Add orderBy and filter to parquetQueryWorker

Author: platypii

eslint.config.js

@@ -9,7 +9,11 @@ import typescript from 'typescript-eslint'
 export default typescript.config(
   { ignores: ['coverage/', 'dist/', 'lib/'] },
   {
-    extends: [javascript.configs.recommended, ...typescript.configs.strictTypeChecked, ...typescript.configs.stylisticTypeChecked],
+    extends: [
+      javascript.configs.recommended,
+      ...typescript.configs.strictTypeChecked,
+      ...typescript.configs.stylisticTypeChecked,
+    ],
     files: ['**/*.{ts,tsx,js}'],
     languageOptions: {
       globals: globals.browser,
@@ -49,6 +53,7 @@ export default typescript.config(
       'func-style': ['error', 'declaration'],
       indent: ['error', 2, { SwitchCase: 1 }],
       'key-spacing': 'error',
+      'keyword-spacing': 'error',
       'no-constant-condition': 'off',
       'no-extra-parens': 'error',
       'no-multi-spaces': 'error',

package.json

@@ -64,25 +64,25 @@
   },
   "devDependencies": {
     "@eslint/js": "9.29.0",
-    "@storybook/react-vite": "9.0.12",
+    "@storybook/react-vite": "9.0.13",
     "@testing-library/react": "16.3.0",
-    "@types/node": "24.0.3",
+    "@types/node": "24.0.4",
     "@types/react": "19.1.8",
     "@types/react-dom": "19.1.6",
-    "@vitejs/plugin-react": "4.5.2",
+    "@vitejs/plugin-react": "4.6.0",
     "@vitest/coverage-v8": "3.2.4",
     "eslint": "9.29.0",
     "eslint-plugin-react": "7.37.5",
     "eslint-plugin-react-hooks": "5.2.0",
     "eslint-plugin-react-refresh": "0.4.20",
-    "eslint-plugin-storybook": "9.0.12",
+    "eslint-plugin-storybook": "9.0.13",
     "globals": "16.2.0",
     "jsdom": "26.1.0",
     "nodemon": "3.1.10",
     "npm-run-all": "4.1.5",
-    "storybook": "9.0.12",
+    "storybook": "9.0.13",
     "typescript": "5.8.3",
-    "typescript-eslint": "8.34.1",
+    "typescript-eslint": "8.35.0",
     "vite": "6.3.5",
     "vitest": "3.2.4"
   },

src/lib/workers/parquetWorker.ts

@@ -51,10 +51,10 @@ self.onmessage = async ({ data }: { data: ClientMessage }) => {
       postErrorMessage({ error: error as Error, queryId })
     }
   } else {
-    const { rowStart, rowEnd, chunks } = data
+    const { rowStart, rowEnd, orderBy, filter, chunks } = data
     const onChunk = chunks ? (chunk: ColumnData) => { postChunkMessage({ chunk, queryId }) } : undefined
     try {
-      const result = await parquetQuery({ metadata, file, rowStart, rowEnd, compressors, onChunk })
+      const result = await parquetQuery({ metadata, file, rowStart, rowEnd, orderBy, filter, compressors, onChunk })
       postResultMessage({ result, queryId })
     } catch (error) {
       postErrorMessage({ error: error as Error, queryId })

src/lib/workers/parquetWorkerClient.ts

@@ -1,7 +1,7 @@
 import ParquetWorker from './parquetWorker?worker&inline'
 /// ^ the worker is bundled with the main thread code (inline) which is easier for users to import
 /// (no need to copy the worker file to the right place)
-import { ColumnData } from 'hyparquet'
+import type { ColumnData } from 'hyparquet'
 import type { Cells, ColumnRanksClientMessage, ColumnRanksWorkerMessage, ColumnRanksWorkerOptions, QueryClientMessage, QueryWorkerMessage, QueryWorkerOptions } from './types.js'
 
 let worker: Worker | undefined
@@ -66,7 +66,7 @@ function getWorker() {
  * Instead of taking an AsyncBuffer, it takes a AsyncBufferFrom, because it needs
  * to be serialized to the worker.
  */
-export function parquetQueryWorker({ metadata, from, rowStart, rowEnd, onChunk }: QueryWorkerOptions): Promise<Cells[]> {
+export function parquetQueryWorker({ metadata, from, rowStart, rowEnd, orderBy, filter, onChunk }: QueryWorkerOptions): Promise<Cells[]> {
   // TODO(SL) Support passing columns?
   return new Promise((resolve, reject) => {
     const queryId = nextQueryId++
@@ -75,7 +75,7 @@ export function parquetQueryWorker({ metadata, from, rowStart, rowEnd, onChunk }
 
     // If caller provided an onChunk callback, worker will send chunks as they are parsed
     const chunks = onChunk !== undefined
-    const message: QueryClientMessage = { queryId, metadata, from, rowStart, rowEnd, chunks, kind: 'query' }
+    const message: QueryClientMessage = { queryId, metadata, from, rowStart, rowEnd, orderBy, filter, chunks, kind: 'query' }
     worker.postMessage(message)
   })
 }

src/lib/workers/types.ts

@@ -1,4 +1,5 @@
-import { ColumnData, FileMetaData } from 'hyparquet'
+import type { ColumnData, FileMetaData } from 'hyparquet'
+import type { ParquetQueryFilter } from 'hyparquet/src/types.js'
 
 // Serializable constructors for AsyncBuffers
 interface AsyncBufferFromFile {
@@ -29,6 +30,8 @@ export interface ErrorMessage extends Message {
 export interface QueryWorkerOptions extends CommonWorkerOptions {
   rowStart?: number,
   rowEnd?: number,
+  orderBy?: string,
+  filter?: ParquetQueryFilter,
   onChunk?: (chunk: ColumnData) => void
 }
 export interface QueryClientMessage extends QueryWorkerOptions, Message {