Download the photos from a news page
Paste a news URL and the scanner reads the page's HTML and lists the photography it carries, at every size the site declares. It reads the markup rather than the viewport, so pictures far below the fold are found without scrolling. One thing to know before you start: news photography is usually licensed from a wire service, and finding a file is not permission to use it.
What this does
- Measured on a live news SECTION FRONT — the index page listing the day's stories, not an individual article — where the scanner found 100% of the images a real browser loads, checked against headless-Chromium ground truth. Article pages were not separately measured, and they carry larger photos than a front page does.
- A news site serves the same photo at a dozen widths. They arrive as one tile standing for the largest, with the rest behind a chip, so the grid shows pictures rather than breakpoints.
- Pictures far below the fold are found without scrolling, because the lazy-loading attributes are read straight from the markup instead of waiting for something to enter a viewport.
- Arriving from this page the grid is filtered to the page's own photography, so the masthead logo, the favicons and the social preview tags are not in the way.
What a news page holds
A news page is unusually dense in responsive imagery. A single photograph is typically declared at eight or more widths so browsers can choose, and the page then defers most of them until you scroll. Both of those defeat a naive reader: one turns a page of forty pictures into three hundred entries, and the other hides everything below the first screen.
Reading the markup rather than the rendered viewport handles the second, and grouping by logical image handles the first. What arrives is a grid of photographs at the count a person would count, each standing for its largest available version.
About the licensing
Most photographs on a news site were licensed by the publisher from an agency, for that publisher’s use. That has no bearing on whether a file downloads — it bears entirely on what you may do with it afterwards. The credit line beside a photo names the agency, and it is the only reliable signal on the page. This tool does not read it, which is exactly why it is worth reading yourself.
What it won't do
- News photography is usually licensed from a wire service — Getty, Reuters, AP. Finding the file is not permission to use it, and the credit line beside the photo is the thing to read before you republish anything.
- A section front carries grid-sized crops; an article page carries the full-size photo. If you want the large version, paste the article URL rather than the front.
- The grid arrives filtered to page images. The masthead logo, the site's icons and its social preview image are one click away in the source filter rather than missing.
- A paywalled or consent-gated article may serve a teaser or an interstitial instead of the story. The scan succeeds, returns very little, and that is the wall answering rather than the parser failing.
- Captions and photo credits are not captured. You get the files and their filenames; the credit lives in the page's text, which this does not read.
- If the site's robots.txt asks automated tools not to read the page, the scan stops and says so. There is no override.
Questions
Section front or article page?
The article, if you want the full-size photograph — a front page carries the crop sized for its grid, which is often a fraction of the width. The front is better for a sweep of the day's imagery at thumbnail size. The 100% measurement on this page was taken on a section front.
Why does one photograph show up as "8 versions"?
News sites generate a set of widths for every picture so browsers can pick one. All of them are found and grouped into a single tile that stands for the largest. Open the chip if you want a specific width.
Can I get the photo credit or the caption?
No. This returns image files and their filenames, not the page's text, so the credit and caption are not part of what it extracts. On a wire photo the credit is the thing that determines whether you can use it at all, so read it on the page before downloading.
Why did a paywalled article return almost nothing?
Because the HTML served to us was the teaser or the consent screen rather than the article. Nothing failed — the page genuinely contained few images. A browser-based tool signed into the same paywall would see more; this reads what an anonymous request is given.
Related
Images belong to their creators. Only download what you have the right to use. See all tools or read what our requests do.