Guide | Setting up a different (image) search engine

MoosMas

Recently Vivaldi added the ability to change the default image search engine from Google to one you like. However, when I first tried to change it I thought it was confusing and difficult to set up without knowing what data to enter. When I figured it out, I decided to make a guide to help people who are also having trouble setting it up. So here's a step-by-step guide to walk you through the process of setting up your preferred image search engine in Vivaldi, and while we're at it we'll also look at setting up a regular search engine.

Decide which search engine you want. For images I'm assuming you don't want Google (as it's already the default). Other popular engines include Yandex and Bing, so in this example I will be setting up Yandex.
For regular search: Vivaldi already comes with a wide variety of engines, but there are plenty of others available.
Go to vivaldi://settings/search/ and click on the '+' button to create a new search engine.
Fill in the details you already know (name, nickname etc.). The nickname is used to quickly select an engine for your search. I personally prefer the first 1 or 2 letters from the engine's name.
Go to this page. This is a 'database' of search engines, and it has all the details you need to enter next.
Press Ctrl + F and type in the name of the (image) search engine you want to use. In this case it's yandex_com. Then do the following:
If you're setting up a regular search engine (not image search), you can skip [2] and [3]. For image search only [4] is optional.

[1] Copy the URL after "search_url" and paste it in the 'URL' field
[2] Copy the URL after "image_url" and paste it in the 'Image Search URL' field
[3] Copy the text after "image_url_post_params" and paste it in the 'Image Search POST Parameters' field
[4] (Optional) Copy the URL after "suggest_url" and paste it in the 'Suggest URL' field

Note: the URLs and text should be copied without the quotes

Now everything is set up, you will need to set the new search engine as default. Select the new search engine from the dropdown at the top of the page:

[1] Default search engine
[2] Default search engine in Private windows
[3] Default image search engine

Thats it! Now you can use the newly configured search engines.

For regular search engines: click the address bar and type in the nickname of one of your search engines followed by a space. You'll now see the icon of a configured search engine, indicating it will be used for your search. You can type your search query and hit enter, and it will use the selected search engine.

For image search engines: when you right-click on an image, the new search engine will appear in the context menu. Clicking on it will open your image search in a new tab.

If it doesn't show up in the context menu, head over to vivaldi://settings/appearance/, scroll down to the 'Menu Customization' section and select 'Image' from the dropdown. Then, from the 'Commands' list drag 'Search for Image' to anywhere in the 'Content' column. Now it should appear in the context menu.

I hope this guide was helpful! Feel free to ask any further questions in the comments and I'll try to answer them. If you have any suggestions for this guide, feel free to let me know so I can update it.

Edit: Updated guide to show how to set up Yandex instead of Bing, as Bing is already the default. Thanks @Pathduck for the feedback!

Pathduck

@MoosMas Very useful to have a guide on how to do this, as it's something many users struggle with.

However, Bing & Google image searches are already defined, and I think for some locales also Yandex images. I think maybe changing the guide to show how to add a non-default image search engine would be better. Or use maybe Yandex as an example from the example Chromium source data, as Bing is already defined for most users.

From what I know, there's not a lot more reverse image search engines out there that would support this anyway, the only ones I can think of are TinEye and possibly Google Lens. But Lens is "impossible" or at least really really hard to make.

For TinEye I have:
URL: https://tineye.com/search?url=%s
Image search URL: https://tineye.com/search?url={google:imageURL}

Note also: You don't need to define a second "Bing Images" search engine, you can add the Image Search URL directly to the existing Bing search. But again, it already exists for Bing...

MoosMas

@Pathduck Thanks for the detailed reply! I'll edit the guide right away! I thought Bing was one of the defaults, but I wasn't sure so I included it. I know I could've edited the default Bing, but wanted to create a new (although duplicate) one for this guide. I'll update it to show how to set up Yandex instead of Bing as you're totally right. Thanks for the feedback!

PowerfulUser

@MoosMas Hello!
Firstly, apreciation for a greate guide!
I wonder if you could help me find the parameters for the COPYSEEKER image search, since its by far my favorite.
Surprisingly, I couldn't find it in the huge list you provided. Any idea how to proceed?

undachen

@MoosMas Very useful,thanks !

MoosMas

@PowerfulUser Thank you! Apologies for the late reply, but I'm happy to help you out. It took me a while to find the right URL and parameters, but I managed to get it working. Here you go:

Name: Copyseeker (can be whatever you like)
Nickname: cs (can also be whaterver you like)
URL: https://copyseeker.net/
Image Search URL: https://copyseeker.net/discovery
Image Search POST Parameters: imageUrl={google:imageURL}

Here is a screenshot of the settings, just in case you prefer it that way.

Don't forget to set it as your default image search engine in the dropdown at the top of the settings page! Now you can right-click any image and search with Copyseeker.

PowerfulUser

@MoosMas I've been able to get it partially working.
However each time I search anything using the right click menu, it oddly searches an older image I've searched before.

Example:

Searched Image:

Copyseeker Result:

From my observation it doesn't seem to be a visual bug.

After clearing website cache, image search results in a error page:

Odd, don't know what could be happening.

Pathduck

@PowerfulUser That site does funny things with POST requests and JSON and cookies/storage. I think making a custom search engine work will be impossible unless they change how their search works.

Besides, when I saw how chock full of horrible ads that site is I figure I'm not going to touch it anyway. Even Google and Bing images is better than that crap.

PowerfulUser

@Pathduck

Yes, the website seems to be "fiddly" to say the least.

uBlock Origin makes it very clean.

However I still find it being the best image search I've tested by quite a margin, I even found it being highly recommended on a Reddit post about Image Search Engines, so it would still be desirable to use it.

MattSing

Every search result can be accessed using the URL https://copyseeker.net/discovery?id=YOUR_SEARCH_ID, for example, https://copyseeker.net/discovery?id=fat0BBaE8WN3zQjL6LxUu4A5PN68UgqdW4u4o5DLofrNZwgcSY9JjD8jkwYhfsgrnGWxhghCgU4S5ZfMIeQ.

I hope that helps.

Pathduck

@MattSing Hello and Welcome to the Vivaldi Community

The problem is not keeping track of the searches.
The problem is the way the site handles the request for an image url, making it impossible to create a custom search engine.

For instance working image search engines I use take a url as request parameter.
TinEye: https://tineye.com/search?url={google:imageURL}
Yandex: https://yandex.com/images/search?rpt=imageview&url={google:imageURL}
Google Lens: https://lens.google.com/uploadbyurl?url={google:imageURL}
Bing: https://www.bing.com/images/searchbyimage?cbir=sbi&imgurl={google:imageURL}

This site however, POSTs a bunch of JSON and loads the results page from there. Before it's changed so it's possible to send urls directly without first visiting the site, no browser can add it as a custom image search.

MattSing

The URL below works perfectly:
https://copyseeker.net/search?imageurl=https://images.unsplash.com/photo-1557766131-dca3a8acae87

Pathduck

@MattSing Thanks for that, it seems to work fine

URL: https://copyseeker.net/search?imageurl=%s
Image Search URL: https://copyseeker.net/search?imageurl={google:imageURL}

The URL value is really only used if you need to do a search for an image and the Copyseeker engine is not your default, so you can do a cs <image-url> for instance.

Don't know why I didn't just try a GET request from the outset - but then again there was no indication in the code the server would support it, and there's no reference to the /search endpoint in the origin page so I didn't even think in that direction - guess I could've just done a blind guess though

astero

@PowerfulUser said in Guide | Setting up a different (image) search engine:

However I still find it being the best image search I've tested by quite a margin

very true. CopySeeker is the best image reverse search I've tried so far. The rest are junk.
For anybody like me who is not satisfied with the current Google Lens BS, here is the working URL for classic-like image search they had before:

https://www.google.com/searchbyimage?image_url=<your_url>&client=app

PowerfulUser

@astero hey, appreciate the feedback. still think its the best one also!