Local authority eng codes #275
Conversation
|
Thanks! There is a better lookup to use if you have a code, e.g. |
|
Oh cool! I use I'll remove 06cc1ff and force push this branch. |
symroe
force-pushed
the
local-authority-eng-codes
branch
from
06cc1ff
to
db08a16
Compare
| @@ -386,7 +386,7 @@ class CodeType(models.Model): | |||
| # This could be extended to a more generic data store of information on an | |||
| # object, perhaps. | |||
|
|
|||
| code = models.CharField(max_length=10, unique=True, help_text="A unique code, eg 'ons' or 'unit_id'") | |||
| code = models.CharField(max_length=100, unique=True, help_text="A unique code, eg 'ons' or 'unit_id'") | |||
There was a problem hiding this comment.
As type.code and code.code both went up to 500 when they were migrated, shall we just do the same here for consistency? And we might as well do nametype.code up to 500 as well, then they all are :)
There was a problem hiding this comment.
Pushed 4357e30 that makes these changes.
| E10000023 local-authority-eng:NYK North Yorkshire | ||
| E07000154 local-authority-eng:NOR Northampton | ||
| E10000021 local-authority-eng:NTH Northamptonshire | ||
| E06000057 local-authority-eng:NBL Northumberland |
There was a problem hiding this comment.
Should this file also have other GSS codes that match to the same local authority? e.g. here, E06000048 is also Northumberland.
There was a problem hiding this comment.
There are 8 councils with multiple GSS codes - Glasgow, East Dunbartonshire, St Albans, Welwyn Hatfield, East Hertfordshire, Stevenage, Northumberland, Gateshead.
There was a problem hiding this comment.
This of course then raises another issue as two areas in MapIt would have the same code, and /code/ currently returns an error when that happens. No reason it couldn't return a list if multiple results, though...
There was a problem hiding this comment.
#127 would presuambly 'solve' that issue, if it allowed the concept to have its own IDs (of which this would certainly be one), separate from its boundaries (which have the GSS IDs).
There was a problem hiding this comment.
I can't reproduce this. I've added a new line to the map file for E06000048/Northumberland/NBL and /code/local-authority-eng/NBL still works.
Of course, I should add the extra GSS->ID maps in the file in the local-authority-eng repo. Do you have a full list of them, or should I read through the mapit_UK_fix* scripts?
There was a problem hiding this comment.
Also Code is unique on ('area', 'type'), so I'm not sure how this would ever work without removing that constraint.
There was a problem hiding this comment.
Ah, right, yes, so the example I gave was the same code in areas of two different types (OLF and OLG). Probably easiest for this for now if we just put them on the "latest" GSS codes (ie. what you have already done) and then raise a new ticket to do something about it? This will have a deadline of as soon as ONS issue a new GSS code for a local authority...
There was a problem hiding this comment.
That's quite soon! But I agree, this should be a different issue. Are you happy to write it, as you will have a better idea of how to express it. I'm happy to have a go at working on it if you like.
There was a problem hiding this comment.
Is it? They don't do it very often, or we'd probably have had to do something properly sooner! :) I've added a comment on #127 whereby we could add parent areas to the councils that remain stable through such boundary changes, then these codes could theoretically be placed on there instead...
There was a problem hiding this comment.
Oh sorry, I'm thinking about wards, where there are 13 counties getting new ward codes next May. That doesn't change the council geography GSS codes though, so we're fine.
This is because the name of the 'local-authority-eng' is longer than the previous 10 characters allowed.
symroe
force-pushed
the
local-authority-eng-codes
branch
from
db08a16
to
fac826a
Compare
|
|
||
| # Add Codes | ||
| data_path = os.path.abspath( | ||
| os.path.join('mapit_gb', 'data', 'gss-to-local-authority-eng.tsv')) |
There was a problem hiding this comment.
This means it only works if you run manage.py from inside its own directory? Could it use __file__ in some way so that it doesn't matter - there are some other uses in this repo.
| area=gss_code.area | ||
| ) | ||
| except Code.DoesNotExist: | ||
| pass |
There was a problem hiding this comment.
Presumably we want to know if we're missing a GSS code? :)
There was a problem hiding this comment.
How do you want to know? Is there some logging system, or is printing to stdout ok?
There was a problem hiding this comment.
Yeah, printing out is find, probably to self.stderr, I imagine?
There was a problem hiding this comment.
Ok, I've made that change :)
symroe
force-pushed
the
local-authority-eng-codes
branch
2 times, most recently
from
ee5c713
to
266ec0b
Compare
symroe
force-pushed
the
local-authority-eng-codes
branch
from
266ec0b
to
9f6f2eb
Compare
This PR adds codes from the local-authority-eng register.
I checked the map file in to the data directory as the remote location of that file isn't stable, even though the content is.
The thing I'm least sure about is 06cc1ff. Guessing at the CodeType from a regex seems to work for the other IDs, but a three letter code isn't unique to
local-authority-eng(for example, we havelocal-authority-wlscoming, with the same ID system).One option would be to convert the query to an
__inquery, with each regex block appending to the list of CodeTypes. This feels like a fairly large change with possible performance implications, so I thought it better to add it as it's own commit that I can remove if it's too tricky to get done in this PR.