Browse Source

Merged master.

pull/68/head
Gregory Gundersen 5 years ago
parent
commit
dc8f227b05
  1. 18
      README.md
  2. 30
      deletefb/deletefb.py
  3. 62
      deletefb/logging_conf.json
  4. 54
      deletefb/tools/archive.py
  5. 61
      deletefb/tools/common.py
  6. 4
      deletefb/tools/config.py
  7. 110
      deletefb/tools/likes.py
  8. 16
      deletefb/tools/login.py
  9. 87
      deletefb/tools/wall.py
  10. 29
      deletefb/types.py
  11. 3
      requirements.txt
  12. 12
      setup.py

18
README.md

@ -19,13 +19,15 @@ Personally, I did this so I would feel less attached to my Facebook profile
## Installation
You have several options to run it.
1) Install from PyPI with `pip install --user delete-facebook-posts`
2) Clone this repo and run `pip install --user .` or do `pip install --user
1) Install from PyPI with `pip3 install --user delete-facebook-posts`
2) Clone this repo and run `pip3 install --user .` or do `pip3 install --user
git+https://github.com/weskerfoot/DeleteFB.git`
3) Set up a Python virtualenv, activate it, and run `pip install -r requirements.txt`, then you can just run `python -m deletefb.deletefb` in the DeleteFB directory.
3) Set up a Python virtualenv, activate it, and run `pip3 install -r requirements.txt`, then you can just run `python -m deletefb.deletefb` in the DeleteFB directory.
## How To Use It
* Make sure that you have a recent version of Python 3.x installed (preferably
3.6 or greater)
* Make sure that you have Google Chrome installed and that it is up to date
* Also install the chromedriver for Selenium. See [here](https://sites.google.com/a/chromium.org/chromedriver/home) for an explanation of what the chromedriver does.
* On Linux, it will be called something like `chromium-chromedriver` or just
@ -37,15 +39,18 @@ git+https://github.com/weskerfoot/DeleteFB.git`
brew cask install chromedriver
```
* Run `deletefb -E "youremail@example.org" -P "yourfacebookpassword" -U "https://www.facebook.com/your.profile.url"`
* Run `deletefb -E 'youremail@example.org' -P 'yourfacebookpassword' -U 'https://www.facebook.com/your.profile.url'`
* The script will log into your Facebook account, go to your profile page, and
start deleting posts. If it cannot delete something, then it will "hide" it
from your timeline instead.
* Be patient as it will take a very long time, but it will eventually clear
everything. You may safely minimize the chrome window without breaking it.
## Login
* The tool will log in using the credentials passed to it. It will wait until the page `https://www.facebook.com/` is loaded in order to avoid any issues with logging in. If you pass a 2FA token explicitly with the `-F` option, then it will try to enter that for you. If there are any issues, it simply pauses indefinitely to allow the user to resolve the problems, and then continues execution.
## 2FA
* It is recommended that you disable Two-Factor Authentication tempoprarily
* It is recommended that you disable Two-Factor Authentication temporarily
while you are running the script, in order to get the best experience.
* If you run into issues with Facebook complaining about your browser,
@ -63,7 +68,8 @@ git+https://github.com/weskerfoot/DeleteFB.git`
## Unlike Pages
* You may use `-M unlike_pages` to unlike all of your pages. The names of the
pages will be archived (unless archival is turned off), and this option
conflicts with the year option.
conflicts with the year option. This will only unlike your *pages* that you
have liked. It will *not* unlike anything else (like books or movies).
## Archival
* The tool will archive everything being deleted by default in `.log` files.

30
deletefb/deletefb.py

@ -1,19 +1,16 @@
#!/usr/bin/env python
from .tools.common import logger
from .tools.config import settings
from .tools.likes import unlike_pages
from .tools.login import login
from .tools.wall import delete_posts
import argparse
import getpass
import json
import os
import sys
from .tools.common import logger
from .tools.login import login
from .tools.wall import delete_posts
from .tools.likes import unlike_pages
LOG = logger("deletefb")
def run_delete():
parser = argparse.ArgumentParser()
@ -92,11 +89,7 @@ def run_delete():
args = parser.parse_args()
if args.archive_off:
os.environ["DELETEFB_ARCHIVE"] = "false"
else:
os.environ["DELETEFB_ARCHIVE"] = "true"
settings["ARCHIVE"] = not args.archive_off
if args.year and args.mode != "wall":
parser.error("The --year option is only supported in wall mode")
@ -111,11 +104,14 @@ def run_delete():
)
if args.mode == "wall":
delete_posts(driver,
args.profile_url,
year=args.year)
delete_posts(
driver,
args.profile_url,
year=args.year
)
elif args.mode == "unlike_pages":
unlike_pages(driver)
unlike_pages(driver, args.profile_url)
else:
print("Please enter a valid mode")
sys.exit(1)

62
deletefb/logging_conf.json

@ -0,0 +1,62 @@
{
"logging": {
"version": 1,
"disable_existing_loggers": true,
"formatters": {
"brief": {
"class": "logging.Formatter",
"style": "{",
"datefmt": "%H:%M:%S",
"format": "{name:s}-{levelname:s}-{asctime:s}-{message:s}"
},
"verbose": {
"class": "logging.Formatter",
"style": "{",
"datefmt": "%Y-%m-%dT%H:%M:%S",
"format": "{name:s}:{levelname:s}:L{lineno:d} {asctime:s} {message:s}"
}
},
"handlers": {
"console": {
"level": "DEBUG",
"class": "logging.StreamHandler",
"formatter": "brief",
"stream": "ext://sys.stdout"
},
"file_handler": {
"level": "INFO",
"class": "logging.handlers.WatchedFileHandler",
"formatter": "verbose",
"filename": "./deletefb.log",
"mode": "a",
"encoding": "utf-8"
}
},
"loggers": {
"root": {
"level": "DEBUG",
"handlers": ["console", "file_handler"]
},
"deletefb": {
"level": "DEBUG",
"handlers": ["console"],
"propagate": false
},
"login": {
"level": "DEBUG",
"handlers": ["file_handler"],
"propagate": false
},
"likes": {
"level": "DEBUG",
"handlers": ["file_handler"],
"propagate": false
},
"wall": {
"level": "DEBUG",
"handlers": ["file_handler"],
"propagate": false
}
}
}
}

54
deletefb/tools/archive.py

@ -0,0 +1,54 @@
from .config import settings
from contextlib import contextmanager
from pathlib import Path
import attr
import json
# Used to avoid duplicates in the log
from pybloom_live import BloomFilter
def make_filter():
return BloomFilter(
capacity=settings["MAX_POSTS"],
error_rate=0.001
)
@attr.s
class Archive:
archive_type = attr.ib()
# We give the Archive class a file handle
archive_file = attr.ib()
_bloom_filter = attr.ib(factory=make_filter)
def archive(self, content):
"""
Archive an object
"""
print("Archiving {0}".format(content))
if content.name not in self._bloom_filter:
self.archive_file.write(json.dumps(attr.asdict(content)) + "\n")
self._bloom_filter.add(content.name)
return
@contextmanager
def archiver(archive_type):
archive_file = open(
(Path(".") / Path(archive_type).name).with_suffix(".log"),
mode="ta",
buffering=1
)
archiver_instance = Archive(
archive_type=archive_type,
archive_file=archive_file
)
try:
yield archiver_instance
finally:
archive_file.close()

61
deletefb/tools/common.py

@ -1,16 +1,14 @@
import json
import logging
import logging.config
import os
from os.path import abspath, relpath, split, isfile
import time
from os.path import isfile
from selenium.common.exceptions import (
NoSuchElementException,
StaleElementReferenceException,
TimeoutException
)
import json
import logging
import logging.config
import os
SELENIUM_EXCEPTIONS = (
NoSuchElementException,
@ -18,13 +16,15 @@ SELENIUM_EXCEPTIONS = (
TimeoutException
)
def try_move(actions, el):
for _ in range(10):
try:
actions.move_to_element(el).perform()
except StaleElementReferenceException:
time.sleep(5)
continue
def click_button(driver, el):
"""
Click a button using Javascript
Args:
driver: seleniumrequests.Chrome Driver instance
Returns:
None
"""
driver.execute_script("arguments[0].click();", el)
def logger(name):
"""
@ -34,7 +34,10 @@ def logger(name):
Returns:
logging.Logger
"""
config_path = "deletefb/logging_conf.json"
# Make sure the path always points to the correct directory
config_path = os.path.dirname(os.path.realpath(__file__)) + "/../logging_conf.json"
if not isfile(config_path): # called from file (deletefb.py)
os.chdir("..")
with open(config_path, "r", encoding="utf-8") as config_file:
@ -42,34 +45,6 @@ def logger(name):
logging.config.dictConfig(config["logging"])
return logging.getLogger(name)
def archiver(category):
"""
Log content to file. Call using `archive("some content")`
Args:
category: str The category of logs you want to log
Returns:
(log_file_handle, archiver)
"""
log_path = "{0}.log".format(abspath(relpath(split(category)[-1], ".")))
log_file = open(log_path, mode="ta", buffering=1)
def log(content, timestamp=False):
if os.environ.get("DELETEFB_ARCHIVE", "true") == "false":
return
structured_content = {
"category" : category,
"content" : content,
"timestamp" : timestamp
}
log_file.write("{0}\n".format(json.dumps(structured_content)))
return (log_file, log)
NO_CHROME_DRIVER = """
You need to install the chromedriver for Selenium\n
Please see this link https://github.com/weskerfoot/DeleteFB#how-to-use-it\n

4
deletefb/tools/config.py

@ -0,0 +1,4 @@
settings = {
"ARCHIVE" : True,
"MAX_POSTS" : 5000
}

110
deletefb/tools/likes.py

@ -1,14 +1,13 @@
from .archive import archiver
from ..types import Page
from .common import SELENIUM_EXCEPTIONS, logger, click_button
from selenium.webdriver.common.by import By
from selenium.webdriver.common.action_chains import ActionChains
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from .common import SELENIUM_EXCEPTIONS, archiver, logger
from selenium.webdriver.support.ui import WebDriverWait
LOG = logger(__name__)
def load_likes(driver):
def load_likes(driver, profile_url):
"""
Loads the page that lists all pages you like
@ -18,69 +17,98 @@ def load_likes(driver):
Returns:
None
"""
driver.get("https://www.facebook.com/pages/?category=liked")
driver.get("{0}/likes_all".format(profile_url))
wait = WebDriverWait(driver, 20)
try:
wait.until(
EC.presence_of_element_located((By.XPATH, "//button/div/div[text()='Liked']"))
)
wait.until(
EC.presence_of_element_located((By.XPATH, "//button/div/i[@aria-hidden=\"true\"]"))
EC.presence_of_element_located((By.CSS_SELECTOR, ".PageLikeButton"))
)
except SELENIUM_EXCEPTIONS:
LOG.exception("Traceback of load_likes")
return
def unlike_pages(driver):
def get_page_links(driver):
"""
Unlike all pages
Gets all of the links to the pages you like
Args:
driver: seleniumrequests.Chrome Driver instance
Returns:
List of URLs to pages
"""
pages = driver.find_elements_by_xpath("//li//div/div/a[contains(@class, 'lfloat')]")
return [page.get_attribute("href").replace("www", "mobile") for page in pages]
def unlike_page(driver, url, archive=None):
"""
Unlikes a page given the URL to it
Args:
driver: seleniumrequests.Chrome Driver instance
url: url string pointing to a page
archive: archiver instance
Returns:
None
"""
like_log, archive_likes = archiver("likes")
driver.get(url)
print("Unliking {0}".format(url))
wait = WebDriverWait(driver, 20)
actions = ActionChains(driver)
try:
wait.until(
EC.presence_of_element_located((By.XPATH, "//*[text()='Liked']"))
)
except SELENIUM_EXCEPTIONS:
# Something went wrong with this page, so skip it
return
load_likes(driver)
button = driver.find_element_by_xpath("//*[text()='Liked']")
pages_list = driver.find_element_by_css_selector("#all_liked_pages")
# Click the "Liked" button to open up "Unlike"
click_button(driver, button)
actions.move_to_element(pages_list).perform()
wait.until(
EC.presence_of_element_located((By.XPATH, "//*[text()='Unlike']"))
)
unlike_buttons = pages_list.find_elements_by_xpath("//button/div/div[text()='Liked']/../..")
# There should be an "Unlike" button now, click it
unlike_button = driver.find_element_by_xpath("//*[text()='Unlike']/..")
while unlike_buttons:
for button in unlike_buttons:
try:
if "Liked" in button.text:
page_name = button.find_element_by_xpath("./../..").text.split("\n")[0]
click_button(driver, unlike_button)
driver.execute_script("arguments[0].click();", button)
if archive:
archive(Page(name=url))
archive_likes(page_name)
def unlike_pages(driver, profile_url):
"""
Unlike all pages
print("{0} was unliked".format(page_name))
Args:
driver: seleniumrequests.Chrome Driver instance
Returns:
None
"""
with archiver("likes") as archive_likes:
load_likes(driver, profile_url)
urls = get_page_links(driver)
while urls:
for url in urls:
unlike_page(driver, url, archive=archive_likes.archive)
try:
load_likes(driver, profile_url)
urls = get_page_links(driver)
except SELENIUM_EXCEPTIONS:
continue
load_likes(driver)
try:
pages_list = driver.find_element_by_css_selector("#all_liked_pages")
actions.move_to_element(pages_list).perform()
unlike_buttons = pages_list.find_elements_by_xpath("//button")
if not unlike_buttons:
# We're done
break
except SELENIUM_EXCEPTIONS:
break
# Explicitly close the log file when we're done with it
like_log.close()

16
deletefb/tools/login.py

@ -1,12 +1,10 @@
import time
import sys
from .common import NO_CHROME_DRIVER
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.chrome.options import Options
from seleniumrequests import Chrome
from .common import NO_CHROME_DRIVER
import sys
import time
def login(user_email_address,
user_password,
@ -47,7 +45,7 @@ def login(user_email_address,
driver.implicitly_wait(10)
driver.get("https://facebook.com")
driver.get("https://www.facebook.com/login/device-based/regular/login/?login_attempt=1&lwv=110")
email = "email"
password = "pass"
@ -99,4 +97,10 @@ def login(user_email_address,
else:
pass
# block until we have reached the main page
# print a message warning the user
while driver.current_url != "https://www.facebook.com/":
print("Execution blocked: Please navigate to https://www.facebook.com to continue")
time.sleep(5)
return driver

87
deletefb/tools/wall.py

@ -1,12 +1,13 @@
import time
from ..types import Post
from .archive import archiver
from .common import SELENIUM_EXCEPTIONS, click_button
from .config import settings
from selenium.webdriver.common.action_chains import ActionChains
from .common import SELENIUM_EXCEPTIONS, archiver
import time
# Used as a threshold to avoid running forever
MAX_POSTS = 15000
MAX_POSTS = settings["MAX_POSTS"]
def delete_posts(driver,
user_profile_url,
@ -31,42 +32,54 @@ def delete_posts(driver,
post_content_sel = "userContent"
post_timestamp_sel = "timestampContent"
wall_log, archive_wall_post = archiver("wall")
button_types = ["FeedDeleteOption", "HIDE_FROM_TIMELINE", "UNTAG"]
while True:
try:
timeline_element = driver.find_element_by_class_name(post_button_sel)
with archiver("wall") as archive_wall_post:
while True:
try:
timeline_element = driver.find_element_by_class_name(post_button_sel)
post_content_element = driver.find_element_by_class_name(post_content_sel)
post_content_ts = driver.find_element_by_class_name(post_timestamp_sel)
post_content_element = driver.find_element_by_class_name(post_content_sel)
post_content_ts = driver.find_element_by_class_name(post_timestamp_sel)
archive_wall_post(post_content_element.text, timestamp=post_content_ts.text)
actions = ActionChains(driver)
actions.move_to_element(timeline_element).click().perform()
# Archive the post
archive_wall_post.archive(
Post(
content=post_content_element.text,
date=post_content_ts.text
)
)
menu = driver.find_element_by_css_selector("#globalContainer > div.uiContextualLayerPositioner.uiLayer > div")
actions.move_to_element(menu).perform()
actions = ActionChains(driver)
actions.move_to_element(timeline_element).click().perform()
menu = driver.find_element_by_css_selector("#globalContainer > div.uiContextualLayerPositioner.uiLayer > div")
actions.move_to_element(menu).perform()
delete_button = None
for button_type in button_types:
try:
delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"{0}\"]".format(button_type))
break
except SELENIUM_EXCEPTIONS:
continue
if not delete_button:
print("Could not find anything to delete")
break
actions.move_to_element(delete_button).click().perform()
confirmation_button = driver.find_element_by_class_name("layerConfirm")
click_button(driver, confirmation_button)
try:
delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"FeedDeleteOption\"]")
except SELENIUM_EXCEPTIONS:
try:
delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"HIDE_FROM_TIMELINE\"]")
except SELENIUM_EXCEPTIONS:
delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"UNTAG\"]")
actions.move_to_element(delete_button).click().perform()
confirmation_button = driver.find_element_by_class_name("layerConfirm")
# Facebook would not let me get focus on this button without some custom JS
driver.execute_script("arguments[0].click();", confirmation_button)
except SELENIUM_EXCEPTIONS:
continue
else:
break
wall_log.close()
# Required to sleep the thread for a bit after using JS to click this button
time.sleep(5)
driver.refresh()
continue
else:
break
# Required to sleep the thread for a bit after using JS to click this button
time.sleep(5)
driver.refresh()

29
deletefb/types.py

@ -0,0 +1,29 @@
import attr
import uuid
import datetime
def timestamp_now():
"""
Returns: a timestamp for this instant, in ISO 8601 format
"""
return datetime.datetime.isoformat(datetime.datetime.now())
# Data type definitions of posts and comments
@attr.s
class Post:
content = attr.ib()
comments = attr.ib(default=[])
date = attr.ib(factory=timestamp_now)
name = attr.ib(factory=lambda: uuid.uuid4().hex)
@attr.s
class Comment:
commenter = attr.ib()
content = attr.ib()
date = attr.ib(factory=timestamp_now)
name = attr.ib(factory=lambda: uuid.uuid4().hex)
@attr.s
class Page:
name = attr.ib()
date = attr.ib(factory=timestamp_now)

3
requirements.txt

@ -1,6 +1,9 @@
attrs==19.1.0
bitarray==0.9.3
certifi==2018.11.29
chardet==3.0.4
idna==2.8
pybloom-live==3.0.0
requests==2.22.0
requests-file==1.4.3
selenium==3.141.0

12
setup.py

@ -5,7 +5,7 @@ with open("README.md", "r") as fh:
setuptools.setup(
name="delete-facebook-posts",
version="1.1.1",
version="1.1.7",
author="Wesley Kerfoot",
author_email="wes@wesk.tech",
description="A Selenium Script to Delete Facebook Posts",
@ -13,10 +13,18 @@ setuptools.setup(
long_description_content_type="text/markdown",
url="https://github.com/weskerfoot/DeleteFB",
packages=setuptools.find_packages(),
include_package_data=True,
requires_python=">=3.6",
package_data={
# Include *json files in the package:
'': ['*.json'],
},
install_requires = [
"selenium",
"selenium-requests",
"requests"
"requests",
"pybloom-live",
"attrs"
],
classifiers= [
"Programming Language :: Python :: 3",

Loading…
Cancel
Save