the starter files and the built-in Web APIs.

You must use the fetch API for AJAX calls.

Can I disable any code style checks?

Do not disable any code style check from any python code style tool (pycodestyle, pydocstyle, pylint), besides the three exceptions specified in the Project 2 spec

Additionally, do not disable any eslint checks in your .jsx files. There are no exceptions to this rule.

Testing

Make sure that you’ve completed the End-to-end testing tutorial.

Several unit tests are published with the starter files. Make sure you’ve copied the tests directory. Note that your files may be slightly different.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ ls tests/
conftest.py    test_rest_api.py  test_slow_server_index.py  utils.py
pytest.ini     test_scripts.py   test_style.py
test_index.py  test_scroll.py    testdata

Rebuild your javascript bundles by running webpack and then run the tests.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-clientside
$ npx webpack
Hash: 94d02a55e475959ef08d
Version: webpack 2.6.1
Time: 4910ms
...
$ pytest -v --log-cli-level=INFO

Note: if you get deprecation warnings from third party libraries, check out the pytest tutorial - deprecation warnings to suppress them.

insta485test

Add JavaScript style checking you insta485test script from project 2. In addition to the tests run in project 2, insta485test should run eslint on all files within the insta485/js/ directory. Refer back to the eslint instructions for direction on how to do this.

$ ./bin/insta485test

Deploy to AWS

You should have already created an AWS account and instance (instructions). Resume the Project 2 AWS Tutorial - Deploy a web app .

After you have deployed your site, download the main page along with a log. Do this from your local machine.

$ pwd
/Users/awdeorio/src/eecs485/p3-insta485-serverside
$ curl \
  --request POST \
  --cookie-jar cookies.txt \
  --form 'username=awdeorio' \
  --form 'password=password' \
  --form 'submit=login' \
  "http://<Public DNS (IPv4)>/accounts/login/"
$ curl -v -b cookies.txt "http://<Public DNS (IPv4)>/" > deployed_index.html 2> deployed_index.log
$ curl -v -b cookies.txt "http://<Public DNS (IPv4)>/static/js/bundle.js" > deployed_bundle.js 2> deployed_bundle.log

Be sure to verify that the output in deployed_index.log and deployed_bundle.log doesn’t include errors like “Couldn’t connect to server”. If it does contain an error like this, it means curl couldn’t successfully connect with your flask app.

Also be sure to verify that the output in deployed_index.html looks like the index.html file you coded while deployed_bundle.js contains Javascript code.

Submitting and grading

One team member should register your group on the autograder using the create new invitation feature.

Submit a tarball to the autograder, which is linked from https://eecs485.org. Include the --disable-copyfile flag only on macOS.

$ tar \
  --disable-copyfile \
  --exclude '*__pycache__*' \
  -czvf submit.tar.gz \
  bin \
  insta485 \
  package-lock.json \
  package.json \
  setup.py sql \
  webpack.config.js \
  deployed_index.html \
  deployed_index.log \
  deployed_bundle.js \
  deployed_bundle.log

The autograder will run pip install -e YOUR_SOLUTION and cd YOUR_SOLUTION && npm install .. The exact library versions in starter_files/setup.py and package.json are cached on the autograder, so be sure not to add extra library dependencies to either one.

Rubric

This is an approximate rubric.

Tests Value
Public Unit tests 50%
Public Python, JS, JSON, and HTML style 15%
Hidden unit tests run after the deadline 35%

FAQ

My JavaScript code doesn’t work. What do I do?

  1. Make sure it’s eslint clean. Instructions here.
  2. Make sure it’s free from exceptions by checking the developer console for exception messages
  3. Try the React Developer tools Chrome extension
  4. Check your assumptions about when React methods are called. This is called the React component lifecycle. Add console.log() messages to each React method (constructor(), render(), etc.).

How do I use the fetch API?

Refer to the Fetch API documentation for a brief tutorial on how to use fetch.

Don’t forget credentials: 'include'! Make sure you aren’t using host:port in your client fetch URL.

Can I use jQuery?

Do not use jQuery for this project. Doing so is likely to cause issues with the Autograder.

The only external libraries that are needed for this project are already included for you in starter_files/package.json.

Can I use XMLHttpRequest?

Do not use XMLHttpRequest to make HTTP requests. Instead use the fetch API detailed above

How do I make a button toggle?

Toggle buttons are useful for the like/like button. See this example for more information.

Do trailing slashes in URLs matter to Flask?

Yes. Use them with the route decorator your REST API. See the “Unique URLs / Redirection Behavior” section in the Flask quickstart. Here’s a good example:

@insta485.app.route("/u/<username_url_slug>/", methods=["GET", "POST"])

Can we use console.log()?

Yes. Ideally you should only log in the case of an error.

eslint Error … “is missing in props validation”

You’ll probably encounter this error while running eslint:

$ eslint --ext jsx insta485/js/
  24:38  error    'url' is missing in props validation  react/prop-types

With prop-types, you’ll get a nice error in the console when a type property is violated at run time. For example,

“Warning: Failed propType: The prop url is marked as required in CommentInput, but its value is undefined. Check the render method of Comments.

More on the prop-types library: https://www.npmjs.com/package/prop-types.

How do I append to an array of mutable state in a React Component?

Be really careful when both reading and writing this.state.MY_VARIABLE. Here’s how.

this.setState(prevState => ({
  posts: prevState.posts.concat(data.results),
}));

Reconciliation and keys

When using a collection of React components, they need to have unique key properties. This enables the fast shadow DOM to real DOM update performed by react. More info here:

How to fix pylint “Similar lines in 2 files”

The REST API shares some code in common with portions of insta485’s static pages that haven’t been modified. For example, both the REST API and the static /u/<postid>/ read the comments and likes from the database. This could lead to pylint detecting copy paste errors.

************* Module insta485.views.user
R:  1, 0: Similar lines in 2 files
==insta485.api.comments:30
==insta485.views.post:42

A nice way to resolve this problem is by adding helper functions to your model. The canonical way to solve this problem is an Object Relational Model (ORM), but we’re simplifying in this project.

Acknowledgements

Original project written by Andrew DeOrio awdeorio@umich.edu, fall 2017. Updated, Winter 2019 485 team, February 2019.