Use Zillow parcel hints for CAD lookup

skills/web-automation/scripts/zillow-identifiers.js

@@ -0,0 +1,81 @@
+#!/usr/bin/env node
+
+import { pathToFileURL } from "node:url";
+
+import {
+  createPageSession,
+  dismissCommonOverlays,
+  fail,
+  gotoListing,
+  parseTarget,
+  runWithOperationTimeout,
+} from "./real-estate-photo-common.js";
+import {
+  extractZillowIdentifierHintsFromNextDataScript,
+  extractZillowIdentifierHintsFromText,
+} from "./zillow-photo-data.js";
+
+export async function extractZillowIdentifierHints(rawUrl, options = {}) {
+  const requestedUrl = parseTarget(rawUrl);
+  const { context, page } = await createPageSession({ headless: process.env.HEADLESS !== "false" });
+  const closeContext = async () => {
+    await context.close().catch(() => {});
+  };
+
+  try {
+    return await runWithOperationTimeout(
+      "Zillow identifier extraction",
+      async () => {
+        await gotoListing(page, requestedUrl);
+        await dismissCommonOverlays(page);
+
+        const [scriptText, bodyText] = await Promise.all([
+          page.locator("#__NEXT_DATA__").textContent().catch(() => null),
+          page.evaluate(() => document.body?.innerText || "").catch(() => ""),
+        ]);
+
+        const structuredHints = extractZillowIdentifierHintsFromNextDataScript(scriptText || "");
+        const textHints = extractZillowIdentifierHintsFromText(bodyText || "");
+        const parcelId = structuredHints.parcelId || textHints.parcelId || null;
+        const apn = structuredHints.apn || textHints.apn || null;
+        const notes = [];
+
+        if (parcelId) {
+          notes.push(`Zillow listing exposed parcel/APN identifier ${parcelId}.`);
+        } else if (apn) {
+          notes.push(`Zillow listing exposed APN/tax identifier ${apn}.`);
+        } else {
+          notes.push("Zillow listing did not expose a parcel/APN identifier that could be extracted reliably.");
+        }
+
+        return {
+          source: "zillow",
+          requestedUrl,
+          finalUrl: page.url(),
+          parcelId,
+          apn,
+          notes,
+        };
+      },
+      {
+        timeoutMs: Number(options.timeoutMs || 0) || undefined,
+        onTimeout: closeContext,
+      }
+    );
+  } finally {
+    await closeContext();
+  }
+}
+
+async function main() {
+  try {
+    const result = await extractZillowIdentifierHints(process.argv[2]);
+    process.stdout.write(`${JSON.stringify(result, null, 2)}\n`);
+  } catch (error) {
+    fail("Zillow identifier extraction failed.", error instanceof Error ? error.message : String(error));
+  }
+}
+
+if (process.argv[1] && import.meta.url === pathToFileURL(process.argv[1]).href) {
+  main();
+}

skills/web-automation/scripts/zillow-photo-data.js

@@ -58,6 +58,92 @@ export function extractZillowStructuredPhotoCandidatesFromNextDataScript(scriptT
   return out;
 }
 
+function collapseIdentifier(value) {
+  return String(value || "").replace(/\s+/g, " ").trim();
+}
+
+function isLikelyIdentifier(value) {
+  return /^[A-Z0-9-]{4,40}$/i.test(collapseIdentifier(value));
+}
+
+function visitForIdentifierHints(node, hints) {
+  if (!node || typeof node !== "object") return;
+  if (Array.isArray(node)) {
+    for (const item of node) {
+      visitForIdentifierHints(item, hints);
+    }
+    return;
+  }
+
+  for (const [key, value] of Object.entries(node)) {
+    const normalizedKey = key.toLowerCase();
+    if ((normalizedKey === "parcelid" || normalizedKey === "parcelnumber") && hints.parcelId == null) {
+      if (typeof value === "string" || typeof value === "number") {
+        const candidate = collapseIdentifier(value);
+        if (isLikelyIdentifier(candidate)) {
+          hints.parcelId = candidate;
+        }
+      }
+    }
+    if ((normalizedKey === "apn" || normalizedKey === "apnnumber" || normalizedKey === "taxparcelid" || normalizedKey === "taxid") && hints.apn == null) {
+      if (typeof value === "string" || typeof value === "number") {
+        const candidate = collapseIdentifier(value);
+        if (isLikelyIdentifier(candidate)) {
+          hints.apn = candidate;
+        }
+      }
+    }
+
+    if (value && typeof value === "object") {
+      visitForIdentifierHints(value, hints);
+    }
+  }
+}
+
+export function extractZillowIdentifierHintsFromNextDataScript(scriptText) {
+  if (typeof scriptText !== "string" || !scriptText.trim()) {
+    return {};
+  }
+
+  let nextData;
+  try {
+    nextData = JSON.parse(scriptText);
+  } catch {
+    return {};
+  }
+
+  const hints = {};
+  visitForIdentifierHints(nextData, hints);
+
+  const cacheText = nextData?.props?.pageProps?.componentProps?.gdpClientCache;
+  if (typeof cacheText === "string" && cacheText.trim()) {
+    try {
+      visitForIdentifierHints(JSON.parse(cacheText), hints);
+    } catch {
+      // Ignore cache parse failures; base next-data parse already succeeded.
+    }
+  }
+
+  return hints;
+}
+
+export function extractZillowIdentifierHintsFromText(text) {
+  const source = typeof text === "string" ? text : "";
+  const hints = {};
+
+  const parcelMatch = source.match(/\b(?:parcel|parcel number|parcel #|tax parcel)(?:\s*(?:number|#|no\.?))?\s*[:#]?\s*([A-Z0-9-]{4,40})\b/i);
+  if (parcelMatch) {
+    hints.parcelId = collapseIdentifier(parcelMatch[1]);
+  }
+
+  const apnMatch = source.match(/\b(?:apn|apn #|apn no\.?|tax id)(?:\s*(?:number|#|no\.?))?\s*[:#]?\s*([A-Z0-9-]{4,40})\b/i);
+  if (apnMatch) {
+    hints.apn = collapseIdentifier(apnMatch[1]);
+  }
+
+  return hints;
+}
+
 const DEFAULT_MINIMUM_TRUSTED_STRUCTURED_PHOTO_COUNT = 12;
 
 export function shouldUseStructuredZillowPhotos(candidates, options = {}) {

skills/web-automation/scripts/zillow-photo-data.test.mjs

@@ -3,6 +3,7 @@ import assert from "node:assert/strict";
 
 import {
   extractZillowStructuredPhotoCandidatesFromNextDataScript,
+  extractZillowIdentifierHintsFromNextDataScript,
   shouldUseStructuredZillowPhotos,
 } from "./zillow-photo-data.js";
 
@@ -119,3 +120,27 @@ test("shouldUseStructuredZillowPhotos returns false for a tiny structured set wh
     false
   );
 });
+
+test("extractZillowIdentifierHintsFromNextDataScript finds parcel and APN style identifiers", () => {
+  const scriptText = JSON.stringify({
+    props: {
+      pageProps: {
+        componentProps: {
+          gdpClientCache: JSON.stringify({
+            SomeQuery: {
+              property: {
+                parcelId: "1234567890",
+                apn: "APN-77-55",
+              },
+            },
+          }),
+        },
+      },
+    },
+  });
+
+  assert.deepEqual(extractZillowIdentifierHintsFromNextDataScript(scriptText), {
+    parcelId: "1234567890",
+    apn: "APN-77-55",
+  });
+});