add multi_apps; update chrome utilities

desktop_env/evaluators/getters/__init__.py

@@ -1,7 +1,6 @@
 from .chrome import get_default_search_engine, get_cookie_data, get_bookmarks, get_open_tabs_info, get_pdf_from_url, \
     get_shortcuts_on_desktop, get_history, get_enabled_experiments, get_chrome_language, get_chrome_font_size, \
-    get_profile_name, \
-    get_number_of_search_results
+    get_profile_name, get_number_of_search_results, get_googledrive_file
 from .file import get_cloud_file, get_vm_file, get_cache_file
 from .general import get_vm_command_line
 from .impress import get_audio_in_slide
@@ -9,4 +8,4 @@ from .info import get_vm_screen_size, get_vm_window_size, get_vm_wallpaper
 from .misc import get_rule, get_accessibility_tree
 from .replay import get_replay
 from .vlc import get_vlc_playing_info, get_vlc_config
-from .vscode import get_vscode_config
+from .vscode import get_vscode_config

desktop_env/evaluators/getters/chrome.py

@@ -2,8 +2,9 @@ import json
 import logging
 import os
 import sqlite3
-from typing import Dict
-
+from typing import Dict, Any
+from pydrive.auth import GoogleAuth
+from pydrive.drive import GoogleDrive, GoogleDriveFileList, GoogleDriveFile
 from playwright.sync_api import sync_playwright
 
 logger = logging.getLogger("desktopenv.getters.chrome")
@@ -461,3 +462,49 @@ def get_number_of_search_results(env, config: Dict[str, str]):
         browser.close()
 
     return actual_count
+
+
+def get_googledrive_file(env, config: Dict[str, Any]) -> str:
+    """ Get the desired file from Google Drive based on config, return the downloaded local filepath.
+    """
+    settings_file = config.get('settings_file', 'evaluation_examples/settings/googledrive/settings.json')
+    auth = GoogleAuth(settings_file=settings_file)
+    drive = GoogleDrive(auth)
+
+    q = config['query']
+    filelist: GoogleDriveFileList = drive.ListFile({'q': q}).GetList()
+    if len(filelist) == 0: # target file not found
+        return None
+    
+    file: GoogleDriveFile = filelist[0] # HACK: if multiple candidates, just download the first one
+    _path = os.path.join(env.cache_dir, config['dest'])
+
+    try:
+        file.GetContentFile(_path, mimetype=file.metadata['mimeType'])
+    except:
+        logger.info('[ERROR]: Failed to download the file from Google Drive')
+        return None
+    return _path
+
+
+def get_googledrive_file(env, config: Dict[str, Any]) -> str:
+    """ Get the desired file from Google Drive based on config, return the downloaded local filepath.
+    """
+    settings_file = config.get('settings_file', 'evaluation_examples/settings/googledrive/settings.json')
+    auth = GoogleAuth(settings_file=settings_file)
+    drive = GoogleDrive(auth)
+
+    q = config['query']
+    filelist: GoogleDriveFileList = drive.ListFile({'q': q}).GetList()
+    if len(filelist) == 0: # target file not found
+        return None
+    
+    file: GoogleDriveFile = filelist[0] # HACK: if multiple candidates, just download the first one
+    _path = os.path.join(env.cache_dir, config['dest'])
+
+    try:
+        file.GetContentFile(_path, mimetype=file.metadata['mimeType'])
+    except:
+        logger.info('[ERROR]: Failed to download the file from Google Drive')
+        return None
+    return _path

desktop_env/evaluators/metrics/__init__.py

@@ -1,5 +1,5 @@
 from .chrome import is_expected_tabs, is_expected_bookmarks, compare_pdfs, is_cookie_deleted, is_shortcut_on_desktop, check_font_size, \
-                     check_enabled_experiments, check_history_deleted
+                     check_enabled_experiments, check_history_deleted, is_expected_search_query
 from .docs import compare_font_names, compare_subscript_contains, has_page_numbers_in_footers, compare_docx_lines
 from .docs import find_default_font, contains_page_break, compare_docx_files, compare_docx_tables, compare_line_spacing, \
     compare_insert_equation, compare_highlighted_text
@@ -20,4 +20,4 @@ from .thunderbird import check_thunderbird_prefs, check_thunderbird_filter
 from .vlc import is_vlc_playing, is_vlc_recordings_folder, is_vlc_fullscreen, compare_images, compare_audios, \
     compare_videos, check_qt_bgcone, check_one_instance_when_started_from_file,check_qt_minimal_view, check_qt_max_volume, \
     check_qt_slider_colours, check_global_key_play_pause
-from .vscode import compare_text_file, compare_config, compare_answer, is_extension_installed, check_json_settings, check_json_keybindings
+from .vscode import compare_text_file, compare_config, compare_answer, is_extension_installed, check_json_settings, check_json_keybindings

desktop_env/evaluators/metrics/chrome.py

@@ -1,4 +1,4 @@
-import logging
+import logging, re
 from typing import Any, Dict, List
 
 import fitz  # PyMuPDF
@@ -44,6 +44,15 @@ def is_expected_bookmarks(bookmarks: List[str], rule: Dict[str, Any]) -> float:
         raise TypeError(f"{rule['type']} not support yet!")
 
 
+def is_expected_search_query(active_tab_info: Dict[str, str], rules: Dict[str, Any]) -> float:
+    expected = rules['expect']
+    pattern = expected['pattern']
+    matched = re.search(pattern, active_tab_info['url'])
+    if matched:
+        return 1.
+    return 0.
+
+
 def compare_pdfs(pdf1_path, pdf2_path):
     """
     Compare two PDF files.
@@ -56,11 +65,14 @@ def compare_pdfs(pdf1_path, pdf2_path):
             for page in pdf:
                 text += page.get_text()
         return text.strip()
+    try:
+        text1 = extract_text_from_pdf(pdf1_path)
+        text2 = extract_text_from_pdf(pdf2_path)
 
-    text1 = extract_text_from_pdf(pdf1_path)
-    text2 = extract_text_from_pdf(pdf2_path)
-
-    return fuzz.ratio(text1, text2) / 100
+        return fuzz.ratio(text1, text2) / 100
+    except Exception as e:
+        logger.info(f"[ERROR]: unexpected error occurred when comparing PDF files: {e}")
+        return 0.0
 
 
 def is_cookie_deleted(cookie_data, rule):