🇬🇧 UK Schema

[AndrewBarber]

Is your feature request related to a problem? Please describe.
It would be great to have UK Breweries on the DB too!

Directory Structure
In the UK we have all the individual countries then individual county areas, so what would be the best directory tree for this?
Would something like:

openbrewerydb/data/uk/scotland/west-dunbartonshire.csv
openbrewerydb/data/uk/scotland/argyll-bute.csv

or

openbrewerydb/data/uk/west-dunbartonshire/west-dunbartonshire.csv   
openbrewerydb/data/uk/argyll-bute/argyll-bute.csv

or

openbrewerydb/data/uk/scotland/west-dunbartonshire/west-dunbartonshire.csv   
openbrewerydb/data/uk/scotland/argyll-bute/argyll-bute.csv

be best?

CSV Structure
Additionally, what would be the best layout for the CSV be? Maybe:
id,name,brewery_type,street,city,**county**,postal_code,website_url,phone,created_at,updated_at,country,longitude,latitude,tags

Info
Unfortunately, SIBA don't provide a great data source, but I'm hoping to start the pull from there, then try and automate some further parts..

[chrisjm]

Awesome! Thanks for working on this @AndrewBarber !

For the directory structure options, I think option 2 or 3 follows the structure and I think I'm leaning towards option 3. What do you think?

Re: the CSV structure, this StackOverflow post (https://stackoverflow.com/questions/929684/is-there-common-street-addresses-database-design-for-all-addresses-of-the-world) suggests county_province which I think will cover more countries.

Additionally, does it make sense to add more address lines? ex. address_2, address_3?

[AndrewBarber]

Dir Structure - Sure 3 looks great!
CSV Structure - I think county_province works for this. I also think that address_2 & address_3 lines would work as optional fields, and would help with address structure.

[chrisjm]

Great! If you're able to work on some CSVs, I'll get the DB prepped for the extra columns.

[chrisjm]

Closed by #43

[chrisjm]

Heya, @AndrewBarber !

As I was updating the automation scripts, I decided to update the directory structure to match what's available in the CSV which made it easier to code. So now the structure is data/scotland/[county]. Also, the script splits on either county_province or state so Argyle and Bute were split.

Look it over and let me know what you think!

[AndrewBarber]

No problem on the restructure.

Re: Argyll & Bute, is one county/province. I'll need to go in and combine them on county_province. Thats a bad choice I made I guess! I'll push an update to fix this soon. 👍