Refactor timestamp parsing

deletefb/tools/common.py

@@ -5,13 +5,12 @@ from selenium.common.exceptions import (
     TimeoutException,
     JavascriptException
 )
-from arrow.parser import ParserError
 
 import json
 import logging
 import logging.config
 import os
-import arrow
+import pendulum
 
 SELENIUM_EXCEPTIONS = (
     NoSuchElementException,
@@ -34,9 +33,6 @@ def scroll_to(driver, el):
     except SELENIUM_EXCEPTIONS:
         return
 
-def parse_ts(text):
-    return arrow.get(text, "DD/M/YYYY")
-
 def logger(name):
     """
     Args:

deletefb/tools/conversations.py

@@ -1,6 +1,6 @@
 from .archive import archiver
 from ..types import Conversation
-from .common import SELENIUM_EXCEPTIONS, logger, parse_ts, ParserError
+from .common import SELENIUM_EXCEPTIONS, logger
 from selenium.webdriver.common.by import By
 from selenium.webdriver.support import expected_conditions as EC
 from selenium.webdriver.support.ui import WebDriverWait
@@ -25,26 +25,32 @@ def get_conversations(driver):
         LOG.exception("No conversations")
         return
 
+    # This function *cannot* be a generator
+    # Otherwise elements will become stale
+    conversations = []
+
     while True:
         for convo in driver.find_elements_by_xpath("//a"):
             url = convo.get_attribute("href")
+
             timestamp = None
 
             if url and "messages/read" in url:
-                try:
-                    timestamp = parse_ts(convo.find_element_by_xpath("../../..//abbr").text)
-                except ParserError:
-                    print("Failed to parse timestamp")
-                    continue
 
+                timestamp = convo.find_element_by_xpath("../../..//abbr").text
                 conversation_name = convo.find_element_by_xpath("../../../div/div/header/h3").text.strip()
 
                 assert(conversation_name)
                 assert(url)
 
-                yield Conversation(url=url,
-                                   name=conversation_name,
-                                   timestamp=timestamp)
+                conversations.append(
+                    Conversation(
+                        url=url,
+                        timestamp=timestamp,
+                        name=conversation_name
+                    )
+                )
+
         try:
             next_url = (driver.find_element_by_id("see_older_threads").
                         find_element_by_xpath("a").
@@ -56,16 +62,16 @@ def get_conversations(driver):
             break
         driver.get(next_url)
 
-def delete_conversations(driver):
+    return conversations
+
+def delete_conversations(driver, older_than=None):
     """
     Remove all conversations within a specified range
     """
 
     driver.get("https://mobile.facebook.com/messages/?pageNum=1&selectable&see_older_newer=1")
 
-    convos = list(get_conversations(driver))
+    convos = get_conversations(driver)
 
     for convo in convos:
-        print(convo.url)
-        print(convo.name)
-        print(convo.timestamp)
+        print(convo)

deletefb/types.py

@@ -1,35 +1,39 @@
 import attr
 import uuid
-import datetime
+import pendulum
 
-def timestamp_now():
-    """
-    Returns: a timestamp for this instant, in ISO 8601 format
-    """
-    return datetime.datetime.isoformat(datetime.datetime.now())
+def convert_timestamp(text):
+    try:
+        return pendulum.from_format(text, "DD/M/YYYY")
+    except ValueError:
+        try:
+            return (pendulum.from_format(text, "DD MMM")
+                    .set(year=pendulum.now().year))
+        except ValueError:
+            return None
 
 # Data type definitions of posts and comments
 @attr.s
 class Post:
     content = attr.ib()
     comments = attr.ib(default=[])
-    date = attr.ib(factory=timestamp_now)
+    date = attr.ib(factory=pendulum.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)
+    date = attr.ib(factory=pendulum.now)
     name = attr.ib(factory=lambda: uuid.uuid4().hex)
 
 @attr.s
 class Conversation:
     url = attr.ib()
     name = attr.ib()
-    timestamp = attr.ib(default=None)
+    timestamp = attr.ib(converter=convert_timestamp)
 
 @attr.s
 class Page:
     name = attr.ib()
-    date = attr.ib(factory=timestamp_now)
+    date = attr.ib(factory=pendulum.now)

requirements.txt

@@ -1,13 +1,25 @@
 attrs==19.1.0
 bitarray==0.9.3
+bleach==3.1.0
 certifi==2018.11.29
 chardet==3.0.4
+docutils==0.14
 idna==2.8
+pendulum==2.0.5
+pkginfo==1.5.0.1
 pybloom-live==3.0.0
+Pygments==2.4.2
+python-dateutil==2.8.0
+pytzdata==2019.2
+readme-renderer==24.0
 requests==2.22.0
 requests-file==1.4.3
+requests-toolbelt==0.9.1
 selenium==3.141.0
 selenium-requests==1.3
 six==1.12.0
 tldextract==2.2.0
+tqdm==4.32.2
+twine==1.13.0
 urllib3==1.25.2
+webencodings==0.5.1

setup.py

@@ -24,7 +24,8 @@ setuptools.setup(
         "selenium-requests",
         "requests",
         "pybloom-live",
-        "attrs"
+        "attrs",
+        "pendulum"
     ],
     classifiers= [
         "Programming Language :: Python :: 3",