Media Fetcher

Problem statement

Solution

media-fetcher performs fetching of images and videos performance from various data sources.

Deliverable (implementation)

media-fetcher is implemented as a:

• library - Use it in your projects with a help of media_fetching.MediaFetchingService class.
• CLI tool - media-fetcher tool is available to be used in the terminal.

Deployment

Prerequisites

• Python 3.10+
• Google Ads API enabled and configured
• YouTube Data API enabled and configured
• (Optional) If using tagging modules - media-tagger configured

Installation

Install media-fetcher with pip install media-fetching command.

Usage

Fetching media

Once media-fetcher is installed you can call it:

media-fetcher \
  --source <MEDIA_SOURCE> \
  --media-type <MEDIA_TYPE> \
  --extra_info <EXTRA_INFO_MODULES> \
  --writer <WRITER_TYPE> \
  --output <OUTPUT_FILE_NAME>

where:

• <SOURCE> - source of media data:
  • googleads - fetch all assets from a Google Ads account / MCC.
  • file - fetch all assets with their tags and metrics from CSV files
  • youtube - fetch public videos from a YouTube channel.
• <MEDIA_TYPE> - type of media (YOUTUBE_VIDEO, VIDEO, IMAGE).
• <WRITER_TYPE> - writer identifier (check available options at garf-io library).
• <OUTPUT_FILE_NAME> - name of the file to store results of tagging (by default media_results).

Source customizations

Source customizations are performed via --source.parameters=value syntax (i.e. --googleads.account=1)

• googleads:

  Mandatory:

  • account=ACCOUNT_ID

  Optional:

  • ads_config_path=PATH-TO-GOOGLE-ADS-YAML
  • campaign-types=CAMPAIGN_TYPE
  • start-date=YYYY-MM-DD
  • end-date=YYYY-MM-DD

• youtube:

  Mandatory:

  • channel=YOUTUBE_CHANNEL_ID

• file:

  Mandatory:

  • path=PATH_TO_FILE

  Optional:

  • media_identifier=IDENTIFIER_OF_MEDIA
  • metric_names=COMMA_SEPARATED_METRIC_NAMES

• bq:

  Mandatory:

  • table=FULLY_QUALIFIED_TABLE_NAME (in project.dataset.table format)

  Optional:

  • media_identifier=IDENTIFIER_OF_MEDIA
  • metric_names=COMMA_SEPARATED_METRIC_NAMES

• sqldb:

  Mandatory:

  • connection_string=DATABASE_CONNECTION_STRING (in SQLAlchemy format)
  • table=TABLE_NAME

  Optional:

  • media_identifier=IDENTIFIER_OF_MEDIA
  • metric_names=COMMA_SEPARATED_METRIC_NAMES

Available extra info modules

Source customizations are performed via --extra-info module.method syntax (i.e. --extra-info tagging.languages,googleads.main_geo)

Currently supported modules:

• googleads:

  • main_geo - identifies main spending country for a media.
  • approval_rate - calculates approval rate (from 0 to 1) for each media.

• tagging:

  • language - identifies language of a media.

• youtube:

  • language - identifies language of YouTube Video based on YouTube Data API.