It provides tools to grab images from a google search by extracting the links of the images and downloading original images.
This module is written for windows 10 or Ubuntu 16.0 and above on 64-bit processor. It uses Selenium to open browser so as to scroll down to get more images than possible otherwise. Thus it needs a browser to work correctly. This is on default set to use chrome browser in case of inability to open it Firefox will be used. The package comes with chromedriver and geckodriver with it.
To install gImageGrabber do as follow:
$ pip install gImageGrabber
There are two python files imgScrape and imgTools.
imgScrape has all the utilities needed to run the script but if you want to have additional control over the functions you could explore imgTools.
To import this module to your script do this :
from gimagegrabber import imgScrape
from gimagegrabber import imgTools
imgScrape.build_url(search)
This is to compose a google search URL for your search term. To specify your search term use search argument of the function to build the URL.
Usage :
from gimagegrabber import imgScrape
searchTerm = "kamikaze eminem"
url = imgScrape.build_url(searchTerm)
print(url) #FOR DEBUG PURPOSE
imgScrape.browser(url, test=False)
This to start a browser windows and scroll down the webpage to let more pictures load.It returns a raw source code data of the webpage encoded in utf-8 format. It takes 2 arguments url and test .
- url is the url of the page it needs to open.
- test is to make the browser scroll down less thus taking less time to return the source code. This is useful when you are writing or debugging something in your script.
It uses Chrome or Firefox to work so make sure you have Google chrome or Firefox installed at their default directory.
Sometimes you might need to click on show more images on webpage to load even more images
If you cant seem to open browser make sure you are on 64-bit OS and that you have chrome or Firefox installed.
If you are on 32-bit processor you need to use Firefox and you also have to download 32 bit driver from here and replace it with the already present geckodriver.exe saved in driver folder of the gImageGrabber Module folder.
Usage :
from gimagegrabber import imgScrape
searchTerm = "kamikaze eminem"
url = imgScrape.build_url(searchTerm)
raw_data = imgScrape.browser(url)
print(raw_data) #FOR DEBUG PURPOSE
imgScrape.imageLink(html)
This extracts the original link of the images from the html(source code) provided. html is the source code of the google image search page. It returns a dict with format [ link : file extension ] . If you want it in [file extension : link] you can use imgTools.invDict()` function from imgTools
Usage :
from gimagegrabber import imgScrape
from gimagegrabber import imgTools
searchTerm = "kamikaze eminem"
debug = False
url = imgScrape.build_url(searchTerm)
raw_data = imgScrape.browser(url,debug)
links = imgScrape.imageLink(raw_data)
print(links) #FOR DEBUG PURPOSE
print(imgTools,invDict(links)) #FOR DEBUG PURPOSE
imgScrape.saveImages(data, name, onlyType)`
This saves all the images given to it in a list of format [link: file extension].
It takes 3 arguments:
- data This is to provide dictionary containing links to images in format [link: file extension].
- name This is to provide the name for the folder under which images will be saved.
- onlyType If you want only a particular file extension then use this mention that. If not, then pass it a empty string or just don't use that argument.
The format in which it saves images is
Root folder
|-- Search Term
|-- file extension(eg 'jpg')
|-- 000001.jpg
|-- 000002.jpg
Usage :
from gimagegrabber import imgScrape
searchTerm = "Kamikaze"
extension = '' #save all types of images
url = imgScrape.build_url(searchTerm)
raw_data = imgScrape.browser(url)
links = imgScrape.imageLink(raw_data)
imgScrape.saveImages(links,searchTerm,extension)
This code is included in the package as simpleScript.py.
from imggrabber import imgScrape
# Search term
search = 'kamikaze eminem'
fType = '' # if you want all the files them make it empty string
debug = False
html = imgScrape.browser(imgScrape.build_url(search), debug)
data = imgScrape.imageLink(html)
imgScrape.saveImages(data, search, fType)
Saksham Sharma