Download Google Image search results with free proxies

pip install gimagedownload 

First step

Create a proxy pkl file using [freeproxydownloader](freeproxydownloader · PyPI)

If you have already installed [site2hdd](site2hdd · PyPI) (before 2023-01-24 , you need to update it)

pip install --upgrade site2hdd

or

pip install --upgrade gimagedownload

from site2hdd import download_url_list,get_proxies,download_webpage



xlsxfile,pklfile = get_proxies(

  save_path_proxies_all_filtered='c:\\newfilepath\\myproxiefile\\proxy', #  path doesn't have to exist, it will be created, last 

 # part (proxy) is the name of the file - pkl and xlsx will be added

 # important: There will be 2 files, in this case: c:\\newfilepath\\myproxiefile\\proxy.pkl and c:\\newfilepath\\myproxiefile\\proxy.xlsx



  http_check_timeout=4, # if proxy can't connect within 4 seconds to wikipedia, it is invalid



  threads_httpcheck=50, # threads to check if the http connection is working



  threads_ping=100 ,  # before the http test, there is a ping test to check if the server exists



  silent=False, # show results when a working server has been found



  max_proxies_to_check=20000, # stops the search at 20000

)

CLI

#You can download the pictures using the command line:

python "C:\Users\Gamer\anaconda3\envs\dfdir\Lib\site-packages\gimagedownload\__init__.py" -p c:\newfilepath\myproxiefile\proxy.pkl -s house,elephant,lion -d f:\googleimgdownload -v 3 -t 50 -r 7 -q 9

#Underlines are converted to space

python "C:\Users\Gamer\anaconda3\envs\dfdir\googleimgs.py" -p c:\newfilepath\myproxiefile\proxy.pkl -s brazilian_food,ferrari,Los_Angeles -d f:\googleimgdownload -v 3 -t 50 -r 7 -q 9

CLI Arguments

#arguments: 

-p / --proxy_pickle_file  

The proxy pkl file created with freeproxydownloader

# pip install freeproxydownloader



-s / --search 

Search terms separated by comma: house,elephant,lion





-d / --download_folder 

Download folder, will be created if it doesn't exist 

default: os.path.join(os.getcwd(), "GOOGLE_IMAGE_DOWNLOADS")



-v / --variations 

Grab links from slightly different search terms

dog/DoG/Dog/doG (only uppercase/lowercase variations) 

with different proxies. Don't exaggerate. It might get very slow, but it helps to get more results. 3 or 4 is a good start 

default = 3



-t / --threads 

How many requests threads

default = 50



-r / --requests_timeout 

Timeout in seconds for requests

default = 7



-q / --thread_timeout 

Timeout in seconds for running thread. It should be higher than the timeout for requests to avoid problems. 

default = 9

Import the function

from gimagedownload import start_image_download

start_image_download(

   ProxyPickleFile=r'c:\newfilepath\myproxiefilexxx\proxy.pkl', # pip install freeproxydownloader

   search_terms=['halloween'],

   download_folder=r'f:\googleimgdownload',

   search_variations=3,

   threadlimit=50,

   RequestsTimeout=7,

   ThreadTimeout=9,

)

Results "brazilian_food"

Results "elephant"

Results "ferrari"

Results "house"

Results "lion"

Results "los_angeles"