Assistance - ChangeDetection Crashes for a Page with Lots of Text - How to Track?

[XBadFellaX]

I am trying to set up notifications when a new AdGuard version is released. But when I add the link to Change Detection, it crashes, probably because the Adguard Release page is Huge.

Does anyone know how I can just track the top of the page?

Link: https://adguard.com/en/versions/windows/nightly.html

[dgtlmoon]

Wow thats a long page, can you define exactly what you mean by "it crashes" ?

[XBadFellaX]

Just tried to crash it for you.

[XBadFellaX]

The logs are now full with these and changedetection seems to be stuck in a loop checking each link again and again.

2025-03-28 10:43:27.809 | WARNING  | changedetectionio.content_fetchers.playwright:run:210 - Page full Height: 27951px longer than 8000px, using 'stitched screenshot method'.
Playwright console: Watch URL: https://adguard.com/en/versions/windows/nightly.html log: Scanning %ELEMENTS% [<JSHandle preview=Scanning %ELEMENTS%>]
Playwright console: Watch URL: https://adguard.com/en/versions/windows/nightly.html log: Returning 'Possibly in stock' - cant' find any useful matching text [<JSHandle preview=Returning 'Possibly in stock' - cant' find any useful matching text>]
2025-03-28 10:43:29.238 | DEBUG    | changedetectionio.content_fetchers.playwright:run:196 - Time to scrape xpath element data in browser 2.12s
2025-03-28 10:43:29.241 | WARNING  | changedetectionio.content_fetchers.playwright:run:210 - Page full Height: 207525px longer than 8000px, using 'stitched screenshot method'.

[XBadFellaX]

Restarted container but didn't help. Had to recreate container to get it to work and to remove the Adguard link from Changedetection.

[dgtlmoon]

thanks for the super nice debug output

207,525px wow yeah.. hmm

So add the URL by using "Edit > watch"

And then under REQUEST tab, select "Basic fast Plaintext/HTTP Client - May not work with all sites, but is a lot faster than Chrome." then click SAVE

Then add the filter i told you .

tell me if it works

[XBadFellaX]

thanks for the super nice debug output

207,525px wow yeah.. hmm

So add the URL by using "Edit > watch"

And then under REQUEST tab, select "Basic fast Plaintext/HTTP Client - May not work with all sites, but is a lot faster than Chrome." then click SAVE

Then add the filter i told you .

tell me if it works

Just tried that & it works. Thanks so much!

[dgtlmoon]

But in any case, just use the 'plaintext' fetcher and a CSS filter of div.article__content > div > div > :nth-child(-n+3)

to get the first 3 items