-
-
Save WilCF/8e858cd3fc2aec1ee495d9e9333c1e71 to your computer and use it in GitHub Desktop.
KEXP show scraper
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import requests | |
import os | |
from datetime import datetime, timedelta | |
destination_path = "shows" | |
# ids from https://api.kexp.org/v2/shows/ | |
programs = [ | |
3, # expansions | |
5, # midnight in a perfect world | |
9, # seek and destroy | |
12, # street sounds | |
24, # pacific notions | |
] | |
def get_shows(start_time, programs): | |
shows_endpoint = "https://api.kexp.org/v2/shows/?expand=hosts&start_time_after={}&limit=200".format(start_time) | |
shows_response = requests.get(shows_endpoint) | |
assert(shows_response.status_code == 200) | |
shows = shows_response.json() | |
return [x for x in shows['results'] if x['program'] in programs] | |
def get_show(show): | |
sg_endpoint = "https://api.kexp.org/get_streaming_url/?bitrate=256×tamp={}".format(show['start_time']) | |
sg_response = requests.get(sg_endpoint) | |
assert(sg_response.status_code == 200) | |
print(sg_response.json()) | |
mp3_url = sg_response.json()["sg-url"] | |
if not mp3_url: | |
return | |
download_url(mp3_url) | |
mp3_url_next = sg_response.json()["sg-url-next"] | |
print("my other url was:", mp3_url_next) | |
if "expansions" in mp3_url_next or "seek-and-destroy" in mp3_url_next: | |
download_url(mp3_url_next) | |
# ex. 20230402210008-22-508-expansions.mp3 to KEXP - Expansions (2023-04-02).mp3 | |
def convert_server_name_to_local_name(filename): | |
y = filename[0:4] | |
m = filename[4:6] | |
d = filename[6:8] | |
show = " ".join( x.capitalize() for x in filename.split("-")[3:] )[:-4] | |
return "KEXP - {} ({}-{}-{}).mp3".format(show, y, m, d) | |
def download_url(mp3_url): | |
print(mp3_url) | |
src_filename = mp3_url.split("/")[-1] | |
dest_filename = convert_server_name_to_local_name(src_filename) | |
if os.path.exists(os.path.join(destination_path, dest_filename)): | |
print("already exists") | |
return | |
file_response = requests.get(mp3_url) | |
with open(os.path.join(destination_path, dest_filename), mode='wb') as localfile: | |
localfile.write(file_response.content) | |
if __name__ == "__main__": | |
look_date = (datetime.now() + timedelta(days=-14)).strftime("%Y-%m-%dT%H:00:00.000Z") | |
for show in get_shows(look_date, programs): | |
get_show(show) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment