Technology we use

From Redazione
Jump to navigation Jump to search

Here is a list of the tools we use to make this platform work. Of course each of them depends in turn by other tools (for instance Quasar Framework depends on Vue.js, and Vue.js is in turn composed by several components, like the router and vuex, a "state management pattern"), however, given a set even arbitrary of the most prominent tools we use, the reader might explore on his or her own both in extension and in depth the vast areas not covered by the following table.

Currently used

software description use
Mediawiki 1.35 The collaboration and documentation platform which inspired and powers Wikipedia. We use it for the back-end side of this site, in order to ensure collaborative creation of quality contents.
Mediawiki extensions Plugins extending the set of features of Mediawiki. Some of the most prominent extensions include: Page forms, Semantic Mediawiki, Visual editor, Ms Upload, MultimediaViewer, and more: we are also using a custom extension created by us, including CIForms!
Quasar framework open-source Vue.js based framework for building apps We use it for all the Front-end part of the site, taking advantage of its server side rendering feature and beautifully crafted components
Cheerio DOM manipulation from an html string We use it to perform some adjustments to the mediawiki html output before recording the latest version of a page to be served through the APIs
nodemw Node.js interface for the Mediawiki APIs We use it to call the mediawiki APIs from a Node.js script to retrieve and structure wiki pages before serving them through the APIs
detectlanguage Language identification JSON webservice We use it to determine the language of a wiki page, in order to filter them by language in the front-end site, or to perform hyphenation
Detecting languages (Google Cloud Translation) Language detect using Google Cloud Translation APIs As above, we use both of them concurrently
Knex Query builder for Node.js We use it to assemble mysql queries in our Node.js environment
Email templates A wrapper for sending email using template engines (like PUG) and transports (like Nodemailer) We use it to send email using PUG templates and Node mailer-Sendgrid transport
Prism Javascript syntax highlighter We use it to format code snippets on the Front-end site
jsonwebtoken Node.js implementation of An implementation of JSON Web Tokens. We use it to authenticate logged-in users
fancytree Javascript library providing interactive trees based on hierarchical data structures We use it to display a structured and editable table of contents of a given set of wiki pages to be used as navigation panel in the related front-end site.
Hyphenopoly "Hyphenation for node and Polyfill for client-side hyphenation." We use it to hyphenate the organization or documentation's pages based on the detected language of each wiki page. Especially important on mobile devices combined with justified columns.
pouchdb client-side open-source JavaScript database inspired by CouchDB We use it to store all the organization or documentation's pages on the client, to minimize loading data from the server and to allow offline navigation
International Telephone Input "JavaScript plugin for entering and validating international telephone numbers" We use it to assist the entry of international telephone numbers in some tables of private wikis
Redis "in-memory data structure store" We use it to record private files on the back-end which can be accessed on the front-end given logged-in user credentials or permissions. In certain circumstances the list could be composed each time that a page is accessed, hence the need of a solution allowing maximum performance. "real-time, bidirectional and event-based communication" We use it to broadcast advertisements and notifications through our internal system, based on user's contributions !
TinyMCE online rich-text editor We use it to empower our integrated, web-based email client
split-html split HTML in balanced HTML fragments We use it to correctly split pages for the Books section of the site, or to insert multiple choice questions, within pages
split-human-name "Split a person's name into first name and last name fields" We use it to fill-in the list of contacts from incoming emails for our web-based email client

Previously used

software description use notes
Esprima ECMAScript parser We use it to securely parse the javascript object produced by the Frontend Index (the link might be broken) storing the structure of the tree The functioning of the the former "Frontend Index" (now renamed "Table of contents") has been reviewed. Now the tree object representing the table of contents is saved as a semantic property directly by a form, rather than by a script invoking the Mediawiki javascript api, in a process technically virtuous but not the simplest.

... and more ! The list is in progress !