- 03/03: Added description for users_all
- 05/03: Reiterated not to modify/move stub files as per week 2 lecture; submission open comment
- 09/03:
- "set handle" minimum length set as 2
- Clarification that all ids uniquely identify an entity
- All "between" references clarified to be inclusive
- Clarity on how search results are sorted
- password reset, message sendlater, message react/pin, profile pic uploads, standups, and permision changes added
- 14/03: PLEASE READ
- Added "reacts, is_pinned" to messages data type
- Added "reacts" data type
- Permission ID's clarified for global permissions
- Ability to login, register if not registered, and log out
- Ability to reset password if forgotten it
- Ability to see a list of channels
- Ability to create a channel, join a channel, invite someone else to a channel, and leave a channel
- Within a channel, ability to view all messages, view the members of the channel, and the details of the channel
- Within a channel, ability to send a message now, or to send a message at a specified time in the future
- Within a channel, ability to edit, remove, pin, unpin, react, or unreact to a message
- Ability to view user anyone's user profile, and modify a user's own profile (name, email, handle, and profile photo)
- Ability to search for messages based on a search string
- Ability to modify a user's privileges: (MEMBER, OWNER)
- Ability to begin a "standup", which is an X minute period where users can send messages that at the end of the period will automatically be collated and summarised to all users
Completed
Completed
Completed
-
Fix any outstanding bugs and complete any components that were not full implemented for the iteration 2 deadline. In doing this, we should ensure that your code is of sufficiently high quality for it to be maintained. Consider principles like DRY, KISS, top-down thinking and encapsulation. Furthermore, ensure all functions are documented with Pydoc using a consistent style.
-
Implement the
/passwordreset/request
and/passwordreset/reset
routes. By doing this, the "Forgot your password" feature of the frontend should now work. -
Modify backend such that it is able to persist and reload its data store. The persistance should happen at regular intervals so that in the event of unexpected program termination (e.g. sudden power outage) a minimal amount of data is lost. We may implement this using whatever method of serialisation you prefer (e.g. pickle, JSON ) or Simply mySQL.
-
Implement the
/user/profile/uploadphoto
route as described in table below. If done correctly, we should be able to set a profile image when using the frontend. Any additional packages you use need to be added torequirements.txt
. -
Add support for slackr owners (Admins) to remove users from slackr. This requires modifying both the backend and frontend. Modify the backend by implementing
/admin/user/remove
in the table below. For the frontend, add an additional entry to the admin menu that provides an interface for removing users. -
Allow users to relax and play a game of Hangman in slackr. If the command
/hangman
is typed into a channel, it should start a game where the users of the channel cooperatively try to guess a word or phrase letter by letter.
After a game of Hangman has been started any user in the channel can type /guess X
where X
is an individual letter. If that letter is contained in the word or phrase they're trying to guess, the app should indicate where it occurs. If it does not occur, more of the hangman is drawn. It can be done only by modifying the backend and relying on messages to communicate the state of the game (e.g. after making a guess, the "Hangman" posts a message with a drawing of the hangman in ASCII/emoji art). Alternatively we can modify the frontend.
The app should use words and phrases from an external source, not just a small handful hardcoded into the app. One suitable source is /usr/share/dict/words
available on Unix-based systems. Alternatively, the python wikiquote module is available via pip and can be used to retrieve quotes and phrases from Wikiquote.
Variable name | Type |
---|---|
named exactly email | string |
named exactly id | integer |
named exactly length | integer |
named exactly password | string |
named exactly token | string |
named exactly message | string |
contains substring name | string |
contains substring code | string |
has prefix is_ | boolean |
has prefix time_ | integer (unix timestamp), check this out |
has suffix _id | integer |
has suffix _url | string |
has suffix _str | string |
has suffix end | integer |
has suffix start | integer |
(outputs only) named exactly user | Dictionary containing u_id, email, name_first, name_last, handle_str, profile_img_url |
(outputs only) named exactly users | List of dictionaries, where each dictionary contains types u_id, email, name_first, name_last, handle_str, profile_img_url |
(outputs only) named exactly messages | List of dictionaries, where each dictionary contains types { message_id, u_id, message, time_created, reacts, is_pinned } |
(outputs only) named exactly channels | List of dictionaries, where each dictionary contains types { channel_id, name } |
(outputs only) name ends in members | List of dictionaries, where each dictionary contains types { u_id, name_first, name_last, profile_img_url } |
(outputs only) name ends in reacts | List of dictionaries, where each dictionary contains types { react_id, u_ids, is_this_user_reacted } where react_id is the id of a react, and u_ids is a list of user id's of people who've reacted for that react. is_this_user_reacted is whether or not the authorised user has been one of the reacts to this post |
For outputs with data pertaining to a user, a profile_img_url is present. When images are uploaded for a user profile, after processing them we should store them on the server such that the server now locally has a copy of the cropped image of the original file linked. Then, the profile_img_url should be a URL to the server, such as http://localhost:5001/imgurl/adfnajnerkn23k4234.jpg (a unique url)
Many of these functions (nearly all of them) need to be called from the perspective of a user who is logged in already. When calling these "authorised" functions, we need to know:
- Which user is calling it
- That the person who claims they are that user, is actually that user
To solve this when a user logs in or registers the backend should return a "token" (an authorisation hash) that the front end will store and pass into most of your functions in future. When these "authorised" functions are called, you can check if a token is valid, and determine the user ID.
For errors to be appropriately raised on the frontend, they must be raised by the following:
if True: # condition here
raise InputError(description='Description of problem')
The types in error.py have been modified appropriately for you.
The only React ID currently associated with the frontend is React ID 1, which is a thumbs up.
- Members in a channel have one of two channel permissions.
-
- Owner of the channel (the person who created it, and whoever else that creator adds)
-
- Members of the channel
-
- Slackr user's have two global permissions
-
- Owners, who can also modify other owners' permissions. (permission_id 1)
-
- Members, who do not have any special permissions. (permission_id 2)
-
- All slackr users are by default members, except for the very first user who signs up, who is an owner
A user's primary permissions are their global permissions. Then the channel permissions are layered on top. For example:
- An owner of slackr has owner privileges in every channel they've joined
- A member of slackr is a member in channels they are not owners of
- A member of slackr is an owner in channels they are owners of
Once standups are finished, all of the messages sent to standup/send are packaged together in one single message posted by the user who started the standup and sent as a message to the channel the standup was started in, timestamped at the moment the standup finished.
The structure of the packaged message is like this:
Standups can be started on the frontend by typing "/standup X", where X is the number of seconds that the standup lasts for, into the message input and clicking send.
AccessError
- For all functions except auth/register, auth/login
- Error thrown when token passed in is not a valid token
The behaviour in which channel_messages returns data is called pagination. It's a commonly used method when it comes to getting theoretially unbounded amounts of data from a server to display on a page in chunks. The chunks sent should be in the size of 50.
- All IDs (e.g. user id, channel id) uniquely identify a data item for its entire existence. Even if that item is removed, the ID cannot be reused for any new or other existing items.
HTTP Route | HTTP Method | Parameters | Return type | Exceptions | Description |
---|---|---|---|---|---|
auth/login | POST | (email, password) | { u_id, token } | InputError when any of:
|
Given a registered users' email and password and generates a valid token for the user to remain authenticated |
auth/logout | POST | (token) | { is_success } | N/A | Given an active token, invalidates the taken to log the user out. If a valid token is given, and the user is successfully logged out, it returns true, otherwise false. |
auth/register | POST | (email, password, name_first, name_last) | { u_id, token } | InputError when any of:
|
Given a user's first and last name, email address, and password, create a new account for them and return a new token for authentication in their session. A handle is generated that is the concatentation of a lowercase-only first name and last name. If the concatenation is longer than 20 characters, it is cutoff at 20 characters. If the handle is already taken, you may modify the handle in any way you see fit to make it unique. |
auth/passwordreset/request | POST | (email) | {} | N/A | Given an email address, if the user is a registered user, send's them a an email containing a specific secret code, that when entered in auth_passwordreset_reset, shows that the user trying to reset the password is the one who got sent this email. |
auth/passwordreset/reset | POST | (reset_code, new_password) | {} | InputError when any of:
|
Given a reset code for a user, set that user's new password to the password provided |
channel/invite | POST | (token, channel_id, u_id) | {} | InputError when any of:
|
Invites a user (with user id u_id) to join a channel with ID channel_id. Once invited the user is added to the channel immediately |
channel/details | GET | (token, channel_id) | { name, owner_members, all_members } | InputError when any of:
|
Given a Channel with ID channel_id that the authorised user is part of, provide basic details about the channel |
channel/messages | GET | (token, channel_id, start) | { messages, start, end } | InputError when any of:
|
Given a Channel with ID channel_id that the authorised user is part of, return up to 50 messages between index "start" and "start + 50" exclusive. Message with index 0 is the most recent message in the channel. This function returns a new index "end" which is the value of "start + 50", or, if this function has returned the least recent messages in the channel, returns -1 in "end" to indicate there are no more messages to load after this return. |
channel/leave | POST | (token, channel_id) | {} | InputError when any of:
|
Given a channel ID, the user removed as a member of this channel |
channel/join | POST | (token, channel_id) | {} | InputError when any of:
|
Given a channel_id of a channel that the authorised user can join, adds them to that channel |
channel/addowner | POST | (token, channel_id, u_id) | {} | InputError when any of:
|
Make user with user id u_id an owner of this channel |
channel/removeowner | POST | (token, channel_id, u_id) | {} | InputError when any of:
|
Remove user with user id u_id an owner of this channel |
channels/list | GET | (token) | { channels } | N/A | Provide a list of all channels (and their associated details) that the authorised user is part of |
channels/listall | GET | (token) | { channels } | N/A | Provide a list of all channels (and their associated details) |
channels/create | POST | (token, name, is_public) | { channel_id } | InputError when any of:
|
Creates a new channel with that name that is either a public or private channel |
message/send | POST | (token, channel_id, message) | { message_id } | InputError when any of:
|
Send a message from authorised_user to the channel specified by channel_id |
message/sendlater | POST | (token, channel_id, message, time_sent) | { message_id } | InputError when any of:
|
Send a message from authorised_user to the channel specified by channel_id automatically at a specified time in the future |
message/react | POST | (token, message_id, react_id) | {} | InputError when any of:
|
Given a message within a channel the authorised user is part of, add a "react" to that particular message |
message/unreact | POST | (token, message_id, react_id) | {} | InputError
|
Given a message within a channel the authorised user is part of, remove a "react" to that particular message |
message/pin | POST | (token, message_id) | {} | InputError when any of:
|
Given a message within a channel, mark it as "pinned" to be given special display treatment by the frontend |
message/unpin | POST | (token, message_id) | {} | InputError when any of:
|
Given a message within a channel, remove it's mark as unpinned |
message/remove | DELETE | (token, message_id) | {} | InputError when any of:
|
Given a message_id for a message, this message is removed from the channel |
message/edit | PUT | (token, message_id, message) | {} | AccessError when none of the following are true:
|
Given a message, update it's text with new text. If the new message is an empty string, the message is deleted. |
user/profile | GET | (token, u_id) | { user } | InputError when any of:
|
For a valid user, returns information about their user id, email, first name, last name, and handle |
user/profile/setname | PUT | (token, name_first, name_last) | {} | InputError when any of:
|
Update the authorised user's first and last name |
/user/profile/setemail | PUT | (token, email) | {} | InputError when any of:
|
Update the authorised user's email address |
/user/profile/sethandle | PUT | (token, handle_str) | {} | InputError when any of:
|
Update the authorised user's handle (i.e. display name) |
/user/profile/uploadphoto | POST | (token, img_url, x_start, y_start, x_end, y_end) | {} | InputError when any of:
|
Given a URL of an image on the internet, crops the image within bounds (x_start, y_start) and (x_end, y_end). Position (0,0) is the top left. |
users/all | GET | (token) | { users} | N/A | Returns a list of all users and their associated details |
/search | GET | (token, query_str) | { messages } | N/A | Given a query string, return a collection of messages in all of the channels that the user has joined that match the query. Results are sorted from most recent message to least recent message |
standup/start | POST | (token, channel_id, length) | { time_finish } | InputError when any of:
|
For a given channel, start the standup period whereby for the next "length" seconds if someone calls "standup_send" with a message, it is buffered during the X second window then at the end of the X second window a message will be added to the message queue in the channel from the user who started the standup. X is an integer that denotes the number of seconds that the standup occurs for |
standup/active | GET | (token, channel_id) | { is_active, time_finish } | InputError when any of:
|
For a given channel, return whether a standup is active in it, and what time the standup finishes. If no standup is active, then time_finish returns None |
standup/send | POST | (token, channel_id, message) | {} | InputError when any of:
|
Sending a message to get buffered in the standup queue, assuming a standup is currently active |
admin/userpermission/change | POST | (token, u_id, permission_id) | {} | InputError when any of:
|
Given a User by their user ID, set their permissions to new permissions described by permission_id |
admin/user/remove | DELETE | (token, u_id) | {} | InputError when:
|
Given a User by their user ID, remove the user from the slackr. |
workspace/reset | POST | () | {} | Resets the workspace state |