Linking to textual content fragments in internet pages – Ole Begemann

Textual content fragments are a manner for internet hyperlinks to specify a phrase or phrase a browser ought to spotlight on the vacation spot web page. Google Chrome added assist for them in model 80 (launched in February 2020).

For instance, opening the hyperlink oleb.web/2020/swift-docker-linux/#:~:textual content=working,container in Chrome ought to spotlight the primary heading of the article:

The plain use case is engines like google: while you click on a hyperlink in a search outcome, the browser would robotically spotlight your search time period(s) on the vacation spot web page.

I’ve all the time needed this characteristic. I usually discover myself visiting a web page from a search engine, solely to instantly hit Cmd+F and re-type my search question into the browser’s Discover on Web page textual content discipline. For sure, this can be a ache on cell units and not using a correct keyboard.

However textual content fragments produce other makes use of past engines like google:

1. Linking to a specific sentence or paragraph of a protracted doc. I’d use this on a regular basis when linking to API documentation or discussion board posts. “Regular” URL fragments solely work for anchors the creator of the vacation spot web page created upfront, and readers often can’t see what anchor tags can be found on a web page.

2. Sharing a particular portion of a web page. Browsers might facilitate this by providing to incorporate a textual content fragment within the URL when sharing a hyperlink to a textual content choice.

Right here’s the pattern URL from above as soon as extra:


https://oleb.web/2020/swift-docker-linux/#:~:textual content=working,container


This half is the textual content fragment:


#:~:textual content=working,container


This fragment finds the primary point out of “working” (case-insensitive) on the web page and highlights all the things from that time till it finds “container”. There are a couple of extra variants of the syntax. Learn the textual content fragments draft spec for particulars.

Search phrases might include delicate data that customers don’t need to share with the vacation spot server. For good motive, engines like google stopped reporting the person’s search phrases within the referer header a very long time in the past as a part of the widespread transfer to HTTPS. It might be dangerous if a brand new characteristic reintroduced this outdated knowledge leak.

The spec considers this difficulty. Textual content fragments are designed to be purely a browser-level characteristic — they’re not uncovered to JavaScript code. This screenshot demonstrates that doc.location.hash is clean as a result of Chromium stripped the textual content fragment away:

I believe this the precise conduct, however the spec authors appear to be contemplating altering it again as a result of it could constrain some official use circumstances and since JavaScript can already decide what parts of a web page a person is studying by monitoring the viewport rect. I don’t know — exposing a delicate search time period appears extra invasive to me than the scroll place.

It’s value noting that the privateness concern exists for browsers that don’t assist textual content fragments: they may deal with the gibberish as a traditional URL fragment, which may simply be parsed with a little bit of JavaScript.

As a precaution, engines like google and comparable websites ought to most likely solely embrace textual content fragments of their hyperlinks if the person’s browser helps the characteristic (window.location.fragmentDirective).

Chrome is presently the one browser with textual content fragment assist. From what I’ve learn, the WebKit and Firefox groups are typically supportive of the concept however have some reservations about particular design selections.

I hope this or one thing prefer it turns into broadly supported within the close to future.