Changelog

GeoLocator DP is versioned with Git tags. A data package declares the version it follows through the package-level $schema:

{
  "$schema": "https://raw.githubusercontent.com/GeoPressure/GeoLocator-DP/v1.1/geolocator-dp-profile.json"
}

GeoLocatoR upgrades a package from any earlier version when it reads one, so an archived package never needs to be rewritten by hand.

This page collects the GitHub releases in one place.

v1.1

18 September 2026 — release

Resources

Move the GeoLocator DP table schema URL from resource.$schema to resource.schema. In the Data Package standard, $schema is the profile of the resource descriptor while schema is the Table Schema describing the columns. schema accepts either a versioned table schema URL or an inline descriptor.
Require type on tabular resources, which must be table. It replaces the Data Package v1 profile: "tabular-data-resource".
Require $schema on tabular resources, which must be https://datapackage.org/profiles/2.0/dataresource.json. Implementations read it to decide which version of the standard a resource follows: without it, frictionless::version() reports a resource as Data Package v1 even inside a package that declares v2.

Table schemas

Write fieldsMatch as a string rather than an array, as the standard requires. The array form came from an error in the published Table Schema profile (frictionlessdata/datapackage#965), fixed in Data Package v2.1.
Set fieldsMatch to partial on tags, so that a package may carry extra columns on it, as pressurepaths already allowed. The other six tables keep superset, which lets a table leave out columns but not add them. fieldsMatch says nothing about which columns are required: on its own it lets a table omit any column it declares, including one carrying a required constraint, because the standard cannot yet express “the required columns must be present, the optional ones may be absent” (frictionlessdata/datapackage#1126). Presence of the required columns is therefore checked separately, by GeoLocatoR::validate_gldp().
Write primaryKey and foreignKeys fields as arrays of strings, the Data Package v2 spelling.
Remove the unique constraint on tags.tag_id, which duplicated the primary key.
Declare categories on every categorical field, giving each allowed value a label. The EURING age_class and catching_method codes were previously only partly documented in prose. measurements.sensor also carries its vocabulary term per category. The enum constraint is kept alongside, as the standard allows, because no implementation enforces categories yet.
Fix observations.sex and pressurepaths.type examples, which were not valid values of their own fields.

Profile

Select the resource schema with if/then on the resource name rather than oneOf. The accepted descriptors are unchanged, but a validation error now names the property at fault instead of reporting that the whole resource matched none of the branches.

Repository

Show primaryKey, foreignKeys, missingValues and fieldsMatch on each table page, list categories with their labels, and link SKOS terms by identifier rather than by URL. Say on each table page which columns may be left out and which must be present, rather than leaving the * marking a required column unexplained.
Fix a broken SKOS link on observations.datetime, and restore _data symlink for the measurements schema so the website renders the schema the standard ships.
Add example/datapackage.json and example/measurements.csv.
Validate the example package, and the spelling of fieldsMatch and of the keys, in tests/validate_profile_and_schemas.py.
Add this changelog page.

v1.0

8 April 2026 — release

Major

Remove all metadata from datapackage.json, which becomes a minimal local manifest containing only the schema and the list of resources.
Rename sensor pitch into mean_acceleration_z. See #24.
Remove paths.ind, paths.interp, staps.include, edges.s and edges.t (1, 2).
Drop support for non-deployed tags.
Review and fix errors in primaryKey and foreignKeys.