opendataphilly / open-data-catalog Goto Github PK

Can Open-Data-Catalog be run on MacOSX? I'm trying to use it with Postgres.app, and getting this error:

$ createlang plpython
createlang: language installation failed: ERROR:  could not open extension control file "/Applications/Postgres.app/Contents/MacOS/share/extension/plpython.control": No such file or directory

Admin form cleans data in between model saves so our m2m fields don't update correct

requirements.txt has PIL (unsupported) package.

[localhost] out: Collecting PIL==1.1.7 (from -r requirements.txt (line 2))
[localhost] out: Could not find a version that satisfies the requirement PIL==1.1.7 (from -r requirements.txt (line 2)) (from versions: )
[localhost] out: No matching distribution found for PIL==1.1.7 (from -r requirements.txt (line 2))

It appears that the stewards of Data.gov have set up ways for city data portals to integrate and have their data sets listed in the Cities community.

http://project-open-data.github.io/catalog/

In essence, integration involves generating a compliant data.json that lives at the root URL of participating sites. It would be awesome if there was a way to generate a data.json file for ODC sites.

There is a bit of data required to seed the database in a meaningful way. Now that South is installed, it would be great for a migration to add those bits of data. The things I've come across are:

• flatpages [/about/, /faq/, ...]
• update frequencies [daily, weekly, monthly, quarterly, ...]
• url types [API, application, data, documentation, image]
• Data types / formats [CSV, CSV with header, JSON, Shapefile, XML]

There may be others.

There are a few places that still use OpenDataPhilly as the site name.

/ # (the Twitter account)
/accounts/register/
/ideas/
/opendata/nominate/
...

Maybe use the Sites framework and pass some site_name variable to the templates to help white-label the platform.

Base on the code changes made here 1ab5825

I see the endpoint just returns an array of datasets as oppose following the structure specified here https://project-open-data.cio.gov/v1.1/schema/ . That breaks implementation on our side cos we're getting a json array instead of the expected json object. Can we make changes to follow the standard structure :

{
  "@context": "https://project-open-data.cio.gov/v1.1/schema/catalog.jsonld",
  "@id": "https://data.chattlibrary.org/data.json",
  "@type": "dcat:Catalog",
  "conformsTo": "https://project-open-data.cio.gov/v1.1/schema",
  "describedBy": "https://project-open-data.cio.gov/v1.1/schema/catalog.json",
  "dataset": []
}

Or if changing will break impl for existing clients, We can add a new endpoint that follows the above format.

With most hosting providers not supporting plpythonu, is there a way to get a different sql function in place?

pycsw requires access to spatial functions in Shapely when performing spatial queries.

When odc is deployed using sqlite3, pycsw binds these inline.

When odc is deployed using PostgreSQL, pycsw requires plpythonu functions to be created in the database (see https://github.com/geopython/pycsw/blob/master/sbin/pycsw-admin.py#L172 for a standalone version).

Hey Azavea -- thanks a ton for this project! It is very nicely designed. I'm wondering if you have the PSDs or equivalent source files for the various nicely designed buttons and icons in the project as they would help when rebranding for other cities.

Cheers,

Max

I don't think this typo has made its way to the Open-Data-Catalog codebase,
but I'm posting this here just in case. I've also emailed [email protected].

http://opendataphilly.org/api-doc/ provides conflicting routes
for the API search endpoint; the two endpoints are

http://www.opendataphilly.org/api/resources/search?qs={keyword}
http://www.opendataphilly.org/api/resource/search/?qs={keyword}

The first of these is correct, and the second is not;
"resources" must be plural, and there may be no trailing slash.