Skip to content

Latest commit

 

History

History
24 lines (23 loc) · 2.56 KB

CHANGELOG.md

File metadata and controls

24 lines (23 loc) · 2.56 KB

CHANGELOG

  • 0.2.8
    • Add a new ExAirtable.Table.from_schema method, which is the reverse of to_schema.
  • 0.2.7
    • Add a new ExAirtable.Table.list_params callback, to allow for all list operations against Airtable to be pre-filtered, pre-sorted, etc. This can do a lot to keep data from coming down the wire, and it's easier to use list_params than to keep having to redefine Table.list and/or Table.list_async each time you want to do that.
  • 0.2.6
    • Add Airtable ID to exported table schemas. See ExAirtable.Table.to_schema and ExAirtable.Airtable.Record.to_schema for more details.
  • 0.2.5
    • Make Table.list_async overrideable in situations where you want to change the default table listing parameters.
    • Clean up params for Table.list and Table.list_async to use Keyword lists in both places (instead of a map in one and a keyword list in the other)
  • 0.2.4
    • Add to_schema() methods, to make it easier to define a conversion between Airtable field names and in-app field names. Realistically, this will likely be used most in apps that use Ecto schemas internally to validate Airtable data. See ExAirtable.Table.to_schema and ExAirtable.Airtable.Record.to_schema for more details.
  • 0.2.3
    • More fixing of bugs around table naming. Everything is picky about weird table names!
  • 0.2.2
    • Fixed a bug that prevented starting a supervisor with multiple tables.
    • Fixed a bug in URI encoding that prevented tables with crazy characters in their names (like "/") from being queried.
  • 0.2.1
    • Upgraded to Elixir 1.11.1, and fixed some bugs that came with that. Added ASDF .tool-versions to the repo to help with developers who want to pitch in.
  • 0.2.0
    • Removed the :delete_on_refresh startup option. The new solution takes inspiration from the hashing approach taken here, and removes any need to worry about the cache staying up to date. It's relatively performant, and will only churn the whole cache if the Airtable result and the local cache are different.
    • Converted the RateLimiter from two GenStage modules to a single GenServer module. This is conceptually simpler, and also runs much faster! To be clear, the problem was my misunderstanding and mis-application of GenStage, not any inherent problem with GenStage.
  • 0.1.0 - Initial commit. The thing works!