imageDownload.py

# João Tinti's script, https://github.com/joaotinti75/Pokemon_Classification/blob/master/google_imgs_downloader_v2.py
# adicionado apenas removedor de arquivos
import requests
import os
from PIL import Image
from bs4 import BeautifulSoup

def clearImages(directory):
	img_list = []
	width_list = []
	height_list = []
	count = 0
	height = int(input('Min height to remove: '))
	width = int(input('Min width to remove: '))
	print('Cleaning images...\n')

	for image in os.listdir(directory):
		img_list.append(image)
		width_list.append(Image.open(directory+image).width)
		height_list.append(Image.open(directory+image).height)
	print('Images found\n')

	for x in width_list:
		if x < width:
			os.remove(directory+img_list[count])
		count += 1
	count = 0
	print('Widht found\n')

	for x in height_list:
		try:
			if x < height:
				os.remove(directory+img_list[count])
			count += 1
		except FileNotFoundError:
			pass
	count = 0
	print('height found\n')
	print('All done\n')

def googleImagesDownload(directory, search, num_of_img):

	search_mod = search.upper()
	search_mod = search_mod.split()

	# first page of google images
	url1 = f'https://www.google.com/search?q={search}&hl=pt-BR&gbv=1&source=lnms&tbm=isch&sa=X&ved=2ahUKEwipwcKTxOfrAhUUDrkGHZ3kB5kQ_AUoAXoECB8QAw&sfr=gws&sei=g8xeX6WWO4KI5OUP1Ne2sAQ'
	i = 0

	response = requests.get(url1)

	soup = BeautifulSoup(response.text, 'html.parser')

	links_list = []
	repeat_img = []

	table = soup.find_all('table', attrs={'class': 'TxbwNb'})

	for links in table:
		links_list.append(links.a.get('href'))

	links_list = [sites[7:sites.index('&')] for sites in links_list]

	for link in links_list:
		if i == num_of_img:
			break
		url = link
		try:
			response = requests.get(url)
			soup = BeautifulSoup(response.text, 'html.parser')

			images = soup.find_all('img')
			for image in images:
				if i == num_of_img:
					break
				elif image.get('src') == None or '.svg' in image.get('src'):
					pass
				elif image.get('src') in repeat_img:
					pass
				else:
					search_comp = [i in image.get('src').upper() for i in search_mod]
					if search_comp.count(True) == len(search_comp) and 'http' in image.get('src'):
						with open(f'{directory}/{search.lower()}_{i}.png', 'wb') as f:
							img_response = requests.get(image.get('src'))
							if img_response.status_code == 200:  # sucess
								if i == num_of_img:
									break
								else:
									#print(link, '-', i)
									print(image.get('src'), '-', i)
									repeat_img.append(image.get('src'))
									f.write(img_response.content)
									i += 1
							else:
								continue
		except:
			pass

	page = 20
	if i < num_of_img:
		while True:
			links_list = []
			img_list = []
			urln = f'https://www.google.com/search?q={search}&hl=pt-BR&gbv=1&tbm=isch&ei=78xeX5PTGc_Z5OUPrsyA0A0&start={page}&sa=N'
			response = requests.get(urln)
			soup = BeautifulSoup(response.text, 'html.parser')
			table = soup.find_all('table', attrs={'class': 'IkMU6e'})
			for links in table:
				links_list.append(links.a.get('href'))
			links_list = [sites[7:sites.index('&')] for sites in links_list]
			for link in links_list:
				if i == num_of_img:
					break
				url = link
				try:
					response = requests.get(url)
					soup = BeautifulSoup(response.text, 'html.parser')

					images = soup.find_all('img')
					for image in images:
						if i == num_of_img:
							break
						elif image.get('src') == None or '.svg' in image.get('src'):
							pass
						elif image.get('src') in repeat_img:
							pass
						else:
							search_comp = [i in image.get('src').upper() for i in search_mod]
							if search_comp.count(True) == len(search_comp) and 'http' in image.get('src'):
								with open(f'{directory}/image_{i}.png', 'wb') as f:
									img_response = requests.get(image.get('src'))
									if img_response.status_code == 200:  # sucess
										if i == num_of_img:
											break
										else:
											#print(link, '-', i)
											print(image.get('src'), '-', i)
											repeat_img.append(image.get('src'))
											f.write(img_response.content)
											i += 1
							else:
								pass
				except:
					pass

			if i < num_of_img:
				page += 20
			else:
				break
	print(f'\n{i} images were downloaded\n')

if __name__ == '__main__':
	while True:
		try:
			directory = input('Directory: ')
			if directory == '':
				if 'images' in os.listdir():
					pass
				else:
					os.mkdir('./images')
				directory = './images/'
			else:
				if directory[-1] != '/':
					directory += '/'
			print(directory)

			#directory = '/home/khalli/Desktop/teste/'
			search = input('What images do you want to download: ')
			num_of_img = int(input('How many images: '))
			clear = input('Clear small images ? [Y/N]\n').upper().strip()
			print('\nDownloading...Check your directory\n')

			googleImagesDownload(directory, search, num_of_img)

			if clear in 'SY':
				clearImages(directory)
			else:
				pass

			exit_question = input('Do you want to quit? [Y/N]').upper().strip()
			if exit_question in 'SY':
				print('Bye!')
				break
			else:
				pass

		except ValueError:
			print('\nWrong value type!\n')