EnviDat Frontend

This is the source code for the frontend of the EnviDat (Environmental Data Repository) platform which provides environmental research data for researchers around the globe. The publication of research data is mainly provided to researchers of the Swiss Federal Institute for Forest, Snow and Landscape. However, collaborations with external researchers are possible and encouraged.

The frontend connects to the backend API to load research datasets, projects and organizations via CKAN based actions (CKAN version 2.9).

This main goal of this custom frontend is to improve the UI / UX for the regular usage of researchers looking for environmental research data and managing their data publications.

Since version 0.6.x includes features for dataset management, by version 0.8.x mainly creating and editing dataset. It doesn't replace all the features of the ckan UI. For admins and system admins the CKAN UI is still necessary.

The frontend is build with vue, vuex, vuetify, vite.

Installation

After cloning the project, use npm install to install all the dependencies.

Local development: npm run serve

Create a build: npm run build or npm run build --modern

Local Storybook: npm run storybook-dev

Usage

You have to change the environment variables in the .env.development / .env.production files. The .env.production is used when creating a build via npm run build.

.env.development (release 0.8.0):

# enabled using local testdata, files from public/testdata/
VITE_USE_TESTDATA=false

# enable listing of error on the /#/report page
VITE_ERROR_REPORTING_ENABLED=true

# url to load the config from
VITE_CONFIG_URL=./testdata/config.json

# static root is used to load markdown & images for blog, services page
VITE_STATIC_ROOT=https://frontend-static.s3-zh.os.switch.ch

# api root is used to connect to the server backend
# adjust for different testing environments
# VITE_API_ROOT=http://localhost:8991
VITE_API_ROOT=https://envidat04.wsl.ch

# root & base & individual action to make get & post requests to backend
VITE_API_BASE_URL=/api/action/

When VITE_USE_TESTDATA=true VITE_API_ROOT variable is ignored and the local testfiles are being used. There has to be json files in the /public/testdata/ folder which resemble the result of the actions. Like the action 'current_package_list_with_resources' for all the datasets or 'package_show' for a single dataset.

For more details about the actions check '*actions.js' files in the respective modules. E.g. './src/modules/meatadata/store/metadataAction.js' for metadata / dataset actions './src/modules/projects/store/projectsAction.js' for projects actions.

.env.production (release 0.8.0):

For a productive build you have to change the VITE_API_ROOT variable to point to your CKAN backend. If the VITE_USE_TESTDATA is still on true, the testdata is being used regardless of being a production build.

  VITE_USE_TESTDATA=false
  VITE_CONFIG_URL=./config.json
  VITE_API_ROOT=https://www.envidat.ch

To use any other backend, you would need to adjust the code in the actions of the different vuex stores to connect to different endpoints and to handle their responses accordingly.

Check the CKAN actions and their details here: https://docs.ckan.org/en/2.9/api/index.html

Proxying CKAN

• Due to CORS and sameSite cookies, the CKAN instance must be accessed from the same domain as the frontend.

• This can be achieved by running both the frontend and CKAN on localhost (see https://gitlabext.wsl.ch/EnviDat/ckan-container.git).

• Alternatively, if the CKAN instance is running on a remote server, the traffic can be proxied to be accessible from localhost.

• Create the file .ckan-proxy.secret:

CKAN_HOST=<DOMAIN_OR_IP>:<PORT>

• Run the proxy:

docker compose -f docker-compose.proxy.yml up -d

• Access CKAN at localhost:8989 or include in .env.development.

• VITE_API_ROOT=http://localhost:8989
  

Config

The config.json can be used to inject configurations and change some behavior of frontend components without rebuilding and deploying it.

That can be a simple as changing the welcome text in the dashboard, to completely disconnect the frontend from the backend and load the datasets (metadata only) from a static file as backup. Helpful when doing maintenance or hotfixes on the server-side. In that sense, the config provides some basic feature flags for some frontend components.

Minimal config.json setup on server side is: { "version": "0.8.0" }

The version is used to check if the user has to reload the frontend to get the latest version available.

Configuration Options (version 0.8.0)

Option	Usage	Type	Required	Default
version	Prompts the user to reload the page when a newer build is available. Make sure to increase along your with the version in the package.json this is extracted into a build and compared with the version of the config.	String	true	-
welcomeInfo	Object with different titles and texts for the Homepage / Landingpage.	Object	false	Defaults are hard coded in the landing page and are only used if nothing is provide from the backend.
aboutInfo	Is a list of json objects which are represented in the about info cards. At least provide strings for card "title" and card "text". Title should be kept short. Text can include html / markdown. Overwrite the "img" via an url to provide a different image. Reference About Page.	Array	false	Default infos are hard coded in the about page and are only used if nothing is provide from the backend. Once something is provide via server side config, only the about card infos from the backend config are used. Default cards are 'Contact', 'Our Mission', 'Concept', 'Community', 'WSL' and 'Team'.
metadataConfig	Contains the major configs for MetadataDetailPage component and any fallback urls.	Object	false	false
.loadLocalFile	Set true to make usage of the metadataConfig.localFileUrl	Boolean	false	false
.localFileUrl	The url which is used to load all the metadata for "static usage" or fallback together with the maintenance mode.	String	false	false
.resourcesConfig.downloadActive	Set true to enable downloading of resources	Boolean	false	true
.authorDetailsConfig.showAuthorInfos	Enable to make the author infos show up (includes, email, ORCID, Affiliation)	Boolean	false	false
.authorDetailsConfig.showDataCredits	Enable to make the data credit list show up	Boolean	false	true
.authorDetailsConfig.showDataCreditScore	Enable to make the data credit score and level show up	Boolean	false	false
projectsConfig.loadLocalFile	Set true to make usage of the projectsConfig.localFileUrl	Boolean	false	false
projectsConfig.localFileUrl	The url which is used to load all the projects data for "static usage" or fallback together with the maintenance mode.	String	false	false
maintenanceConfig	Contains the details for the maintenance / message.	Object	false	false
.signinDisabled	Disable sign in links to prevent the user from using any signed in functionalities.	Boolean	false	false
.messageActive	Enables the message banner for the maintenance mode message.	Boolean	false	false
.message	The actual message shown on the banner.	String	""	false
newsConfig.newsActive	If set to true the LandingPage will show the "news" entries in the list.	Boolean	false	false
newsConfig.entries	A list of news entries (object with title, text and an image url).	Array	false	[]
effectsConfig	Contains the details for the shown effects.	Object	false	false
.landingPageParticles	Enables polygon particles showing on the lower part of the landing page for an "dynamic forest analysis" effect.	Boolean	false	true
.decemberParticles	Enables showing snow particles falling down in the background of all pages during december.	Boolean	false	true
userEditMetadataConfig	Configurations of the Editing / Creation workflow	Object	false
.publicationYearsList	Amount of years in the selection for the publication year	Number	false	30
.publicationMaxFunders	Max amount of possible funder entries	Number	false	55
.datasetCreationActive	Enables the creation workflow (dataset creation in the local storage), if false users will be redirected to the CKAN UI for dataset creation	Boolean	false	true
.resourceEditingActive	Enables the editing of resources in the editing workflow	Boolean	false	true
.editingRestrictingActive	Enables dataset owners restrict access to a resource to users which aren't in the same organization and to give access for specific users	Boolean	false	true
.customFieldsMax	Max amount of possible custom fields	Number	false	10
.keywordsListWordMax	Max amount words a keyword can consist of (this is avoid long multiple word keywords which filter to narrow immediately)	Number	false	2
.keywordsCountMin	Minimum of keywords necessary for a dataset	Number	false	5
userDashboardConfig	Configurations of user dashboard when a user is signed in	Object	false
.dashboardRedirect	Enable it to redirect the user to the CKAN UI instead of this one (for maintenance)	Boolean	false	false
.introText	The introText & feedbackText are combined to provide an introduction text at the top of the dashboard	String	false	Implemented in the introductionCard.vue component
.feedbackText	Max amount of possible funder entries	Number	false	55
.userEditingEnabled	When enabled the users can change it's infos	Boolean	false	true
.useTokenSignin	When enabled the signin is done via modern JWT token exchange provided by the CKAN backend via the passwordless plugin	Boolean	false	true
.organizationRolesText	Object with different texts which describe the different roles a user can have. This is important for new users which don't have editor rights, as it provides a short instruction how to get the editor rights.	Object	false	true
.showOldDashboardUrl	When enabled the introduction text shows a url to change to the CKAN UI, if users can't find a specific feature prodivded by the EnviDat Frontend	Boolean	false	true

Example Config

{
  "version": "0.6.921",
  "aboutInfo": [
    {
      "title": "Contact",
      "text": "Contact the EnviDat team by <a href='mailto:[email protected]'>[email protected]</a> for support, quesitons or feedback."
    },
    {
      "title": "Concept",
      "text": "EnviDat supports the user-friendly registration, documentation, storage, publication, search and retrieval of data sets from the environmental domain. We provide various services to our users and we follow a set of principles as summarized in our concept image below. Additional detailed information can be found in our <a href='https://www.dora.lib4ri.ch/wsl/islandora/object/wsl:18703' target='_blank' rel='noopener noreferrer' onclick='event.stopPropagation();' >concept paper on DORA</a>."
    },
    {
      "title": "Team",
      "widthClass": "col-12 col-sm-6 col-md-8"
    }
  ],
  "metadataConfig": {
    "publicationsConfig": {
      "resolveIds": true,
      "idPrefix": "*",
      "idDelimiter": ":",
      "resolveBaseUrl": "https://www.envidat.ch/dora/"
    },
    "authorDetailsConfig": {
      "showAuthorInfos": true,
      "showDataCredits": true,
      "showDataCreditScore": false
    },
    "resourcesConfig": {
      "downloadActive": false
    },
    "loadLocalFile": true,
    "localFileUrl": "./testdata/packagelist.json"
  },
  "projectsConfig": {
    "loadLocalFile": true,
    "localFileUrl": "./testdata/projects.json"
  },
  "maintenanceConfig": {
    "signinDisabled": true,
    "messageActive": true,
    "message": "We are currently doing maintenance on the server and EnviDat is now in <strong>read-only mode</strong>. Data download, upload and <strong>user data management functionalities are currently disabled</strong>."
  }
}

Missing Major Features vs CKAN UI

• Organizations-Page which lists all the organizations
• User-Page which shows the details of user
• There are various minor features which are missing

Known issues

• When using the text search on the BrowsePage (route /#/browse), the 'query' action which is being called, isn't a standard ckan action it's custom built. That has to be replace manually to the 'package_search' action from ckan, with the respective parameters for a solr query and the repsonse maybe have to be handled differently.