The Chrome Extension API documentation for topSites is not informative about how Chrome calculates a "topSite." Even with a clean profile, it's unknown how the mechanism work.
I'm looking for ways to predict whether or not a given site match the criteria to be considered a "top site."
I also tried looking at https://cs.chromium.org as Chromium also has this API built-in, but no luck finding relevant info.
Related
Greetings fellow developers,
I would like to ask for help regarding the following problem: Is there a way to request removal of stored website data from search engines? Most of the links that show up when searching my domain are old and non-existent.
What I've found from personal research regarding this question/problem:
From my personal research I have found that removal requests can be made individually to the well-known search engines such as Google, Yahoo and Bing, but this is not what I am looking for, since I am well-aware that it would take a lot of time for the requests to be processed and the removal of the data to be done. Also, I wasn't able to find this "removal-request" webpage for the other search engines.
To be more precise/clear...
... I want to request this website-data-removal to all (most) search engines at once, so that when I upload my new website (to the same domain), working and functional links (URLs) would be displayed. Can this be anyhow achieved and, if so, how? Also, how much time would it take for this removal to be finished?
Hope my question is clear enough, and any answer/help would be very much appreciated.
No, there is not a way to do this for all search engines at once. You will have to request it from each site individually. As for the smaller search engines you can try and find any contact information or customer support however their is a chance they will ignore your request (heck, some sites ignore the robot.txt file and just search your site anyways... it's just a part of being on the web).
I am trying to develop a contextual gadget, but not getting any documentation for it. Google provides a document which is very old and not updated from long time. Process explained in that documentation to develop a gadget is deprecated.
Please if anyone have solution, help me.
You may want to check full documentation in Gmail Contextual Gadgets which was last updated June 29, 2016.
To develop a Gmail Contextual Gadgets, you may want to first check the given implementation parts also discussed in the documentation. Then, you may go through these summary of steps:
Use JQuery, or write JavaScript that conforms to ECMAScript 5 Strict Mode.
Note: You need to be using the correct development frameworks to provide an extra layer of protection between your gadget's potential vulnerabilities and your end users. To find out why, see Using the right frameworks for security.
Choose one or more pre-canned extractors. This determines which type of content will trigger your gadget.
Write a manifest for the gadget.
Write the gadget spec. This determines what the gadget will do when it is triggered.
Publish the gadget spec to a location which is accessible on the public Internet. An intranet will not work. Your hard drive will not work. (Why? Google's servers need to download the gadget. If they can't reach it, then Gmail can't display it.)
Install the gadget.
Test the gadget by sending yourself some email. The gadget should appear in Gmail whenever you read an email that contains the right sort of content. For more tips on testing gadgets, see Publishing Your Gadget in the gadgets API site.
It will really help if you go through the documentation as there are best practices, limitations and important details that you should note.
This related SO post might also help.
Say, I want to develop an alternative spellcheck module to google docs.
This means that I have to get corrections from my backend, and color the misspelled text's background, and do a small popup bubble when user hovers over it, where I'd display the correction. (please mind that spellcheck is not the actual goal of my project, but it does address my problems in a more simplified way)
What are my options? Any ideas how to do this?
Few possible solutions I came up with:
Chrome extension vs Apps script
Chrome extension
pros: user has to grant permissions once, can freely traverse and append anything to dom via content script
cons: is a "hacky" way, if google changes classnames or js source, it would stop working, and also, reverse engineering google docs's editor engine is impossible
Apps script
pros: supported by google: if it works, I dont need to be afraid of docs updates
cons: it seems to me that I can't just fiddle with the dom (because of Caja compiler), has very limited support (if any) for custom highlighting or hover functionality.
As I see it, neither of these are perfect solutions for this project. What do you think? Any suggestions are very welcomed.
I know this is an old question, but I have recently gotten into the same problem, and believe I have a solution. So for future Googler's I will post my answer here.
My solution was to create a Chrome Extension and understand how the Google Docs DOM's are structures to interact with it.
You can find my code to work with the Google Doc DOM's here
In Apps Script you can't "fiddle" with the DOM and you won't be able to implement hover functionality. Also, a lame Highlighting would involve changing the current document itself (which would go to revision history, undo queue, etc)
Therefore, your only altertive is the Chrome Extension. But I agree with you on the cons. It is a super hard task that could break at any minute without notice.
I'm making an app and I plan to have some cloud happening with it, but I do not want to create a user data base and have the users need to remember their username and password.
Since it will be distributed through the chrome app store it's basically guaranteed that the user will have a google account. All I want to do is:
Get the user's email through the google account stuff. If I get it through there, well then their email is all the authentication I need to get that user's data.
If I end up putting the application on something other than chrome browser, I'll just have the user use their email to request a validation link, and then I'll send them a validation code for that account, they put the validation code into the application, it takes that as a verified user, so that's secure and easy as well.
EDIT: I'm looking into this. So far I have:
OAuth 2
Google API
But I have a problem that I don't know what to set as my javascript origins in the Google API and there isn't too much info on this abroad. If anyone can tell me what javascript origins I need to set for a chrome extension to access google api it would be a great help.
PS: Thanks for down vote, this is why I love resorting to stack exchange.
Hmmm, I think the only reason this was voted down is the fact that this question may be been asked somewhere on the site already (but I'll help you and give give a 1up).
So what you are wanting to use the Google OpenID. You will have to register your application with Google so they can provide OAuth2 tokens for you application. I have not done this with Google but with other services and it is pretty easy, just search around.
In terms of obtaining OAuth2 for your application in the chrome extension - this can be a pain since the extension is sandboxed and Google's example uses OAuth not OAuth2.
Here is solution I host on GitHub for this - I also use this in my extension GitHub Repositories:
https://github.com/jjNford/oauth2-chrome-extension
Hope this helps in some way. Don't get discourages with StackOverflow, it is a great resource with many great contributors.
Good luck!
I had to up vote you too as I'm tracking down a related issue so here is what I've found that may help.
According to these directions - http://code.google.com/p/google-api-javascript-client/wiki/Authentication - "In the "Authorized JavaScript Origins" box, enter the protocol and domain for your site." This should be the protocol (http:// or https://) followed by any optional subdomain followed by your domain name and no trailing slash. Nothing after the domain name.
This prevents certain kind of security attacks, see: http://en.wikipedia.org/wiki/Same_origin_policy.
There are some related questions here that I found that may help:
Problems with Google Picker API and selecting Google Drive items and google apis console 'Javascript origins'.
Now with all of that said, I am still trying to track down what values to put in there for one of my sites hosted as a Google Site, as none of the obvious values are working for me. So there may be some subtlety there that I have missed in this explanation.
I want to write a mobile app which takes a picture and searches google images for similar pictures and then displays the results.
However, with google image search I can only search for text strings, and with the search API it seems there's no possibility to search for similar pictures; this feature seems to be available only through the web interface.
Any idea how I can solve this problem?
thanks,
Christoph
There is a way you can do this now, but its not officially supported, and there are probably some restrictions on the number of queries you can perform. Update
http://images.google.com/searchbyimage?hl=en&biw=1060&bih=766&gbv=2&site=search&image_url={{URL To your image}}&sa=X&ei=H6RaTtb5JcTeiALlmPi2CQ&ved=0CDsQ9Q8
There is also a google image search API, which is being officially deprecated, but it will work for now.
http://code.google.com/apis/imagesearch/
The Google Vision API.
https://cloud.google.com/vision/
This is very simple and easy and powerful.
I had done something like that recently for a mobile app, this is the code for it, it uses google search by image feature, and returns the "best guess" or the whole page
you could use that and modify it to do what you want, but once you get the best guess of the image you could search for any image with that title, etc
https://github.com/hbattat/search-by-image
I don't think it's possible. If you click the link to find similar images from the images result page you get a link with the original query included:
google.com/images?q=ORIGINAL_QUERY&imgtype=i_similar&sa=...
If you remove that GET param manually, the search does not work, it only shows the images search form.
I dont think it is possible to find similar images with google if you do not know what's on it.
I was looking for an answer to this some time ago, and found tineye. You have to pay for it, though. Currently (Jan 2012) USD300 for 5K searches, USD1.5K for 30K searches...
SerpAPI enables to search through Google Images and returns a clean JSON.
URL example:
https://serpapi.com/search.json?q=Apple&tbm=isch&ijn=0
Documentation:
https://serpapi.com/images-results
This service is integrated with most of the programming languages: python, php, java, golang, nodejs...
Google limit the number of search per day. but this service provides unlimited searches...