Merge pull request #53 from weskerfoot/refactor

Refactor wall module & refactor archiver

deletefb/deletefb.py

@@ -1,17 +1,14 @@
 #!/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.config import settings
-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():

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()

deletefb/tools/common.py

@@ -1,21 +1,15 @@
-import json
+from os.path import isfile
-import logging
-import logging.config
-import os
-import time
-
-from .config import settings
-
-# Used to avoid duplicates in the log
-from pybloom_live import BloomFilter
-
-from os.path import abspath, relpath, split, isfile
 from selenium.common.exceptions import (
     NoSuchElementException,
     StaleElementReferenceException,
     TimeoutException
 )
 
+import json
+import logging
+import logging.config
+import os
+
 SELENIUM_EXCEPTIONS = (
     NoSuchElementException,
     StaleElementReferenceException,
@@ -42,8 +36,7 @@ def logger(name):
     """
 
     # Make sure the path always points to the correct directory
-    config_path = os.path.dirname(
+    config_path = os.path.dirname(os.path.realpath(__file__)) + "/../logging_conf.json"
-                    os.path.realpath(__file__)) + "/../logging_conf.json"
 
     if not isfile(config_path):  # called from file (deletefb.py)
         os.chdir("..")
@@ -52,46 +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)
-
-    bfilter = BloomFilter(
-            capacity=settings["MAX_POSTS"],
-            error_rate=0.001
-    )
-
-    def log(content, timestamp=False):
-        if not settings["ARCHIVE"]:
-            return
-
-        if content in bfilter:
-            # This was already archived
-            return
-
-        structured_content = {
-            "category" : category,
-            "content" : content,
-            "timestamp" : timestamp
-        }
-
-        log_file.write("{0}\n".format(json.dumps(structured_content)))
-
-        bfilter.add(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

deletefb/tools/likes.py

@@ -1,9 +1,9 @@
+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 selenium.webdriver.support.ui import WebDriverWait
-from .common import SELENIUM_EXCEPTIONS, archiver, logger, click_button
 
 LOG = logger(__name__)
 
@@ -41,8 +41,6 @@ def get_page_links(driver):
     """
     pages = driver.find_elements_by_xpath("//li//div/div/a[contains(@class, 'lfloat')]")
 
-    actions = ActionChains(driver)
-
     return [page.get_attribute("href").replace("www", "mobile") for page in pages]
 
 def unlike_page(driver, url, archive=None):
@@ -64,8 +62,6 @@ def unlike_page(driver, url, archive=None):
 
     wait = WebDriverWait(driver, 20)
 
-    actions = ActionChains(driver)
-
     try:
         wait.until(
             EC.presence_of_element_located((By.XPATH, "//*[text()='Liked']"))
@@ -89,7 +85,7 @@ def unlike_page(driver, url, archive=None):
     click_button(driver, unlike_button)
 
     if archive:
-        archive(url)
+        archive(Page(name=url))
 
 def unlike_pages(driver, profile_url):
     """
@@ -102,21 +98,17 @@ def unlike_pages(driver, profile_url):
         None
     """
 
-    like_log, archive_likes = archiver("likes")
+    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)
+                unlike_page(driver, url, archive=archive_likes.archive)
             try:
                 load_likes(driver, profile_url)
                 urls = get_page_links(driver)
             except SELENIUM_EXCEPTIONS:
                 # We're done
                 break
-
-    # Explicitly close the log file when we're done with it
-    like_log.close()

deletefb/tools/login.py

@@ -1,12 +1,10 @@
-import time
+from .common import NO_CHROME_DRIVER
-import sys
-
 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,

deletefb/tools/wall.py

@@ -1,8 +1,10 @@
-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 .config import settings
+import time
-from .common import SELENIUM_EXCEPTIONS, archiver, click_button
 
 # Used as a threshold to avoid running forever
 MAX_POSTS = settings["MAX_POSTS"]
@@ -30,8 +32,9 @@ 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"]
 
+        with archiver("wall") as archive_wall_post:
             while True:
                 try:
                     timeline_element = driver.find_element_by_class_name(post_button_sel)
@@ -39,7 +42,14 @@ def delete_posts(driver,
                     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)
+
+                    # Archive the post
+                    archive_wall_post.archive(
+                        Post(
+                            content=post_content_element.text,
+                            date=post_content_ts.text
+                        )
+                    )
 
                     actions = ActionChains(driver)
                     actions.move_to_element(timeline_element).click().perform()
@@ -47,13 +57,18 @@ def delete_posts(driver,
                     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=\"FeedDeleteOption\"]")
+                            delete_button = menu.find_element_by_xpath("//a[@data-feed-option-name=\"{0}\"]".format(button_type))
-                except SELENIUM_EXCEPTIONS:
+                            break
-                    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\"]")
+                            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")
@@ -64,7 +79,6 @@ def delete_posts(driver,
                     continue
                 else:
                     break
-        wall_log.close()
 
             # Required to sleep the thread for a bit after using JS to click this button
             time.sleep(5)

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)