Changed formatting, added some comments, moved config file location

.gitignore

@@ -1,3 +1,7 @@
 *.conf
 *.txt
-log/
+log/
+*.bak
+bin/
+lib/
+*.cfg

discorss.py

@@ -12,35 +12,48 @@
 
 import requests
 import feedparser
+import hashlib
 from pathlib import Path
 import json
 import time
 import os
 import re
 
-config_file_path = r"/etc/discorss.conf"
+config_dir = os.environ.get('XDG_CONFIG_HOME')
-# config_file_path = r"discorss.conf"
+if config_dir is None:
+    config_file_path = r"~/.config/discorss/discorss.conf"
+    config_dir = r"~/.config/discorss"
+else:
+    config_file_path = config_dir + r"/discorss/discorss.conf"
 log_file_path = r"/var/log/discorss"
 # log_file_path = r"./log"
 log_file_name = r"/app.log"
 # Yes, I know you "can't parse HTML with regex", but
 # just watch me.
 html_filter = re.compile(r"\<\/?([A-Za-z \"\=])*\>")
-success_codes = ['200', '201', '202', '203', '204', '205', '206']
+success_codes = ["200", "201", "202", "203", "204", "205", "206"]
 
 
+# This function gets and formats the brief excerpt that goes in the embed
+# Different feeds put summaries in different fields, so we pick the best
+# one and limit it to 150 characters.
+# TODO: make the character limit smarter, as to split at a natural point
 def get_description(feed):
     try:
         temporary_string = str(feed.entries[0]["summary_detail"]["value"])
         temporary_string = html_filter.sub("", temporary_string)
         desc = (
-            temporary_string[:150] if len(temporary_string) > 150 else temporary_string
+            temporary_string[:150]
+            if len(temporary_string) > 150
+            else temporary_string
         )
     except KeyError:
         temporary_string = str(feed.entries[0]["description"])
         temporary_string = html_filter.sub("", temporary_string)
         desc = (
-            temporary_string[:150] if len(temporary_string) > 150 else temporary_string
+            temporary_string[:150]
+            if len(temporary_string) > 150
+            else temporary_string
         )
     return desc
 
@@ -48,24 +61,41 @@ def get_description(feed):
 def main():
     os.environ["TZ"] = "America/Toronto"
     time.tzset()
+    # Check for log and config files/paths, create empty directories if needed
     try:
         Path(log_file_path).mkdir(parents=True, exist_ok=True)
     except FileExistsError:
-        print("This path already exists and is not a directory!")
+        print("The logfile path {} already exists and is not a directory!".format(log_file_path))
-    # Load and read the config file
     if not Path(config_file_path).exists():
-        print("No config file! Snarf.  Directories were created for you.")
+        print("No config file at {}! Snarf.\n{} was created for you.".format(config_file_path, config_dir))
+        Path(config_file_path).mkdir(parents=True, exist_ok=True)
         return
     with open(config_file_path, "r") as config_file:
         app_config = json.load(config_file)
     now = time.mktime(time.localtime())
-    last_check = app_config["lastupdate"]
+    try:
-    for hook in app_config["feeds"]:
+        last_check = app_config["lastupdate"]
+    except KeyError:
+        last_check = (
+            now - 21600
+        )  # first run, no lastupdate, check up to 6 hours ago
+    for i, hook in enumerate(app_config["feeds"]):
         # Get the feed
         feed = feedparser.parse(hook["url"])
         published_time = time.mktime(feed.entries[0]["published_parsed"])
         published_time = published_time + hook["offset"]
         print("Parsing feed {}...".format(hook["name"]))
+        # Hash the title of the latest post and use that to determine if it's been posted
+        new_hash = hashlib.sha3_512(
+            bytes(feed.entries[0]["title"], "utf-8")
+        ).hexdigest()
+        try:
+            if hook["lasthash"] != new_hash:
+                app_config["feeds"][i]["lasthash"] = new_hash
+            else:
+                continue
+        except KeyError:
+            app_config["feeds"][i]["lasthash"] = new_hash
         # Generate the webhook
         webhook = {
             "embeds": [
@@ -77,7 +107,10 @@ def main():
                         "name": "DiscoRSS",
                         # "url": "https://git.frzn.dev/amr/discorss",
                     },
-                    "author": {"name": str(hook["name"]), "url": str(hook["siteurl"])},
+                    "author": {
+                        "name": str(hook["name"]),
+                        "url": str(hook["siteurl"]),
+                    },
                     "fields": [
                         {
                             "name": "Excerpt from post:",
@@ -93,8 +126,8 @@ def main():
             "content-type": "application/json",
         }
         webhook_string = json.dumps(webhook)
-        # print(webhook_string)
+
-        if published_time > last_check and published_time < now:
+        if published_time > last_check:
             r = requests.post(
                 hook["webhook"], data=webhook_string, headers=custom_header
             )