Update expert ui/api to show unlabeled websites from elements

[qjhart]

@UcDust this PR #428 adds these unlabeled websites to the expert record

[UcDust]

Here's an example of some links from Quinn.

[UcDust]

@wrenaria

Could you weigh in on the display of longer links? @Vensberg had mentioned in slack

I asked my students to input correct URLs into Elements that are now supposed to appear under "Links". None of the URLs have labels, and tend to be pretty long. Any recommendations on how to display them?

You can see an example of some links for Quinn above, the longest being 85 characters in length.

One possibility is we could display links over a certain threshold with the domain followed by ..., ie https://scholar.google.com/....

[wrenaria]

@UcDust Can you hide the https:// and https://www. portion of website addresses? It's unnecessary to show in this context and will help make scanning the list easier for users.

I am also open to the ellipses solution for super long URLs in addition to the above, but would want to be long enough to go beyond .com, .edu, etc. so that if there are multiple URLs from the same domain it's clear that they are not the same link. I'm not sure what the character count threshold for that would be.

[qjhart]

CDL elements allows you to identify your link type with the following types: Blog,Company,,Department,figshare,Google Scholar,Laboratory,LinkedIn,Mendeley,Other,Personal,Portfolio,ResearchGate,RSS Feed,Twitter

These are carried through into the experts document via additional types for the links:

"hasURL": [
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-4",
"@type": [
"URL",
"ucdlib:URL_department"
],
"rank": 4,
"url": "https://library.ucdavis.edu/"
},
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-0",
"@type": "URL",
"rank": 0,
"name": "Evaporation (Meteorology)",
"url": "http://id.worldcat.org/fast/917080"
},
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-5",
"@type": [
"URL",
"ucdlib:URL_company"
],
"rank": 5,
"url": "https://ucdavis.edu/"
},
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-1",
"@type": [
"URL",
"ucdlib:URL_department"
],
"rank": 1,
"name": "Library Bio",
"url": "https://library.ucdavis.edu/person/quinn-hart/"
},
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-2",
"@type": "URL",
"rank": 2,
"url": "https://library.ucdavis.edu/online-strategies"
},
{
"@id": "expert/66356b7eec24c51f01e757af2b27ebb8#vcard-oap-1-web-3",
"@type": [
"URL",
"ucdlib:URL_googlescholar"
],
"rank": 3,
"url": "https://scholar.google.com/citations?user=Od5weOkAAAAJ&hl=en&inst=9081605168663377197"
}
]

I don't believe that there is any checking of the URLS on elements size, eg. I could call https://ucdavis.edu my GoogleScholar Link.

So in addition to the defined type, if we should check the URL makes sense, for example if it doesn't look like a google scholar URL, then change type back to other, we need to have some URL regex testers in place.

[UcDust]

We're going to add custom icons for the following link types (if they exist):

Font Awesome: Google Scholar: https://fontawesome.com/icons/google-scholar?f=brands&s=solid LinkedIn: https://fontawesome.com/icons/linkedin?f=brands&s=solid Mendeley: https://fontawesome.com/icons/mendeley?f=brands&s=solid RSS Feed: https://fontawesome.com/icons/square-rss?f=classic&s=solid ResearchGate: https://fontawesome.com/icons/researchgate?f=brands&s=solid Twitter: https://fontawesome.com/icons/x-twitter?f=brands&s=solid

Academicons has one for figshare