feat: emit only nameIdentifier on creators[].nameIdentifiers[] (0.9.2)

Stop writing nameIdentifierScheme and schemeURI onto name identifiers; the schema requires only nameIdentifier and both derive from the identifier URL. ORCID detection and dedup now key off the orcid.org URL (orcid.py, identifiers.py, extract.py), so name + affiliation ORCID enrichment is unchanged.

Author: jimnoneill

CHANGELOG.md

@@ -5,6 +5,12 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.9.2] - 2026-06-05
+
+### Changed
+
+- **`creators[].nameIdentifiers[]` now carry only `nameIdentifier`.** poster2json no longer writes `nameIdentifierScheme` or `schemeURI` onto name identifiers. The schema requires only `nameIdentifier`, and both fields are derivable from the identifier URL downstream. ORCID detection and dedup now key off the `orcid.org` URL instead of the scheme field, so name + affiliation ORCID enrichment is unchanged. Any scheme/schemeURI the model emits on name identifiers is stripped.
+
 ## [0.9.1] - 2026-06-05
 
 ### Fixed

poster2json/extract.py

@@ -1946,7 +1946,10 @@ def _needs_orcid_enrichment(creators) -> bool:
         if not has_aff:
             continue
         has_orcid = any(
-            isinstance(ni, dict) and ni.get("nameIdentifierScheme") == "ORCID"
+            isinstance(ni, dict) and (
+                ni.get("nameIdentifierScheme") == "ORCID"
+                or "orcid.org" in str(ni.get("nameIdentifier", "")).lower()
+            )
             for ni in c.get("nameIdentifiers") or []
         )
         if not has_orcid:

poster2json/identifiers.py

@@ -235,18 +235,15 @@ def _enrich_existing_identifiers(data: dict) -> dict:
             if not isinstance(ni, dict):
                 continue
             value = ni.get("nameIdentifier", "")
-            result = infer_identifier_scheme(value)
-            if result:
-                scheme, uri = result
-                if "nameIdentifierScheme" not in ni or not ni["nameIdentifierScheme"]:
-                    ni["nameIdentifierScheme"] = scheme
-                if "schemeURI" not in ni or not ni["schemeURI"]:
-                    ni["schemeURI"] = uri
-                # Fix 1: bare ORCIDs -> full URL format
-                if scheme == "ORCID":
-                    m = ORCID_RE.search(value)
-                    if m and not value.startswith("http"):
-                        ni["nameIdentifier"] = f"https://orcid.org/{m.group(1)}"
+            # nameIdentifiers carry only `nameIdentifier`; drop any scheme /
+            # schemeURI the model emitted (schema requires only nameIdentifier).
+            ni.pop("nameIdentifierScheme", None)
+            ni.pop("schemeURI", None)
+            # Normalize a bare ORCID to full URL form.
+            if isinstance(value, str):
+                m = ORCID_RE.search(value)
+                if m and not value.startswith("http"):
+                    ni["nameIdentifier"] = f"https://orcid.org/{m.group(1)}"
 
     # creators[*].affiliation[*].affiliationIdentifier -- normalize bare ROR IDs
     for creator in data.get("creators", []):
@@ -358,8 +355,6 @@ def _add_extracted_identifiers(
                         creators[i].setdefault("nameIdentifiers", [])
                         creators[i]["nameIdentifiers"].append({
                             "nameIdentifier": f"https://orcid.org/{orcid}",
-                            "nameIdentifierScheme": "ORCID",
-                            "schemeURI": "https://orcid.org",
                         })
             else:
                 # Attach all to first creator
@@ -368,8 +363,6 @@ def _add_extracted_identifiers(
                     for orcid in new_orcids:
                         creators[0]["nameIdentifiers"].append({
                             "nameIdentifier": f"https://orcid.org/{orcid}",
-                            "nameIdentifierScheme": "ORCID",
-                            "schemeURI": "https://orcid.org",
                         })
 
     # --- Crossref Funder IDs → fundingReferences[*].funderIdentifier ---

poster2json/orcid.py

@@ -198,7 +198,14 @@ def lookup(
 
 def _creator_has_orcid(creator: dict) -> bool:
     for ni in creator.get("nameIdentifiers", []):
-        if isinstance(ni, dict) and ni.get("nameIdentifierScheme") == "ORCID":
+        if not isinstance(ni, dict):
+            continue
+        # Detect by the identifier URL, since we no longer write
+        # nameIdentifierScheme onto nameIdentifiers.
+        if ni.get("nameIdentifierScheme") == "ORCID":
+            return True
+        nid = ni.get("nameIdentifier", "")
+        if isinstance(nid, str) and "orcid.org" in nid.lower():
             return True
     return False
 
@@ -234,11 +241,7 @@ def enrich_creators_orcid(creators: list, client: OrcidClient) -> list:
             continue
         creator.setdefault("nameIdentifiers", [])
         creator["nameIdentifiers"].append(
-            {
-                "nameIdentifier": f"https://orcid.org/{orcid}",
-                "nameIdentifierScheme": "ORCID",
-                "schemeURI": "https://orcid.org",
-            }
+            {"nameIdentifier": f"https://orcid.org/{orcid}"}
         )
     return creators
 

pyproject.toml

@@ -1,7 +1,7 @@
 [tool.poetry]
 
 name = "poster2json"
-version = "0.9.1"
+version = "0.9.2"
 description = "Convert scientific posters (PDF/images) to structured JSON metadata using Large Language Models"
 
 packages = [{ include = "poster2json" }]