From 161636840695476a330d3d6e13aeba6f7e6ec74d Mon Sep 17 00:00:00 2001 From: riyap03 Date: Sun, 31 Aug 2025 16:36:41 +0530 Subject: [PATCH] Clarify IRI comparison for identity tokens using Simple String Comparison (#1623) --- specs/jsonschema-core.md | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/specs/jsonschema-core.md b/specs/jsonschema-core.md index 4d629d3f..98282304 100644 --- a/specs/jsonschema-core.md +++ b/specs/jsonschema-core.md @@ -978,9 +978,18 @@ To differentiate between schemas in a vast ecosystem, schema resources are identified by [absolute IRIs](https://www.rfc-editor.org/rfc/rfc3987.html#section-2.2) (without fragments). These identifiers are used to create references between -schema resources. When comparing IRIs for the purposes of resource -identification, implementations SHOULD first follow the IRI normalization -procedures defined in [RFC 3987][rfc3987], section 5.3. +schema resources. ### IRI comparison for identity tokens + +When IRIs are used as **identity tokens** (not locators), implementations **MUST use Simple String Comparison** as defined in RFC 3987 section 5.3.1. +Normalization or scheme-based comparison is **not required** in this context. + +Example: + +- `https://example.com/schema/α` +- `https://example.com/schema/α` + +These two IRIs are considered identical only if their strings match exactly, without performing additional normalization. + Several keywords can accept a relative IRI reference, or a value used to construct a relative IRI reference. For these keywords, it is necessary