Translator failure on LinkedIn

Rintze

URL: https://www.linkedin.com/posts/andy-gombert-20077935_saccharomyces-cerevisiae-strains-used-industrially-activity-6813171015716286464-Tuy8/
Debug ID: D1283566853
Popup: "An error occurred while saving this item. See Troubleshooting Translator Issues for more information."
Translator: "Save to Zotero (DOI)" (folder icon)

dstillman

I'm not seeing a DOI on that page at all. Where on the page is that showing up for you? Maybe the logged-in page is different, or some browser extension you have is adding it?

The DOI is being detected as "10.1042/EBC20200160/229078/Saccharomyces-cerevisiae-strains-used-industrially#.YNIrcFfdiF4.linkedin", when the actual DOI only goes through "078". That's what's causing the error.

Rintze

In Chrome, in a session where I'm not logged into LinkedIn, I get:

<a class="mini-card__title-link " data-tracking-control-name="public_post-content_share-article_title" data-tracking-will-navigate="" href="https://www.linkedin.com/redir/redirect?url=https%3A%2F%2Fportlandpress%2Ecom%2Fessaysbiochem%2Farticle%2Fdoi%2F10%2E1042%2FEBC20200160%2F229078%2FSaccharomyces-cerevisiae-strains-used-industrially%23%2EYNIrcFfdiF4%2Elinkedin&amp;urlhash=cuEQ&amp;trk=public_post-content_share-article_title">

Saccharomyces cerevisiae strains used industrially for bioethanol production

</a>

In Firefox, in a session where I'm logged in (and for which I reported the error and debug ID), I get:

<a class="app-aware-link feed-shared-article__meta flex-grow-1 full-width tap-target" target="_blank" href="https://portlandpress.com/essaysbiochem/article/doi/10.1042/EBC20200160/229078/Saccharomyces-cerevisiae-strains-used-industrially#.YNIrcFfdiF4.linkedin">
<div class="display-flex">
<div class="display-flex flex-column flex-grow-1">
<div>
<h2 class="feed-shared-article__title break-words t-14 t-bold t-black">
<span dir="ltr">Saccharomyces cerevisiae strains used industrially for bioethanol production</span>
</h2>

<h3 class="feed-shared-article__subtitle t-12 t-normal t-black--light">
portlandpress.com • 2 min read
</h3>
</div>
<!----> </div>
<!----> </div>
</a>

dstillman

Yeah, there's not really much we can do about that. I guess we could consider progressively deleting slash-or-hash-delimited components from the end of a DOI within a URL after it fails to resolve and retrying, but that would be a pretty messy and inefficient solution.

Part of the reason DOI is always a folder icon is that we can't guarantee in advance that something that looks like a DOI is actually a valid DOI, or that it will represent the intended item if so. Assuming you want to save the original article here, you'd really want to just click through to the publisher site, which is fully supported.