add pg vector and embed

prisma/migrations/20250822104301_init/migration.sql

@@ -0,0 +1,26 @@
+-- Enable pgvector extension
+CREATE EXTENSION IF NOT EXISTS vector;
+
+-- CreateTable
+CREATE TABLE "icd_codes" (
+    "id" TEXT NOT NULL,
+    "code" TEXT NOT NULL,
+    "display" TEXT NOT NULL,
+    "version" TEXT NOT NULL,
+    "category" TEXT NOT NULL,
+    "embedding" vector(1536),
+    "createdAt" TIMESTAMP(3) NOT NULL DEFAULT CURRENT_TIMESTAMP,
+    "updatedAt" TIMESTAMP(3) NOT NULL,
+
+    CONSTRAINT "icd_codes_pkey" PRIMARY KEY ("id")
+);
+
+-- Create unique index on code
+CREATE UNIQUE INDEX "icd_codes_code_key" ON "icd_codes"("code");
+
+-- Create ivfflat index for fast vector similarity search
+CREATE INDEX "icd_codes_embedding_idx" ON "icd_codes" USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);
+
+-- Add comments for documentation
+COMMENT ON COLUMN "icd_codes"."embedding" IS 'Vector embedding for semantic search using pgvector (1536 dimensions)';
+COMMENT ON INDEX "icd_codes_embedding_idx" IS 'IVFFlat index for fast cosine similarity search with 100 lists';

prisma/migrations/20250822104302_add_pgvector/migration.sql

@@ -0,0 +1,27 @@
+-- Migration: Add pgvector support to icd_codes table
+
+-- Enable pgvector extension
+CREATE EXTENSION IF NOT EXISTS vector;
+
+-- Add embedding column with pgvector type
+ALTER TABLE "icd_codes" ADD COLUMN IF NOT EXISTS "embedding" vector(1536);
+
+-- Add metadata column for LangChain pgvector
+ALTER TABLE "icd_codes" ADD COLUMN IF NOT EXISTS "metadata" JSONB;
+
+-- Add content column for LangChain pgvector
+ALTER TABLE "icd_codes" ADD COLUMN IF NOT EXISTS "content" TEXT;
+
+-- Create ivfflat index for fast vector similarity search
+CREATE INDEX IF NOT EXISTS "icd_codes_embedding_idx" ON "icd_codes" 
+USING ivfflat (embedding vector_cosine_ops) WITH (lists = 100);
+
+-- Create index on metadata for fast JSON queries
+CREATE INDEX IF NOT EXISTS "icd_codes_metadata_idx" ON "icd_codes" USING GIN (metadata);
+
+-- Add comments for documentation
+COMMENT ON COLUMN "icd_codes"."embedding" IS 'Vector embedding for semantic search using pgvector (1536 dimensions)';
+COMMENT ON COLUMN "icd_codes"."metadata" IS 'JSON metadata for LangChain pgvector operations';
+COMMENT ON COLUMN "icd_codes"."content" IS 'Text content for LangChain pgvector operations';
+COMMENT ON INDEX "icd_codes_embedding_idx" IS 'IVFFlat index for fast cosine similarity search with 100 lists';
+COMMENT ON INDEX "icd_codes_metadata_idx" IS 'GIN index for fast JSON metadata queries';

prisma/migrations/migration_lock.toml

@@ -0,0 +1,3 @@
+# Please do not edit this file manually
+# It should be added in your version-control system (e.g., Git)
+provider = "postgresql"

prisma/schema.prisma

@@ -1,12 +1,5 @@
-// This is your Prisma schema file,
-// learn more about it in the docs: https://pris.ly/d/prisma-schema
-
-// Looking for ways to speed up your queries, or scale easily with your serverless or edge functions?
-// Try Prisma Accelerate: https://pris.ly/cli/accelerate-init
-
 generator client {
   provider = "prisma-client-js"
-  output   = "../generated/prisma"
 }
 
 datasource db {
@@ -15,13 +8,18 @@ datasource db {
 }
 
 model IcdCode {
-  id        String   @id @default(uuid())
-  code      String   @unique
+  id        String                 @id @default(uuid())
+  code      String                 @unique
   display   String
   version   String
-  category  String   // "ICD9" or "ICD10"
-  createdAt DateTime @default(now())
-  updatedAt DateTime @updatedAt
+  category  String
+  embedding Unsupported("vector")?
+  createdAt DateTime               @default(now())
+  updatedAt DateTime               @updatedAt
+  metadata  Json?
+  content   String?
 
+  @@index([embedding])
+  @@index([metadata], type: Gin)
   @@map("icd_codes")
 }