Bump urllib3 from 1.25.2 to 1.25.8 (#162)

Bumps [urllib3](https://github.com/urllib3/urllib3) from 1.25.2 to 1.25.8.
- [Release notes](https://github.com/urllib3/urllib3/releases)
- [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst)
- [Commits](https://github.com/urllib3/urllib3/compare/1.25.2...1.25.8)

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

FUNDING.yml

@@ -0,0 +1,2 @@
+github: [weskerfoot]
+custom: "39qHYvjVcMCNFr3RPAVetci9mKjzYGTQPz"

README.md

@@ -1,3 +1,7 @@
+WARNING:
+This currently only works for English language Facebook accounts, due to the lack of a usable API.
+Also, year by year deletion is currently broken. Feel free to fork or make pull requests.
+
 ## Why?
 
 I needed a simple and reliable way to delete Facebook posts. There are
@@ -18,7 +22,7 @@ Personally, I did this so I would feel less attached to my Facebook profile
 (and hence feel the need to use it less).
 
 ## Dependencies
-- This tool requires at least Python 3.6 in order to run.
+- This tool requires at least Python 3.7 in order to run.
 - A recent copy of Chrome or Chromium installed and available in your `$PATH`
 
 ## Installation
@@ -57,7 +61,7 @@ optional arguments:
 ```
 
 * Make sure that you have a recent version of Python 3.x installed (preferably
-  3.6 or greater)
+  3.7 or greater)
 * Make sure that you have Google Chrome installed and that it is up to date
 * The tool will attempt to automatically install chromedriver for Selenium. See [here](https://sites.google.com/a/chromium.org/chromedriver/home) for an explanation of what the chromedriver does. You may have to manually install it if auto-install fails.
   * On Linux, it will be called something like `chromium-chromedriver` or just

deletefb/tools/chrome_driver.py

@@ -2,6 +2,7 @@ from ..exceptions import UnknownOSException, ChromeError
 from .common import NO_CHROME_DRIVER
 from clint.textui import puts, colored
 from selenium import webdriver
+from selenium.common.exceptions import WebDriverException
 from shutil import which
 from subprocess import check_output
 from urllib.request import urlretrieve

deletefb/tools/common.py

@@ -12,6 +12,7 @@ import json
 import logging
 import logging.config
 import os
+import urllib.parse as urlparse
 
 SELENIUM_EXCEPTIONS = (
     NoSuchElementException,
@@ -19,13 +20,20 @@ SELENIUM_EXCEPTIONS = (
     TimeoutException
 )
 
-
 def click_button(driver, el):
     """
     Click a button using Javascript
     """
     driver.execute_script("arguments[0].click();", el)
 
+def scroll_until_element_exists(driver, xpath_expr):
+    while True:
+        driver.execute_script("window.scrollTo(0, document.body.scrollHeight)")
+        try:
+            element = driver.find_element_by_xpath(xpath_expr)
+        except SELENIUM_EXCEPTIONS:
+            continue
+        break
 
 def scroll_to(driver, el):
     """
@@ -36,7 +44,6 @@ def scroll_to(driver, el):
     except SELENIUM_EXCEPTIONS:
         return
 
-
 def logger(name):
     """
     Args:
@@ -56,17 +63,29 @@ def logger(name):
         logging.config.dictConfig(config["logging"])
     return logging.getLogger(name)
 
-
+def wait_xpath(driver, expr, timeout=20):
-def wait_xpath(driver, expr):
     """
-    Takes an XPath expression, and waits at most 20 seconds until it exists
+    Takes an XPath expression, and waits at most `timeout` seconds until it exists
     """
-    wait = WebDriverWait(driver, 20)
+    wait = WebDriverWait(driver, timeout)
     try:
         wait.until(EC.presence_of_element_located((By.XPATH, expr)))
     except SELENIUM_EXCEPTIONS:
         return
 
+def force_mobile(url):
+    """
+    Force a url to use the mobile site.
+    """
+    parsed = urlparse.urlparse(url)
+    # Ensure a protocol is given (needed by selenium).
+    scheme = parsed.scheme or "https"
+    return urlparse.urlunparse((scheme,
+                                "mobile.facebook.com",
+                                parsed.path,
+                                parsed.params,
+                                parsed.query,
+                                parsed.fragment))
 
 NO_CHROME_DRIVER = """
 You need to manually install the chromedriver for Selenium\n

deletefb/tools/conversations.py

@@ -6,6 +6,7 @@ from selenium.webdriver.common.action_chains import ActionChains
 from selenium.webdriver.support.ui import Select
 from pendulum import now
 from json import loads
+from time import sleep
 
 import lxml.html as lxh
 
@@ -29,12 +30,13 @@ def get_conversations(driver):
             date = None
 
             if url and "messages/read" in url:
-
+                try:
-                date = convo.find_element_by_xpath("../../..//abbr").text
+                    date = convo.find_element_by_xpath("../../..//abbr").text
-                conversation_name = convo.find_element_by_xpath("../../../div/div/header/h3").text.strip()
+                    conversation_name = convo.find_element_by_xpath("../../../div/div/header/h3").text.strip()
-
+                    assert(conversation_name)
-                assert(conversation_name)
+                    assert(url)
-                assert(url)
+                except (SELENIUM_EXCEPTIONS + (AssertionError,)):
+                    continue
 
                 conversations.append(
                     Conversation(
@@ -49,7 +51,10 @@ def get_conversations(driver):
                         find_element_by_xpath("a").
                         get_attribute("href"))
 
-        except SELENIUM_EXCEPTIONS:
+            print("next_url", next_url)
+
+        except SELENIUM_EXCEPTIONS as e:
+            print(e)
             break
         if not next_url:
             break

deletefb/tools/likes.py

@@ -1,6 +1,6 @@
 from .archive import archiver
 from ..types import Page
-from .common import SELENIUM_EXCEPTIONS, logger, click_button
+from .common import SELENIUM_EXCEPTIONS, logger, click_button, wait_xpath, force_mobile, scroll_to, scroll_until_element_exists
 from selenium.webdriver.common.by import By
 from selenium.webdriver.support import expected_conditions as EC
 from selenium.webdriver.support.ui import WebDriverWait
@@ -18,17 +18,25 @@ def load_likes(driver, profile_url):
         None
     """
 
-    driver.get("{0}/likes_all".format(profile_url))
+    likes_link_xpath = "//div[normalize-space(text())='Likes']/../..//a[contains(@href, 'app_section')]"
 
-    wait = WebDriverWait(driver, 20)
+    all_likes_link_xpath = "//div[normalize-space(text())='All Likes']/../../../..//a[contains(@href, 'app_collection')]"
 
-    try:
+    driver.get(force_mobile("{0}/about".format(profile_url)))
-        wait.until(
+
-            EC.presence_of_element_located((By.CSS_SELECTOR, ".PageLikeButton"))
+    scroll_until_element_exists(driver, "//div[text()='Likes']")
-        )
+
-    except SELENIUM_EXCEPTIONS:
+    likes_link_el = driver.find_element_by_xpath(likes_link_xpath)
-        LOG.exception("Traceback of load_likes")
+
-        return
+    driver.get(likes_link_el.get_attribute("href"))
+
+    wait_xpath(driver, all_likes_link_xpath)
+
+    all_likes_link_el = driver.find_element_by_xpath(all_likes_link_xpath)
+
+    all_likes_link = all_likes_link_el.get_attribute("href")
+
+    driver.get(all_likes_link)
 
 def get_page_links(driver):
     """
@@ -39,8 +47,7 @@ def get_page_links(driver):
     Returns:
         List of URLs to pages
     """
-    pages = driver.find_elements_by_xpath("//li//div/div/a[contains(@class, 'lfloat')]")
+    pages = driver.find_elements_by_xpath("//header/..//a")
-
     return [page.get_attribute("href").replace("www", "mobile") for page in pages]
 
 def unlike_page(driver, url, archive=None):
@@ -60,7 +67,7 @@ def unlike_page(driver, url, archive=None):
 
     print("Unliking {0}".format(url))
 
-    wait = WebDriverWait(driver, 20)
+    wait = WebDriverWait(driver, 5)
 
     try:
         wait.until(
@@ -70,19 +77,14 @@ def unlike_page(driver, url, archive=None):
         # Something went wrong with this page, so skip it
         return
 
-    button = driver.find_element_by_xpath("//*[text()='Liked']")
+    driver.find_element_by_xpath("//*[text()='Liked']/../../../..").click()
-
-    # Click the "Liked" button to open up "Unlike"
-    click_button(driver, button)
 
     wait.until(
         EC.presence_of_element_located((By.XPATH, "//*[text()='Unlike']"))
     )
 
     # There should be an "Unlike" button now, click it
-    unlike_button = driver.find_element_by_xpath("//*[text()='Unlike']/..")
+    driver.find_element_by_xpath("//*[text()='Unlike']/..").click()
-
-    click_button(driver, unlike_button)
 
     if archive:
         archive(Page(name=url))

deletefb/tools/wall.py

@@ -1,6 +1,6 @@
 from ..types import Post
 from .archive import archiver
-from .common import SELENIUM_EXCEPTIONS, click_button, wait_xpath
+from .common import SELENIUM_EXCEPTIONS, click_button, wait_xpath, force_mobile
 from .config import settings
 from selenium.webdriver.common.action_chains import ActionChains
 
@@ -24,23 +24,43 @@ def delete_posts(driver,
     if year is not None:
         user_profile_url = "{0}/timeline?year={1}".format(user_profile_url, year)
 
+    user_profile_url = force_mobile(user_profile_url)
+
     driver.get(user_profile_url)
 
-    for _ in range(MAX_POSTS):
+    finished = False
-        post_button_sel = "_4xev"
+
+    with archiver("wall") as archive_wall_post:
+        for _ in range(MAX_POSTS):
+            if finished:
+                break
+            post_button_sel = "_4s19"
+
+            post_content_sel = "userContent"
+            post_timestamp_sel = "timestampContent"
+
+            confirmation_button_exp = "//div[contains(@data-sigil, 'undo-content')]//*/a[contains(@href, 'direct_action_execute')]"
 
-        post_content_sel = "userContent"
+            # Cannot return a text node, so it returns the parent.
-        post_timestamp_sel = "timestampContent"
+            # Tries to be pretty resilient against DOM re-organizations
+            timestamp_exp = "//article//*/header//*/div/a[contains(@href, 'story_fbid')]//text()/.."
 
-        button_types = ["FeedDeleteOption", "HIDE_FROM_TIMELINE", "UNTAG"]
+            button_types = ["Delete post", "Remove tag", "Hide from timeline", "Hide from profile"]
 
-        with archiver("wall") as archive_wall_post:
             while True:
                 try:
-                    timeline_element = driver.find_element_by_class_name(post_button_sel)
+                    try:
+                        timeline_element = driver.find_element_by_xpath("//div[@data-sigil='story-popup-causal-init']/a")
+                    except SELENIUM_EXCEPTIONS:
+                        print("Could not find any posts")
+                        finished = True
+                        break
+
+                    post_content_element = driver.find_element_by_xpath("//article/div[@class='story_body_container']/div")
+                    post_content_ts = driver.find_element_by_xpath(timestamp_exp)
 
-                    post_content_element = driver.find_element_by_class_name(post_content_sel)
+                    if not (post_content_element or post_content_ts):
-                    post_content_ts = driver.find_element_by_class_name(post_timestamp_sel)
+                        break
 
                     # Archive the post
                     archive_wall_post.archive(
@@ -53,15 +73,18 @@ def delete_posts(driver,
                     actions = ActionChains(driver)
                     actions.move_to_element(timeline_element).click().perform()
 
-                    wait_xpath(driver, "//*[@id='feed_post_menu']/..")
+                    # Wait until the buttons show up
-
+                    wait_xpath(driver, "//*[contains(@data-sigil, 'removeStoryButton')]")
-                    menu = driver.find_element_by_xpath("//*[@id='feed_post_menu']/..")
 
                     delete_button = None
 
+                    # Search for visible buttons in priority order
+                    # Delete -> Untag -> Hide
                     for button_type in button_types:
                         try:
-                            delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"{0}\"]".format(button_type))
+                            delete_button = driver.find_element_by_xpath("//*[text()='{0}']".format(button_type))
+                            if not delete_button.is_displayed():
+                                continue
                             break
                         except SELENIUM_EXCEPTIONS as e:
                             print(e)
@@ -72,8 +95,10 @@ def delete_posts(driver,
                         break
 
                     click_button(driver, delete_button)
-                    confirmation_button = driver.find_element_by_class_name("layerConfirm")
+                    wait_xpath(driver, confirmation_button_exp)
+                    confirmation_button = driver.find_element_by_xpath(confirmation_button_exp)
 
+                    print(confirmation_button)
                     click_button(driver, confirmation_button)
 
                 except SELENIUM_EXCEPTIONS as e:

requirements.txt

@@ -1,20 +1,20 @@
 appdirs==1.4.3
 args==0.1.0
-attrs==19.1.0
+attrs==20.3.0
 bitarray==0.9.3
-bleach==3.1.2
+bleach==3.3.0
-cattrs-3.8==0.9.1
+cattrs==1.1.2
 certifi==2018.11.29
 chardet==3.0.4
 clint==0.5.1
 docutils==0.14
 idna==2.8
-lxml==4.4.0
+lxml==4.6.3
 pendulum==2.0.5
 pkginfo==1.5.0.1
 progressbar==2.5
 pybloom-live==3.0.0
-Pygments==2.4.2
+Pygments==2.7.4
 python-dateutil==2.8.0
 pytzdata==2019.2
 readme-renderer==24.0
@@ -28,5 +28,5 @@ tldextract==2.2.0
 tqdm==4.32.2
 twine==1.13.0
 typing==3.7.4
-urllib3==1.25.2
+urllib3==1.25.8
 webencodings==0.5.1

setup.py

@@ -5,7 +5,7 @@ with open("README.md", "r") as fh:
 
 setuptools.setup(
     name="delete-facebook-posts",
-    version="1.1.13",
+    version="1.1.17",
     author="Wesley Kerfoot",
     author_email="wes@wesk.tech",
     description="A Selenium Script to Delete Facebook Posts",
@@ -24,8 +24,8 @@ setuptools.setup(
         "selenium-requests",
         "requests",
         "pybloom-live",
-        "attrs",
+        "attrs>=20.3.0",
-        "cattrs-3.8",
+        "cattrs>=1.1.2",
         "lxml",
         "pendulum",
         "clint",