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

Robust Evaluation, Blocking File Open, Grader Sensitivity, and LibreOffice Writer Fixes (#217)

Browse Source 
* Refactor evaluator structure in LibreOffice Writer example JSON to support multiple expected and result files, enhancing evaluation flexibility.

* Update instance type to t3.large and add VNC access URL logging for allocated VMs, enhancing remote access capabilities.

* Update instance type to t3.large and add VNC access URL logging for allocated VMs, enhancing remote access capabilities.

* Update time format in get_vm_file function to include hours, minutes, and seconds for more precise file naming with time suffix.

* More delay for 936321ce-5236-426a-9a20-e0e3c5dc536f; support one more potential solutions.

* Enhance SetupController with configurable retry limit and improved error handling for file opening requests. Introduce new function to compare unique training records, and update logging for better debugging. Adjust JSON examples for evaluation to support multiple expected and result files.

* Clean debug code

---------

Co-authored-by: yuanmengqi <yuanmengqi@mail.ustc.edu.cn>
This commit is contained in:
Tianbao Xie
2025-06-16 21:37:19 +08:00
committed by GitHub
parent 347238e17e
commit 4e11eafd1d
13 changed files with 523 additions and 135 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
desktop_env/controllers/setup.py
View File

@@ -36,6 +36,8 @@ FILE_PATH = os.path.dirname(os.path.abspath(__file__))
init_proxy_pool(PROXY_CONFIG_FILE) # initialize the global proxy pool
MAX_RETRIES = 20
class SetupController:
def __init__(self, vm_ip: str, server_port: int = 5000, chromium_port: int = 9222, vlc_port: int = 8080, cache_dir: str = "cache"):
self.vm_ip: str = vm_ip
@@ -64,16 +66,16 @@ class SetupController:
# make sure connection can be established
logger.info(f"try to connect {self.http_server}")
retry = 0
while retry < 50:
while retry < MAX_RETRIES:
try:
_ = requests.get(self.http_server + "/terminal")
break
except:
time.sleep(5)
retry += 1
logger.info(f"retry: {retry}/50")
logger.info(f"retry: {retry}/{MAX_RETRIES}")
if retry == 50:
if retry == MAX_RETRIES:
return False
@@ -219,13 +221,14 @@ class SetupController:
# send request to server to open file
try:
response = requests.post(self.http_server + "/setup" + "/open_file", headers=headers, data=payload)
if response.status_code == 200:
logger.info("Command executed successfully: %s", response.text)
else:
logger.error("Failed to open file. Status code: %s", response.text)
# The server-side call is now blocking and can take time.
# We set a timeout that is slightly longer than the server's timeout (1800s).
response = requests.post(self.http_server + "/setup" + "/open_file", headers=headers, data=payload, timeout=1810)
response.raise_for_status() # This will raise an exception for 4xx and 5xx status codes
logger.info("Command executed successfully: %s", response.text)
except requests.exceptions.RequestException as e:
logger.error("An error occurred while trying to send the request: %s", e)
logger.error(f"Failed to open file '{path}'. An error occurred while trying to send the request or the server responded with an error: {e}")
raise Exception(f"Failed to open file '{path}'. An error occurred while trying to send the request or the server responded with an error: {e}") from e
def _launch_setup(self, command: Union[str, List[str]], shell: bool = False):
if not command: