initial commit

.gitignore

@@ -0,0 +1,2 @@
+/dist
+/node_modules

.npmignore

@@ -0,0 +1,3 @@
+/jest.config.json
+**/*.test.ts
+/.gitignore

.prettierrc

@@ -0,0 +1,4 @@
+{
+  "singleQuote": true,
+  "trailingComma": "all"
+}

README.md

@@ -0,0 +1,80 @@
+# Incremental CSV Parser
+
+A fast & lightweight CSV parser with an incremental parsing api.  No
+dependencies; browser compatible.  Includes ESM and CJS builds.  All column
+values are strings (numbers and dates will **not** automatically be parsed).
+
+Fully compatible with [rfc4180](https://www.ietf.org/rfc/rfc4180.txt).
+
+
+## Basic parsing
+
+```js
+import { parse } from 'incremental-csv-parser';
+
+
+const results = await parse(`a,b,c,d
+1,2,3,4
+5,6,7,8
+`);
+
+console.log(results);
+// [
+//   { a: '1', b: '2', c: '3', d: '4' },
+//   { a: '5', b: '6', c: '7', d: '8' },
+// ]
+```
+
+## Incremental parsing
+
+```js
+import { CSVParser } from 'incremental-csv-parser';
+
+const results = [];
+
+const parser = new CSVParser((row) => {
+  results.push(row);
+});
+
+parser.process('a,b,c,d\n');
+parser.process('1,2,3,4\n');
+console.log(results.shift()); // { a: '1', b: '2', c: '3', d: '4' }
+parser.process('5,6,7,8');
+parser.flush();
+console.log(results.shift()); // { a: '5', b: '6', c: '7', d: '8' }
+```
+
+## Typescript support
+
+If column names are known ahead of time, they can be passed in via generics.
+
+```ts
+import { CSVParser } from 'incremental-csv-parser';
+
+type ColumnName = 'a' | 'b' | 'c' | 'd';
+const results: Record<ColumnName, string> = [];
+
+const parser = new CSVParser<ColumName>((row) => {
+  results.push(row);
+});
+```
+
+## CSV files without pre-defined headers
+
+Both the `parse` function and `CSVParser` class have an optional second argument
+for explicitly providing column names for a csv.
+
+```ts
+import { CSVParser } from 'incremental-csv-parser';
+
+const columns = ['a', 'b', 'c', 'd'];
+
+const results = parse(`1,2,3,4
+5,6,7,8`, columns);
+
+console.log(results);
+// [
+//   { a: '1', b: '2', c: '3', d: '4' },
+//   { a: '5', b: '6', c: '7', d: '8' },
+// ]
+```

jest.config.json

@@ -0,0 +1,5 @@
+{
+  "transform": {
+    "^.+\\.ts$": ["esbuild-jest", { "sourcemap": true }]   
+  }
+}

package.json

@@ -0,0 +1,35 @@
+{
+  "name": "incremental-csv-parser",
+  "version": "1.0.0",
+  "description": "A simple, browser compatible, incremental CSV parser.",
+  "main": "dist/esm/index.js",
+  "repository": "https://github.com/AllAwesome497/incremental-csv-parser",
+  "author": "Wade Bourne <[email protected]>",
+  "license": "MIT",
+  "scripts": {
+    "build": "yarn build:types && yarn build:esm && yarn build:cjs",
+    "build:types": "tsc",
+    "build:esm": "esbuild src/*.ts --outdir=dist/esm --format=esm --out-extension:.js=.mjs --sourcemap",
+    "build:cjs": "esbuild src/*.ts --outdir=dist/cjs --format=cjs --sourcemap",
+    "test": "jest",
+    "prepack": "yarn test && yarn build"
+  },
+  "exports": {
+    ".": {
+      "import": "./dist/esm/index.mjs",
+      "require": "./dist/cjs/index.js",
+      "types": "./dist/types/index.d.ts"
+    },
+    "./package.json": "./package.json"
+  },
+  "devDependencies": {
+    "@babel/preset-typescript": "^7.18.6",
+    "@jest/globals": "^29.3.1",
+    "esbuild": "^0.15.15",
+    "esbuild-jest": "^0.5.0",
+    "jest": "^29.3.1",
+    "ts-jest": "^29.0.3",
+    "typescript": "^4.9.3",
+    "prettier": "^2.7.1"
+  }
+}

src/__tests__/index.test.ts

@@ -0,0 +1,121 @@
+/* eslint-env jest */
+import { expect, describe, it } from '@jest/globals';
+import { CSVParser } from '../index';
+
+interface CSVParserTest {
+  name: string;
+  headers?: Array<string>;
+  chunks: Array<string>;
+  expected: Array<Record<string, string>>;
+}
+
+describe('SimpleCSVParser', () => {
+  it.each<CSVParserTest>([
+    {
+      name: 'basic',
+      chunks: [
+        `a,b,c,d
+1,2,3,4
+5,6,7,8
+`,
+      ],
+      expected: [
+        { a: '1', b: '2', c: '3', d: '4' },
+        { a: '5', b: '6', c: '7', d: '8' },
+      ],
+    },
+    {
+      name: 'mixed quotes',
+      chunks: [
+        `a,b,c,"d"
+"1",2,"3","4"
+5,6,7,8
+`,
+      ],
+      expected: [
+        { a: '1', b: '2', c: '3', d: '4' },
+        { a: '5', b: '6', c: '7', d: '8' },
+      ],
+    },
+    {
+      name: 'quoted newlines',
+      chunks: [
+        `a,b,c,d
+"1
+2",3,4,5
+`,
+      ],
+      expected: [{ a: '1\n2', b: '3', c: '4', d: '5' }],
+    },
+    {
+      name: 'empty values',
+      chunks: [
+        `a,b,c,d
+,,,`,
+      ],
+      expected: [{ a: '', b: '', c: '', d: '' }],
+    },
+    {
+      name: 'carriage returns + newlines',
+      chunks: [
+        `a,b,c,d\r
+1,2,3,4\r
+5,6,7,8
+`,
+      ],
+      expected: [
+        { a: '1', b: '2', c: '3', d: '4' },
+        { a: '5', b: '6', c: '7', d: '8' },
+      ],
+    },
+    {
+      name: 'commas in quoted text',
+      chunks: [
+        `a,b,c,d
+"1,2",3,4,5
+`,
+      ],
+      expected: [{ a: '1,2', b: '3', c: '4', d: '5' }],
+    },
+    {
+      name: 'basic chunks',
+      chunks: ['a,b,c,d\n', '1,', '2,', '3,', '4\n'],
+
+      expected: [{ a: '1', b: '2', c: '3', d: '4' }],
+    },
+    {
+      name: 'chunks with newlines',
+      chunks: ['a,b,c\n', '"1\n', '\n', '\n', '"', ',2,3\n'],
+      expected: [{ a: '1\n\n\n', b: '2', c: '3' }],
+    },
+    {
+      name: 'chunks with commas and quotes',
+      chunks: ['a,b,c\n', ',', '",,,', '",'],
+      expected: [{ a: '', b: ',,,', c: '' }],
+    },
+    {
+      name: 'manually set headers',
+      chunks: [
+        `1,2,3,4
+5,6,7,8`,
+      ],
+      headers: ['a', 'b', 'c', 'd'],
+      expected: [
+        { a: '1', b: '2', c: '3', d: '4' },
+        { a: '5', b: '6', c: '7', d: '8' },
+      ],
+    },
+  ])('$name', ({ chunks, expected, headers }) => {
+    const results: Array<Record<string, string>> = [];
+
+    const parser = new CSVParser((data) => results.push(data), headers);
+
+    for (const chunk of chunks) {
+      parser.process(chunk);
+    }
+
+    parser.flush();
+
+    expect(results).toEqual(expected);
+  });
+});