feat: detect common free/disposable email providers locally (#1096)

This PR regenerates the WASM to pull in some updates for our email validation primitive.
Notably it also provides functionality to detect the five most common free email providers locally, preventing a call to the Arcjet API.
This won't result in any behaviour change for email validation beyond potentially fewer api requests.

Closes #1095

analyze/edge-light.ts

@@ -6,12 +6,21 @@ import type {
   EmailValidationConfig,
   BotDetectionResult,
   BotType,
+  EmailValidationResult,
 } from "./wasm/arcjet_analyze_js_req.component.js";
 
 import componentCoreWasm from "./wasm/arcjet_analyze_js_req.component.core.wasm?module";
 import componentCore2Wasm from "./wasm/arcjet_analyze_js_req.component.core2.wasm?module";
 import componentCore3Wasm from "./wasm/arcjet_analyze_js_req.component.core3.wasm?module";
 
+const FREE_EMAIL_PROVIDERS = [
+  "gmail.com",
+  "yahoo.com",
+  "hotmail.com",
+  "aol.com",
+  "hotmail.co.uk",
+];
+
 interface AnalyzeContext {
   log: ArcjetLogger;
   characteristics: string[];
@@ -43,6 +52,23 @@ async function init(context: AnalyzeContext) {
         log.error(msg);
       },
     },
+    "arcjet:js-req/email-validator-overrides": {
+      isFreeEmail(domain) {
+        if (FREE_EMAIL_PROVIDERS.includes(domain)) {
+          return "yes";
+        }
+        return "unknown";
+      },
+      isDisposableEmail() {
+        return "unknown";
+      },
+      hasMxRecords() {
+        return "unknown";
+      },
+      hasGravatar() {
+        return "unknown";
+      },
+    },
   };
 
   try {
@@ -97,14 +123,23 @@ export async function isValidEmail(
   context: AnalyzeContext,
   candidate: string,
   options?: EmailValidationConfig,
-) {
+): Promise<EmailValidationResult> {
   const analyze = await init(context);
+  const optionsOrDefault = {
+    requireTopLevelDomain: true,
+    allowDomainLiteral: false,
+    blockedEmails: [],
+    ...options,
+  };
 
   if (typeof analyze !== "undefined") {
-    return analyze.isValidEmail(candidate, options);
+    return analyze.isValidEmail(candidate, optionsOrDefault);
   } else {
-    // TODO: Fallback to JS if we don't have WASM?
-    return true;
+    // Skip the local evaluation of the rule if WASM is not available
+    return {
+      validity: "valid",
+      blocked: [],
+    };
   }
 }
 

analyze/index.ts

@@ -6,12 +6,21 @@ import type {
   EmailValidationConfig,
   BotDetectionResult,
   BotType,
+  EmailValidationResult,
 } from "./wasm/arcjet_analyze_js_req.component.js";
 
 import { wasm as componentCoreWasm } from "./wasm/arcjet_analyze_js_req.component.core.wasm?js";
 import { wasm as componentCore2Wasm } from "./wasm/arcjet_analyze_js_req.component.core2.wasm?js";
 import { wasm as componentCore3Wasm } from "./wasm/arcjet_analyze_js_req.component.core3.wasm?js";
 
+const FREE_EMAIL_PROVIDERS = [
+  "gmail.com",
+  "yahoo.com",
+  "hotmail.com",
+  "aol.com",
+  "hotmail.co.uk",
+];
+
 interface AnalyzeContext {
   log: ArcjetLogger;
   characteristics: string[];
@@ -57,6 +66,23 @@ async function init(context: AnalyzeContext) {
         log.error(msg);
       },
     },
+    "arcjet:js-req/email-validator-overrides": {
+      isFreeEmail(domain) {
+        if (FREE_EMAIL_PROVIDERS.includes(domain)) {
+          return "yes";
+        }
+        return "unknown";
+      },
+      isDisposableEmail() {
+        return "unknown";
+      },
+      hasMxRecords() {
+        return "unknown";
+      },
+      hasGravatar() {
+        return "unknown";
+      },
+    },
   };
 
   try {
@@ -111,14 +137,23 @@ export async function isValidEmail(
   context: AnalyzeContext,
   candidate: string,
   options?: EmailValidationConfig,
-) {
+): Promise<EmailValidationResult> {
   const analyze = await init(context);
+  const optionsOrDefault = {
+    requireTopLevelDomain: true,
+    allowDomainLiteral: false,
+    blockedEmails: [],
+    ...options,
+  };
 
   if (typeof analyze !== "undefined") {
-    return analyze.isValidEmail(candidate, options);
+    return analyze.isValidEmail(candidate, optionsOrDefault);
   } else {
-    // TODO: Fallback to JS if we don't have WASM?
-    return true;
+    // Skip the local evaluation of the rule if WASM is not available
+    return {
+      validity: "valid",
+      blocked: [],
+    };
   }
 }
 

analyze/wasm/arcjet_analyze_js_req.component.d.ts

@@ -18,18 +18,33 @@ export interface BotDetectionResult {
   botType: BotType,
   botScore: number,
 }
+/**
+* # Variants
+* 
+* ## `"valid"`
+* 
+* ## `"invalid"`
+*/
+export type EmailValidity = 'valid' | 'invalid';
+export interface EmailValidationResult {
+  validity: EmailValidity,
+  blocked: string[],
+}
 export interface EmailValidationConfig {
-  requireTopLevelDomain?: boolean,
-  allowDomainLiteral?: boolean,
+  requireTopLevelDomain: boolean,
+  allowDomainLiteral: boolean,
+  blockedEmails: string[],
 }
+import { ArcjetJsReqEmailValidatorOverrides } from './interfaces/arcjet-js-req-email-validator-overrides.js';
 import { ArcjetJsReqLogger } from './interfaces/arcjet-js-req-logger.js';
 export interface ImportObject {
+  'arcjet:js-req/email-validator-overrides': typeof ArcjetJsReqEmailValidatorOverrides,
   'arcjet:js-req/logger': typeof ArcjetJsReqLogger,
 }
 export interface Root {
   detectBot(headers: string, patternsAdd: string, patternsRemove: string): BotDetectionResult,
   generateFingerprint(request: string, characteristics: string[]): string,
-  isValidEmail(candidate: string, options: EmailValidationConfig | undefined): boolean,
+  isValidEmail(candidate: string, options: EmailValidationConfig): EmailValidationResult,
 }
 
 /**