Coder Academy Group Project 2018 by Drew, Maxi, Shun and Winter - Team Painframe
đ "Chicken-In" đ is a Timesheet and Shift Tracking web app designed specifically to our client's specifications. Employees can log in on a mobile-first site design, and quickly fill out their worked hours, whilst Managers can log in to a full desktop dashboard, where they can review, approve, and get information about the shifts logged.
Designed to be smooth, seamless and easy to use, Chicken-In gets employees Chicken-In, and Chicken-Out. đ !
(We've added some Dev instructions down the Very End of this document: Dev instructions )
Manager/Employee Logins:
Username:
[email protected]Password:
password
The demo account will update React state, but NOT store information to the database.
Table of Contents generated with DocToc
- Presentation Slides
- The Client Brief
- Application Design
- Project Management
- Tools & Methodologies
- Future Development
Winter did an amazing job of the design for this project, as well as fully custom slides for our presentation. They're worth a look, if nothing else for a chuckle! The slides can be found HERE
Our client, Edwin Huang, owns and operates a number of charcoal chicken shops around Victoria, under the brand name âRed Rocks Charcoal Chickenâ. He splits his time between working in the stores as a manager, and doing all the administrative/entrepreneurial work as the owner.
Red Rocks Charcoal Chicken is a small group of chicken shops all owned by the same owner. They aren't a franchise, more a small group owned by the same group. A lot of their regular employees are students, and as they aren't a very technologically invested company, they are still doing their time-sheets and overtime calculations manually.
Red Rocks Charcoal Chicken have little to no branding (Besides the colours Red and White and various "chicken" mascot logos) to speak of, and almost no internet presence, so we did not have to worry about linking into an existing system. This was a challenge but also provided us with an opportunity to work on branding and style.
One of our client's most significant challenges is keeping track of employee hours worked, calculating overtime and calculating how much he owes in wages to each employee. Currently, his employees record how many hours theyâve worked on a paper form, and he manually updates his excel spreadsheet at the end of the week. Itâs time-consuming work, and his employees often make mistakes which means he has to follow up with each employee if there are any discrepancies in the shifts they've logged.
We saw a lot of potential for being able to help him with his business needs- He suggested we could develop him a front-facing website that his clients could use to look up their locations and menus, but we felt that it wouldn't suit the scope of our project. In the end due to time and scope constraints, we decided with Edwin to work on a small app to take track of weekly hours worked by his staff, and calculate the number of overtime hours.
Big considerations in this project are the differences in the ways in which different users will use the product. Employees will likely wish for a simple, quick, easy interaction with the software, most likely on their mobile phones. Managers, however, will want a lot more control, data display and interactivity with their end of the process.
In both cases, the aim of this project will be to help both the employees, and the managers, spend less time on the administrative chores of timesheets. The mobile app for Employees, will need to be simple, responsive, and easy to use- and the app for the managers will need to do a lot of automatic calculation, organization and filtering, to make the managers' jobs easier and simpler. This is about giving them something they will want to use, not something that scares them away from the technology.
We had our first meeting with the client, Ed, on July 5, when we discussed the web-app's requirements. Ed gave us an overview of the functionality he would like the app to have, incuding having the ability to:
- Approve or reject shift submissions
- View a summary of all the shifts by week
- Automatically calculate the pay based on the hours worked and day worked
- Add or remove employees from the system
We also proposed some ideas, including developing for mobile first, and creating a user-friendly interface for his employees. We built our app based on the requirements discussed in this meeting since the concept didn't change much throughout the development process.
Addressing a key consideration from our client, we decided to implement a mobile-first approach for the employee facing app, where each user is able to log in, enter, track, and view the total standard/overtime/double time hours of each shift input.
We found that this stripped back interface would best serve the needs of the client's team, who found their paper-based system inefficient and time-consuming.
In the Employee section, the app allows employees to log in, enter their shifts, and the app automatically calculates the standard and overtime hours based on the information. The app then sends this information back to our mongodb database for later retrieval by the Manager side of the app. The Employee is also able to see the status of their entered shifts, from pending, to approved and rejected.
This app has 2 main views (pages):
- Add Shift
- Employee Settings (Update password, etc)
On successful manager login, the user is redirected to the manager dashboard, where they have access to several core functions of the app.
In the Manager section, Ed or his managers, can add/remove employees from the system and review the submissions before approving them. The dashboard will also include an overview of past wages paid, which he filters by various fields (dates, locations, employee, etc).
This section has 5 main views (pages):
- Reporting / Filtering past shifts
- Data Visualization
- Review/Approval
- Team Management
- Business Settings (ie overtime rates, store locations etc)
After our initial meeting with the client, our team decided that we could not integrate mobile-first design for the manager dashboard without sacrificing the ability to display and easily read shift data. As such, the manager dashboard focuses on desktop functionality, with mobile-driven added in by converting all shifts to "cards". This is a difficult way to use the app, but is still possible.
We focussed on providing a number of basic user profiles based on the types of users we are likely to encounter. We were informed there are a number of students and graduates who are employed in the shops, and the managers vary in age, the youngest being 24. From this, we created some basic profiles, and decided to focus on how the app may be used rather than fleshing out too many details about the characters.
Employee user stories helped us to really understand that the employee will be looking for a fast, quick, easy option. Nobody likes waiting around after their shift to fill in paperwork, and nobody wants many details stored on our service. Quick login, with a quick entry of shifts, and that's that. They're likely to be on mobile, so this will be a mobile first approach.
This was where we realised the bulk of our app's features would take place. The main "employer" section requires us to make multiple pages for administration as well as information the employer can get out of the app. In the future, we'd love to add exporting/ integration into business management software, but for now we decided to focus on a "dashboard", web version first.
You can find the full list of our insights and user stories HERE!
We initially came together to make a first pass wireframe- which we won't inflict on you in this document, but feel free to click the link and have a look:
Link To: First Pass Wireframes on Figma
You can find our beautiful, Final Design Wireframes on Figma HERE
To create our initial wireframe we used the insights gained through our user stories to rapidly prototype our apps features and general design. From there, we moved on to more specific design wireframes, which can be seen below:
We decided to implement a mobile-first design for our guest and employee login pages as they would be the most frequent users of the Chicken-in Web app. The single page design allows the users to rapidly digest the purpose of the app with a summary of its full functionality in the about section.
Iterating on the single page design, Red Rocks employees are redirected to a lightweight rostering app with the immediate ability to add and review past/pending/rejected shifts.
On this page, the manager is given an overview of the weekly shifts that have been approved or are pending approval. A stretch goal is to integrate a "print report" function that produces a print ready pdf for physical documentation. An additional consideration is to include data visualisation within this dashboard further support the reporting functionality.
On this page, managers have the ability to view all pending shifts input from employees and approve/reject them. Stretch goals are to include a comment popup when rejecting a shift to give further clarification to an employee as to why the shift has been denied.
We decided that a full-scale HR app that stored and protected large amounts of employee data was not necessary for our client scale. Designed for ease of use, we kept the interface lean, with the ability to quickly edit, remove (archive), and add new employees within one view.
Rather than hard code the values for overtime and double time, we decided to give our client the option to change the multiplier according to the current award rate in hospitality, as this is prone to change.
You can find our Design Wireframes on Figma HERE
Given the nature of the Data, we have 4 mongodb Schemas for use in our app: Employee, Manager, Shift, and Business.
-Employee
- id: String (Auto generated by mongoose)
- firstName: String
- lastName: String
- fullName: String
- email: String
- password: String (hashed and salted value)
- location: [ array of Strings ]
- standardRate: Number (cents)
- business [ references
businessid]- status: Boolean (default: true) (for tracking employee status)
employee has no references. Their location is an array, selected on the frontend by pulling in the available options from business
-Manager
- id: String (Auto generated by mongoose)
- email: String
- password: String (hashed and salted value)
- business: [ references
businessid]
business references the ID of business, tying the manager to the business
-Shift
- id: String (Auto generated by mongoose)
- employee: [ references
employeeid]- date: Date
- location: [ Array of Strings ]
- startTime: Number
- endTime: Number
- standardMinutes: Number
- overtimeMinutes: Number
- doubleTimeMinutes: Number
- totalPay: Number (cents)
- status: String ('pending', 'approved', 'rejected', 'archived')
- business: [ references
businessid]
employee references the ID of employee, as each shift is owned by an employee
-Business
- id: String (Auto generated by mongoose)
- name: String
- address: String
- location: [ Array of Strings ]
- overtimeMultiplier: Number
- doubleTimeMultiplier: Number
(Initial ideation / decision making process for our 3-app system, and other ideas)
/
-- /backend/
/models/
/Employee.js
/Manager.js
/Shifts.js
/Business.js
/routes/ ## routes start here for each
/authRouter.js ## /api/route/ namespace
/employeeRouter.js
/...etc
/controllers/ ## ...and refer to 'controllers'
/employeeRouter.js ## for their methods/logic
/...etc
/middleware/
/authMiddleware.js
/logger.js
/...etc
/tests/
/unit/
/example.test.js
/integration/
/example.test.js
/route/
/example.test.js
server.js ## Server Entry Point
-- /frontend/
/dist/ ## webpack bundled production files
/src/
/employeeApp/
/dashboardPage/
/exampleComponent/
exampleComponent.js
exampleComponent.scss
/router/ ## Shared Components live in router
/exampleComponent/
exampleComponent.js
exampleComponent.scss
/AppRouter.js ## Main React Entry Point
/settingsPage/
## components go here
/index.html ## Entry Point served by backend
/index.js
/styles.scss ## Overall Page styles
/managerApp/
## Same structure as above
/guestApp/
## Same structure as above
/img/
/tests/
-- /node_modules/
-- /package.json
-- /README.md
-- /webpack.config.js
-- (other files ie .gitignore, .eslint.rc etc etc)
The overarching Architecture of our App is that it sits on one NodeJS Express App Backend, which serves 3 smaller React 'apps' in separate index.js entry points.
All of these higher level components live in the same root folder structure under their own subfolders- with the package.lock and node_modules in the root- able to run the entire app with npm run start (and other dev commands).
This single-backend containing multiple-frontends approach allows for the entire app to be easily containerized and to have all our apps run from the same root using npm run commands. Each frontend app will be able to be dev served using npm run <app name> for development purposes.
You can read all about our Design and Architecture decisions in our initial Part A Design Docs HERE
We had five formal mettings with our client:
- Thursday, 5th July: This was our first formal meeting with Ed. We met to discuss the web-app's requirements, including core functionality and styling requirements. Ed gave us an overview of how he wanted the app to work, including having the ability to:
-
Approve or reject shift submissions
-
View a summary of all the shifts by week
-
Automatically calculate the pay based on the hours worked and day worked
-
Add or remove employees from the system
We also proposed some ideas, including developing for mobile first, and creating a user-friendly interface for his employees. We went back-and-forth discussing these requirements for about an hour. At that point, we decided we had enough information to develop a basic wireframe so we ended the metting and arranged to meet early next week to discuss our draft wireframe.
-
Tuesday 10th July: We met for approximately 30 minutes to discuss the draft wireframe that we had put together. The wireframe included all the core functionality that we discussed the previous week, but did not have all the styling completed. After some discussion, Ed was happy with the functionality we proposed and didn't ask for any major changes. We ended the meeting and arranged to meet again after the wireframes were fully styled.
-
Friday, 13th July: We met for about 20 minutes to discuss the fully styled wireframes. Ed didn't have any preference for the styling or branding, so this was his first time seeing what we had in mind for styling. He liked the logo and color scheme so didn't propose any major styling changes. Now that he could visualize the final app, he was quite excited about it.
-
Friday, 20th July: We met for an hour to show Ed the progress we had made. We had built the functionality for most of the pages, but a lot of the styling work still needed to be done. Ed was a little concernend that there wasn't enough time to finish the app, so he proposed to reduce some of the functionality. We tried to assure him that even though it looked rough, a lot of the functionality had already been done and that we would have enough time to finish the whole thing. Ed tested some of the functionality and made some recommendations about how the filtering and sorting should work. We agreed to change to functionality to align with what he wanted.
-
Wednesday, 25th July: By this stage, we had finished 95% of the app's functionality and styling. We met with Ed to show him the app, and he was impressed with what we had done. We walked him through various aspects of the app, and he tested out some of the functionality as we went along. He was happy with how everything worked, and was glad that it worked on his mobile phone. Afterwards, we discussed next steps, inducing deployment, data and ongoing maintenance. Ed told us he would continue to maintain the app with the help of his nephew, who is a web developer. We told him we would hand over our project to him on Friday, after we finished our presentation.
When tackling this opportunity, our team wanted to develop a lightweight shift tracking app, that merged user centric design with the MERN for intuative communication between the front and backend codebase.
While a competitive marketplace, we found that a disproportionate number of employee and shift management services offered online are built ontop of models with a high barrier of entry for small to medium sized businesses, such as our clients.
In addition to varying financial costs, users are forced to make the choice between larger companies, which can offer customisation, and smaller companies, which while leaner and easier to use, lack the ability to be tailored to their business.
Identify colour trends and flat designs in competing businesses, we were able to develop a visual identity for our clients app to create a sense modernity while complementing the brands heritage (see bottom right).
When picking colours, our aim was to use contrast to our advantage to direct the user's attention towards import buttons and visually guide them through in-app processes.
To ensure we stayed on track, our team implemented a series of system, from agile methodologies to project management apps, to ensure we kept to our project timeline.
Initial Projections had the core planning stages to be completed within the first week and second week would be attributed to the brunt of the codebase.
In conjunction with Trello, we also utilised timeline app create by one of our own team for our initial build. Below you can see the main legs for our stages of development.
We got in and moving on our project timeline really early. We actually got to use one of Maxi's previous projects, a timeline tracking app, to lay out and timeline our journey.
We initially just planned out the design phase for week one, starting with ideation and moving right through to design work.
We reassessed at week 2, plotting out our planned course for the development phase of the app. This began with skeleton structure of the entire app, and then broke out into fleshing out both back and front-end sections.
We were happy to see that for the most part, we kept on schedule or sooner! Using our Trello board in combination with the timeline we were able to keep our heads and feel good about the journey.
While we were fortunate to have a diverse array of proficiencies within our team, we decided that as this was primarily a learning experience for us as junior developers, we would not pigeon hole any indivudal into a single front or backend task.
Instead we devided our work through feature and each were individually responsible for the front or backend logic ascociated.
When it came to breaking out individual tasks and allocations, we decided instead to work on an Agile model, breaking down the work into cards, organising the cards at regular intervals, and then breaking out into sections we felt we could achieve, or wanted to work on.
As a high level overview, we ended up seeing everybody working well together on everything, but in particular, we saw the team working on:
-
Maxi :
- Webpack and Front-end Design Structure
- Deployment and Production Environment
- Backend Controller Logic, Routes, Models
- Front-End React Pages in Guest, Employee, and Manager
- CSS Styling
- ...And more!
-
Drew :
- Back End Design and Structure
- Backend Routes Logic, Models
- Git Workflow and Github oversight
- Scrum Board Management
- Front-End React Pages in Manager
- Testing, Development and Production Environment Setup
- Basic Intergration Testing
- ...And more!
-
Winter :
- Layout, Component and App Visual Design
- Component Design and CSS
- Backend Routes Logic, Models
- Front-End React Pages in Guest and Manager
- CSS Styling
- ...And more!
-
Shun :
- Front-End Logic for Filter methods
- Front-End Logic for Sorting methods
- Employee Management React Front-End
- Backend Routes Logic, Models
- CSS Styling
- ...And more!
Most of my pain points came around deployment. Whilst we did set up the ability to work in development and production modes, avoiding the need to build with webpack for every single small edit, having to switch between the two at production time was a definite hassle. We would do that differently next time. Having to run build and reset pm2 every time we wanted to deploy was also a bit of a pain...I'm sure there's a more streamlined way to do it than the way we did implemented it. Dealing with changing requirements from the client added challenges, and as the main point of contact I experienced the brunt of this. I found it a big challenge in the midst of a 2 week assignment, as interesting as it was.
I loved getting in and getting this done. I found myself immediately drawn to the project management and team organization aspects of working in a team of coders- and threw myself into git, github, trello and all of those things. Git is an interesting challenge, and we had a few stumbles early on, but mostly corrected them by sticking to our git workflow.
Project Scope was probably the biggest challenge. It wasn't far in that I realised how much we had to achieve. As a result, we had to opt out of TDD (time didn't seem to allow for it), and that was a real shame, to me. If I could do over, I'd choose a smaller project, and full test it as I go.
I struggled with React State, a little, early on- but I ended up learning more about React in this project than in the entirety of my prior learning. By the end of it, I had a great understanding of how that all works, and I'm keen to expand my knowledge into Redux, etc.
I had some challenges with Route and Model testing, so much so that I had to leave the work half-finished by the end. This was a shame, but given the scope of the project I had to let it slide in order to get the core app functionality right. The little testing I did do, brought to light some issues immediately which we were able to resolve. The next project I work on will be Tested From The Start!
The greatest challenge I personally encountered in this project was handling the axios request and linking front-end logic to our applications database.
While focussing largely on front-end development in this project, this was an area that I actively sought to hone. From this, I garnered a greater appreciation and understanding of back-end logic, database architecture and how to effectively interpret other developers code.
Guest app front-end
- Change states by user input (Manager/Employee)
Manager app manage-employee page front-end
- Create modal(react-modal)
- Styling grid on modal
- Implement sorting function for employee cards
- Display previous states on edit employee form modal
- Display/hide location checkbox in form modal
- Link employee data to backend by axios and lifecycle methods
Task tracking and breaking out tasks to individual features was really important to our team dynamic. Early on, we had very frank and open discussions about the fact that we had a diverse team, with varying levels of time outside of class and confidence in abilities. We decided to work hard to implement agile practices and break down tasks into achievable chunks, so as to best work together to achieve our goals.
To that end, from the outset we implemented a main Scrum-style Trello board, split into a number of sections:
Link To Main Scrum Trello Board
Ideas / Stretch Goals
All our ideas started here - straight from our very first planning session after our initial client brief. New ideas land here first, are discussed and formed if agreed to, where they move to âTasksâ.
Tasks
All our main work starts here, first. âTasksâ are usually larger chunks of work, including documentation, planning, and research as well as the obvious âApp Featureâ tasks. If necessary (mostly during development), these Tasks will be broken down into smaller âFeaturesâ, where they can be worked on by individuals. An example might be âUser System (Model, Login, Logout, Auth)â
Features
Mostly used in the Development stage, âFeaturesâ are generally front or back-end pieces of code, that come together to make one functional piece of the App. An example might be âFrontend: User Login Formâ
To-Do (Next Few Days)
The title is fairly self-explanatory, this is just a staging area for us to further breakdown our workflow. From here, weâll decide at standup meetings who is going to work on which card.
In Progress
Self-explanatory- This is where weâll hold the cards weâre currently working on. No card gets here without assigning a team member to it.
Blocked
If something is holding us up from work, we put the card here, and comment as to what the blocker is. The intention here is for us to help one another overcome the blockage, or rethink the card scope or timing.
Pending Review
Somewhere to put cards if we need another team member to check on the work or review the code.
Completed
Completed cards go here! Yay!
Access / Resources
A collection of links we can use to quickly get around.
The board was designed to be self-documenting, and make sense, so as not to be too confusing. We decided against using many, smaller boards, in favour of one large board, where we can follow a card from original idea through to completion.
From the outset of this project, we decided to take a lot from the Agile methodology. Given that each of us has our own unique strengths, we have embraced the idea that we will be far more powerful as a team, and the Agile (Specifically, Scrum) practices offer a lot in the way of organization.
- Scrum Board (See Scrum section for breakdown)
- StandUp Meetings
- Ideation sessions (multiple for design, architecture, initial planning)
- Review (Retro)
- Code Reviews
We kicked it off with a big Ideation whiteboard session. This was great in laying out what we needed and wanted to achieve given the clientâs needs. Further meetings were had later in this vein for the Architecture section, and again for Design. Iterations continued with each meeting, and the product started to take form. From this, we started up our Trello board âIdeasâ column, and the flow progressed from there.
(We also had fun. Winter got really into his post-it notes :D)
We also agreed to daily stand-ups, so as to get one another on the same page as far as what weâre all working on. We had some big conversations about our plans to break down tasks and work together. We all agreed that whilst itâs important to lean on our strengths, nobody should have to feel as if they are âCarryingâ any aspect of the project - from design, through to code. To this end, regular stand-ups, and breaking things down into achievable chunks were very important to all of us.
Adding in a âBlockedâ step to the scrum board was another big part of this. We agreed that if somethingâs holding us up, we can park the card here and move on to something else, after notifying the team or finding someone to help. We plan to hold weekly reviews- Perhaps not full âretrospectivesâ, as we may not have time, but at the very least borrow from the âsafe, blameless spaceâ ethos while holding these reviews.
Our process was handled largely through Trello, with each completed being sent to our "Pending Review" list prior to deployment.
If a feature/page/branch encountered an error that could not be immediately fixed, a member of our team would place it in the "Blocked" column. This signalled for one or more team members to jump on the issue and try to find a solution.
We used annotations and TODO: lists in code to identify problem areas or necessary fixes.
For version and source control, we used Git, with a main repository stored on Github. We went in with production in mind, and designed the flow with a âProtectedâ master branch for pushing to production, and a develop development branch for working between releases.
This actually worked out really well, allowing us to separate out our releases toward the end of our development cycle.
We have taken a lot of inspiration from this model: https://nvie.com/posts/a-successful-git-branching-model/
Master (Protected): The master branch is âProtectedâ against any git push commands, and only able to be merged via Pull Request on Github. The idea behind this is that master is only to be used for stable releases that have been tested and checked thoroughly.
For individual features, we used separate feature branches cloned off of the develop branch. These branches will be merged into the develop branch, after changes are pulled and conflicts resolved. We intend to push these feature branches as well, to keep a record of our progress, though this isnât really necessary for a project of this scale.
We began by pushing these smaller branches to github, but when it became clear that we'd be polluting github with over 100+ branches, we decided to keep those local.
Creating a feature branch (Example, âLogin Formâ)
Step 1: Make sure youâre in âdevelopâ
git checkout develop
Step 2: Pull the changes from the main repo, getting you up to date
git pull origin develop
Step 3: Create your new feature branch - This is important! Otherwise weâll be conflicting up the wazoo on develop :P And try to name your branch smart! Use the name on the Trello card youâre working on, if possible.
git checkout -b login-form develop
(note the âdevelopâ on the end, makes sure it creates the branch off of âdevelopâ, not âmasterâ)
Step 4: Make your changes! Do your code!
<-- đ -->
**Step 5: Add and commit your changes to your branchv
git add .
git commit -m âA descriptive and useful commit message please!â
Step 6: Push your branch to github (for history's sake)
git push origin login-form
Step 7: Change back to âdevelopâ
git checkout develop
Step 8: Do a git pull (on âdevelopâ branch) - this is to make sure that if someone else has gone ahead and pushed since you last worked on it, youâre all up to speed. ALWAYS GIT PULL BEFORE PUSH! :D
git pull -u origin develop
(or just
git pullif youâve already set your upstream)
Step 9: Merge your branch back in (Again: Donât forget to pull, first!)
git merge --no-ff login-form develop
Step 10: Push your change to develop
git push origin develop
We had finished about 95% of the app's functionality and styling by July 25. We met with Ed to show him the app, and he was impressed with what we had done. We walked him through various aspects of the app, and he tested out some of the functionality as we went along. He was happy with how everything worked, and was glad that it worked on his mobile phone. Afterwards, we discussed next steps, indlucing deployment, data and ongoing maintenence. Ed told us he would continue to maintain the app with the help of his nephew, who is a web developer. We told him we would hand over our project to him on Friday, after we finished our presentation.
We put together a short questionaire for our client, to gauge how well they thought we completed their brief, and get feedback on the process.
- How satisfied are you with our app?
- How well did we communicate with you throughout the process?
- How easy do you find it to navigate through our app?
- How likely are you to use our app for your business?
- Which feature did you like the most?
- Which feature do you think needs the most improvement?
As of the time of submitting this documentation, we haven't heard back - but We're eager to hear back from our client about what he thinks!
This was a fun, but challenging project. We all had a chance to work on various aspects of the app, including both the front-end and back-end. This was the first time any of us had developed an app as a team - we had to quickly learn how to work in a team efficiently and find a balance between pushing out work and making sure people got to work on what they wanted. Overall, it was a great experoence and we became better developers because of it.
We went into this project knowing that the scope was wide. As a result, we made a decision not to use Test Driven Development, as we felt that we'd have enough challenges as juniors working out our methods, before we could test them.
As a result, a large portion of the codebase is untested- and we had to rely on fairly extensive User Testing to get by. This worked fine for us, in the end, and we did add a few basic intergration tests to see us through, but it was a shame that given the scope of the project we couldn't fully test.
We all did see the value of testing frameworks and test coverage, though- especially toward the end of the development process. Reviewing the codebase, we can now see how much we could have benefited from having a Test First structure, as the code would end up being much more cohesive and robust.
We did use both Unit and Intergration testing, and we did all get a chance to see, and learn how that works. On our next projects we all look forward to being afforded the time to be able to take testing on.
Given the scope and scale of the project, our team was cognizant of the need to track and monitor our progress with each stage of development. To best support the team dynamic and ensure continual development we used the following tools:
For developing the wireframes. We had a two-stage design process- An initial group-planned mock wireframe, and then more fleshed-out "design" wireframes.
For general project management. We leaned heavily on this! We've multiple boards for resources and internal info, but it all flows back to our main trello board (example below).
Scrum board.
For version control. See below for more detail about our Git workflow.
Given the limited time-frame of our project there were several features that did not make it to production:
⢠Display user avatars or names on login to personalize the user experience and help managers quickly identify employees.
⢠Animation to create a more fluid interface and help guide and reward user interaction.
⢠Add additional admin status to include sub-managers with limited admin control/ dashboard control.
⢠"Print PDF" functionality for the Reports and Data Visualisation pages, so that the client can easily export/store data.
In addition to features for future inclusion, our team considered several key requirements for handover and ongoing support:
⢠Legiable source code and annotations to allow for future independant developemnt.
⢠Initial login details handed to the client.
Welcome, Nerds! We're impressed that you got this far. If you're interested in toying with our code - Here's a quick overview of what you can do to get started.
- NodeJS 8.0+ (with NPM)
- MongoDB (running on localhost, default port)
Once you've installed the necessary pre-requisites, you can go ahead and clone the repo:
~/ $ git clone https://github.com/oheydrew/chicken-in.git
~/ $ cd chicken-in
~/chicken-in/ $ npm install
First, you'll need to create a .env file inside /chicken-in 's root.
~/chicken-in/ $ touch .env
Open this file, and add the following configuration variables:
.env:
SERVER_PORT=3000
MONGO_URL=mongodb://localhost
MONGO_PORT=27017
MANAGER_PASSWORD='password'
EMPLOYEE_PASSWORD='password'
Next, open up the hostUrl.js file, to configure it back to development mode.
~/chicken-in/ $ cd frontend/src/
Uncomment the localhost:3000 line, and comment out the undefined line.
~/chicken-in/frontend/src/hostUrl.js:
const hostURL = 'localhost:3000'
//const hostURL = undefined
export { hostURL }
~/chicken-in/ $ npm run seedData
Then, from the ~/chicken-in/ root:
~/chicken-in/ $ npm run start:dev
For the Guest (Login) app:
~/chicken-in/ $ npm run guest
For the Manager app:
~/chicken-in/ $ npm run manager
For the Employee app:
~/chicken-in/ $ npm run employee
Open chrome to http://localhost:8080/ and you're away!
You'll need to set the hostUrl file back to how it was when you cloned the repo.
Then: ~/chicken-in/ $ npm run build
Run the prod server with: ~/chicken-in/ $ npm run start
Open chrome to http://localhost:3000/ and you're away!
Brought to you by Painframe LLC