Upload schemas/form.json with huggingface_hub

schemas/form.json

@@ -0,0 +1,66 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "german-ocr-3/schemas/form.json",
+  "title": "GermanOCR3 Form (Formular)",
+  "description": "Generisches Schema fuer ausgefuellte deutsche Formulare. Felder werden als Liste von Label/Value-Paaren geliefert, plus optionale Checkbox-Liste und Unterschriften.",
+  "type": "object",
+  "additionalProperties": false,
+  "required": ["document_type", "language", "fields"],
+  "properties": {
+    "document_type": {"const": "form"},
+    "language": {"type": "string", "default": "de"},
+    "form_title": {"type": ["string", "null"]},
+    "form_id": {"description": "z.B. Antragsnummer / Formularkennung", "type": ["string", "null"]},
+    "issuing_authority": {"type": ["string", "null"]},
+
+    "fields": {
+      "type": "array",
+      "description": "Erkannte Label/Wert-Paare in Lesereihenfolge.",
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["label", "value"],
+        "properties": {
+          "label": {"type": "string"},
+          "value": {"type": ["string", "number", "boolean", "null"]},
+          "section": {"type": ["string", "null"]},
+          "page": {"type": ["integer", "null"]}
+        }
+      }
+    },
+
+    "checkboxes": {
+      "type": "array",
+      "default": [],
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "required": ["label", "checked"],
+        "properties": {
+          "label": {"type": "string"},
+          "checked": {"type": "boolean"},
+          "section": {"type": ["string", "null"]}
+        }
+      }
+    },
+
+    "signatures": {
+      "type": "array",
+      "default": [],
+      "items": {
+        "type": "object",
+        "additionalProperties": false,
+        "properties": {
+          "by": {"type": ["string", "null"]},
+          "place": {"type": ["string", "null"]},
+          "date": {"type": ["string", "null"]},
+          "present": {"type": "boolean"}
+        }
+      }
+    },
+
+    "raw_text": {"type": ["string", "null"]},
+    "confidence": {"type": ["number", "null"], "minimum": 0, "maximum": 1},
+    "notes": {"type": "array", "items": {"type": "string"}, "default": []}
+  }
+}