example: https://alpha.geocodes.earthcube.org/dataset.html?o=/ssdb/d8a2ee48474e4b4d2ee1b7fe05858f373f200e48.jsonld

Dataset should have unique schema:name. Currently see
"http://schema.org/name": "NO_MATCH_ADO_FILE_NAME",

Noticed several 'norway' results that are flipped,
One way to check w/'location'=Norwegian in:
pd.read_html(f'https://www.geonames.org/search.html?q={location}&country=')
taking top similar results, and seeing if closer to the flipped
This would be last resort check

Look at adding pychacl as a part of integration testing.

Push some data from your local developers machine for gleaner/glcon….
Define a gihub project with a developers/test configuration. Or just add to geocodes project.

read sitemap
read set of sitemaps from csv/google doc.

Two Tabs with Test setup in the sources spreadsheet:
https://docs.google.com/spreadsheets/d/1G7Wylo9dLlq3tmXe8E8lZDFNKFDuoIEeEZd3epS0ggQ/edit#gid=1340502269

We need to be sure that items get pushed.

For a record or two here, and for a record or two of the ecrr_submitted form. can we get a sparql query

maybe just an ask query that we could template as a last step.

servers to test:
graph. geocodes. earthcube .org
graph. geodex . org
the fuskei triplestore

fake a headless... can we create github page with a jsonld that pulls in a "application/json+ld" file for headless testing.

duplicate of: gleanerio/gleaner#134

Sitemaps from a GitHub directory.
creating an action, would be a great step

https://github.com/marketplace/actions/generate-sitemap
seems to be for a website, and not a GitHub repository

Can we utilize the Research Object Create as a way send information to tools, aka a collection of tools and data

If there are multiple versions of a resource, newer versions should have 'related item' link to versions that it supersedes

An 'organization' is a 'sitemap' and each row a google sheet references a sitemap... so we need multiple directories to be testable sitemaps.
Need a couple of CSV descriptions for the HTML directory, and Dataset.

• What terms can be used to retrieve data,
• properties that should be there (name, keywords, distribution counts) .
• what tools might match)

Future: {Gleaner|nabu}/Unit Test not based on sitemap, but individual files

Test: html directory to test for extracting JSONLD

• Single JSONLD block
• Multiple JSONLD Block
• no JSONLD

Test: Dataset Directory

• files that should load.
• Files that should fail
• FIles that should be corrected (http https)

Test Resource Directory

Link to #6

pass/fail
html/json
what search would retrieve the result
is there a tool linkage that would be expected
others