Add donna hay

[mlduff]

Resolves #1150

No schema support. Most functions are supported (except for times).

Worked on collaboratively by myself, @heathrampazis , @a1831319 and @Mooree003.

[jayaddison]

Thanks all! And apologies for taking a while here; I plan to review this within the next 24h or so.

[jayaddison]

This is looking pretty good! I have two requests after reading through the code:

• Could we try retrieving the recipe title from one of the other elements on the page, and filtering out the pipe (|) and subsequent content from that? I think that would make for more readable recipe titles.
• To confirm that the ingredient_groups functionality works as expected, could we add another test case for a recipe that involves ingredient groupings?

[a1831319]

• To confirm that the ingredient_groups functionality works as expected, could we add another test case for a recipe that involves ingredient groupings?

Can do, I'll use https://www.donnahay.com.au/recipes/snacks-and-sides/smoky-eggplant-dip-with-hand-cut-potato-chips as the target if that works?

[jayaddison]

Sounds good - thanks, @a1831319!

[jayaddison]

* To confirm that the `ingredient_groups` functionality works as expected, could we add another test case for a recipe that involves ingredient groupings?

Resolved, thank you @a1831319 @mlduff!

* Could we try retrieving the recipe `title` from one of the other elements on the page, and filtering out the pipe (`|`) and subsequent content from that?  I think that would make for more readable recipe titles.

This isn't completely resolved yet - could we use the HTML <title> element, or og:title from a meta tag to retrieve the recipe title instead?

[jayaddison]

Suggested change

	return self.soup.find("title").text.split("|")[0].strip()
	html_title = self.soup.find("title")
	recipe_title, _, _ = html_title.text.partition("|")
	return recipe_title.strip()

Edit: call str.partition instead of str.rpartition

[jayaddison]

Ah, nope.. not quite correct. rpartition would return an empty result when | is not found in the string.

[jayaddison]

(updated/fixed to use str.partition instead)

[jayaddison]

Looks good to me! Thank you @a1831319 @heathrampazis @mlduff @Mooree003!

Ready to merge once the merge conflict in __init__.py is resolved; the str.partition usage suggestion is optional.

[jayaddison]

This pull request is generally ready I think - just some merge conflicts to resolve.

There's a small cleanup opportunity here too - after #1174 we don't need these allow-untyped-defs mypy directives, so this can be removed from the file header.

[jayaddison]

Please note: we've begun checking for a preferred ordering of the JSON key names (not alphabetical; more like priority/review-aid based).

After merging recent changes into your branch, one of the unit tests may begin complaining about the JSON files because of that. There is however a script provided that can automatically fix them -- running python scripts/reorder_json_keys.py should do that for you.