Merge branch 'main' into feature/strict-mypy-checks

.github/workflows/ghcr-build.yml

@@ -42,7 +42,7 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.3.0
+        uses: docker/setup-qemu-action@v3.4.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR
@@ -91,7 +91,7 @@ jobs:
       - name: Checkout
         uses: actions/checkout@v4
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.3.0
+        uses: docker/setup-qemu-action@v3.4.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR

frontend/package.json

@@ -50,7 +50,7 @@
   "scripts": {
     "dev": "npm run make-i18n && cross-env VITE_MOCK_API=false react-router dev",
     "dev:mock": "npm run make-i18n && cross-env VITE_MOCK_API=true react-router dev",
-    "build": "npm run make-i18n && tsc && react-router build",
+    "build": "npm run make-i18n && npm run typecheck && react-router build",
     "start": "npx sirv-cli build/ --single",
     "test": "vitest run",
     "test:e2e": "playwright test",

openhands/agenthub/codeact_agent/codeact_agent.py

@@ -303,7 +303,6 @@ def get_observation_message(
                 and len(obs.set_of_marks) > 0
                 and self.config.enable_som_visual_browsing
                 and self.llm.vision_is_active()
-                and self.llm.is_visual_browser_tool_supported()
             ):
                 text += 'Image: Current webpage screenshot (Note that only visible portion of webpage is present in the screenshot. You may need to scroll to view the remaining portion of the web-page.)\n'
                 message = Message(

openhands/core/config/utils.py

@@ -3,6 +3,7 @@
 import pathlib
 import platform
 import sys
+from ast import literal_eval
 from types import UnionType
 from typing import Any, MutableMapping, get_args, get_origin
 from uuid import uuid4
@@ -72,6 +73,9 @@ def set_attr_from_env(sub_config: BaseModel, prefix=''):
                     # Attempt to cast the env var to type hinted in the dataclass
                     if field_type is bool:
                         cast_value = str(value).lower() in ['true', '1']
+                    # parse dicts like SANDBOX_RUNTIME_STARTUP_ENV_VARS
+                    elif get_origin(field_type) is dict:
+                        cast_value = literal_eval(value)
                     else:
                         cast_value = field_type(value)
                     setattr(sub_config, field_name, cast_value)

openhands/llm/llm.py

@@ -75,16 +75,6 @@
     'o3-mini',
 ]
 
-# visual browsing tool supported models
-# This flag is needed since gpt-4o and gpt-4o-mini do not allow passing image_urls with role='tool'
-VISUAL_BROWSING_TOOL_SUPPORTED_MODELS = [
-    'claude-3-5-sonnet',
-    'claude-3-5-sonnet-20240620',
-    'claude-3-5-sonnet-20241022',
-    'o1-2024-12-17',
-]
-
-
 REASONING_EFFORT_SUPPORTED_MODELS = [
     'o1-2024-12-17',
     'o1',
@@ -495,15 +485,6 @@ def is_function_calling_active(self) -> bool:
         """
         return self._function_calling_active
 
-    def is_visual_browser_tool_supported(self) -> bool:
-        return (
-            self.config.model in VISUAL_BROWSING_TOOL_SUPPORTED_MODELS
-            or self.config.model.split('/')[-1] in VISUAL_BROWSING_TOOL_SUPPORTED_MODELS
-            or any(
-                m in self.config.model for m in VISUAL_BROWSING_TOOL_SUPPORTED_MODELS
-            )
-        )
-
     def _post_completion(self, response: ModelResponse) -> float:
         """Post-process the completion response.
 

openhands/resolver/patching/apply.py

@@ -94,12 +94,17 @@ def apply_diff(diff, text, reverse=False, use_patch=False):
                     hunk=hunk,
                 )
             if lines[old - 1] != line:
-                raise HunkApplyException(
-                    'context line {n}, "{line}" does not match "{sl}"'.format(
-                        n=old, line=line, sl=lines[old - 1]
-                    ),
-                    hunk=hunk,
-                )
+                # Try to normalize whitespace by replacing multiple spaces with a single space
+                # This helps with patches that have different indentation levels
+                normalized_line = ' '.join(line.split())
+                normalized_source = ' '.join(lines[old - 1].split())
+                if normalized_line != normalized_source:
+                    raise HunkApplyException(
+                        'context line {n}, "{line}" does not match "{sl}"'.format(
+                            n=old, line=line, sl=lines[old - 1]
+                        ),
+                        hunk=hunk,
+                    )
 
     # for calculating the old line
     r = 0

openhands/runtime/action_execution_server.py

@@ -21,6 +21,7 @@
 from pathlib import Path
 from zipfile import ZipFile
 
+import psutil
 from fastapi import Depends, FastAPI, HTTPException, Request, UploadFile
 from fastapi.exceptions import RequestValidationError
 from fastapi.responses import JSONResponse, StreamingResponse
@@ -108,6 +109,22 @@ def __init__(
         self.last_execution_time = self.start_time
         self._initialized = False
 
+        if _override_max_memory_gb := os.environ.get('RUNTIME_MAX_MEMORY_GB', None):
+            self.max_memory_gb = int(_override_max_memory_gb)
+            logger.info(
+                f'Setting max memory to {self.max_memory_gb}GB (according to the RUNTIME_MAX_MEMORY_GB environment variable)'
+            )
+        else:
+            # Get available system memory
+            total_memory_gb = psutil.virtual_memory().total / (
+                1024 * 1024 * 1024
+            )  # Convert to GB
+            self.max_memory_gb = int(max(0.5, total_memory_gb - 1.0))
+            # Reserve 1GB as head room, minimum of 0.5GB
+            logger.info(
+                f'Total memory: {total_memory_gb}GB, setting limit to {self.max_memory_gb}GB (reserved 1GB for action execution server, minimum 0.5GB)'
+            )
+
     @property
     def initial_cwd(self):
         return self._initial_cwd
@@ -120,8 +137,10 @@ async def ainit(self):
             no_change_timeout_seconds=int(
                 os.environ.get('NO_CHANGE_TIMEOUT_SECONDS', 30)
             ),
+            max_memory_mb=self.max_memory_gb * 1024,
         )
         self.bash_session.initialize()
+
         await wait_all(
             (self._init_plugin(plugin) for plugin in self.plugins_to_load),
             timeout=30,

openhands/runtime/impl/action_execution/action_execution_client.py

@@ -142,11 +142,13 @@ def copy_from(self, path: str) -> Path:
                 stream=True,
                 timeout=30,
             ) as response:
-                temp_file = tempfile.NamedTemporaryFile(delete=False)
-                for chunk in response.iter_content(chunk_size=8192):
-                    if chunk:  # filter out keep-alive new chunks
-                        temp_file.write(chunk)
-                return Path(temp_file.name)
+                with tempfile.NamedTemporaryFile(delete=False) as temp_file:
+                    total_length = 0
+                    for chunk in response.iter_content(chunk_size=8192):
+                        if chunk:  # filter out keep-alive new chunks
+                            total_length += len(chunk)
+                            temp_file.write(chunk)
+                    return Path(temp_file.name)
         except requests.Timeout:
             raise TimeoutError('Copy operation timed out')
 

openhands/runtime/impl/remote/remote_runtime.py

@@ -212,13 +212,17 @@ def _start_runtime(self):
             plugins=self.plugins,
             app_config=self.config,
         )
+        environment = {
+            'DEBUG': 'true'
+            if self.config.debug or os.environ.get('DEBUG', 'false').lower() == 'true'
+            else {},
+        }
+        environment.update(self.config.sandbox.runtime_startup_env_vars)
         start_request = {
             'image': self.container_image,
             'command': command,
             'working_dir': '/openhands/code/',
-            'environment': {'DEBUG': 'true'}
-            if self.config.debug or os.environ.get('DEBUG', 'false').lower() == 'true'
-            else {},
+            'environment': environment,
             'session_id': self.sid,
             'resource_factor': self.config.sandbox.remote_runtime_resource_factor,
         }

openhands/runtime/utils/bash.py

@@ -175,25 +175,32 @@ def __init__(
         work_dir: str,
         username: str | None = None,
         no_change_timeout_seconds: int = 30,
+        max_memory_mb: int | None = None,
     ):
         self.NO_CHANGE_TIMEOUT_SECONDS = no_change_timeout_seconds
         self.work_dir = work_dir
         self.username = username
         self._initialized = False
+        self.max_memory_mb = max_memory_mb
 
     def initialize(self):
         self.server = libtmux.Server()
-        window_command = '/bin/bash'
+        _shell_command = '/bin/bash'
         if self.username in ['root', 'openhands']:
             # This starts a non-login (new) shell for the given user
-            window_command = f'su {self.username} -'
+            _shell_command = f'su {self.username} -'
         # otherwise, we are running as the CURRENT USER (e.g., when running LocalRuntime)
+        if self.max_memory_mb is not None:
+            window_command = (
+                f'prlimit --as={self.max_memory_mb * 1024 * 1024} {_shell_command}'
+            )
+        else:
+            window_command = _shell_command
 
+        logger.debug(f'Initializing bash session with command: {window_command}')
         session_name = f'openhands-{self.username}-{uuid.uuid4()}'
         self.session = self.server.new_session(
             session_name=session_name,
-            window_name='bash',
-            window_command=window_command,
             start_directory=self.work_dir,
             kill_session=True,
             x=1000,
@@ -207,6 +214,7 @@ def initialize(self):
         # We need to create a new pane because the initial pane's history limit is (default) 2000
         _initial_window = self.session.attached_window
         self.window = self.session.new_window(
+            window_name='bash',
             window_shell=window_command,
             start_directory=self.work_dir,
         )