Using the CSS Selector Complex Bridge How does one use nth-child when no CSS Selector is present

[miguelarios]

I have an article where the timestamp doesn't have a CSS Selector and is just in a

on its own and the author also is on its on

. The timestamp is the nth-child(2) for this container and I can't select.

[HolgerAusB]

Would be much easier, if we know the article URL.

So we can just wild guessing:

//div[@class='header']/div[2]

while I am not using RSS-Bridge at the moment, so I don't know, if this is in the right format for this bridge.

[miguelarios]

This is the site https://www.dallasnews.com/news/, and this is the CSS selector for the elementsdiv.dmnc_generic-cards-card-module__FcjA3 and for the individual links it is div>h2>a the article itself the lowest CSS selector I could find was div.fusion-app. From this article page I am using the boolean of Load article from page, and extracting each item there mostly bc the timestamp is wrong on the articles when using the more basic one.

[miguelarios]

@HolgerAusB so i just tried it by using //div/section/div/div/p[2] and div/section/div/div/p[2] and while it works in the dev tools to find the timestamp it doesn't work in the rss-bridge. I get this error.

[miguelarios]

The + is that it works in the dev tools search html bar which is how I test my CSS selectors usually.

[miguelarios]

I checked this further and using some AI chatbot I do think my selector of section.dmnc_layouts-content-right-rail-with-gap-module__gezss>div.dmnc_layouts-content-right-rail-with-gap-module__sZ7QP>div.dmnc_generic-article-byline-byline-module__4T6-C>p[2] does work to extract the timestamp of 7:30 AM on Jul 17, 2024 CDT. I added a format string for parsing time as g:i A \o\n M d Y T and it still doesn't work it gives me

Can anyone help?

[HolgerAusB]

Sorry, RSS Bridge using a different format. I am using XPath-1.0 selectors (//div[condition]) with a different product. CSS selectors seem to be a bit more difficult for getting complex parts. I was not able to even get the author from the article page. So I am not sure if the date-selector is throwing back something that could be parsed with your format string.

[HolgerAusB]

OK, my problem was, that I was not using #fusion-app as the selector for article element but one for the core article. But all the following selectors are searched within this part and not in the whole html content, which is, in my opinion, not useful.

After setting #fusion app I could manage to get author with
.dmnc_generic-article-byline-byline-module__4T6-C p span a
spaces instead of >

If You put the following line into the author selector field (just for testing), you get the time string from article there.
.dmnc_generic-article-byline-byline-module__4T6-C p[2]

If you I set this into the Time selector field the bridge raises an error. I did not find a valid format string for parsing date/time. BTW you forgot a comma after d

Maybe @LarsStegman can help here?

[miguelarios]

Hmm i kept so far everything I have above and just changed the time selector to what you just gave me but it still gives me a error 400.

[HolgerAusB]

it's awkward. Everything was fine with that. But testing now, I got the 400 too. Replace the * with + => g:i A \o\n M d, Y+
But as said, the date is not transferred into the feed

[miguelarios]

Try .body-text

Why would this work just curious if you can point me to something educational since I realize this is not an ID in the html itself.

NVM i tested it out. You are using a class used lots of elements. I actually found you can use div[@itemprop="articleBody"] to target this whole box, but I think this selector since it filters out all the other sections you don't get the title or author anymore.

[HolgerAusB]

It is a class name. While you can search for an ID with # you can look-up classnames with the dot. So you can use div.body-text or the short version .body-text That matches also, if the class field has more than one class:
<div class="body-text dmnc_features-article-body-body-module__Ldbl- mb-8 false">

but I think this selector since it filters out all the other sections you don't get the title or author anymore.

Yes, that is exactly why I think, that it is not wise to look-up selectors for author, title, datetime only in the previously selected article-body scope. They should be searched within the whole html content. That should be changed by the bridge maintainer @LarsStegman.

[miguelarios]

@HolgerAusB yup I understand what you mean. I thought of the same thing that the content of the article should be match separately from the other elements.