As title says I'm trying to create a custom search engine (using Google's "Custom search") and was wondering if I can use google's image search within my custom one. For instance i have one website allocated to my custom engine "www.topman.com" and would want a user to be able to upload an image of a tshirt and find a link within topman that has a similiar styled tshirt. Thankyou in advance for any help.
No, Google Custom Search does not have "search by image" functionality.
Related
I could set custom picture and caption for link. Now it is impossible for v2.9. Is there some kind of workaround without specifying og: tags.
When using Facebook Graph v2.8, you could post a link and display a custom image and not the "og:image".
This was done by adding the picture parameter to the hook.
https://graph.facebook.com/v2.9/me/feed?link=https%3A%2F%2Fwww.google.com%2F&picture=https%3A%2F%2Fi.ytimg.com%2Fvi%2FtntOCGkgt98%2Fmaxresdefault.jpg&name=test&message=this%20is%20a%20test&access_token=**********
The graph API is at 2.9 and as per the documentations this is still supposed to work Facebook API doc
However, it keeps getting overridden by the scraped image. Is anyone else experiencing this problem or have a workaround.? Facebook API Explorer
picture, name, caption, thumbnail, description are all set to be deprecated.
https://developers.facebook.com/docs/apps/changelog#v2_9_90_day_deprecations
You can go ahead and use Open Graph for posting images with custom description and title.
You can get an insight of the same from this link -> https://developers.facebook.com/docs/sharing/webmasters?hc_location=ufi
I want to create a custom google search for my site. My requirements are as follows:
I need a text box for search and two radio buttons(one for searching my site and other for searching web).
Depending on the radio button selected the results should be displayed.
I have a separate search results page ('/search').
I know how to create a simple google custom search. But this requirement of providing an option to search the site or web is a little tricky for me.
Please help
I was able to solve this issue.
I first created a search engine in https://cse.google.com/cse/ for my site and configured it to search the web in Basics tab-> Sites to Search option.
With this the search engine would search the entire web.
The code that I placed on the search results page had something this:
"gcse:searchresults-only as_sitesearch='site to search'"
This is a part of the code that google gives you.
Here the parameter 'as_sitesearch' will control which site is to be searched. When the radio button of Searching the web was selected I did not pass any value to 'as_sitesearch' and when the radio button of Searching my site was selected I passed the site url to 'as_sitesearch'.
This worked for me
I would like to use google's 'search by image' feature to search a defined set of images (rather than the entire web) for matches. Has anyone done this previously or do they know if it is possible?
Yes, this is possible.
If you enter the image url, the search will be performed directly, however afterwards you can just type in additional keywords, or for example a domain.
If you want to search in a set of images that typically cannot be found by google, then it will probably not be possible.
I'm just beginning with programming, but i wanted to know if it's possible to use google docs api to make documents on another site using the google docs text editor?
Is there some sort of way i can put the google docs text editor onto a website so that we can use that for document creation instead of tiny mce?
Basically the functionality needed would be documents created, openly shared, a postable version of it (take html code) -- so it can go on the document display page, and
Of course there would be google login and everything, but i just wanted to see if this would work.
No, that is not possible, sorry.
Similar to Facebook's UI, I am attempting at generating a preview image from an external linked website. So that when a user types in a url he is linking, the UI will by default, scan that site for an img and scrape a preview thumb.
Is there a specific name for this technique? Or can anyone point me in the direction of learning this?
Thanks so much!
Its called scraping. There is a library called scrAPI.
Here is a code example http://crunchlife.com/articles/2007/08/13/code-snippet-ruby-image-scraper
There are a couple different options when it comes to page scraping. Another one to check out would be nokogiri, http://nokogiri.org/. You can find tutorials on how to use it at http://nokogiri.org/tutorials.
Instead of grabbing an image from the site, why not grab the image of the entire page? You could make use of a free screenshot service like http://www.websnapr.com/ or http://www.thumbshots.com/ among others. In one application, I use that for my preview image, and use nokogiri to scrape the page title and description. Just an idea.