chore: Changed warning text, and some logging values

Now if you want to add something to the end of the description,
you can pass it via the addons parameter. Also moved a line
that was duplicated to reduce the function length.

README.md

@@ -23,7 +23,7 @@ Logging was recently enabled. Make sure that the user running the script (especi
 
 To configure the script, create ~/.config/discorss/discorss.conf with the following structure:
 
-```
+```json
 {
     "feeds": [
         {
@@ -54,7 +54,7 @@ To automate feed posting, create a systemd service and timer to execute the scri
 
 Use the command `systemctl --user edit --full --force discorss.service` and then paste in something like this:
 
-```
+```systemd
 [Unit]
 Description=Discord RSS feeder
 Wants=discorss.timer
@@ -68,7 +68,7 @@ WantedBy=default.target
 ```
 
 Make sure to edit the ExecStart to point to the correct location. Then we need a systemd timer to automatically fire the script. Run `systemctl --user edit --full --force discorss.timer` and then paste in this:
-```
+```systemd
 [Unit]
 Description=Timer for DiscoRSS
 Requires=discorss.service

discorss.py

@@ -32,28 +32,39 @@ log_dir = r"/var/log/discorss"
 log_file_path = 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"]
+html_filter = re.compile(r"\<\/?([A-Za-z0-9 \:\.\-\/\"\=])*\>")
+success_codes = [200, 201, 202, 203, 204, 205, 206]
 app_config = {}
 
+# IDEA: Consider making this into a class-based program
+# This would solve a couple issues around global variables and generally
+# make things a bit neater
+
 
 # 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 250 characters.
-# TODO: make the character limit smarter, as to split at a natural point
-def get_description(feed):
+def get_description(feed, length=250, min_length=150, addons=None):
     try:
         temporary_string = str(feed["summary_detail"]["value"])
         temporary_string = html_filter.sub("", temporary_string)
-        desc = (
-            temporary_string[:250] if len(temporary_string) > 250 else temporary_string
-        )
+        while length > min_length:
+            if temporary_string[length - 1 : length] == " ":
+                break
+            else:
+                length -= 1
     except KeyError:
         temporary_string = str(feed["description"])
         temporary_string = html_filter.sub("", temporary_string)
-        desc = (
-            temporary_string[:250] if len(temporary_string) > 250 else temporary_string
-        )
+        while length > min_length:
+            if temporary_string[length - 1 : length] == " ":
+                break
+            else:
+                length -= 1
+
+    desc = temporary_string[:length]
+    if addons is not None:
+        desc = desc + str(addons)
     return desc
 
 
@@ -78,7 +89,7 @@ def setupPaths():
             Path(config_dir).mkdir(parents=True, exist_ok=True)
         except FileExistsError:
             print(
-                "The config dir {} already exists and is not a directory! Please fix manually.".format(
+                "The config dir {} already exists and is not a directory! Please fix manually. Quitting!".format(
                     config_dir
                 )
             )
@@ -108,9 +119,8 @@ def main():
         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
-        logger.info("Parsing feed %s...", hook["name"])
+    for i, hook in enumerate(app_config["feeds"]):  # Feed loop start
+        logger.debug("Parsing feed %s...", hook["name"])
         feeds = feedparser.parse(hook["url"])
         latest_post = []
         prev_best = 0
@@ -128,12 +138,15 @@ def main():
             else:
                 continue
         if bad_time is True:
-            logger.warning(
+            logger.debug(
                 "Feed %s doesn't supply a published time, using updated time instead",
                 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(latest_post["title"], "utf-8")).hexdigest()
+        # Hash the title and time of the latest post and use that to determine if it's been posted
+        # Yes, SHA3-512 is totally unnecessary for this purpose, but I love SHA3
+        new_hash = hashlib.sha3_512(
+            bytes(latest_post["title"] + str(published_time), "utf-8")
+        ).hexdigest()
         try:
             if hook["lasthash"] != new_hash:
                 app_config["feeds"][i]["lasthash"] = new_hash
@@ -184,10 +197,14 @@ def main():
         r = requests.post(hook["webhook"], data=webhook_string, headers=custom_header)
         if r.status_code not in success_codes:
             logger.error(
-                "Error %d while trying to post %s", r.status_code, hook["webhook"]
+                "Error %d while trying to post %s", r.status_code, hook["name"]
             )
+        else:
+            logger.debug("Got %d when posting %s", r.status_code, hook["name"])
 
     # End of feed loop
+
+    # Dump updated config back to json file
     app_config["lastupdate"] = now
     with open(config_file_path, "w") as config_file:
         json.dump(app_config, config_file, indent=4)