fix chrome

desktop_env/evaluators/getters/chrome.py

@@ -1200,6 +1200,32 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]):
             elements = target_page.query_selector_all(selector)
             return [element.text_content().strip() for element in elements if element]
 
+        def safely_get_direct_text_nodes_playwright(selector):
+            """
+            Extract all direct text node contents under the specified selector element (excluding text inside child div, span, etc.).
+            Returns a list of lists, each sublist contains the direct text nodes of one element.
+            Suitable for structures like: <div>SEA<div class="aura-separator"></div>NYC</div>
+            """
+            elements = target_page.query_selector_all(selector)
+            results = []
+            for element in elements:
+                texts = element.evaluate('''
+                    (node) => Array.from(node.childNodes)
+                        .filter(n => n.nodeType === Node.TEXT_NODE)
+                        .map(n => n.textContent.trim())
+                        .filter(Boolean)
+                ''')
+                results.append(texts)
+            return results[0]
+        
+        def safely_get_direct_li_playwright(selector):
+            elements = target_page.query_selector_all(selector + " li.catAllProducts")
+            return [element.query_selector('span').inner_text().strip() for element in elements if element.query_selector('span')]
+
+        def safely_get_only_child_text_content(selector):
+            elements = target_page.query_selector_all(selector)
+            return [element.query_selector('h3').text_content().strip() for element in elements if element.query_selector('h3')]
+
         if config["category"] == "class":
             class_multiObject = config.get("class_multiObject", {})
             for class_name, object_dict in class_multiObject.items():
@@ -1208,6 +1234,41 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]):
                     index = int(order_key)
                     if len(elements_texts) > index:
                         return_json[key] = elements_texts[index]
+                        
+            class_multiObject_child = config.get("class_multiObject_child", {})
+            for class_name, object_dict in class_multiObject_child.items():
+                elements_texts = safely_get_direct_text_nodes_playwright("." + class_name)
+                for order_key, key in object_dict.items():
+                    index = int(order_key)
+                    if len(elements_texts) > index:
+                        return_json[key] = elements_texts[index]
+            
+            class_multiObject_only_child = config.get("class_multiObject_only_child", {})
+            for class_name, object_dict in class_multiObject_only_child.items():
+                elements_texts = safely_get_only_child_text_content("." + class_name)
+                for order_key, key in object_dict.items():
+                    index = int(order_key)
+                    if len(elements_texts) > index:
+                        return_json[key] = elements_texts[index]
+
+            class_multiObject_search_exist = config.get("class_multiObject_search_exist", {})
+            for class_name, object_list in class_multiObject_search_exist.items():
+                elements_texts = safely_get_text_content("." + class_name)
+                for each_object in object_list:
+                    if each_object == "is_other_exist":
+                        continue
+                    if each_object in elements_texts:
+                        return_json[each_object] = True
+                    else:
+                        return_json[each_object] = False
+                if "is_other_exist" in object_list:
+                    for each_element in elements_texts:
+                        if each_element not in object_list:
+                            return_json["is_other_exist"] = True
+                            break
+                    if "is_other_exist" not in return_json.keys():
+                        return_json["is_other_exist"] = False
+                
 
             class_singleObject = config.get("class_singleObject", {})
             for class_name, key in class_singleObject.items():
@@ -1236,6 +1297,55 @@ def get_active_tab_html_parse(env, config: Dict[str, Any]):
                 inputs = target_page.locator(f"xpath={xpath}")
                 if inputs.count() > 0:
                     return_json[key] = inputs.first.input_value().strip()
+        
+        elif config["category"] == "class&url":
+            class_multiObject = config.get("class_multiObject", {})
+            for class_name, object_list in class_multiObject.items():
+                elements_texts = safely_get_text_content("." + class_name)
+                for each_key in object_list:
+                    if any(each_key.lower() == text.lower() for text in elements_texts):
+                        return_json[each_key.lower()] = True
+
+                for each_key in elements_texts:
+                    # each_key.lower() not in object_list.lower():
+                    if all(each_key.lower() not in item.lower() for item in object_list):
+                        return_json["is_other_exist"] = True
+                        break
+                if "is_other_exist" not in return_json.keys():
+                    return_json["is_other_exist"] = False
+            
+            class_multiObject_li = config.get("class_multiObject_li", {})
+            for class_name, object_list in class_multiObject_li.items():
+                elements_texts = safely_get_direct_li_playwright("." + class_name)
+                for each_key in object_list:
+                    if any(each_key.lower() == text.lower() for text in elements_texts):
+                        return_json[each_key.lower()] = True
+                
+                for each_key in elements_texts:
+                    # each_key.lower() not in object_list.lower():
+                    if all(each_key.lower() not in item.lower() for item in object_list):
+                        return_json["is_other_exist"] = True
+                        break
+                if "is_other_exist" not in return_json.keys():
+                    return_json["is_other_exist"] = False
+
+            url_include_expected = config.get("url_include_expected", [])
+            for key in url_include_expected:
+                if key.lower() in target_page.url.lower():
+                    if key.lower() not in return_json.keys():
+                        return_json[key.lower()] = True
+                else:
+                    if key.lower() not in return_json.keys():
+                        return_json[key.lower()] = False
+            
+            url_include_expected_multichoice = config.get("url_include_expected_multichoice", {})
+            for key, value in url_include_expected_multichoice.items():
+                if key.lower() in target_page.url.lower():
+                    if value.lower() not in return_json.keys():
+                        return_json[value.lower()] = True
+                else:
+                    if value.lower() not in return_json.keys():
+                        return_json[value.lower()] = False
 
         browser.close()
     return return_json
@@ -1274,13 +1384,14 @@ def get_gotoRecreationPage_and_get_html_content(env, config: Dict[str, Any]):
             browser = p.chromium.connect_over_cdp(remote_debugging_url)
         page = browser.new_page()
         page.goto("https://www.recreation.gov/")
-        page.fill("input#hero-search-input", "Albion Basin")
+        page.fill("input#hero-search-input", "Diamond")
         page.click("button.nav-search-button")
         print("after first click")
-        time.sleep(2)
+        time.sleep(10)
         # Assuming .search-result-highlight--success leads to a new page or requires page load
         with page.expect_popup() as popup_info:
             page.click(".search-result-highlight--success")
+        time.sleep(30)
         print("after second click")
         newpage = popup_info.value
         newpage.wait_for_load_state()
@@ -1362,3 +1473,57 @@ def get_url_dashPart(env, config: Dict[str, str]):
         return dash_part
     elif config["returnType"] == "json":
         return {config["key"]: dash_part}
+
+
+def get_url_path_parse(env, config: Dict[str, str]):
+    """
+    Parse Macy's product url path, extract:
+    - mens_clothing: true if 'mens-clothing' in path, else None
+    - t_shirts: true if any key 'Top_style' or 'Product_department' value is 'T-shirts', else None
+    - Men_regular_size_t, Price_discount_range (as list), Sleeve_length: as before, None if not found
+    All fields are None if not found for robustness.
+    """
+    from urllib.parse import urlparse, unquote
+    result = {}
+    # 1. Parse URL
+    active_tab_url = get_active_url_from_accessTree(env, config)
+    if active_tab_url is None:
+        return None
+    parsed = urlparse(active_tab_url)
+    path = unquote(parsed.path)
+    result = {}
+    # mens_clothing
+    result['mens_clothing'] = True if 'mens-clothing' in path else None
+    # key-value
+    path_parts = path.strip('/').split('/')
+    key_value_json = {}
+    tshirts_flag = False
+    if "mens-t-shirts" in path:
+        tshirts_flag = True
+    for i in range(len(path_parts)-1):
+        if ',' in path_parts[i] and ',' in path_parts[i+1]:
+            keys = [k.strip() for k in path_parts[i].split(',')]
+            values = [v.strip() for v in path_parts[i+1].split(',')]
+            for k, v in zip(keys, values):
+                if k == "Price_discount_range":
+                    key_value_json[k] = [item.strip() for item in v.split('|')] if v else None
+                else:
+                    key_value_json[k] = v if v else None
+                if (k == 'Top_style' or k == 'Product_department') and (v == 'T-shirts' or v == 'T-Shirts' or v == 'T-Shirt'):
+                    tshirts_flag = True
+            break
+    for field in ['Men_regular_size_t', 'Price_discount_range', 'Sleeve_length']:
+        if field not in key_value_json:
+            key_value_json[field] = None
+    result['t_shirts'] = tshirts_flag if tshirts_flag else None
+    # parse_keys
+    for key in config["parse_keys"]:
+        if key in key_value_json:
+            if key == "Price_discount_range":
+                if '50_PERCENT_ off & more' in key_value_json[key] and not '30_PERCENT_ off & more' in key_value_json[key] and not '20_PERCENT_ off & more' in key_value_json[key]:
+                    result[key] = '50_PERCENT_ off & more'
+                else:
+                    result[key] = 'not_50_PERCENT_ off & more'
+            else:
+                result[key] = key_value_json[key]
+    return result