Improve SQL Beautify: use sql-formatter and support bind variables (#2071)

Co-authored-by: GCHQDeveloper581 <63102987+GCHQDeveloper581@users.noreply.github.com> - (additional tests)
2026-03-02 18:25:54 +05:30 · 2026-03-02 18:25:54 +05:30 · 813a543baf
commit 813a543baf
parent b2ae8c186e
6 changed files with 160 additions and 10 deletions
--- a/package-lock.json
+++ b/package-lock.json
@ -92,6 +92,7 @@
        "snackbarjs": "^1.1.0",
        "sortablejs": "^1.15.7",
        "split.js": "^1.6.5",
+        "sql-formatter": "^15.6.5",
        "ssdeep.js": "0.0.3",
        "stream-browserify": "^3.0.0",
        "tesseract.js": "^6.0.1",
@ -8161,6 +8162,11 @@
      "integrity": "sha512-fGTi3gxV/23FTYdAoUtLYp6qySe2KE3teyZitipKNRuVYcBkoP/bB3guXN/XVKUe9mxCHXnc9C4ocyz8OmgN0g==",
      "license": "MIT"
    },
+    "node_modules/discontinuous-range": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/discontinuous-range/-/discontinuous-range-1.0.0.tgz",
+      "integrity": "sha512-c68LpLbO+7kP/b1Hr1qs8/BJ09F5khZGTxqxZuhzxpmwJKOgRFHJWIb9/KmqnqHhLdO55aOxFH/EGBvUQbL/RQ=="
+    },
    "node_modules/dns-packet": {
      "version": "5.6.1",
      "resolved": "https://registry.npmjs.org/dns-packet/-/dns-packet-5.6.1.tgz",
@ -13204,6 +13210,11 @@
        "node": "*"
      }
    },
+    "node_modules/moo": {
+      "version": "0.5.2",
+      "resolved": "https://registry.npmjs.org/moo/-/moo-0.5.2.tgz",
+      "integrity": "sha512-iSAJLHYKnX41mKcJKjqvnAN9sf0LMDTXDEvFv+ffuRR9a1MIuXLjMNL6EsnDHSkKLTWNqQQ5uo61P4EbU4NU+Q=="
+    },
    "node_modules/more-entropy": {
      "version": "0.0.7",
      "resolved": "https://registry.npmjs.org/more-entropy/-/more-entropy-0.0.7.tgz",
@ -13309,6 +13320,32 @@
      "dev": true,
      "license": "MIT"
    },
+    "node_modules/nearley": {
+      "version": "2.20.1",
+      "resolved": "https://registry.npmjs.org/nearley/-/nearley-2.20.1.tgz",
+      "integrity": "sha512-+Mc8UaAebFzgV+KpI5n7DasuuQCHA89dmwm7JXw3TV43ukfNQ9DnBH3Mdb2g/I4Fdxc26pwimBWvjIw0UAILSQ==",
+      "dependencies": {
+        "commander": "^2.19.0",
+        "moo": "^0.5.0",
+        "railroad-diagrams": "^1.0.0",
+        "randexp": "0.4.6"
+      },
+      "bin": {
+        "nearley-railroad": "bin/nearley-railroad.js",
+        "nearley-test": "bin/nearley-test.js",
+        "nearley-unparse": "bin/nearley-unparse.js",
+        "nearleyc": "bin/nearleyc.js"
+      },
+      "funding": {
+        "type": "individual",
+        "url": "https://nearley.js.org/#give-to-nearley"
+      }
+    },
+    "node_modules/nearley/node_modules/commander": {
+      "version": "2.20.3",
+      "resolved": "https://registry.npmjs.org/commander/-/commander-2.20.3.tgz",
+      "integrity": "sha512-GpVkmM8vF2vQUkj2LvZmD35JxeJOLCwJ9cUkugyk2nuhbv3+mJvpLYYt+0+USMxE+oj+ey/lJEnhZw75x/OMcQ=="
+    },
    "node_modules/negotiator": {
      "version": "0.6.3",
      "resolved": "https://registry.npmjs.org/negotiator/-/negotiator-0.6.3.tgz",
@ -15110,6 +15147,23 @@
        "url": "https://github.com/sponsors/sindresorhus"
      }
    },
+    "node_modules/railroad-diagrams": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/railroad-diagrams/-/railroad-diagrams-1.0.0.tgz",
+      "integrity": "sha512-cz93DjNeLY0idrCNOH6PviZGRN9GJhsdm9hpn1YCS879fj4W+x5IFJhhkRZcwVgMmFF7R82UA/7Oh+R8lLZg6A=="
+    },
+    "node_modules/randexp": {
+      "version": "0.4.6",
+      "resolved": "https://registry.npmjs.org/randexp/-/randexp-0.4.6.tgz",
+      "integrity": "sha512-80WNmd9DA0tmZrw9qQa62GPPWfuXJknrmVmLcxvq4uZBdYqb1wYoKTmnlGUchvVWe0XiLupYkBoXVOxz3C8DYQ==",
+      "dependencies": {
+        "discontinuous-range": "1.0.0",
+        "ret": "~0.1.10"
+      },
+      "engines": {
+        "node": ">=0.12"
+      }
+    },
    "node_modules/randombytes": {
      "version": "2.1.0",
      "resolved": "https://registry.npmjs.org/randombytes/-/randombytes-2.1.0.tgz",
@ -15534,6 +15588,14 @@
        "node": ">=8"
      }
    },
+    "node_modules/ret": {
+      "version": "0.1.15",
+      "resolved": "https://registry.npmjs.org/ret/-/ret-0.1.15.tgz",
+      "integrity": "sha512-TTlYpa+OL+vMMNG24xSlQGEJ3B/RzEfUlLct7b5G/ytav+wPrplCpVMFuwzXbkecJrb6IYo1iFb0S9v37754mg==",
+      "engines": {
+        "node": ">=0.12"
+      }
+    },
    "node_modules/retry": {
      "version": "0.13.1",
      "resolved": "https://registry.npmjs.org/retry/-/retry-0.13.1.tgz",
@ -16440,6 +16502,18 @@
      "dev": true,
      "license": "BSD-3-Clause"
    },
+    "node_modules/sql-formatter": {
+      "version": "15.6.5",
+      "resolved": "https://registry.npmjs.org/sql-formatter/-/sql-formatter-15.6.5.tgz",
+      "integrity": "sha512-fr4TyM1udCSrOHOmouotwUi8dxIDhSLpYNmPePGFVzxq8/i8jd828IapE49QXG7Gzkswxo5WwdAGnYX4YpKoTg==",
+      "dependencies": {
+        "argparse": "^2.0.1",
+        "nearley": "^2.20.1"
+      },
+      "bin": {
+        "sql-formatter": "bin/sql-formatter-cli.cjs"
+      }
+    },
    "node_modules/ssdeep.js": {
      "version": "0.0.3",
      "resolved": "https://registry.npmjs.org/ssdeep.js/-/ssdeep.js-0.0.3.tgz",
--- a/package.json
+++ b/package.json
@ -175,6 +175,7 @@
    "snackbarjs": "^1.1.0",
    "sortablejs": "^1.15.7",
    "split.js": "^1.6.5",
+    "sql-formatter": "^15.6.5",
    "ssdeep.js": "0.0.3",
    "stream-browserify": "^3.0.0",
    "tesseract.js": "^6.0.1",
--- a/src/core/operations/SQLBeautify.mjs
+++ b/src/core/operations/SQLBeautify.mjs
@ -3,8 +3,7 @@
 * @copyright Crown Copyright 2016
 * @license Apache-2.0
 */
-
-import vkbeautify from "vkbeautify";
+import { format } from "sql-formatter";
 import Operation from "../Operation.mjs";

 /**
@ -39,7 +38,26 @@ class SQLBeautify extends Operation {
     */
    run(input, args) {
        const indentStr = args[0];
-        return vkbeautify.sql(input, indentStr);
+        // Extract and replace bind variables like :Bind1 with __BIND_0__
+        const bindRegex = /:\w+/g;
+        const bindMap = {};
+        let bindCounter=0;
+        const placeholderInput = input.replace(bindRegex, (match) => {
+            const placeholder = `__BIND_${bindCounter++}__`;
+            bindMap[placeholder] = match;
+            return placeholder;
+        });
+        // Format the SQL with chosen options
+        let formatted= format(placeholderInput, {
+            language: "mysql", // Use MySQL as the default dialect for better compatibility with real-world SQL
+            useTabs: indentStr==="\t", // true if tab, false if spaces
+            tabWidth: indentStr.length || 4,     // fallback if empty
+            indentStyle: "standard"              // fine for most SQL
+        });
+        // Replace placeholders back with original bind variables
+        formatted = formatted.replace(/__BIND_\d+__/g, match => bindMap[match] || match);
+
+        return formatted;
    }

 }
--- a/tests/node/tests/operations.mjs
+++ b/tests/node/tests/operations.mjs
@ -867,13 +867,15 @@ pCGTErs=
    }),

    it("SQL Beautify", () => {
-        const result = chef.SQLBeautify(`SELECT MONTH, ID, RAIN_I, TEMP_F
-FROM STATS;`);
-        const expected = `SELECT MONTH,
-         ID,
-         RAIN_I,
-         TEMP_F
-FROM STATS;`;
+        const result = chef.SQLBeautify(`SELECT MONTH, ID, RAIN_I, TEMP_F FROM STATS;`);
+        const expected =
+`SELECT
+  MONTH,
+  ID,
+  RAIN_I,
+  TEMP_F
+FROM
+  STATS;`;
        assert.strictEqual(result.toString(), expected);
    }),

--- a/tests/operations/index.mjs
+++ b/tests/operations/index.mjs
@ -153,6 +153,7 @@ import "./tests/SIGABA.mjs";
 import "./tests/SM2.mjs";
 import "./tests/SM4.mjs";
 // import "./tests/SplitColourChannels.mjs"; // Cannot test operations that use the File type yet
+import "./tests/SQLBeautify.mjs";
 import "./tests/StrUtils.mjs";
 import "./tests/StripIPv4Header.mjs";
 import "./tests/StripTCPHeader.mjs";
--- a/tests/operations/tests/SQLBeautify.mjs
+++ b/tests/operations/tests/SQLBeautify.mjs
@ -0,0 +1,54 @@
+/**
+ * SQLBeautify tests.
+ *
+ * @author GCHQDeveloper581
+ * @copyright Crown Copyright 2026
+ * @license Apache-2.0
+ */
+import TestRegister from "../../lib/TestRegister.mjs";
+
+TestRegister.addTests([
+    {
+        name: "SQL Beautify - basic",
+        input: "SELECT MONTH, ID, RAIN_I, TEMP_F FROM STATS;",
+        expectedOutput:
+`SELECT
+  MONTH,
+  ID,
+  RAIN_I,
+  TEMP_F
+FROM
+  STATS;`,
+        recipeConfig: [
+            {
+                op: "SQL Beautify",
+                args: ["  "],
+            },
+        ],
+    },
+    {
+        name: "SQL Beautify - upsert",
+        input: "INSERT INTO Table1 SELECT * FROM (SELECT :Bind1 as Field1, :Bind2 as Field2, :id as id) as new_data ON DUPLICATE KEY UPDATE Field1 = new_data.Field1, Field2 = new_data.Field2;",
+        expectedOutput:
+`INSERT INTO
+  Table1
+SELECT
+  *
+FROM
+  (
+    SELECT
+      :Bind1 as Field1,
+      :Bind2 as Field2,
+      :id as id
+  ) as new_data
+ON DUPLICATE KEY UPDATE
+  Field1 = new_data.Field1,
+  Field2 = new_data.Field2;`,
+        recipeConfig: [
+            {
+                op: "SQL Beautify",
+                args: ["  "],
+            },
+        ],
+    },
+]);