-
Notifications
You must be signed in to change notification settings - Fork 11
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
GoogleDWIMages breaks #17
Comments
I can't reproduce this (yes, it's possible they're A/B testing or still rolling this out). In order to investigate, I'll need more details (this applies to any issue report), e.g.:
My guess is that it's a conflict with another addon (e.g.), so I'd rule that out first. |
Am not sure what you mean by platform, so I am going to assume Google? So based on this, we know it's not a conflict with another script and/or add-on. EDIT: Would you like a copy of the HTML, in case it is A/B testing that's affecting both my logged in and incognito mode? |
I've changed it to Operating System.
Any relevant settings in Firefox, Tampermonkey, or the script.
Sure, you can email it to me at the address in my profile. Just incognito, though. Please don't send me anything which could expose your account details. |
Ah okay. Windows 10 Education Edition version 1709 (incapable of updating further due to functionality issues.) I would just list all the Firefox setting changes I made, but there's quite a bit about:configs that are different and it would be very difficult to list them all. I will email you the incognito HTML in a bit then. I will have everything disabled in order for you to apply the script and see what's wrong. EDIT: The file has been sent. |
Thanks, got it. Try this. |
Hey nice work! Looks like it's working fine! If I may ask, what changed in the A/B testing to break the script so... selectively? I am still a programmer, just not familiar with javascript. Would love to expand my knowledge a bit, hence why I'm asking. |
The metadata appears to have switched from (simplified): Mine[
{ type: "image", index: 0, url: "https://example.com/1.jpg" },
{ type: "relatedSearches", links: [...] }, // no index, but consumes 1
{ type: "image", index: 2, url: "https://example.com/2.jpg" },
] to: Yours[
{ type: "image", index: 0, url: "https://example.com/1.jpg" },
{ type: "advert", data: [...] }, // no index, and doesn't consume 1
{ type: "image", index: 1, url: "https://example.com/2.jpg" },
] - which means an array can no longer be used to look up the metadata for an image based on its index. I switched to a Map keyed on the As for why your metadata/results are different than mine: I'm not sure. It looks like there are some advertising-related nodes in your metadata that aren't in mine. I can't tell from the HTML and it might not be visible if you're using an ad-blocker. New advertising widgets, and possibly distinguishing the metadata that populates them from the metadata for real results, seems like the sort of thing an advertising company might experiment with, particularly when searching for something like "lawn mower". |
That's curious. I gave you the complete HTML with all addons and userscripts disabled, so I am unsure as to why this is the case. I wanted the raw to be as clean as possible. If there was some way to make it cleaner, I'm all ears on that one so we may better find the root of the cause. Curiously, uBlock does block the elements on both local and online, so if you're using an adblocker yourself, this may be why you don't see it. I mention this specifically as I can see the ads myself on my copy of the local but only with uBlock disabled. None the less, I appreciate you heavily for fixing it! Thanks for keeping this up to date! |
Heyo! So, sorry to bother, but I was doing some image searching yesterday with this script installed. Something appears to have changed on google's end as I'm getting this wonderful error in console:
Hopefully this isn't a matter of google slowly pushing out the update and it's applied to everyone, but essentially due to this error, what happens is that the images it couldn't process are not added to the list. So, the corresponding direct links to images steps back. Meaning after the first failure, the link to the image is for the previous image behind it. This worsens with every error it encounters. You can see it yourself by searching "Lawn Mower" and clicking the one with the granny pushing the mower. That one will be an image that failed to link and everything after links to the wrong image.
The text was updated successfully, but these errors were encountered: