can not parse image from <picture> tag

[DophinL]

fetch('https://r.jina.ai/https://www.cnbc.com/2024/10/29/sam-altmans-reddit-stake-now-worth-over-1-billion-after-earnings-pop.html', {
  method: 'GET',
  headers: {
    "Accept": "application/json",
    "X-With-Images-Summary": "true",
      "X-Target-Selector": ".ArticleBody-articleBody"
  },
})

You can see from the execution result that the main image has not been retrieved.

Please support the tag. I'm a paid user, thanks 🫡

[nomagick]

Thanks for reporting.

I have just updated the service to expose the image in <picture>.

[DophinL]

Thanks for reporting.

I have just updated the service to expose the image in <picture>. Amazing speed, thank you so much!

[DophinL]

@nomagick seems not work on this site, didn't get the main picture:

fetch('https://r.jina.ai/https://www.foxsports.com/stories/mlb/chris-sale-aaron-judge-shohei-ohtani-highlight-all-mlb-2024-awardees', {
  method: 'GET',
  headers: {
    "Accept": "application/json",
    "X-With-Images-Summary": "true",
      "X-Target-Selector": ".story-content-main"
  },
}).then(res => res.json()).then(res => console.log('jina resp',res))

[DophinL]

This website is in the same situation:

fetch('https://r.jina.ai/https://cryptodaily.co.uk/2024/11/crypto-price-analysis-11-14-bitcoin-btc-ethereum-eth-solana-sol-ripple-xrp-dogecoin-doge-uniswap-uni-immutable-imx', {
  method: 'GET',
  headers: {
    "Accept": "application/json",
    "X-With-Images-Summary": "true",
      "X-Target-Selector": ".news-details-layout1"
  },
}).then(res => res.json()).then(res => console.log('jina resp',res))

[nomagick]

This was because it only has a srcset and does not have an src.

Now I've modified Reader a bit to consider srcset.

[DophinL]

This was because it only has a srcset and does not have an src.

Now I've modified Reader a bit to consider srcset.

Thanks for the quick response! It seems it hasn't taken effect yet. Has it not been released?

[nomagick]

This was because it only has a srcset and does not have an src. Now I've modified Reader a bit to consider srcset.

Thanks for the quick response! It seems it hasn't taken effect yet. Has it not been released?

It had been deployed days ago. The image did not show up for this particular page because the Readability lib we use intentionally removed the block while trying to clean up the page. Try the markdown return format to inform Reader not to use Readability:

curl https://r.jina.ai/https://www.foxsports.com/stories/mlb/chris-sale-aaron-judge-shohei-ohtani-highlight-all-mlb-2024-awardees \
  -H 'x-return-format: markdown' \
  -H 'x-target-selector: .story-content-main'