Add support for handling description and long_description

[benoit74]

The following logic could be shared among Python scrapers for handling description and long_description length and the distinction between user-provided and default values:

• three parameters are provided to the function:
  • a default description (typically retrieved online from some metadata)
  • an optional user-defined description (typically a CLI arg)
  • an optional user-defined long_description (typically a CLI arg)
• the function returns:
  • a description
  • an optional long_description
• if set, the two user-defined values are checked to not exceed ZIM maximum length ; an error is thrown otherwise ; if ok, they are returned
• if user-defined description is not set, it is computed based on the default description (truncated if needed)
• if user-defined long_description is not set and default description has been truncated (because it was too long for the description field), the long_description is computed based on the default description (truncated if needed), otherwise it is not set

Code is almost already ready in kolibri

[rgaudin]

Where would this land in the API?

[benoit74]

In a function named compute_descriptions in src/zimscraperlib/inputs.py?

[rgaudin]

👍