Closed pgulley closed 6 months ago
My thinking was that it was better to have the endpoint accept some version of the url
instead of the ID itself, since we don't include that hash in any document produced by the api and the user would have to generate it themselves otherwise- as I said in #48. Since the endpoint accepts a URL instead of the ID, putting it in a query parameter avoids having to muck about with encoding the URL directly- although that might not be the headache I imagine it to be.
Ah, I've talked with @Evan-Leon about adding an id
column to the CSV story list download, because the research team (and the overall MC team) is used to having a unique ID as a shortcut for referring to stories under discussion. So the ID will be user-facing. We've found that helps various types of detail oriented tasks. I think this will be a stronger and more useful design if stick with this idea that there is a unique machine-oriented id that is used.
@philbudne Any blockers?
Endpoint for retrieving individual articles by their url, plus associated test. #48