Add Frieze.js #1259
Add Frieze.js #1259
Conversation
ready for review
|
@owcz -- much preferred to get item type right in `detectWeb`. Among other things, otherwise tests fail.
|
|
||
|
|
||
| function detectWeb(doc, url) { | ||
| if (url.includes("/article/") { // does not handle /event/ or /media/ pages, which EM alone can handle |
There was a problem hiding this comment.
For those /event and media pages, you should be able to exclude them from the EM fixes you make in scrape, right? That way they'd work correctly when imported as part of search results (currently that'd e.g. all turn them into blogposts)
There was a problem hiding this comment.
The event/media pages are basically listings, so I didn't think anyone would want to cite them like we wouldn't handle classified ads in a newspaper (hence why I excluded them from the multis)
Let me know if you'd still want them
| let href = rows[i].href; | ||
| let title = ZU.trimInternal(rows[i].textContent); | ||
| if (!href || !title) continue; | ||
| if (/\/(event|media)\//.test(href)) continue; // scrap items that link to /event/ or /media/ pages |
There was a problem hiding this comment.
if what I say above is right, you could then remove this restriction.
|
Thanks! |
Art magazine https://frieze.com/