Draft work:
"""
This script contains functions for counting participant participation and generating tweets.
"""
import glob
from collections import defaultdict
import yaml
def get_participants_by_year(years: list) -> dict:
"""
Returns a dictionary of sets of participants for each year in the input list.
Args:
years (list): A list of years to retrieve participants for.
Returns:
dict: A dictionary where the keys are the input years as strings and the values are sets of participants for each year.
"""
participant_sets = {}
for year in years:
episode_files = glob.glob(f"_episodes/{year}*.md")
year_participants = set()
for episode_file in episode_files:
with open(episode_file, "r") as file:
content = file.read()
front_matter = content.split("---")[1]
data = yaml.safe_load(front_matter)
badges = data.get("badges", [])
for badge in badges:
if badge["participant"]:
participant = badge["participant"]
year_participants.add("@" + participant)
participant_sets[str(year)] = tuple(year_participants)
return participant_sets
def generate_tweet(year) -> str:
"""
Generates a tweet for a given year.
Args:
year (int): The year to generate a tweet for.
Returns:
str: A tweet for the given year.
"""
participants = get_participants_by_year([year])
handles = ", ".join(participants[str(year)])
tweet = f"Hey Community Service Hour friends, it's badge-dropping time! You participated in at least one hour in {year}, so we're dropping you a free POAP badge proving your participation.\n\nLet us know where to drop it by adding your ethereum address to https://github.com/community-service/hour.gg/edit/main/_data/participants.yml. Let @fulldecent or I know if you have any trouble. Thanks, and happy collecting.\n\n{handles}"
return tweet
# TODO: This seems to double-count somewhere. For instance, @fullddecent returned 50 partipacations for 2023 even before there has been 50 weeks in 2023.
# def count_participation(years: list) -> dict:
# """
# Count the number of times each participant stayed to the end of an episode.
# Parameters
# ----------
# years : list
# The years to count participation for.
# Returns
# -------
# dict
# A dictionary of participant names and the number of times they stayed to the end of an episode.
# """
# # Initialize a dictionary to store participant counts
# participant_counts: defaultdict = defaultdict(list)
# # Iterate over each year
# for year in years:
# # Use glob to get a list of all episode files from the year
# episode_files = glob.glob(f"_episodes/{year}*.md")
# # Initialize a dictionary to store participant counts for the year
# year_participant_counts: dict = {}
# # Iterate over each episode file
# for episode_file in episode_files:
# with open(episode_file, "r") as file:
# # Read the file content
# content = file.read()
# # Extract the YAML front matter
# front_matter = content.split("---")[1]
# # Load the YAML as a Python dictionary
# data = yaml.safe_load(front_matter)
# # Extract the list of badges
# badges = data.get("badges", [])
# # Iterate over each badge
# for badge in badges:
# # If participant is present, increment the participant's count
# if badge["participant"]:
# participant = badge["participant"]
# if participant in year_participant_counts:
# year_participant_counts[participant] += 1
# else:
# year_participant_counts[participant] = 1
# # Add the year's participant counts to the overall participant counts
# for participant, count in year_participant_counts.items():
# participant_counts[participant].append((year, count))
# return dict(participant_counts)
if __name__ == "__main__":
year = user_input = input("Enter a year: ")
print(generate_tweet(year))
# Simply add a line to the file with your ethereum address.
# dtedesco1:
# x: dtedesco1
# name: Daniel Tedesco
# ethereum: 0x072408eA37972B83720693D158a85D98A8316340Another approach like:
for episode in _episodes/2023-*.md; do yq --front-matter=extract '.badges[].participant' $episode; done | sort -u
# Will get easier after https://github.com/mikefarah/yq/issues/1737Easiest to reuse the chapter markers in the existing audio files. Then repull audio from the mixed video files.
#!/bin/bash
# Loop over episode numbers
for i in {3..97}
do
# Find the corresponding video file for the episode, whether it is .mov or .m4v
VIDEO_FILE=$(find . -type f \( -name "*-episode-$i.m4v" -o -name "*-episode-$i.mp4" \) -print -quit)
# Find the corresponding audio file for the episode
AUDIO_FILE=$(find . -type f -name "*-episode-$i.m4a" -print -quit)
# If both files exist, process them
if [[ -n $VIDEO_FILE && -n $AUDIO_FILE ]]; then
# Extract the base name for the output file
BASE_NAME=$(basename "$VIDEO_FILE")
BASE_NAME="${BASE_NAME%.*}"
# Define the output file
OUTPUT_FILE="${BASE_NAME}-REMASTER.m4a"
# Execute FFmpeg command with increased thread_queue_size
ffmpeg -thread_queue_size 512 -i "$VIDEO_FILE" -vn -acodec aac -ac 2 -ar 44100 -b:a 160k -af "loudnorm=I=-16:TP=-1:LRA=11:print_format=json" -f matroska - | ffmpeg -thread_queue_size 512 -i - -i "$AUDIO_FILE" -map_metadata 1 -codec copy "$OUTPUT_FILE"
echo "Processed $VIDEO_FILE and $AUDIO_FILE into $OUTPUT_FILE"
else
echo "Missing files for episode $i, skipping..."
fi
doneUpload to media.phor.net.
Update feed.
# Loop over each .m4a file in the current directory
for file in *.m4a; do
# Extract the base name without the extension for the markdown file
EPISODE="${file%.m4a}"
# Get the size of the episode file
SIZE=$(stat -f%z "$file")
# Update the markdown file with the new size
sed -i '' "s/enclosure-length:.*/enclosure-length: $SIZE/" ~/Sites/hour.gg/_episodes/"$EPISODE.md"
# Get the duration of the episode file, cutting off the decimal part
DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$file" | cut -d. -f1)
# Update the markdown file with the new duration
sed -i '' "s/itunes-duration:.*/itunes-duration: $DURATION/" ~/Sites/hour.gg/_episodes/"$EPISODE.md"
donemv _episodes/2023-10-31-episode-100 _episodes/2023-11-07-episode-100 sed -i '' 's/2023-10-31/2023-11-07/g' _episodes/2023-11-07-episode-100.md sed -i '' 's/31 Oct 2023/07 Nov 2023/g' _episodes/2023-11-07-episode-100.md