@intuned/browser-dev 0.1.15-dev.1 → 0.1.16-dev.1

package/.prettierrc

@@ -0,0 +1,3 @@
+{
+  "trailingComma": "es5"
+}

package/dist/ai/tests/extractStructuredDataDomMatchingCache.spec.js

@@ -0,0 +1,87 @@
+"use strict";
+
+var _extendedTest = require("../../common/extendedTest");
+var _ = require("..");
+var _neverthrow = require("neverthrow");
+function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
+const DATA_SCHEMA = {
+  type: "object",
+  required: ["phone"],
+  properties: {
+    name: {
+      type: "string",
+      description: "the contact name"
+    },
+    phone: {
+      type: "string",
+      description: "the contact phone"
+    }
+  }
+};
+const CONTACT_HTML = `<div class="contact"><p>Andreas Jansson</p><p>PH: (727) 471-4768</p></div>`;
+function makeOptions(page) {
+  return {
+    source: page,
+    dataSchema: DATA_SCHEMA,
+    strategy: "HTML",
+    enableDomMatching: true,
+    enableCache: true,
+    model: "claude-sonnet-4-20250514"
+  };
+}
+async function mockExtractStructuredDataUsingAi(result) {
+  const aiModule = await Promise.resolve().then(() => _interopRequireWildcard(require("../extractStructuredDataUsingAi")));
+  return _extendedTest.vi.spyOn(aiModule, "extractStructuredDataUsingAi").mockResolvedValue((0, _neverthrow.ok)({
+    result,
+    xpathMapping: {}
+  }));
+}
+async function useInMemoryCache() {
+  const cacheModule = await Promise.resolve().then(() => _interopRequireWildcard(require("../../intunedServices/cache/cache")));
+  const store = new Map();
+  _extendedTest.vi.spyOn(cacheModule.cache, "get").mockImplementation(async key => store.has(key) ? store.get(key) : null);
+  _extendedTest.vi.spyOn(cacheModule.cache, "set").mockImplementation(async (key, value) => {
+    store.set(key, value);
+  });
+}
+(0, _extendedTest.describe)("extractStructuredData - DOM matching cache (enableDomMatching: true)", () => {
+  (0, _extendedTest.afterEach)(() => {
+    _extendedTest.vi.restoreAllMocks();
+  });
+  (0, _extendedTest.test)("reuses the cache on a second call, even when a value is a partial match", async ({
+    page
+  }) => {
+    const aiResult = {
+      name: "Andreas Jansson",
+      phone: "(727) 471-4768"
+    };
+    const aiSpy = await mockExtractStructuredDataUsingAi(aiResult);
+    await useInMemoryCache();
+    await page.setContent(CONTACT_HTML);
+    const first = await (0, _.extractStructuredData)(makeOptions(page));
+    (0, _extendedTest.expect)(first).toEqual(aiResult);
+    (0, _extendedTest.expect)(aiSpy).toHaveBeenCalledTimes(1);
+    const second = await (0, _.extractStructuredData)(makeOptions(page));
+    (0, _extendedTest.expect)(second).toEqual(first);
+    (0, _extendedTest.expect)(aiSpy).toHaveBeenCalledTimes(1);
+  });
+  (0, _extendedTest.test)("re-extracts when an extracted value's DOM text changes", async ({
+    page
+  }) => {
+    const aiResult = {
+      name: "Andreas Jansson",
+      phone: "(727) 471-4768"
+    };
+    const aiSpy = await mockExtractStructuredDataUsingAi(aiResult);
+    await useInMemoryCache();
+    await page.setContent(CONTACT_HTML);
+    await (0, _.extractStructuredData)(makeOptions(page));
+    (0, _extendedTest.expect)(aiSpy).toHaveBeenCalledTimes(1);
+    await page.evaluate(() => {
+      const ps = document.querySelectorAll("div.contact p");
+      if (ps[1]) ps[1].textContent = "PH: (999) 999-9999";
+    });
+    await (0, _.extractStructuredData)(makeOptions(page));
+    (0, _extendedTest.expect)(aiSpy).toHaveBeenCalledTimes(2);
+  });
+});

package/dist/ai/tests/testMatching.spec.js

@@ -260,6 +260,44 @@ var _matching = require("../../common/matching/matching");
       });
       (0, _extendedTest.expect)(result === null || result === void 0 ? void 0 : result.matchText).toBe("Hello");
     });
+    (0, _extendedTest.test)("should prefer a full match over a partial match", async () => {
+      const matches = [{
+        matched_value: "Hello",
+        match_mode: "partial",
+        fuzzy_distance: 0,
+        match_source: "text_content",
+        xpath: "/html/body/div[1]"
+      }, {
+        matched_value: "Hello World",
+        match_mode: "full",
+        fuzzy_distance: 0,
+        match_source: "text_content",
+        xpath: "/html/body/div[2]"
+      }];
+      const result = (0, _matching.selectBestMatch)({
+        original: "Hello World",
+        matches
+      });
+      (0, _extendedTest.expect)(result === null || result === void 0 ? void 0 : result.matchText).toBe("Hello World");
+      (0, _extendedTest.expect)(result === null || result === void 0 ? void 0 : result.matchXpath).toBe("/html/body/div[2]");
+    });
+    (0, _extendedTest.test)("should preserve the attribute name for attribute matches", async () => {
+      const matches = [{
+        matched_value: "https://example.com/x.pdf",
+        match_mode: "full",
+        fuzzy_distance: 0,
+        match_source: "attribute",
+        attribute: "href",
+        xpath: "/html/body/a[1]"
+      }];
+      const result = (0, _matching.selectBestMatch)({
+        original: "https://example.com/x.pdf",
+        matches
+      });
+      (0, _extendedTest.expect)(result === null || result === void 0 ? void 0 : result.matchType).toEqual({
+        attribute: "href"
+      });
+    });
   });
   (0, _extendedTest.describe)("matchStringsWithDomContent", () => {
     let browser;

package/dist/ai/tests/testValidateMatchesMapping.spec.js

@@ -262,4 +262,62 @@ const IFRAME_WITH_EXTRA_CONTENT = `
     (0, _extendedTest.expect)(isValid2).toBe(true);
     (0, _extendedTest.expect)(isValid3).toBe(true);
   });
+  (0, _extendedTest.test)("PARTIAL match (value is a substring of the element text, PH: prefix) stays valid", async () => {
+    await page.setContent(`<div class="contact"><p>PH: (727) 471-4768</p></div>`);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      phone: "(727) 471-4768"
+    });
+    (0, _extendedTest.expect)(cachedMapping["(727) 471-4768"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(true);
+  });
+  (0, _extendedTest.test)("value sitting among sibling direct text nodes (br-separated block) stays valid", async () => {
+    await page.setContent(`
+      <table><tbody><tr><td class="vendor">
+        Acme Corp<br/>
+        Email: sales@acme.com<br/>
+        FAX: (630) 904-4118
+      </td></tr></tbody></table>
+    `);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      phone: "(630) 904-4118",
+      email: "sales@acme.com"
+    });
+    (0, _extendedTest.expect)(cachedMapping["(630) 904-4118"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(cachedMapping["sales@acme.com"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(true);
+  });
+  (0, _extendedTest.test)("href / attribute match (relative href resolved to absolute) stays valid", async () => {
+    await page.setContent(`<!DOCTYPE html><html><head><base href="https://example.com/files/"></head><body>` + `<div class="doc"><a href="annual-report.pdf">Annual Report</a></div></body></html>`);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      href: "annual-report.pdf",
+      name: "Annual Report"
+    });
+    (0, _extendedTest.expect)(cachedMapping["annual-report.pdf"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(true);
+  });
+  (0, _extendedTest.test)("FUZZY match (extracted value has a small typo) stays valid", async () => {
+    await page.setContent(`<div class="vendor"><span class="name">International Cleaning Services</span></div>`);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      name: "Internationl Cleaning Servces"
+    });
+    (0, _extendedTest.expect)(cachedMapping["Internationl Cleaning Servces"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(true);
+  });
+  (0, _extendedTest.test)("irregular internal whitespace stays valid", async () => {
+    await page.setContent(`<div class="vendor"><span>Audioconferencing  systems</span></div>`);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      category: "Audioconferencing systems"
+    });
+    (0, _extendedTest.expect)(cachedMapping["Audioconferencing systems"].length).toBeGreaterThan(0);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(true);
+  });
+  (0, _extendedTest.test)("invalidates when a partial-matched value's text changes", async () => {
+    await page.setContent(`<div class="contact"><p>PH: (727) 471-4768</p></div>`);
+    const cachedMapping = await (0, _matching.createMatchesMapping)(page, {
+      phone: "(727) 471-4768"
+    });
+    (0, _extendedTest.expect)(cachedMapping["(727) 471-4768"].length).toBeGreaterThan(0);
+    await page.setContent(`<div class="contact"><p>PH: (999) 999-9999</p></div>`);
+    (0, _extendedTest.expect)(await (0, _matching.validateMatchesMapping)(page, cachedMapping)).toBe(false);
+  });
 });

package/dist/common/matching/matching.js

@@ -122,12 +122,12 @@ function selectBestMatch({
 }) {
   const exactMatches = matches.filter(match => match.match_mode !== "fuzzy");
   if (exactMatches.length > 0) {
-    const bestMatch = exactMatches[0];
+    const bestMatch = exactMatches.find(match => match.match_mode === "full") ?? exactMatches[0];
     return {
       matchText: bestMatch.matched_value,
       matchXpath: bestMatch.xpath,
       matchType: bestMatch.match_source === "direct_text_node" ? "direct-text" : bestMatch.match_source === "attribute" ? {
-        attribute: "unknown"
+        attribute: bestMatch.attribute ?? "unknown"
       } : "all-text"
     };
   }
@@ -148,7 +148,7 @@ function selectBestMatch({
       matchText: bestMatch.matched_value,
       matchXpath: bestMatch.xpath,
       matchType: bestMatch.match_source === "direct_text_node" ? "direct-text" : bestMatch.match_source === "attribute" ? {
-        attribute: "unknown"
+        attribute: bestMatch.attribute ?? "unknown"
       } : "all-text"
     };
   }

package/dist/common/xpathMapping.js

@@ -6,13 +6,15 @@ Object.defineProperty(exports, "__esModule", {
 exports.createXPathMapping = createXPathMapping;
 exports.validateXPathMapping = validateXPathMapping;
 var _ensureBrowserScripts = require("./ensureBrowserScripts");
+var _utils = require("../optimized-extractors/common/matching/utils");
 async function validateXPathMapping(page, cachedMapping, prefix) {
   try {
     for (const [expectedText, entries] of Object.entries(cachedMapping)) {
       for (const entry of entries) {
         const {
           xpath,
-          matchType
+          matchType,
+          sourceText
         } = entry;
         const elementExists = await page.evaluate(({
           xpath,
@@ -27,34 +29,45 @@ async function validateXPathMapping(page, cachedMapping, prefix) {
         if (!elementExists) {
           return false;
         }
-        const actualText = await page.evaluate(({
+        const actualTexts = await page.evaluate(({
           xpath,
           prefix,
           matchType
         }) => {
           const element = document.evaluate(prefix ? `${prefix}/${xpath}` : xpath, document, null, XPathResult.FIRST_ORDERED_NODE_TYPE, null).singleNodeValue;
-          if (!element) return "";
+          if (!element) return [];
           if (matchType === "direct-text") {
-            let directText = "";
+            const nodes = [];
             for (const child of element.childNodes) {
               if (child.nodeType === Node.TEXT_NODE) {
-                directText += child.textContent || "";
+                const t = (child.textContent || "").trim();
+                if (t) nodes.push(t);
               }
             }
-            return directText.trim();
+            return nodes;
           } else if (matchType === "all-text") {
             var _element$textContent;
-            return ((_element$textContent = element.textContent) === null || _element$textContent === void 0 ? void 0 : _element$textContent.trim()) || "";
+            return [((_element$textContent = element.textContent) === null || _element$textContent === void 0 ? void 0 : _element$textContent.trim()) || ""];
           } else if (typeof matchType === "object" && matchType.attribute) {
-            return element.getAttribute(matchType.attribute) || "";
+            const attr = matchType.attribute;
+            const raw = element.getAttribute(attr) || "";
+            const candidates = [raw];
+            const reflected = element[attr];
+            if (typeof reflected === "string" && reflected) {
+              candidates.push(reflected);
+            }
+            return candidates;
           }
-          return "";
+          return [];
         }, {
           xpath,
           prefix,
           matchType
         });
-        if (actualText !== expectedText) {
+        const expectedFull = sourceText ?? expectedText;
+        const normExpected = (0, _utils.normalizeSpacing)(expectedFull);
+        const matched = actualTexts.some(t => (0, _utils.normalizeSpacing)(t) === normExpected);
+        if (!matched) {
           return false;
         }
       }

package/dist/helpers/downloadFile.js

@@ -38,6 +38,9 @@ const downloadFile = async input => {
     trigger,
     timeoutInMs
   } = input;
+  if (!trigger) {
+    throw new Error("trigger is required");
+  }
   let pageToDownloadFrom = page;
   let shouldClosePage = false;
   let downloadPromise;

package/dist/helpers/saveFileToS3.js

@@ -30,6 +30,9 @@ const saveFileToS3 = async input => {
   if (!page || typeof page.goto !== "function") {
     throw new Error("page must be a playwright Page object");
   }
+  if (!trigger) {
+    throw new Error("trigger is required");
+  }
   const download = await (0, _downloadFile.downloadFile)({
     page,
     trigger,

package/dist/helpers/tests/testDownloadFile.spec.js

@@ -279,6 +279,22 @@ _extendedTest.describe.skip("TestNotInGeneration", () => {
       (0, _extendedTest.expect)(cancelled).toContain("canceled");
     }
   });
+  (0, _extendedTest.test)("should throw when trigger is missing (wrong param names)", async () => {
+    await page.setContent(content);
+    await (0, _extendedTest.expect)((0, _.downloadFile)({
+      page,
+      url: "https://example.com/file.pdf",
+      fileName: "test.pdf"
+    })).rejects.toThrow("trigger is required");
+  });
+  (0, _extendedTest.test)("saveFileToS3 should throw when trigger is missing (wrong param names)", async () => {
+    await page.setContent(content);
+    await (0, _extendedTest.expect)((0, _.saveFileToS3)({
+      page,
+      url: "https://example.com/file.pdf",
+      fileName: "test.pdf"
+    })).rejects.toThrow("trigger is required");
+  });
 });
 const SIGNED_URL = "https://bucket.r2.cloudflarestorage.com/img.jpg?X-Amz-Signature=abc";
 (0, _extendedTest.describe)("SignedUrlAttachmentInTypedArray", () => {

package/dist/optimized-extractors/common/aiModelsValidations.js

@@ -3,30 +3,11 @@
 Object.defineProperty(exports, "__esModule", {
   value: true
 });
-exports.SUPPPORTED_GPT_MODELS = exports.SUPPPORTED_CLAUDE_MODELS = exports.SUPPORTED_MODELS = exports.MODELS_MAPPINGS = exports.MAX_TOKENS_OVERRIDES = exports.GPT_MODELS = exports.GOOGLE_MODELS = exports.CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_MODELS_MAPPINGS = exports.CLAUDE_MODELS = void 0;
-const CLAUDE_MODELS = exports.CLAUDE_MODELS = ["claude-opus-4-20250514", "claude-sonnet-4-20250514", "claude-3-7-sonnet-20250219", "claude-3-5-sonnet-20241022", "claude-3-5-sonnet-20240620", "claude-3-5-haiku-20241022", "claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307"];
+exports.MAX_TOKENS_OVERRIDES = void 0;
 const MAX_TOKENS_OVERRIDES = exports.MAX_TOKENS_OVERRIDES = {
   "claude-3-5-sonnet-20240620": 8192,
   "gemini-1.5-pro-002": 8192,
   "gemini-1.5-flash-8b-002": 8192,
   "gemini-1.5-flash-002": 8192,
   "gemini-2.0-flash-exp": 8192
-};
-const CLAUDE_VISION_SUPPORTED_MODELS = exports.CLAUDE_VISION_SUPPORTED_MODELS = [...CLAUDE_MODELS];
-const CLAUDE_MODELS_MAPPINGS = exports.CLAUDE_MODELS_MAPPINGS = {
-  "claude-3-haiku": "claude-3-haiku-20240307",
-  "claude-3-5-haiku": "claude-3-5-haiku-20241022",
-  "claude-3-opus": "claude-3-opus-20240229",
-  "claude-3-sonnet": "claude-3-sonnet-20240229",
-  "claude-3.5-sonnet": "claude-3-5-sonnet-20241022",
-  "claude-4-sonnet": "claude-sonnet-4-20250514",
-  "claude-4-opus": "claude-opus-4-20250514"
-};
-const GPT_MODELS = exports.GPT_MODELS = ["gpt-4.1", "gpt-4.1-mini", "gpt-4.1-nano", "gpt-4o", "gpt-4o-mini", "gpt-4o-audio-preview", "gpt-4-turbo", "gpt-4", "gpt-3.5-turbo", "o1", "o1-mini", "o1-preview", "o3-mini", "o3", "o4-mini", "chatgpt-4o-latest", "gpt4-turbo"];
-const GOOGLE_MODELS = exports.GOOGLE_MODELS = ["gemini-2.5-pro", "gemini-2.5-flash", "gemini-2.5-flash-lite", "gemini-2.5-flash-lite-preview-06-17", "gemini-2.0-flash", "gemini-1.5-pro", "gemini-1.5-pro-latest", "gemini-1.5-flash", "gemini-1.5-flash-latest", "gemini-1.5-flash-8b", "gemini-1.5-flash-8b-latest"];
-const MODELS_MAPPINGS = exports.MODELS_MAPPINGS = {
-  ...CLAUDE_MODELS_MAPPINGS
-};
-const SUPPPORTED_CLAUDE_MODELS = exports.SUPPPORTED_CLAUDE_MODELS = ["claude-3-5-haiku-20241022", "claude-3-5-haiku-latest", "claude-3-5-sonnet-20240620", "claude-3-5-sonnet-20241022", "claude-3-5-sonnet-latest", "claude-3-7-sonnet-20250219", "claude-3-7-sonnet-latest", "claude-3-haiku-20240307", "claude-4-opus-20250514", "claude-4-sonnet-20250514", "claude-opus-4-1", "claude-opus-4-1-20250805", "claude-opus-4-20250514", "claude-sonnet-4-20250514"];
-const SUPPPORTED_GPT_MODELS = exports.SUPPPORTED_GPT_MODELS = ["gpt-3.5-turbo", "gpt-3.5-turbo-0125", "gpt-3.5-turbo-0301", "gpt-3.5-turbo-0613", "gpt-3.5-turbo-1106", "gpt-3.5-turbo-16k", "gpt-3.5-turbo-16k-0613", "gpt-3.5-turbo-instruct", "gpt-3.5-turbo-instruct-0914", "gpt-4", "gpt-4-0314", "gpt-4-0613", "gpt-4-32k", "gpt-4-32k-0314", "gpt-4-32k-0613", "gpt-4-turbo", "gpt-4-turbo-2024-04-09", "gpt-4.1", "gpt-4.1-2025-04-14", "gpt-4.1-mini", "gpt-4.1-mini-2025-04-14", "gpt-4.1-nano", "gpt-4.1-nano-2025-04-14", "gpt-4o", "gpt-4o-2024-05-13", "gpt-4o-2024-08-06", "gpt-4o-2024-11-20", "gpt-4o-mini", "gpt-4o-mini-2024-07-18", "gpt-5", "gpt-5-2025-08-07", "gpt-5-chat", "gpt-5-chat-latest", "gpt-5-mini", "gpt-5-mini-2025-08-07", "gpt-5-nano", "gpt-5-nano-2025-08-07", "o1", "o1-2024-12-17", "o1-mini", "o1-mini-2024-09-12", "o1-pro", "o1-pro-2025-03-19", "o3", "o3-2025-04-16", "o3-deep-research", "o3-deep-research-2025-06-26", "o3-mini", "o3-mini-2025-01-31", "o3-pro", "o3-pro-2025-06-10", "o4-mini", "o4-mini-2025-04-16", "o4-mini-deep-research", "o4-mini-deep-research-2025-06-26"];
-const SUPPORTED_MODELS = exports.SUPPORTED_MODELS = [...SUPPPORTED_CLAUDE_MODELS, ...SUPPPORTED_GPT_MODELS];
+};

package/dist/optimized-extractors/common/extractStructuredDataUsingClaude.js

@@ -56,8 +56,7 @@ async function extractStructuredDataUsingClaude(input) {
   const anthropic = (0, _anthropicModel.createAnthropicInstance)({
     apiKey
   });
-  const modelName = _aiModelsValidations.CLAUDE_MODELS_MAPPINGS[model] ?? model;
-  const maxTokens = _aiModelsValidations.MAX_TOKENS_OVERRIDES[modelName] ?? 4096;
+  const maxTokens = _aiModelsValidations.MAX_TOKENS_OVERRIDES[model] ?? 4096;
   const response = await (0, _neverthrow.fromPromise)(anthropic.messages.create({
     max_tokens: maxTokens,
     temperature: 0,
@@ -66,7 +65,7 @@ async function extractStructuredDataUsingClaude(input) {
       role: "user",
       content
     }],
-    model: modelName,
+    model,
     tools: [{
       input_schema: processedJsonSchema,
       name: toolName,

package/dist/optimized-extractors/common/extractStructuredDataUsingGoogle.js

@@ -15,10 +15,7 @@ async function extractStructuredDataUsingGoogle(input) {
   if (!input.apiKey) {
     return (0, _neverthrow.err)(Errors.invalidInput("Google AI is only supported with a custom API key. Please provide it or use a different AI provider."));
   }
-  let model = input.model;
-  if (input.model in _aiModelsValidations.MODELS_MAPPINGS) {
-    model = _aiModelsValidations.MODELS_MAPPINGS[input.model];
-  }
+  const model = input.model;
   const googleGenAi = (0, _google.createGoogleGenerativeAI)({
     apiKey: input.apiKey
   });

package/dist/optimized-extractors/common/extractStructuredDataUsingOpenAi.js

@@ -8,7 +8,6 @@ var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 var _utils = require("./utils");
 var _Logger = require("../../common/Logger");
-var _aiModelsValidations = require("../common/aiModelsValidations");
 var _openaiModel = require("../models/openaiModel");
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
 async function extractStructuredDataUsingOpenAi(input) {
@@ -50,7 +49,7 @@ async function extractStructuredDataUsingOpenAi(input) {
     }));
     content.push(...imageContent);
   }
-  const modelName = _aiModelsValidations.MODELS_MAPPINGS[input.model] ?? input.model;
+  const modelName = input.model;
   const toolName = `extract_${entityName}`;
   const openAiInstance = (0, _openaiModel.createOpenAIInstance)({
     apiKey

package/dist/optimized-extractors/common/findTableHeaders.js

@@ -10,7 +10,7 @@ var _imageSize = require("image-size");
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
-async function getTableHeadersUsingAi(handle, identifier) {
+async function getTableHeadersUsingAi(handle, model) {
   var _response$error;
   let image;
   try {
@@ -57,7 +57,7 @@ async function getTableHeadersUsingAi(handle, identifier) {
         }
       }]
     }],
-    model: "claude-3-haiku-20240307",
+    model,
     tools: [{
       input_schema: {
         type: "object",

package/dist/optimized-extractors/common/index.js

@@ -12,15 +12,15 @@ var _extractStructuredDataUsingClaude = require("./extractStructuredDataUsingCla
 var _extractStructuredDataUsingOpenAi = require("./extractStructuredDataUsingOpenAi");
 var _utils = require("./utils");
 var _extractStructuredDataUsingGoogle = require("./extractStructuredDataUsingGoogle");
-var _aiModelsValidation = require("../types/aiModelsValidation");
+var _getModelProvider = require("../../common/getModelProvider");
 function isClaudeModel(model) {
-  return _aiModelsValidation.SUPPORTED_CLAUDE_MODELS.includes(model);
+  return (0, _getModelProvider.getModelProvider)(model) === "anthropic";
 }
 function isGoogleModel(model) {
-  return _aiModelsValidation.SUPPORTED_GOOGLE_MODELS.includes(model);
+  return (0, _getModelProvider.getModelProvider)(model) === "google_vertexai";
 }
 function isOpenAiModel(model) {
-  return _aiModelsValidation.SUPPORTED_GPT_MODELS.includes(model);
+  return (0, _getModelProvider.getModelProvider)(model) === "openai";
 }
 async function extractStructuredDataUsingAi(input) {
   let extractionResult;

package/dist/optimized-extractors/common/isTableHeaderOrFooter.js

@@ -9,7 +9,7 @@ var _zod = require("zod");
 var _neverthrow = require("neverthrow");
 var Errors = _interopRequireWildcard(require("../types/errors"));
 function _interopRequireWildcard(e, t) { if ("function" == typeof WeakMap) var r = new WeakMap(), n = new WeakMap(); return (_interopRequireWildcard = function (e, t) { if (!t && e && e.__esModule) return e; var o, i, f = { __proto__: null, default: e }; if (null === e || "object" != typeof e && "function" != typeof e) return f; if (o = t ? n : r) { if (o.has(e)) return o.get(e); o.set(e, f); } for (const t in e) "default" !== t && {}.hasOwnProperty.call(e, t) && ((i = (o = Object.defineProperty) && Object.getOwnPropertyDescriptor(e, t)) && (i.get || i.set) ? o(f, t, i) : f[t] = e[t]); return f; })(e, t); }
-async function isTableHeaderOrFooter(content) {
+async function isTableHeaderOrFooter(content, model) {
   var _response$error;
   if (!content) {
     return (0, _neverthrow.ok)({
@@ -39,7 +39,7 @@ async function isTableHeaderOrFooter(content) {
             a table header is a row that contains labels for table columns, and footer usually has pagination information or summary of the table`
       }, itemContent]
     }],
-    model: "claude-3-haiku-20240307",
+    model,
     tools: [{
       input_schema: {
         type: "object",

package/dist/optimized-extractors/common/matching/utils.js

@@ -102,7 +102,8 @@ function selectBestMatch(original, matches) {
     return {
       matchText: bestMatch.matchText,
       matchXpath: bestMatch.matchXpath,
-      matchType: bestMatch.matchType
+      matchType: bestMatch.matchType,
+      sourceText: bestMatch.sourceText
     };
   }
   const fuzzyMatches = matches.filter(match => match.isFuzzy).map(match => [match, rankMatch(original, match)]).filter(([_, rank]) => rank === "HIGH");
@@ -112,7 +113,8 @@ function selectBestMatch(original, matches) {
     return {
       matchText: bestMatch.matchText,
       matchXpath: bestMatch.matchXpath,
-      matchType: bestMatch.matchType
+      matchType: bestMatch.matchType,
+      sourceText: bestMatch.sourceText
     };
   }
   return null;

package/dist/optimized-extractors/common/modelStringSupport.test.js

@@ -0,0 +1,82 @@
+"use strict";
+
+var _vitest = require("vitest");
+var _validators = require("../validators");
+var _index = require("./index");
+(0, _vitest.describe)("strategy model accepts any string", () => {
+  (0, _vitest.it)("accepts a non-enum text model for the HTML strategy", () => {
+    const result = _validators.strategySchema.safeParse({
+      model: "gpt-4.1",
+      type: "HTML"
+    });
+    (0, _vitest.expect)(result.success).toBe(true);
+    if (result.success) {
+      (0, _vitest.expect)(result.data.model).toBe("gpt-4.1");
+    }
+  });
+  (0, _vitest.it)("accepts a non-enum vision model for the IMAGE strategy", () => {
+    const result = _validators.strategySchema.safeParse({
+      model: "gemini-2.5-pro",
+      type: "IMAGE"
+    });
+    (0, _vitest.expect)(result.success).toBe(true);
+  });
+  (0, _vitest.it)("accepts a brand new claude model name", () => {
+    const result = _validators.strategySchema.safeParse({
+      model: "claude-sonnet-4-5",
+      type: "HTML"
+    });
+    (0, _vitest.expect)(result.success).toBe(true);
+  });
+  (0, _vitest.it)("rejects an empty model", () => {
+    const result = _validators.strategySchema.safeParse({
+      model: "",
+      type: "HTML"
+    });
+    (0, _vitest.expect)(result.success).toBe(false);
+  });
+  (0, _vitest.it)("rejects an unknown strategy type", () => {
+    const result = _validators.strategySchema.safeParse({
+      model: "gpt-4o",
+      type: "FOO"
+    });
+    (0, _vitest.expect)(result.success).toBe(false);
+  });
+  (0, _vitest.it)("falls back to the default strategy when omitted", () => {
+    const result = _validators.strategySchema.safeParse(undefined);
+    (0, _vitest.expect)(result.success).toBe(true);
+    if (result.success) {
+      (0, _vitest.expect)(result.data).toEqual({
+        model: "claude-haiku-4-5-20251001",
+        type: "HTML"
+      });
+    }
+  });
+});
+(0, _vitest.describe)("provider routing handles arbitrary model strings", () => {
+  (0, _vitest.it)("routes OpenAI models by prefix", () => {
+    for (const model of ["gpt-4.1", "gpt-5", "gpt-4o", "o3", "o4-mini"]) {
+      (0, _vitest.expect)((0, _index.isOpenAiModel)(model)).toBe(true);
+      (0, _vitest.expect)((0, _index.isGoogleModel)(model)).toBe(false);
+      (0, _vitest.expect)((0, _index.isClaudeModel)(model)).toBe(false);
+    }
+  });
+  (0, _vitest.it)("routes Google models by prefix", () => {
+    for (const model of ["gemini-2.5-pro", "gemini-2.0-flash"]) {
+      (0, _vitest.expect)((0, _index.isGoogleModel)(model)).toBe(true);
+      (0, _vitest.expect)((0, _index.isOpenAiModel)(model)).toBe(false);
+      (0, _vitest.expect)((0, _index.isClaudeModel)(model)).toBe(false);
+    }
+  });
+  (0, _vitest.it)("routes Claude models by prefix", () => {
+    for (const model of ["claude-sonnet-4-5", "claude-3-7-sonnet-latest", "claude-3-haiku-20240307"]) {
+      (0, _vitest.expect)((0, _index.isClaudeModel)(model)).toBe(true);
+      (0, _vitest.expect)((0, _index.isOpenAiModel)(model)).toBe(false);
+      (0, _vitest.expect)((0, _index.isGoogleModel)(model)).toBe(false);
+    }
+  });
+  (0, _vitest.it)("no longer silently misroutes new OpenAI/Google models to Claude", () => {
+    (0, _vitest.expect)((0, _index.isOpenAiModel)("gpt-4.1")).toBe(true);
+    (0, _vitest.expect)((0, _index.isGoogleModel)("gemini-2.5-pro")).toBe(true);
+  });
+});