Merge pull request #198 from pixlie/feature/180_named_entity_extraction_with_gliner

Named entity extraction with gliner

Author: brainless

admin/src/api_types/APIEdges.ts

@@ -1,4 +1,7 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
 import type { APINodeEdges } from "./APINodeEdges";
 
+/**
+ * A map of node IDs to their outgoing edges.
+ */
 export type APIEdges = { [key in number]?: APINodeEdges };

admin/src/api_types/APIMatch.ts

@@ -1,5 +1,9 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
 
+/**
+ * A list of all outgoing edges of a node, with the ID of the node and the label of the edge.
+ * The UNIX timestamp represents when a node's edge list was last written to.
+ */
 export type APINodeEdges = {
   edges: Array<[number, string]>;
   written_at: bigint;

admin/src/api_types/APINodeEdges.ts

@@ -3,10 +3,28 @@ import type { APINodeFlags } from "./APINodeFlags";
 import type { APIPayload } from "./APIPayload";
 import type { NodeLabel } from "./NodeLabel";
 
+/**
+ * Schema for a node in any API response.
+ */
 export type APINodeItem = {
+  /**
+   * The ID of the node
+   */
   id: number;
+  /**
+   * The labels of the node. A node can have multiple labels, like tags, indexed by relevance.
+   */
   labels: Array<NodeLabel>;
+  /**
+   * The payload of the node. This can be a link, text, or a tree.
+   */
   payload: APIPayload;
+  /**
+   * The flags of the node. This can be used to indicate if the node is processed, requesting, etc.
+   */
   flags: Array<APINodeFlags>;
+  /**
+   * Unix timestamp of when the node was last written to.
+   */
   written_at: bigint;
 };

admin/src/api_types/APINodeItem.ts

@@ -1,11 +1,18 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
+import type { Classification } from "./Classification";
+import type { ClassifierSettings } from "./ClassifierSettings";
+import type { CrawlerSettings } from "./CrawlerSettings";
+import type { EntityName } from "./EntityName";
+import type { ExtractedEntity } from "./ExtractedEntity";
 import type { Link } from "./Link";
-import { WebMetadata } from "./WebMetadata";
-import type { TableRow } from "./TableRow";
 import type { ProjectSettings } from "./ProjectSettings";
-import type { CrawlerSettings } from "./CrawlerSettings";
-import type { ClassifierSettings } from "./ClassifierSettings";
+import type { TableRow } from "./TableRow";
+import type { WebMetadata } from "./WebMetadata";
 
+/**
+ * All nodes contain a data payload.
+ * APIPayload is the schema for this payload and contains the type of payload and the data.
+ */
 export type APIPayload =
   | { type: "Link"; data: Link }
   | { type: "WebMetadata"; data: WebMetadata }
@@ -14,4 +21,7 @@ export type APIPayload =
   | { type: "TableRow"; data: TableRow }
   | { type: "ProjectSettings"; data: ProjectSettings }
   | { type: "CrawlerSettings"; data: CrawlerSettings }
-  | { type: "ClassifierSettings"; data: ClassifierSettings };
+  | { type: "ClassifierSettings"; data: ClassifierSettings }
+  | { type: "Classification"; data: Classification }
+  | { type: "NamedEntitiesToExtract"; data: Array<EntityName> }
+  | { type: "ExtractedNamedEntities"; data: Array<ExtractedEntity> };

admin/src/api_types/APIPayload.ts

@@ -0,0 +1,7 @@
+// This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
+
+export type Classification = {
+  is_relevant: boolean;
+  reason: string;
+  insight_if_classified_as_relevant: string | null;
+};

admin/src/api_types/Classification.ts

@@ -1,7 +1,5 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
 
 export type ClassifierSettings = {
-  query_to_classify_content_as_relevant_or_irrelevant_to_objective:
-    | string
-    | null;
+  prompt_to_classify_content_as_relevant_to_objective_or_not: string | null;
 };

admin/src/api_types/ClassifierSettings.ts

@@ -11,6 +11,4 @@ export type EdgeLabel =
   | "Suggests"
   | "SuggestedFor"
   | "Classifies"
-  | "ClassifiedFor"
-  | "Matches"
-  | "MatchedFor";
+  | "ClassifiedFor";

admin/src/api_types/EdgeLabel.ts

@@ -1,15 +1,20 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
 import type { APIEdges } from "./APIEdges";
-import { APIMatch } from "./APIMatch";
 import type { APINodeItem } from "./APINodeItem";
 import type { Explore } from "./Explore";
 
+/**
+ * Engine's response for an API request.
+ *
+ * API requests for a project are sent to its engine.
+ * The engine responds with an `EngineResponsePayload`, which is directly passed on as
+ * the API response.
+ */
 export type EngineResponsePayload =
   | { type: "NodeCreatedSuccessfully"; data: number }
   | { type: "EdgeCreatedSuccessfully" }
   | { type: "Nodes"; data: Array<APINodeItem> }
-  | { type: "Labels"; data: Array<string> }
   | { type: "Edges"; data: APIEdges }
+  | { type: "Labels"; data: Array<string> }
   | { type: "Explore"; data: Explore }
-  | { type: "Matches"; data: Array<APIMatch> }
   | { type: "Error"; data: string };

admin/src/api_types/EngineResponsePayload.ts

@@ -1,3 +1,3 @@
 // This file was generated by [ts-rs](https://github.com/Aleph-Alpha/ts-rs). Do not edit this file manually.
 
-export type DateTimeWrapper = string;
+export type EntityName = "Person" | "Organization" | "Date" | "Place";