search

opengeospatial / NamingAuthority

Primary repo for the OGC Naming Authority
6 stars 12 forks source link

Internal Server Errors for definition server #227

Closed jerstlouis closed 11 months ago

jerstlouis commented 1 year ago

These result in 500 Internal Server Errors:

https://defs.opengis.net/vocprez/object?uri=http%3A//www.opengis.net/def/derivedCRSType/OGC/1.1/vertical https://defs.opengis.net/vocprez/object?uri=http%3A//www.opengis.net/def/verticalDatumType/OGC/1.0/barometric

rob-metalinkage commented 1 year ago

We will check the new version doesn't fail this way. And fix the data. Can we identify the custodian of thus content?

jerstlouis commented 1 year ago

@rob-metalinkage Roger Lott actually mentioned that the "barometric" datum type has been abandoned in favor of the use of parameteric CRSs:

A vertical subtype barometric was included in early drafts of Topic 2 (before about 2005, maybe even earlier) but was removed as a vertical subtype and generalised into parametric before the final adoption of Topic 2 and ISO 19111 in 2007. Vertical CRS now is explicitly a geometric coordinate system in the earth's gravity vector.

I am not sure however if that is in line with the EDR approach, which introduced a "vrs" type and most likekly uses this for pressure levels. Related to #228 .

@chris-little @m-burgoyne

rob-metalinkage commented 1 year ago

Note that the Naming Authority manages the registry ( function ) so we will fix the errors. The Register "control body" needs to decide on content issues. First we need to identify the control body and initiate the decision making process.

jerstlouis commented 1 year ago

@rob-metalinkage Yes this issue was specifically about the 500 "function" error. The CRS SWG I believe would be the relevant control body in this case.

KathiSchleidt commented 1 year ago

A short note that you also have some issues on the dev version. I was discussing the GloSIS w3ids Raul Palma recently included in the dev-def server at the edge of EGU this week, we noticed that:

Seems the : must be encoded, to my feeling should be either the entirety of the passed URI, or none of it