scrapy-podcast-rss

This package provides a Scrapy pipeline and items to generate a podcast RSS feed from scraped information. It also allows to save the content locally or in an S3 bucket. You can then point your podcast player to the URL of the file and listen to its content.

Installation

Install scrapy-podcast-rss using pip:

$ pip install scrapy-podcast-rss

Configuration

1. You need to define OUTPUT_URI in your settings.py file, this will determine where your feed will be stored. For example:

   OUTPUT_URI = './my-podcast.xml'  # Local file.
   OUTPUT_URI = 's3://my-bucket/my-podcast.xml'  # S3 bucket (read note on S3 storage).
   

2. Add PodcastPipeline in ITEM_PIPELINES in your settings.py file:

   ITEM_PIPELINES = {
       'scrapy_podcast_rss.pipelines.PodcastPipeline': 300,
   }
   

Usage

scrapy-podcast-rss defines two special items:

• PodcastDataItem: Stores information about the podcast.
• PodcastEpisodeItem: Stores information about each episode of the podcast.

You must yield one PodcastDataItem and one PodcastEpisodeItem for each episode you want to export, before your spider closes.

Here is the information you can currently store in each item (you need to use the same names):

• PodcastDataItem:
  • title: Title of the podcast.
  • description: Description of the podcast.
  • url: URL referencing the podcast.
  • image_url: Main image of the podcast.
• PodcastDataItem:
  • title: Title of the episode.
  • description: Description of the episode.
  • publication_date: Date of publication (datetime object with timezone).
  • audio_url: URL of the audio.
  • guid: Unique identifier of the episode.

Example

You can find a minimal example of a spider using this package here: scrapy-podcast-rss-example.
You can also find an example of the package being used in an AWS Lambda function here: scrapy-podcast-rss-serverless.

Spider example

import datetime
import scrapy
import pytz
from scrapy_podcast_rss import PodcastEpisodeItem, PodcastDataItem


class SimpleSpider(scrapy.Spider):
    name = "simple_spider"
    start_urls = ['http://example.com/']
    custom_settings = {
        'OUTPUT_URI': './my-podcast.xml',
        'ITEM_PIPELINES': {'scrapy_podcast_rss.pipelines.PodcastPipeline': 300, }
    }

    def parse(self, response):
        podcast_data_item = PodcastDataItem()
        podcast_data_item['title'] = "Podcast title"
        podcast_data_item['description'] = "Description of the podcast."
        podcast_data_item['url'] = "Podcast's URL"
        podcast_data_item['image_url'] = "https://live.staticflickr.com/4211/35400224382_9edcb984e5_c.jpg"  # Sample image
        yield podcast_data_item

        episode_item = PodcastEpisodeItem()
        episode_item['title'] = "Episode title"
        episode_item['description'] = "Episode description"
        pub_date_tz = datetime.datetime.strptime("01/01/2020", "%m/%d/%Y").replace(tzinfo=pytz.UTC)
        episode_item['publication_date'] = pub_date_tz  # Publication date NEEDS to have a TIME ZONE.
        episode_item['guid'] = "Episode guid"  # Simulated identifier.
        episode_item['audio_url'] = "https://ia801803.us.archive.org/13/items/MOZARTSerenadeEineKleineNachtmusikK." \
                                    "525-NEWTRANSFER01.I.Allegro/01.I.Allegro.mp3 "  # Sample audio url.
        yield episode_item

Note on using S3 as storage

To use S3 storage locations, you can install scrapy-podcast-rss by doing:

$ pip install scrapy-podcast-rss[s3_storage]

This will simply include boto3 in the dependencies.
Once installed, you will need to have your credentials configured (boto3 quickstart guide).