acryl-datahub-cloud 0.3.11rc0__py3-none-any.whl → 0.3.16.1rc0__py3-none-any.whl

acryl_datahub_cloud/metadata/schemas/SemanticContent.avsc

@@ -0,0 +1,123 @@
+{
+  "type": "record",
+  "Aspect": {
+    "name": "semanticContent"
+  },
+  "name": "SemanticContent",
+  "namespace": "com.linkedin.pegasus2avro.common",
+  "fields": [
+    {
+      "type": {
+        "type": "map",
+        "values": {
+          "type": "record",
+          "name": "EmbeddingModelData",
+          "namespace": "com.linkedin.pegasus2avro.common",
+          "fields": [
+            {
+              "type": "string",
+              "name": "modelVersion",
+              "doc": "Full model identifier including provider.\nExamples: bedrock/cohere.embed-english-v3, openai/text-embedding-ada-002"
+            },
+            {
+              "type": "long",
+              "name": "generatedAt",
+              "doc": "Timestamp when embeddings were generated (milliseconds since epoch)."
+            },
+            {
+              "type": [
+                "null",
+                "string"
+              ],
+              "name": "chunkingStrategy",
+              "default": null,
+              "doc": "Description of the chunking strategy used.\nExamples: sentence_boundary_400t, fixed_512_chars, paragraph"
+            },
+            {
+              "type": "int",
+              "name": "totalChunks",
+              "doc": "Total number of chunks."
+            },
+            {
+              "type": [
+                "null",
+                "int"
+              ],
+              "name": "totalTokens",
+              "default": null,
+              "doc": "Estimated total token count across all chunks."
+            },
+            {
+              "type": {
+                "type": "array",
+                "items": {
+                  "type": "record",
+                  "name": "EmbeddingChunk",
+                  "namespace": "com.linkedin.pegasus2avro.common",
+                  "fields": [
+                    {
+                      "type": "int",
+                      "name": "position",
+                      "doc": "Zero-based position/index of this chunk within the document."
+                    },
+                    {
+                      "type": {
+                        "type": "array",
+                        "items": "float"
+                      },
+                      "name": "vector",
+                      "doc": "The embedding vector for this chunk.\nDimensionality depends on the model (e.g., 1024 for Cohere v3)."
+                    },
+                    {
+                      "type": [
+                        "null",
+                        "int"
+                      ],
+                      "name": "characterOffset",
+                      "default": null,
+                      "doc": "Character offset from start of source text.\nUsed for highlighting and navigation. Optional for privacy."
+                    },
+                    {
+                      "type": [
+                        "null",
+                        "int"
+                      ],
+                      "name": "characterLength",
+                      "default": null,
+                      "doc": "Character length of the chunk in source text.\nOptional for privacy-sensitive use cases."
+                    },
+                    {
+                      "type": [
+                        "null",
+                        "int"
+                      ],
+                      "name": "tokenCount",
+                      "default": null,
+                      "doc": "Estimated token count for this chunk."
+                    },
+                    {
+                      "type": [
+                        "null",
+                        "string"
+                      ],
+                      "name": "text",
+                      "default": null,
+                      "doc": "Original text of this chunk.\nOPTIONAL: May be omitted for privacy-sensitive data sources\nwhere only embeddings should be stored, not source text."
+                    }
+                  ],
+                  "doc": "A single chunk of text with its embedding vector.\n\nChunks enable semantic search over long documents by breaking them\ninto smaller, semantically coherent pieces that fit within model\ncontext windows."
+                }
+              },
+              "name": "chunks",
+              "doc": "Individual chunks with their embedding vectors."
+            }
+          ],
+          "doc": "Embedding data for a specific model.\nContains metadata about the embedding generation and the chunked vectors."
+        }
+      },
+      "name": "embeddings",
+      "doc": "Map of embedding model name to embedding data.\nKey is the model identifier (e.g., cohere_embed_v3, openai_ada_002).\nAllows storing embeddings from multiple models simultaneously."
+    }
+  ],
+  "doc": "Semantic content for enabling vector similarity search.\n\nThis aspect stores chunked text and embedding vectors for any entity that supports semantic search. Generation of the data for this aspect should be built in tight collaboration with the embedding generator during semantic search query processing. Chunk determination and generation can happen somewhat independently\nThe data in this aspect is directly passed along to the semantic search index.\n\nDesign notes:\n- Supports multiple embedding models (e.g., different providers or versions)\n- Supports chunked content for long documents\n- Text field is optional to support privacy-sensitive use cases where\n  only embeddings (not source text) are shared with DataHub"
+}

acryl_datahub_cloud/metadata/schemas/StructuredProperties.avsc

@@ -94,6 +94,75 @@
               "name": "lastModified",
               "default": null,
               "doc": "Audit stamp containing who last modified this relationship edge and when"
+            },
+            {
+              "Searchable": {
+                "/actor": {
+                  "fieldName": "structuredPropertyAttributionActors",
+                  "fieldType": "URN",
+                  "queryByDefault": false
+                },
+                "/source": {
+                  "fieldName": "structuredPropertyAttributionSources",
+                  "fieldType": "URN",
+                  "queryByDefault": false
+                },
+                "/time": {
+                  "fieldName": "structuredPropertyAttributionDates",
+                  "fieldType": "DATETIME",
+                  "queryByDefault": false
+                }
+              },
+              "type": [
+                "null",
+                {
+                  "type": "record",
+                  "name": "MetadataAttribution",
+                  "namespace": "com.linkedin.pegasus2avro.common",
+                  "fields": [
+                    {
+                      "type": "long",
+                      "name": "time",
+                      "doc": "When this metadata was updated."
+                    },
+                    {
+                      "java": {
+                        "class": "com.linkedin.pegasus2avro.common.urn.Urn"
+                      },
+                      "type": "string",
+                      "name": "actor",
+                      "doc": "The entity (e.g. a member URN) responsible for applying the assocated metadata. This can\neither be a user (in case of UI edits) or the datahub system for automation.",
+                      "Urn": "Urn"
+                    },
+                    {
+                      "java": {
+                        "class": "com.linkedin.pegasus2avro.common.urn.Urn"
+                      },
+                      "type": [
+                        "null",
+                        "string"
+                      ],
+                      "name": "source",
+                      "default": null,
+                      "doc": "The DataHub source responsible for applying the associated metadata. This will only be filled out\nwhen a DataHub source is responsible. This includes the specific metadata test urn, the automation urn.",
+                      "Urn": "Urn"
+                    },
+                    {
+                      "type": {
+                        "type": "map",
+                        "values": "string"
+                      },
+                      "name": "sourceDetail",
+                      "default": {},
+                      "doc": "The details associated with why this metadata was applied. For example, this could include\nthe actual regex rule, sql statement, ingestion pipeline ID, etc.\nAlso can include flags like 'propagated'=true or 'inferred'=true."
+                    }
+                  ],
+                  "doc": "Information about who, why, and how this metadata was applied"
+                }
+              ],
+              "name": "attribution",
+              "default": null,
+              "doc": "Information about who, why, and how this metadata was applied"
             }
           ]
         }

acryl_datahub_cloud/metadata/schemas/StructuredPropertyDefinition.avsc

@@ -7,13 +7,22 @@
   "namespace": "com.linkedin.pegasus2avro.structured",
   "fields": [
     {
-      "Searchable": {},
+      "Searchable": {
+        "fieldType": "KEYWORD",
+        "searchLabel": "qualifiedName",
+        "searchTier": 1
+      },
       "type": "string",
       "name": "qualifiedName",
       "doc": "The fully qualified name of the property. e.g. io.acryl.datahub.myProperty"
     },
     {
-      "Searchable": {},
+      "Searchable": {
+        "enableAutocomplete": true,
+        "fieldType": "WORD_GRAM",
+        "searchLabel": "entityName",
+        "searchTier": 1
+      },
       "type": [
         "null",
         "string"
@@ -306,7 +315,8 @@
       "Searchable": {
         "/time": {
           "fieldName": "createdTime",
-          "fieldType": "DATETIME"
+          "fieldType": "DATETIME",
+          "searchLabel": "createdAt"
         }
       },
       "type": [
@@ -364,7 +374,8 @@
       "Searchable": {
         "/time": {
           "fieldName": "lastModified",
-          "fieldType": "DATETIME"
+          "fieldType": "DATETIME",
+          "searchLabel": "lastModifiedAt"
         }
       },
       "type": [

acryl_datahub_cloud/metadata/schemas/StructuredPropertySettings.avsc

@@ -33,6 +33,15 @@
       "default": false,
       "doc": "Whether or not this asset should be displayed in the asset sidebar"
     },
+    {
+      "Searchable": {
+        "fieldType": "BOOLEAN"
+      },
+      "type": "boolean",
+      "name": "hideInAssetSummaryWhenEmpty",
+      "default": false,
+      "doc": "Whether or not this asset should be hidden in the asset sidebar (showInAssetSummary should be enabled)\nwhen its value is empty"
+    },
     {
       "Searchable": {
         "fieldType": "BOOLEAN"

acryl_datahub_cloud/metadata/schemas/SubscriptionInfo.avsc

@@ -15,7 +15,7 @@
       },
       "type": "string",
       "name": "actorUrn",
-      "doc": "Actor Urn",
+      "doc": "Actor Urn who will be notified ",
       "Urn": "Urn"
     },
     {
@@ -54,6 +54,12 @@
       "doc": "The subscription types."
     },
     {
+      "Searchable": {
+        "/time": {
+          "fieldName": "createdOn",
+          "fieldType": "DATETIME"
+        }
+      },
       "type": {
         "type": "record",
         "name": "AuditStamp",
@@ -119,7 +125,7 @@
       ],
       "name": "entityUrn",
       "default": null,
-      "doc": "Optional entity Urn",
+      "doc": "Optional entity Urn - top level for now (Dataset, container etc. )",
       "Urn": "Urn"
     },
     {
@@ -196,7 +202,7 @@
                     "TAG_REMOVED",
                     "TAG_PROPOSED"
                   ],
-                  "doc": "Enum to define change types that can trigger a subscription."
+                  "doc": "Enum to define change types that can trigger a notification for a subscription."
                 },
                 "name": "entityChangeType",
                 "doc": "Change type that triggers a notification for the subscription."
@@ -263,13 +269,15 @@
                           "type": "enum",
                           "symbolDocs": {
                             "EMAIL": "Email target type.",
-                            "SLACK": "Slack target type."
+                            "SLACK": "Slack target type.",
+                            "TEAMS": "Microsoft Teams target type."
                           },
                           "name": "NotificationSinkType",
                           "namespace": "com.linkedin.pegasus2avro.event.notification",
                           "symbols": [
                             "SLACK",
-                            "EMAIL"
+                            "EMAIL",
+                            "TEAMS"
                           ],
                           "doc": "A type of sink / platform to send a notification to."
                         }
@@ -335,6 +343,129 @@
                       "default": null,
                       "doc": "Email Notification Settings"
                     },
+                    {
+                      "type": [
+                        "null",
+                        {
+                          "type": "record",
+                          "name": "TeamsNotificationSettings",
+                          "namespace": "com.linkedin.pegasus2avro.event.notification.settings",
+                          "fields": [
+                            {
+                              "type": [
+                                "null",
+                                {
+                                  "type": "record",
+                                  "name": "TeamsUser",
+                                  "namespace": "com.linkedin.pegasus2avro.settings.global",
+                                  "fields": [
+                                    {
+                                      "type": [
+                                        "null",
+                                        "string"
+                                      ],
+                                      "name": "teamsUserId",
+                                      "default": null,
+                                      "doc": "The Teams user ID (internal Teams identifier).\nThis is the primary identifier used for Teams messaging."
+                                    },
+                                    {
+                                      "Searchable": {
+                                        "fieldType": "KEYWORD",
+                                        "queryByDefault": false
+                                      },
+                                      "type": [
+                                        "null",
+                                        "string"
+                                      ],
+                                      "name": "azureUserId",
+                                      "default": null,
+                                      "doc": "The Azure AD user ID (Azure Active Directory object ID).\nThis is the primary identifier for Azure AD operations and Microsoft Graph API calls."
+                                    },
+                                    {
+                                      "type": [
+                                        "null",
+                                        "string"
+                                      ],
+                                      "name": "email",
+                                      "default": null,
+                                      "doc": "The user's email address (fallback identifier).\nUsed when other IDs are not available or for email-based user resolution."
+                                    },
+                                    {
+                                      "type": [
+                                        "null",
+                                        "string"
+                                      ],
+                                      "name": "displayName",
+                                      "default": null,
+                                      "doc": "The user's display name (cached value, may be stale).\nRetrieved from Microsoft Graph API and cached for performance."
+                                    },
+                                    {
+                                      "type": [
+                                        "null",
+                                        "long"
+                                      ],
+                                      "name": "lastUpdated",
+                                      "default": null,
+                                      "doc": "The timestamp when the cached display name was last updated.\nUsed for TTL-based cache invalidation (milliseconds since epoch)."
+                                    }
+                                  ],
+                                  "doc": "Teams user information with cached display name and multiple ID support.\nFollows the same caching pattern as TeamsChannel for consistency."
+                                }
+                              ],
+                              "name": "user",
+                              "default": null,
+                              "doc": "Optional structured user information with cached metadata.\nSupports both Azure AD user ID and Teams user ID with cached display name."
+                            },
+                            {
+                              "type": [
+                                "null",
+                                {
+                                  "type": "array",
+                                  "items": {
+                                    "type": "record",
+                                    "name": "TeamsChannel",
+                                    "namespace": "com.linkedin.pegasus2avro.settings.global",
+                                    "fields": [
+                                      {
+                                        "type": "string",
+                                        "name": "id",
+                                        "doc": "The Teams channel ID (internal Teams identifier)."
+                                      },
+                                      {
+                                        "type": [
+                                          "null",
+                                          "string"
+                                        ],
+                                        "name": "name",
+                                        "default": null,
+                                        "doc": "The Teams channel display name (cached value, may be stale)."
+                                      },
+                                      {
+                                        "type": [
+                                          "null",
+                                          "long"
+                                        ],
+                                        "name": "lastUpdated",
+                                        "default": null,
+                                        "doc": "The timestamp when the cached name was last updated.\nUsed for TTL-based cache invalidation."
+                                      }
+                                    ],
+                                    "doc": "Teams channel information with cached display name."
+                                  }
+                                }
+                              ],
+                              "name": "channels",
+                              "default": null,
+                              "doc": "Optional list of structured channels with cached metadata"
+                            }
+                          ],
+                          "doc": "Microsoft Teams Notification settings for an actor."
+                        }
+                      ],
+                      "name": "teamsSettings",
+                      "default": null,
+                      "doc": "Microsoft Teams Notification Settings"
+                    },
                     {
                       "type": [
                         "null",

acryl_datahub_cloud/metadata/schemas/SubscriptionKey.avsc

@@ -5,7 +5,8 @@
     "keyForEntity": "subscription",
     "entityCategory": "core",
     "entityAspects": [
-      "subscriptionInfo"
+      "subscriptionInfo",
+      "status"
     ]
   },
   "name": "SubscriptionKey",

acryl_datahub_cloud/metadata/schemas/SystemMetadata.avsc

@@ -80,6 +80,67 @@
       "name": "version",
       "default": null,
       "doc": "Aspect version\n   Initial implementation will use the aspect version's number, however stored as\n   a string in the case where a different aspect versioning scheme is later adopted."
+    },
+    {
+      "type": [
+        "null",
+        {
+          "type": "record",
+          "name": "AuditStamp",
+          "namespace": "com.linkedin.pegasus2avro.common",
+          "fields": [
+            {
+              "type": "long",
+              "name": "time",
+              "doc": "When did the resource/association/sub-resource move into the specific lifecycle stage represented by this AuditEvent."
+            },
+            {
+              "java": {
+                "class": "com.linkedin.pegasus2avro.common.urn.Urn"
+              },
+              "type": "string",
+              "name": "actor",
+              "doc": "The entity (e.g. a member URN) which will be credited for moving the resource/association/sub-resource into the specific lifecycle stage. It is also the one used to authorize the change.",
+              "Urn": "Urn"
+            },
+            {
+              "java": {
+                "class": "com.linkedin.pegasus2avro.common.urn.Urn"
+              },
+              "type": [
+                "null",
+                "string"
+              ],
+              "name": "impersonator",
+              "default": null,
+              "doc": "The entity (e.g. a service URN) which performs the change on behalf of the Actor and must be authorized to act as the Actor.",
+              "Urn": "Urn"
+            },
+            {
+              "type": [
+                "null",
+                "string"
+              ],
+              "name": "message",
+              "default": null,
+              "doc": "Additional context around how DataHub was informed of the particular change. For example: was the change created by an automated process, or manually."
+            }
+          ],
+          "doc": "Data captured on a resource/association/sub-resource level giving insight into when that resource/association/sub-resource moved into a particular lifecycle stage, and who acted to move it into that specific lifecycle stage."
+        }
+      ],
+      "name": "aspectCreated",
+      "default": null,
+      "doc": "When the aspect was initially created and who created it, detected by version 0 -> 1 change"
+    },
+    {
+      "type": [
+        "null",
+        "com.linkedin.pegasus2avro.common.AuditStamp"
+      ],
+      "name": "aspectModified",
+      "default": null,
+      "doc": "When the aspect was last modified and the actor that performed the modification"
     }
   ],
   "doc": "Metadata associated with each metadata change that is processed by the system"

acryl_datahub_cloud/metadata/schemas/TagProperties.avsc

@@ -13,7 +13,9 @@
         "fieldNameAliases": [
           "_entityName"
         ],
-        "fieldType": "WORD_GRAM"
+        "fieldType": "WORD_GRAM",
+        "searchLabel": "entityName",
+        "searchTier": 1
       },
       "type": "string",
       "name": "name",

acryl_datahub_cloud/metadata/schemas/TestInfo.avsc

@@ -25,7 +25,8 @@
     },
     {
       "Searchable": {
-        "fieldType": "TEXT"
+        "fieldType": "TEXT",
+        "searchTier": 2
       },
       "type": [
         "null",

acryl_datahub_cloud/metadata/schemas/UpstreamLineage.avsc

@@ -94,6 +94,7 @@
               "Searchable": {
                 "fieldName": "upstreams",
                 "fieldType": "URN",
+                "hasValuesFieldName": "hasUpstreams",
                 "queryByDefault": false
               },
               "java": {
@@ -199,6 +200,14 @@
                 "doc": "The type of upstream entity"
               },
               {
+                "Searchable": {
+                  "/*": {
+                    "fieldName": "fineGrainedUpstreams",
+                    "fieldType": "URN",
+                    "hasValuesFieldName": "hasFineGrainedUpstreams",
+                    "queryByDefault": false
+                  }
+                },
                 "type": [
                   "null",
                   {

acryl_datahub_cloud/metadata/schemas/UsageFeatures.avsc

@@ -160,6 +160,16 @@
       "doc": "Overall rank of the dataset based on the the last 30 days insert/update operation count on a platform\nIt returns one plus the number of rows proceeding or equals to the current row in the ordering. "
     },
     {
+      "Searchable": {
+        "/*": {
+          "addToFilters": true,
+          "fieldName": "topUsersLast30Days",
+          "fieldType": "URN",
+          "filterNameOverride": "Top users last 30 days",
+          "hasValuesFieldName": "hasTopUsersLast30Days",
+          "queryByDefault": false
+        }
+      },
       "type": [
         "null",
         {