lzy/sci-gui-agent-benchmark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'origin/main'

Browse Source 
# Conflicts:
#	mm_agents/gpt_4v_agent.py
This commit is contained in:
Timothyxxx
2024-02-02 14:37:23 +08:00
parent 068c6f5769 3184f091a8
commit e07a3d52ce
25 changed files with 1141 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

54
desktop_env/controllers/setup.py
View File

@@ -7,6 +7,7 @@ import uuid
import tempfile
from typing import Any, Union, Optional
from typing import Dict, List
import os
import requests
from pydrive.auth import GoogleAuth
@@ -114,6 +115,7 @@ class SetupController:
if not os.path.exists(cache_path):
max_retries = 3
downloaded = False
e = None
for i in range(max_retries):
try:
response = requests.get(url, stream=True)
@@ -128,7 +130,7 @@ class SetupController:
break
except requests.RequestException as e:
logger.error(f"Failed to download {url}. Retrying... ({max_retries - i - 1} attempts left)")
logger.error(f"Failed to download {url} caused by {e}. Retrying... ({max_retries - i - 1} attempts left)")
if not downloaded:
raise requests.RequestException(f"Failed to download {url}. No retries left. Error: {e}")
@@ -344,39 +346,49 @@ class SetupController:
port = 9222 # fixme: this port is hard-coded, need to be changed from config file
remote_debugging_url = f"http://{host}:{port}"
with sync_playwright() as p:
logger.info("Connect to Chrome @: %s", remote_debugging_url)
logger.debug("PLAYWRIGHT ENV: %s", repr(os.environ))
for attempt in range(15):
if attempt>0:
time.sleep(5)
browser = None
for attempt in range(15):
with sync_playwright() as p:
try:
browser = p.chromium.connect_over_cdp(remote_debugging_url)
break
#break
except Exception as e:
if attempt < 14:
logger.error(f"Attempt {attempt + 1}: Failed to connect, retrying. Error: {e}")
time.sleep(1)
#time.sleep(10)
continue
else:
logger.error(f"Failed to connect after multiple attempts: {e}")
raise e
if not browser:
return
if not browser:
return
for i, url in enumerate(urls_to_open):
# Use the first context (which should be the only one if using default profile)
if i == 0:
context = browser.contexts[0]
logger.info("Opening %s...", urls_to_open)
for i, url in enumerate(urls_to_open):
# Use the first context (which should be the only one if using default profile)
if i == 0:
context = browser.contexts[0]
page = context.new_page() # Create a new page (tab) within the existing context
page.goto(url, timeout=60000)
logger.info(f"Opened tab {i + 1}: {url}")
page = context.new_page() # Create a new page (tab) within the existing context
try:
page.goto(url, timeout=60000)
except:
logger.warning("Opening %s exceeds time limit", url) # only for human test
logger.info(f"Opened tab {i + 1}: {url}")
if i == 0:
# clear the default tab
default_page = context.pages[0]
default_page.close()
if i == 0:
# clear the default tab
default_page = context.pages[0]
default_page.close()
# Do not close the context or browser; they will remain open after script ends
return browser, context
# Do not close the context or browser; they will remain open after script ends
return browser, context
def _chrome_close_tabs_setup(self, urls_to_close: List[str]):
time.sleep(5) # Wait for Chrome to finish launching
@@ -552,4 +564,4 @@ class SetupController:
else:
raise NotImplementedError
return browser, context
return browser, context