royshil/export_twitter_image.py

## export_twitter_image.py
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC

import logging
import os
import sys
import argparse
import urllib.parse
import html
import gradio as gr
from PIL import Image

parser = argparse.ArgumentParser(description="Create a screenshot from twitter embed")
# optional argument for multiple urls to get an image for (only used in headless mode)
parser.add_argument(
    "--urls", nargs="+", help="URLs to get images for (only used in headless mode)"
)
# allow running as gradio interface, add an argument to opt for headless mode
parser.add_argument(
    "--headless", action="store_true", help="Run in headless mode (no GUI)"
)
args = parser.parse_args()

logging.getLogger().setLevel(logging.INFO)

script_directory = os.path.dirname(__file__)


def get_photo_from_tweet(tweet_url):
    options = webdriver.ChromeOptions()
    options.add_argument("--allow-insecure-localhost")
    options.add_experimental_option(
        "excludeSwitches", ["ignore-certificate-errors", "enable-automation"]
    )
    options.add_argument("--disable-blink-features=AutomationControlled")
    options.add_argument("--disable-web-security")
    options.add_argument("--autoplay-policy=no-user-gesture-required")
    options.add_argument("--headless")
    options.add_argument("--nogpu")
    options.add_argument("--disablegpu")
    options.add_argument("--window-size=1280,1280")
    options.add_experimental_option("useAutomationExtension", False)
    options.add_experimental_option(
        "prefs",
        {
            "download.default_directory": script_directory,
            "profile.default_content_setting_values.automatic_downloads": 2,
        },
    )
    options.add_argument("--mute-audio")

    logging.info("Create environment")
    driver = webdriver.Chrome(options=options)

    logging.info("Loading embed tool page")
    # replace x.com with twitter.com
    url_for_embed_tool = tweet_url.replace("x.com", "twitter.com")
    # add the embed url to the query string, url-escape it
    driver.get(
        "https://publish.twitter.com/?query=" + urllib.parse.quote(url_for_embed_tool)
    )
    logging.info("Waiting for page to load")
    # wait for the 'EmbedCode-code' element to be loaded
    embed_code = WebDriverWait(driver, 10).until(
        EC.presence_of_element_located((By.CLASS_NAME, "EmbedCode-code"))
    )
    if embed_code is None:
        logging.error("Embed code not found")
        sys.exit(1)

    # get the embed code
    embed_code_html = embed_code.get_attribute("innerHTML")
    logging.info("Embed code loaded")
    # decode html escaping
    embed_code_html = html.unescape(embed_code_html)
    logging.info("Embed code decoded")

    logging.info("Create embed code HTML")
    embedding_html = f"""<html>
<body>
{embed_code_html}
</body>
</html>
"""

    # write the embed code to a file
    with open(f"{script_directory}/embed_twitter.html", "w") as f:
        f.write(embedding_html)

    # load the embed code
    driver.get(f"file://{script_directory}/embed_twitter.html")

    logging.info("Creating video...")

    #log the current html
    logging.info(driver.page_source)

    # wait until the iframe is loaded
    iframe = WebDriverWait(driver, 10).until(
        EC.presence_of_element_located((By.TAG_NAME, "iframe"))
    )
    # switch to the iframe
    driver.switch_to.frame(iframe)
    # wait untill all images are loaded
    images = WebDriverWait(driver, 10).until(
        EC.presence_of_all_elements_located((By.TAG_NAME, "img"))
    )
    # switch back to the main frame
    driver.switch_to.default_content()

    logging.info("Done.")

    # capture screenshot
    driver.save_screenshot("screenshot.png")

    driver.close()
    driver.quit()

    logging.info("Converting...")

    # rename url_for_embed_tool to file-safe name
    output_filename = url_for_embed_tool.replace("https://twitter.com/", "")
    # replace any non alphanumeric characters with underscores
    output_filename = "".join(
        [c if c.isalnum() else "_" for c in output_filename]
    )  # https://stackoverflow.com/a/295146/149721

    # trim whitespace with imagemagick
    os.system(f"convert -trim screenshot.png {output_filename}.png")

    logging.info("Done.")

    # delete the embed code file
    os.remove(f"{script_directory}/embed_twitter.html")

    # read the screenshot with PIL
    im = Image.open(f"{output_filename}.png")

    return im


if __name__ == "__main__":
    if args.headless:
        # headless mode
        if args.urls is None:
            logging.error("URL not specified")
            sys.exit(1)
        else:
            for url in args.urls:
                get_photo_from_tweet(url)
    else:
        # interactive mode - build Gradio interface
        iface = gr.Interface(
            fn=get_photo_from_tweet,
            inputs="text",
            outputs="image",
            title="Twitter Embed Screenshot",
            description="Create a screenshot from a Twitter embed",
            allow_flagging=False,
            allow_screenshot=False,
            allow_download=False,
        )

        iface.launch()
	from selenium import webdriver
	from selenium.webdriver.common.by import By
	from selenium.webdriver.support.ui import WebDriverWait
	from selenium.webdriver.support import expected_conditions as EC

	import logging
	import os
	import sys
	import argparse
	import urllib.parse
	import html
	import gradio as gr
	from PIL import Image

	parser = argparse.ArgumentParser(description="Create a screenshot from twitter embed")
	# optional argument for multiple urls to get an image for (only used in headless mode)
	parser.add_argument(
	"--urls", nargs="+", help="URLs to get images for (only used in headless mode)"
	)
	# allow running as gradio interface, add an argument to opt for headless mode
	parser.add_argument(
	"--headless", action="store_true", help="Run in headless mode (no GUI)"
	)
	args = parser.parse_args()

	logging.getLogger().setLevel(logging.INFO)

	script_directory = os.path.dirname(__file__)


	def get_photo_from_tweet(tweet_url):
	options = webdriver.ChromeOptions()
	options.add_argument("--allow-insecure-localhost")
	options.add_experimental_option(
	"excludeSwitches", ["ignore-certificate-errors", "enable-automation"]
	)
	options.add_argument("--disable-blink-features=AutomationControlled")
	options.add_argument("--disable-web-security")
	options.add_argument("--autoplay-policy=no-user-gesture-required")
	options.add_argument("--headless")
	options.add_argument("--nogpu")
	options.add_argument("--disablegpu")
	options.add_argument("--window-size=1280,1280")
	options.add_experimental_option("useAutomationExtension", False)
	options.add_experimental_option(
	"prefs",
	{
	"download.default_directory": script_directory,
	"profile.default_content_setting_values.automatic_downloads": 2,
	},
	)
	options.add_argument("--mute-audio")

	logging.info("Create environment")
	driver = webdriver.Chrome(options=options)

	logging.info("Loading embed tool page")
	# replace x.com with twitter.com
	url_for_embed_tool = tweet_url.replace("x.com", "twitter.com")
	# add the embed url to the query string, url-escape it
	driver.get(
	"https://publish.twitter.com/?query=" + urllib.parse.quote(url_for_embed_tool)
	)
	logging.info("Waiting for page to load")
	# wait for the 'EmbedCode-code' element to be loaded
	embed_code = WebDriverWait(driver, 10).until(
	EC.presence_of_element_located((By.CLASS_NAME, "EmbedCode-code"))
	)
	if embed_code is None:
	logging.error("Embed code not found")
	sys.exit(1)

	# get the embed code
	embed_code_html = embed_code.get_attribute("innerHTML")
	logging.info("Embed code loaded")
	# decode html escaping
	embed_code_html = html.unescape(embed_code_html)
	logging.info("Embed code decoded")

	logging.info("Create embed code HTML")
	embedding_html = f"""<html>
	<body>
	{embed_code_html}
	</body>
	</html>
	"""

	# write the embed code to a file
	with open(f"{script_directory}/embed_twitter.html", "w") as f:
	f.write(embedding_html)

	# load the embed code
	driver.get(f"file://{script_directory}/embed_twitter.html")

	logging.info("Creating video...")

	#log the current html
	logging.info(driver.page_source)

	# wait until the iframe is loaded
	iframe = WebDriverWait(driver, 10).until(
	EC.presence_of_element_located((By.TAG_NAME, "iframe"))
	)
	# switch to the iframe
	driver.switch_to.frame(iframe)
	# wait untill all images are loaded
	images = WebDriverWait(driver, 10).until(
	EC.presence_of_all_elements_located((By.TAG_NAME, "img"))
	)
	# switch back to the main frame
	driver.switch_to.default_content()

	logging.info("Done.")

	# capture screenshot
	driver.save_screenshot("screenshot.png")

	driver.close()
	driver.quit()

	logging.info("Converting...")

	# rename url_for_embed_tool to file-safe name
	output_filename = url_for_embed_tool.replace("https://twitter.com/", "")
	# replace any non alphanumeric characters with underscores
	output_filename = "".join(
	[c if c.isalnum() else "_" for c in output_filename]
	) # https://stackoverflow.com/a/295146/149721

	# trim whitespace with imagemagick
	os.system(f"convert -trim screenshot.png {output_filename}.png")

	logging.info("Done.")

	# delete the embed code file
	os.remove(f"{script_directory}/embed_twitter.html")

	# read the screenshot with PIL
	im = Image.open(f"{output_filename}.png")

	return im


	if __name__ == "__main__":
	if args.headless:
	# headless mode
	if args.urls is None:
	logging.error("URL not specified")
	sys.exit(1)
	else:
	for url in args.urls:
	get_photo_from_tweet(url)
	else:
	# interactive mode - build Gradio interface
	iface = gr.Interface(
	fn=get_photo_from_tweet,
	inputs="text",
	outputs="image",
	title="Twitter Embed Screenshot",
	description="Create a screenshot from a Twitter embed",
	allow_flagging=False,
	allow_screenshot=False,
	allow_download=False,
	)

	iface.launch()