hacksu / kenthackenough Goto Github PK
View Code? Open in Web Editor NEWThe world's best hackathon website.
License: MIT License
The world's best hackathon website.
License: MIT License
I create a shortened URL like so:
short: /example
long: http://www.example.com
When I go to:
http://API_URL/go/example, I get a 502 Bad Gateway error.
When I refresh the page, it redirects exactly like it should.
Add the ability for users to reset their passwords.
getting this set up to work under any name but Kent Hack Enough is a struggle, there are a lot of files to change and directories to note when changing, including a list of all the files where you need to change directory/name notation in the setup wiki would be wonderful.
Partial List:
provision.sh
config/dev.conf
config/api.khe.conf
I'm sure there are others, but I'm not sure what they are.
Also, the setup doesn't say, nor does the provision file move api.khe.conf to /etc/nginx/sites-vailable nor link it to sites-enabled. It only does anything with dev if there is anything down.
All of these are also the case if for kenthackenough-ui-main and kenthackenough-ui-staff
Make sure that schools returned from the /stats/schools
route are being ordered before slicing the top 10.
Send GCMs like this:
{
"to": "/topics/messages",
"data": {
"action": "create|update|delete|notify",
"body": {}
}
}
I can then stop storing device IDs on the server side, and the GCM API will be similar to the Socket.IO API.
I'm not quite sure what went wrong with resumes this year, but it should have generated an error on our end. We also need to either integrate our own fault tracker or use an existing one so that we get a heads up if a lot of API calls are failing and can start to investigate why.
When I send a new email, it shows a cross-origin request error in the developer console.
The email does not get sent, however, it still gets added to the database successfully. I'm suspicious of something in the actual mailer.
You should be able to query users by properties of their application like this:
User
.find()
.populate({
path: 'application',
match: req.query
})
But something mysterious and weird is going on here.
This isn't a crucial feature but it'd be nice to have fixed.
Right now the client may submit an application where the resume field doesn't identify an actual resume. We need to confirm that the resume exists
Ideas for refactor:
let
It used to be working, but now it's not. Could have something to do with environment variables in production. Needs debugged. Will be annoying to debug.
Move a lot of logic into models, especially for statistics
Also, for socket.io and GCM
title says it all
On FireFox 37.0.1 and 36.0.4 If you try to click a link on khe.io the background changes to blue and the links are unable to be clicked. The background changes to blue no matter where you click on the screen.
Also, the email field is not getting scope.
Display nicknames on mass emails instead of a list of email addresses.
Right now we're throwing away everything from last year and starting fresh to set up the site for the next year. This seem inefficient, leads to filling up the hard drive and takes resources from the server and personnel to maintain.
We either need a way to archive old content, or tagging each year's records so we can switch between them
"The bot sends an email as a string to the API, the API returns true if the email was used to register, and false if it wasn't" - Josh
https://api.khe.io/verify/email/:email
If Email was used to register:
{"valid":true}
Otherwise:
{"valid":false}
If possible can we on the server side prevent double adding device IDs? I've got code on my side now to prevent double adding but while I was testing I double resisted and now I get four notifications every time a change is made. A bit of Googling leads me to mongoose's documentation which seems to imply we could change https://github.com/hacksu/kenthackenough/blob/master/app/devices/model.js#L9-L11 to
var Device = mongoose.model('Device', {
id: {type: String, index: { unique: true }}
});
but I don't feel confident this is actually the case
A user's token should need refreshed every day or two. We can fix this once the front-end refreshes tokens seamlessly.
Add them!
title also says it all
Based on the url shorten-er tool, generate new 5 character code short urls
calling DELETE /tickets/ID after the ticket has been removed crashes the server with the exception
TypeError: Cannot read property '_id' of null
at /var/www/kenthackenough/app/tickets/controller.js:78:22
at Query.<anonymous> (/var/www/kenthackenough/node_modules/mongoose/lib/query.js:2120:28)
at /var/www/kenthackenough/node_modules/mongoose/node_modules/kareem/index.js:177:19
at /var/www/kenthackenough/node_modules/mongoose/node_modules/kareem/index.js:109:16
at doNTCallback0 (node.js:407:9)
at process._tickCallback (node.js:336:13)
After a user's application is deleted, it cannot be submitted again.
Replicate this error by following these steps:
curl KEY:TOKEN@API_URL/users/application -X POST -F APPLICATION_DATA...
curl KEY:TOKEN@API_URL/users/me/application -X DELETE
You will receive the following error:
HTTP/1.1 500 Internal Server Error
Server: nginx/1.7.10
Date: Tue, 24 Mar 2015 22:11:29 GMT
Content-Type: application/json; charset=utf-8
Content-Length: 69
Connection: keep-alive
X-Powered-By: Express
Access-Control-Allow-Origin: *
Vary: Origin, Accept-Encoding
{
"errors": [
"You have already submitted an application"
]
}
This is obviously a rarely encountered error, but let's take care of it anyway.
The current API response codes are pretty meaningless..we should update them to a more RESTful standard.
Right now, all resumes get placed in a 'tmp' directory. Use https://nodejs.org/docs/latest/api/fs.html#fs_fs_rename_oldpath_newpath_callback to move them into a directory (set in config file?). ie. All resumes from 2018 would be in a resumes-2018 folder
It seems that when new messages are submitted their created timestamp is set a minute or more in the future.
Right now, our list of IPs is * for debugging purposes. We should whitelist our production server's IP address.
I'm not really sure what's causing this. Probably something I did when trying to install and setup libraries but I'm not sure.
All unit tests use '/v1.0' as the nginx prefix.
Proposed Solution: Use the prefix from the config/config.js file
There should be a route to map user IDs to usernames for display.
/user/name/:userID
Returns: the name of that user
Need to implement a token expiration/refresh system.
Implement push notifications for certain calendar events.
The /users/application/search?q=String
route works on Vagrant machines, but on development and production servers, it does not.
Needs to use the new v2 API and API keys.
Not sure where to store them yet, but this would be useful.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.