maksympylypenko/google-scraper

scapes google search (web & photo)

Google search scraping in Electron

Simple image & web scraper by using regex expressions

Setup

Install NodeJS then :

npm install
npm run start

Image scraper

Looking for low & original resolutions of the image.

/\["https:\/\/(.*)",([1-9][0-9]*),([1-9][0-9]*)]\s,\["https:\/\/(.*)",([1-9][0-9]*),([1-9][0-9]*)]/gm

e.g:

["https://encrypted-tbn0.gstatic.com/images?q\u003dtbn:ANd9GcQntO1ba63A1B1v0r6ilfBDcnfArFpIRcmyYg\u0026usqp\u003dCAU",183,275]
,["https://i.cbc.ca/1.5448015.1580501469!/fileImage/httpImage/cat-in-pain-facial-expression-scale.JPG",853,1280]

Scrolling down increases the number of images (in the DOM), however they are staying in low resolution. The original link is revealed (unencrypted?) when the user clicks on the image.

Web scraper

Looking for URL, Title and Description.

/<div class="yuRUbf"><a href="(.*?)"|<h3 class="LC20lb DKV0Md"><span>(.*?)<\/span>|<span class="aCOpRe"><span>(.*?)<\/span>/gm

Note, that some descriptions have their date of publication included in another <span>. That is why I used 1 more group:

<span class="aCOpRe"><span class="f">(.*?)<\/span><span>(.*?)<\/span>

Other issues

• Some links to original images are deleted.
• Google sometimes crops very large images to 1x1 ratio. This may break a full screen view, because the gallery requires original width & height during the initialization.