feat: AR-House initial commit

2026-07-03 12:24:58 -04:00
commit 047c05287a
216 changed files with 127552 additions and 0 deletions
@@ -0,0 +1,141 @@
+"""Probe Duval Property Appraiser detail page — map ALL extractable fields.
+
+Test address: 2352 Scenic View Ct, Jacksonville, FL 32218 (user's bug report).
+"""
+from pathlib import Path
+import time
+
+
+def probe():
+    from playwright.sync_api import sync_playwright
+
+    out_dir = Path(__file__).parent.parent / "_probe_out" / "duval_pa"
+    out_dir.mkdir(parents=True, exist_ok=True)
+
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=True)
+        ctx = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 Chrome/131",
+        )
+        page = ctx.new_page()
+
+        # Step 1: search page
+        print("[1] Loading search page...")
+        page.goto("https://paopropertysearch.coj.net/Basic/Search.aspx",
+                  wait_until="networkidle", timeout=20000)
+        print(f"    URL: {page.url}")
+
+        # Step 2: Fill address — 2352 SCENIC VIEW CT
+        print("[2] Filling form (2352 SCENIC VIEW CT 32218)...")
+        page.locator("#ctl00_cphBody_tbStreetNumber").fill("2352")
+        # No prefix
+        page.locator("#ctl00_cphBody_tbStreetName").fill("SCENIC VIEW")
+        # Suffix CT
+        try:
+            page.locator("#ctl00_cphBody_ddStreetSuffix").select_option(value="CT")
+        except Exception:
+            try:
+                page.locator("#ctl00_cphBody_ddStreetSuffix").select_option(label="CT")
+            except Exception:
+                pass
+        try:
+            page.locator("#ctl00_cphBody_tbZipCode").fill("32218")
+        except Exception:
+            pass
+
+        # Submit
+        page.locator("#ctl00_cphBody_bSearch").click()
+        page.wait_for_timeout(4000)
+        print(f"[3] After submit URL: {page.url}")
+        body_text = page.inner_text("body")[:500]
+        print(f"    Body preview: {body_text[:400].encode('ascii', 'replace').decode('ascii')}")
+        (out_dir / "01_results.html").write_text(page.content(), encoding="utf-8")
+        page.screenshot(path=str(out_dir / "01_results.png"), full_page=True)
+
+        # If results table, click first row to get detail page
+        results_table = page.locator("table:has(tr:has(td))").first
+        try:
+            tables = page.locator("table").all()
+            for t in tables[:10]:
+                rows = t.locator("tr").count()
+                if rows < 2:
+                    continue
+                hdrs = [(h.inner_text() or "").strip().lower() for h in t.locator("tr").first.locator("th, td").all()]
+                if any("re" in h or "parcel" in h or "owner" in h or "address" in h for h in hdrs):
+                    # First data row → click first link
+                    first_row = t.locator("tr").nth(1)
+                    link = first_row.locator("a").first
+                    if link.count() > 0:
+                        href = link.get_attribute("href")
+                        link_text = link.inner_text()
+                        print(f"[4] Clicking result link: text={link_text!r} href={href}")
+                        link.click()
+                        page.wait_for_timeout(5000)
+                        break
+        except Exception as e:
+            print(f"    Click result error: {e}")
+
+        print(f"\n[5] Detail page URL: {page.url}")
+        (out_dir / "02_detail.html").write_text(page.content(), encoding="utf-8")
+        page.screenshot(path=str(out_dir / "02_detail.png"), full_page=True)
+
+        # Dump ALL element IDs that have text
+        elements = page.evaluate("""
+            () => {
+                const out = [];
+                const all = document.querySelectorAll('[id]');
+                for (const el of all) {
+                    const txt = (el.textContent || '').trim();
+                    if (txt && txt.length < 300 && el.children.length < 4) {
+                        out.push({id: el.id, text: txt.substring(0, 200)});
+                    }
+                }
+                return out;
+            }
+        """)
+
+        print(f"\n[6] Elements with text content: {len(elements)}\n")
+        for e in elements:
+            tid = e["id"]
+            # Skip nav/utility
+            if tid.startswith("uw-") or tid.startswith("__"):
+                continue
+            txt_safe = e['text'][:120].encode('ascii', 'replace').decode('ascii')
+            print(f"  #{tid:50s} = {txt_safe!r}")
+
+        # All tables data
+        print("\n\n===== TABLES =====")
+        tables_data = page.evaluate("""
+            () => {
+                const out = [];
+                document.querySelectorAll('table').forEach((tbl, idx) => {
+                    const rows = [];
+                    for (const tr of tbl.querySelectorAll('tr')) {
+                        const cells = [];
+                        for (const c of tr.querySelectorAll('td, th')) {
+                            cells.push((c.textContent || '').trim());
+                        }
+                        if (cells.some(c => c && c.length > 0)) rows.push(cells);
+                    }
+                    if (rows.length > 0) out.push({idx, rows});
+                });
+                return out;
+            }
+        """)
+
+        for t in tables_data:
+            rows = t["rows"]
+            if not rows or len(rows) < 1:
+                continue
+            print(f"\n--- Table {t['idx']} ({len(rows)} rows) ---")
+            for r in rows[:8]:
+                line = " | ".join(c[:50] for c in r[:8])[:200]
+                line_safe = line.encode('ascii', 'replace').decode('ascii')
+                print(f"  {line_safe}")
+
+        browser.close()
+        print(f"\n[OK] saved to {out_dir}/")
+
+
+if __name__ == "__main__":
+    probe()