A web project using the AngularJS framework and based on the functionality of Web Article Library rewritten with .NET Core on the server side and Type Script on the client. It implements simple logic for creating, storing and maintaining articles. Some features:
- Authentication (through the session storage and cookies)
- A simple role model
- An administration module
- User settings
- Rating articles, leaving comments and making complaints
- A notification system (SignalR)
- SQL as a default DB model provider (Entity Framework) (read more)
Administrators can review articles if they have taken them over as assignment. Approving the articles makes them available for other users to read.
Administrators are also responsible for dealing with complaints (either about comments or articles).
Administrators can ban other users or make them into administrators or conversely into regular users.
The application is dependent on .NET Core 2.2 (read more) along with SQL Server (read more).
The first user registered in the application will be granted administrative rights.
- Entity Framework Core 2.2.6
- Angular 1.6.4
- Bootstrap 3.4.1
- Font Awesome 4.7.0
- jQuery 2.0.3
- Bootstrap-wysiwyg 1.04
- Hotkeys 0.8
- SignalR 2.4.1
- Select2
- Microsoft .NET Core SDK 2.2.207
- The project was developed in Visual Studio Code (the product page)
- Node.JS 14.17.5 (along with Npm 7.20.5)
The developer computer has to have an access to MS SQL Server installed to deploy the database (read more).
To restore depdendencies of the project run npm install. VS Code suggests restoring packages for all .NET projects on the first run, otherwise it can be done manually by running dotnet restore <project_file_path>
The stages to deploy the application in Docker by command prompt:
- To build an image for the application run a command in the root folder of the solution where the Dockerfile file is available (including the dot at the end): docker build -t webart .
- To build an image for the database layer run a command in the Docker/MsSql folder where the second Dockerfile file is present (including the dot at the end): docker build -t webart_mssqlsrv .
- Providing there is no swarm, it must be initialised: docker swarm init
- The application relies on secret values. The list of all possible secrets can be found in the Docker/docker-compose.yml file, including a secret DB_SA_PASSWORD for storing a password for the SA user of MS SQL Server. An environment variable ASPNETCORE_URLS is necessary to bind ASP.NET Core to certain URLs (0.0.0.0 means all IPv4 addresses on a local machine). The rest of the environment variables and secrets in the file represent the configuration parameters of the application where a section name is separated from its children with two underscores __ . The latter are interchangeable and can be placed either as secrets or environment variables. The secrets are entered the next way: echo "SECRET_VALUE" | docker secret create SECRET_NAME -
- In the Docker directory where the docker-compose.yml file lies run a command: docker stack deploy -c docker-compose.yml webart
appsettings.json lying in the root catalogue of the main project serves as the general file for configuration. The next parameters can be set there:
- Host, Port, UserName, Password in the SMTP section for configuring a mail address which will be used to send messages to the users when registering, resetting password, changing email addresses or their statuses being altered
- AboutUs, Fax, Phone, Mail, YoutubeLink, FacebookLink in the ContactInfo section for storing additional contact information shown in the bottom section of the site
- The Kestrel->Certificates->Default section contains parameters Path and Password for setting up an SSL certificate. The section can be removed for debugging.
There is also one more imperative parameter in the same file, but in a section ConnectionStrings, where a connection string to the database should be added (or changed).
The api module is available through the base url address by adding the api postfix and a respective controller along with a method and its parameters. For instance, http://localhost/WebArticleLibrary/api/Article/GetArticleTitles. The external public api methods (unnecessary parameters are marked with ?):
- POST Authentication/LogIn with an object { name, password }, returns a logged in user's data { status, id, name, firstName, lastName, patronymicName, email, photo, showPrivateInfo, insertDate }
- POST Authentication/Register with the object { name, password, firstName, lastName, patronymicName?, email }
- GET Authentication/LogOut
- GET Authentication/Confirm with a parameter confirmationId, which is a GUID number formed and sent in a link to a new user's email when registering in the application
- GET ContactInfo returns the company's contact data { fax, phone, mail, youtubeLink, facebookLink } (how to configure)
- GET ContactInfo/AboutUs returns a short description from the AboutUs page { aboutUs }
- GET Article/Categories returns an array of names of default article categories existing as constant strings in the system
- GET Article/Titles produces two dictionaries and an object { articles: { id, authorId, name, tags, insertDate, estimate }, userNames: [] }, where userNames is a dictionary with user identificators as keys along with their names as values
- GET Article/{id} accepts two parameters: id, userId?. userId indicates that the outcome should contain additional information such as related comments, photos, current article rating. The result is { article: { id, authorId, assignedToId, name, description, tags, insertDate, status, content }, updatedDate, comments: [{ id, authorId, responseToId, articleId, content, insertDate, status }], userNames: [], userPhotos: [], estimate, curEstimate }, where userNames and userPhotos are dictionaries with user identificators as keys along with the respective objects as values
- GET Article searches for articles by means of the next parameters: author?, tags? (a part of the string containing categories), text? (it's either a part of the description or name field), dateStart?, dateEnd? (a date range of when articles were created), page = 1 (a result page number; the number of items per page is the constant of 10), colIndex = 7 (an index for a sorting column; possible values: NAME = 0, AUTHOR = 1, TAGS = 2, ASSIGNED_TO = 3, STATUS = 4, TEXT = 5, ID = 6, DATE = 7), asc=false (a direction of sorting results: ascending or otherwise descending). It returns the object { articles: [{ id, name, tags, insertDate, status, description, authorId }], articleCount, pageLength, userNames: [], estimates: [] }, whereas userNames and estimates are dictionaries with user identificators as keys along with the respective objects as values
- POST UserInfo/ReplacePassword accepts an object { newPassword, confirmationId }
- GET UserInfo/ResetPassword requires a parameter email as an email address where a message for resetting password will be sent if the respective user exists in the database
- GET UserInfo/ConfirmEmail requires a parameter confirmationId
The project database is based upon SQL Server 2012; later versions of the product are also possible to make use of.
The database model is built with Entity Framework Core 2.2.6. As long as the database does not exist, it will be created automatically when running the application.
There are 8 tables altogether:
- User stores users' data such as: preferences, email, personal data, etc.
- Article keeps all information related to articles
- UserComment, UserComplaint, UserEstimate are all connected to both of the mentioned tables and contain respective information (comments for articles, complaints related to either an article or a comment, ratings for articles)
- Amendment stores amendments corresponding to an article being reviewed
- ArticleHistory contains descriptions of all events related to a particular article
- UserNotification is made up of data related to events from ArticleHistory and connected to a particular user (for instance, notifications for administrators about a new article being waited for a review)
