Fix multiline environment variable parsing in .env files (#928)

* Initial plan

* Fix multiline environment variable parsing in .env files

Co-authored-by: rchiodo <19672699+rchiodo@users.noreply.github.com>

* Improve escape sequence handling in environment parser

Co-authored-by: rchiodo <19672699+rchiodo@users.noreply.github.com>

* Update comment for regex 's' flag usage

Co-authored-by: rchiodo <19672699+rchiodo@users.noreply.github.com>

* Fix linter and prettier formatting issues

Co-authored-by: rchiodo <19672699+rchiodo@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: rchiodo <19672699+rchiodo@users.noreply.github.com>
Co-authored-by: Rich Chiodo <rchiodo@users.noreply.github.com>

Author: Copilot

src/extension/common/variables/environment.ts

@@ -94,16 +94,80 @@ export function appendPaths(
 export function parseEnvFile(lines: string | Buffer, baseVars?: EnvironmentVariables): EnvironmentVariables {
     const globalVars = baseVars ? baseVars : {};
     const vars: EnvironmentVariables = {};
-    lines
-        .toString()
-        .split('\n')
-        .forEach((line, _idx) => {
-            const [name, value] = parseEnvLine(line);
-            if (name === '') {
-                return;
+    const content = lines.toString();
+
+    // State machine to handle multiline quoted values
+    let currentLine = '';
+    let inQuotes = false;
+    let quoteChar = '';
+    let afterEquals = false;
+
+    for (let i = 0; i < content.length; i++) {
+        const char = content[i];
+
+        // Track if we've seen an '=' sign (indicating we're in the value part)
+        if (char === '=' && !inQuotes) {
+            afterEquals = true;
+            currentLine += char;
+            continue;
+        }
+
+        // Handle quote characters - need to check for proper escaping
+        if ((char === '"' || char === "'") && afterEquals) {
+            // Count consecutive backslashes before this quote
+            let numBackslashes = 0;
+            let j = i - 1;
+            while (j >= 0 && content[j] === '\\') {
+                numBackslashes++;
+                j--;
+            }
+
+            // Quote is escaped if there's an odd number of backslashes before it
+            const isEscaped = numBackslashes % 2 === 1;
+
+            if (!isEscaped) {
+                if (!inQuotes) {
+                    // Starting a quoted section
+                    inQuotes = true;
+                    quoteChar = char;
+                } else if (char === quoteChar) {
+                    // Ending a quoted section
+                    inQuotes = false;
+                    quoteChar = '';
+                }
+            }
+            currentLine += char;
+            continue;
+        }
+
+        // Handle newlines
+        if (char === '\n') {
+            if (inQuotes) {
+                // We're inside quotes, preserve the newline
+                currentLine += char;
+            } else {
+                // We're not in quotes, this is the end of a line
+                const [name, value] = parseEnvLine(currentLine);
+                if (name !== '') {
+                    vars[name] = substituteEnvVars(value, vars, globalVars);
+                }
+                // Reset for next line
+                currentLine = '';
+                afterEquals = false;
             }
+        } else {
+            currentLine += char;
+        }
+    }
+
+    // Handle the last line if there's no trailing newline
+    if (currentLine.trim() !== '') {
+        const [name, value] = parseEnvLine(currentLine);
+        if (name !== '') {
             vars[name] = substituteEnvVars(value, vars, globalVars);
-        });
+        }
+    }
+
     return vars;
 }
 
@@ -112,7 +176,8 @@ function parseEnvLine(line: string): [string, string] {
     //   https://github.com/motdotla/dotenv/blob/master/lib/main.js#L32
     // We don't use dotenv here because it loses ordering, which is
     // significant for substitution.
-    const match = line.match(/^\s*(_*[a-zA-Z]\w*)\s*=\s*(.*?)?\s*$/);
+    // Modified to handle multiline values by using 's' flag so $ matches before newlines in multiline strings
+    const match = line.match(/^\s*(_*[a-zA-Z]\w*)\s*=\s*(.*?)?\s*$/s);
     if (!match) {
         return ['', ''];
     }

src/test/unittest/common/environment.unit.test.ts

@@ -0,0 +1,103 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+'use strict';
+
+import { expect } from 'chai';
+import { parseEnvFile } from '../../../extension/common/variables/environment';
+
+suite('Environment File Parsing Tests', () => {
+    test('Should parse simple environment variables', () => {
+        const content = 'VAR1=value1\nVAR2=value2';
+        const result = parseEnvFile(content);
+
+        // eslint-disable-next-line @typescript-eslint/naming-convention
+        expect(result).to.deep.equal({
+            VAR1: 'value1',
+            VAR2: 'value2',
+        });
+    });
+
+    test('Should parse single-quoted multiline values', () => {
+        const content = "EXAMPLE_VAR='very long value\nwith new line , we need to get all the lines'";
+        const result = parseEnvFile(content);
+
+        expect(result.EXAMPLE_VAR).to.equal('very long value\nwith new line , we need to get all the lines');
+    });
+
+    test('Should parse double-quoted multiline values', () => {
+        const content = 'EXAMPLE_VAR="very long value\nwith new line , we need to get all the lines"';
+        const result = parseEnvFile(content);
+
+        expect(result.EXAMPLE_VAR).to.equal('very long value\nwith new line , we need to get all the lines');
+    });
+
+    test('Should parse escaped newlines in single-quoted values', () => {
+        const content = "VAR='line1\\nline2'";
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('line1\nline2');
+    });
+
+    test('Should parse escaped newlines in double-quoted values', () => {
+        const content = 'VAR="line1\\nline2"';
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('line1\nline2');
+    });
+
+    test('Should handle multiple variables with multiline values', () => {
+        const content = "VAR1='multiline\nvalue1'\nVAR2='multiline\nvalue2'";
+        const result = parseEnvFile(content);
+
+        expect(result.VAR1).to.equal('multiline\nvalue1');
+        expect(result.VAR2).to.equal('multiline\nvalue2');
+    });
+
+    test('Should handle unquoted values', () => {
+        const content = 'VAR=value_without_quotes';
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('value_without_quotes');
+    });
+
+    test('Should handle empty values', () => {
+        const content = 'VAR=';
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('');
+    });
+
+    test('Should ignore lines without equals sign', () => {
+        const content = 'VAR1=value1\nInvalid line\nVAR2=value2';
+        const result = parseEnvFile(content);
+
+        // eslint-disable-next-line @typescript-eslint/naming-convention
+        expect(result).to.deep.equal({
+            VAR1: 'value1',
+            VAR2: 'value2',
+        });
+    });
+
+    test('Should handle multiline value with multiple newlines', () => {
+        const content = "VAR='line1\nline2\nline3\nline4'";
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('line1\nline2\nline3\nline4');
+    });
+
+    test('Should parse environment file as Buffer', () => {
+        const content = Buffer.from("VAR='multiline\nvalue'");
+        const result = parseEnvFile(content);
+
+        expect(result.VAR).to.equal('multiline\nvalue');
+    });
+
+    test('Should handle whitespace around variable names and equals', () => {
+        const content = "  VAR1  =  value1  \n  VAR2='multiline\nvalue'";
+        const result = parseEnvFile(content);
+
+        expect(result.VAR1).to.equal('value1');
+        expect(result.VAR2).to.equal('multiline\nvalue');
+    });
+});