Greasy Fork

I don't know why or how but right after you fixed it, it stopped working again today. Damn google...

Can you send/upload a .html save of the page you're getting? I'm not seeing which ever changes made it break again for you :/

I will, sorry for the late reply, it seems my gmail notifier has failed also lately, haha. Hmm not sure how to send the .html save, I will try to upload it somewhere.
https://ufile.io/o2z8v76b is this what you want?

Can you send/upload a .html save of the page you're getting?
I'm not seeing which ever changes made it break again for you :/

Sorry, I messed up and uploaded only the htm.file, I guess you want the folder created also. Let me reupload. https://ufile.io/az8vphqi

No, the first one was fine. (In fact, saving as "HTML Only" may sometimes be more useful, that would let me see where all the links were originally pointing. Saving a "complete" doc replaces most references to local ones)

Sadly, it looks like bad news.
All the top images are data URLs (instead of linking to a source image file, the image data is encoded in text form directly in the URL, usually as base64 PNGs, like: src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUg[ ...lots more characters ]AAAABJRU5ErkJggg==".
They now sit in an anchor tag without reference to the original link (used to be that they provided a reference in the internal link like /imgres?q=test&imgurl=https%3A%2F%2Fwww.link-to-image.file&etc...)
This means that we no longer have a direct way of getting to the actual image for these.
I can try making a v3 that attempts to load Google's details sidebar thing and/or open the target site in a new tab and find the image there and then present it to the user. (At this point, with lazy loading and bot protection, I have little faith in just fetching the target site being successful more than 75% of the time -- if the source page isn't just an ever changing ephemeral feed anyway, with no hope of finding the image anymore... but that's another story)

I'm not patient/skilled/insane enough to try and hook into what ever Google has going on in their obfuscated scripts and hundreds of event handlers probably leading to functions with security features to get the larger image for the details view, so if they decide to make smaller versions of larger images or cached images in general that don't point to the original source, it's curtains for us. Right now, I think the sidebar thing is our only hope.
It'll be messy and slower, and it'll take me some time to make it work well.
😩 Ugh, we can't have nice things anymore 😔

I'll try to get on it tomorrow. Wish me luck 🤞

Good luck man, damn they are cruel for doing all this nonsense. I got a new gmail notifier at least, immediate reply. But yeah, screw google. Do your best.

¡Vaya, Benjamin! 😅 Google siempre encuentra nuevas formas de complicar las cosas, ¿eh?
La idea de una versión v3 que cargue el panel lateral suena prometedora, aunque sí, probablemente será más lenta y difícil de mantener estable.

Por cierto, mientras probaba algunos scripts similares, encontré un artículo interesante sobre cómo optimizar extensiones y herramientas para búsquedas visuales — lo vi en https://spin-mama-es.com/. Tal vez te dé algunas ideas para mejorar el rendimiento o el manejo de URLs dinámicas.

¿Planeas mantener la extensión pública en GitHub cuando termines la nueva versión? Sería genial seguir el progreso y probar las actualizaciones.

I really wish Google used the system Duck Duck Go has for image searching, it's really more sensible.

@613 oh man, so sorry. Haven't made much progress, but I haven't forgotten. Meanwhile, depending on how you feel about a Russian service: I've also had good experiences using Yandex image search. They have a button with image size and direct link already on there: https://yandex.ru/images/search?text=test

@maurita Oh, I completely forgot about GitHub 😬 I'll look into that once I have this working. The site is currently giving me a 502 error.

I know about Yandex, no problem using it, I had never really tried their image search by itself, but it is worth a try. My only concern is due to sanctions and bans, they might not show as many results as other engines. But yeah, we don't get a ton of chinese site results on google for example either, right? Thanks again, I'll be patient in case you come up with something.