Sync catalog metadata

faceberg.yml

@@ -17,3 +17,8 @@ HuggingFaceFW:
     type: dataset
     repo: HuggingFaceFW/fineweb-edu
     config: default
+google-research-datasets:
+  mbpp:
+    type: dataset
+    repo: google-research-datasets/mbpp
+    config: full

google-research-datasets/mbpp/metadata/19bfe5b3-bcfd-49e7-a415-7b05a18e7d95.avro

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ce34348eb5ece72aae48d3e6ed2a55adaa4e681ea06392a6d9d43702e428d42
+size 5873

google-research-datasets/mbpp/metadata/snap-1-0-17bc0e1e-0627-4234-b4da-505d28fe6082.avro

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84643b8f77b99e971f2ccb1045e966ff5f61b8038045581c2bb8973490cddba2
+size 1781

google-research-datasets/mbpp/metadata/v1.metadata.json

@@ -0,0 +1,136 @@
+{
+  "location": "hf://spaces/kszucs/catalog/google-research-datasets/mbpp",
+  "table-uuid": "d654f8c3-6c7b-4ebc-94a3-ab1f7ca70158",
+  "last-updated-ms": 4160531913,
+  "last-column-id": 9,
+  "schemas": [
+    {
+      "type": "struct",
+      "fields": [
+        {
+          "id": 1,
+          "name": "split",
+          "type": "string",
+          "required": false
+        },
+        {
+          "id": 2,
+          "name": "task_id",
+          "type": "int",
+          "required": false
+        },
+        {
+          "id": 3,
+          "name": "text",
+          "type": "string",
+          "required": false
+        },
+        {
+          "id": 4,
+          "name": "code",
+          "type": "string",
+          "required": false
+        },
+        {
+          "id": 5,
+          "name": "test_list",
+          "type": {
+            "type": "list",
+            "element-id": 8,
+            "element": "string",
+            "element-required": false
+          },
+          "required": false
+        },
+        {
+          "id": 6,
+          "name": "test_setup_code",
+          "type": "string",
+          "required": false
+        },
+        {
+          "id": 7,
+          "name": "challenge_test_list",
+          "type": {
+            "type": "list",
+            "element-id": 9,
+            "element": "string",
+            "element-required": false
+          },
+          "required": false
+        }
+      ],
+      "schema-id": 0,
+      "identifier-field-ids": []
+    }
+  ],
+  "current-schema-id": 0,
+  "partition-specs": [
+    {
+      "spec-id": 0,
+      "fields": [
+        {
+          "source-id": 1,
+          "field-id": 1000,
+          "transform": "identity",
+          "name": "split"
+        }
+      ]
+    }
+  ],
+  "default-spec-id": 0,
+  "last-partition-id": 1000,
+  "properties": {
+    "write.parquet.compression-codec": "snappy",
+    "write.py-location-provider.impl": "faceberg.catalog.HfLocationProvider",
+    "write.data.path": "hf://datasets/google-research-datasets/mbpp/full",
+    "hf.dataset.repo": "google-research-datasets/mbpp",
+    "hf.dataset.config": "full",
+    "hf.dataset.revision": "4bb6404fdc6cacfda99d4ac4205087b89d32030c",
+    "hf.write.pattern": "{split}-{uuid}-iceberg.parquet",
+    "hf.write.split": "train",
+    "schema.name-mapping.default": "[{\"names\":[\"split\"],\"field-id\":1},{\"names\":[\"task_id\"],\"field-id\":2},{\"names\":[\"text\"],\"field-id\":3},{\"names\":[\"code\"],\"field-id\":4},{\"names\":[\"test_list\"],\"field-id\":5,\"fields\":[{\"names\":[\"element\"],\"field-id\":8}]},{\"names\":[\"test_setup_code\"],\"field-id\":6},{\"names\":[\"challenge_test_list\"],\"field-id\":7,\"fields\":[{\"names\":[\"element\"],\"field-id\":9}]}]"
+  },
+  "current-snapshot-id": 1,
+  "snapshots": [
+    {
+      "snapshot-id": 1,
+      "sequence-number": 0,
+      "timestamp-ms": 4160531913,
+      "manifest-list": "hf://spaces/kszucs/catalog/google-research-datasets/mbpp/metadata/snap-1-0-17bc0e1e-0627-4234-b4da-505d28fe6082.avro",
+      "summary": {
+        "operation": "append",
+        "added-files-size": "236069",
+        "added-data-files": "4",
+        "added-records": "974",
+        "changed-partition-count": "4",
+        "total-data-files": "4",
+        "total-delete-files": "0",
+        "total-records": "974",
+        "total-files-size": "236069",
+        "total-position-deletes": "0",
+        "total-equality-deletes": "0"
+      },
+      "schema-id": 0
+    }
+  ],
+  "snapshot-log": [],
+  "metadata-log": [],
+  "sort-orders": [
+    {
+      "order-id": 0,
+      "fields": []
+    }
+  ],
+  "default-sort-order-id": 0,
+  "refs": {
+    "main": {
+      "snapshot-id": 1,
+      "type": "branch"
+    }
+  },
+  "statistics": [],
+  "partition-statistics": [],
+  "format-version": 2,
+  "last-sequence-number": 0
+}

google-research-datasets/mbpp/metadata/version-hint.text

@@ -0,0 +1 @@
+1