Grabbing your starred episodes from the Overcast OPML data export

overcastXMLExtract.py

#! /bin/env/python

import xml.etree.ElementTree as ET
from datetime import datetime
import urllib.parse
import webbrowser


# Strongly recommend you run `grep userRecommendedDate overcast.opml | wc -l` to know how many entries you are about to make...

# CHANGE BELOW AS NEED BE
# Last run date will allow you to skip processing anything from before this date
LAST_RUN_DATE=datetime.fromisoformat('2022-01-07')
# Change overcast.opml to the file name
tree = ET.parse('overcast.opml')

starredEpisodes = tree.iterfind("./body/outline[@text='feeds']/outline/outline[@userRecommendedDate]")

for episode in starredEpisodes:
    recommendDate = datetime.strptime(episode.attrib['userRecommendedDate'], '%Y-%m-%dT%H:%M:%S%z')
    if recommendDate.date() < LAST_RUN_DATE.date():
        continue

    episodeTitle = episode.attrib['title']

    overcastURL = episode.attrib['overcastUrl']
    overcastID = episode.attrib['overcastId']

    # Now we need to backtrack to the parent xml to get:
        # htmlURL = episode URL
        # title = episode Name
    podcast = tree.find("./body/outline[@text='feeds']/outline/outline[@overcastId='%s']/.." % overcastID)

    podcastTitle = podcast.attrib['title']

    podcastURL = ''
    if episode.attrib['url']:
        podcastURL = episode.attrib['url']
    else:
        podcastURL = podcast.attrib['htmlUrl']

    hyphenatedName = podcastTitle.replace(" ", "-").lower()

    # Construct the Text for Bear
    noteText = """
<Things I Liked>
---
###### %s - %s

%s
%s

#draft #public #podcast #%s
""" % (podcastTitle, episodeTitle, overcastURL, podcastURL, hyphenatedName)

    noteTitle = "%s - %s" % (podcastTitle, episodeTitle)
    bearURL = "bear://x-callback-url/create?title=%s&open_note=no&show_window=no&text=%s" % (urllib.parse.quote(noteTitle), urllib.parse.quote(noteText))
    print(bearURL)

    webbrowser.open(bearURL, new=0)