Merge pull request #24 from codeflash-ai/VSC-workspace-integration

all roads lead to VSC

Author: KRRT7

.github/workflows/pr_agent.yml

@@ -0,0 +1,20 @@
+on:
+  pull_request:
+    types: [opened, reopened, ready_for_review]
+  issue_comment:
+jobs:
+  pr_agent_job:
+    if: ${{ github.event.sender.type != 'Bot' }}
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+      contents: write
+    name: Run pr agent on every pull request, respond to user comments
+    steps:
+      - name: PR Agent action step
+        id: pragent
+        uses: qodo-ai/pr-agent@main
+        env:
+          OPENAI_KEY: ${{ secrets.OPENAI_KEY }}
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

codeflash.code-workspace

@@ -0,0 +1,80 @@
+{
+    "folders": [
+        {
+            "path": ".",
+            "name": "codeflash",
+            "extensions": [
+                "charliermarsh.ruff",
+                "ms-python.python",
+            ]
+        }
+    ],
+    "settings": {
+        "python.defaultInterpreterPath": "~/miniforge3/envs/codeflash312/bin/python",
+        "python.terminal.activateEnvironment": true,
+        "python.testing.pytestEnabled": true,
+        "python.testing.pytestArgs": ["tests/", "-vv"],
+    },
+    "launch": {
+        "version": "0.2.0",
+        "configurations": [
+            {
+                "name": "bubble_sort",
+                "type": "debugpy",
+                "request": "launch",
+                "program": "${workspaceFolder:codeflash}/codeflash/main.py",
+                "args": [
+                    "--file",
+                    "code_to_optimize/bubble_sort.py",
+                    "--module-root",
+                    "${workspaceFolder:codeflash}",
+                    "--function",
+                    "sorter",
+                    "--test-framework",
+                    "pytest",
+                    "--tests-root",
+                    "code_to_optimize/tests/pytest"
+                ],
+                "cwd": "${workspaceFolder:codeflash}",
+                "console": "integratedTerminal",
+                "env": {
+                    "PYTHONUNBUFFERED": "1"
+                },
+            },
+            {
+                "name": "bubble_sort -all",
+                "type": "debugpy",
+                "request": "launch",
+                "program": "${workspaceFolder:codeflash}/codeflash/main.py",
+                "args": [
+                    "--all",
+                    "--test-framework",
+                    "pytest",
+                    "--tests-root",
+                    "code_to_optimize/tests/pytest",
+                    "--module-root",
+                    "code_to_optimize"
+                ],
+                "cwd": "${workspaceFolder:codeflash}",
+                "console": "integratedTerminal",
+                "env": {
+                    "PYTHONUNBUFFERED": "1"
+                },
+            },
+            {
+                "name": "bubble_sort --file bubble_sort.py (MBR)",
+                "type": "debugpy",
+                "request": "launch",
+                "program": "${workspaceFolder:codeflash}/codeflash/main.py",
+                "args": [
+                    "--all",
+                ],
+                "cwd": "/Users/krrt7/Desktop/work/my-best-repo",
+                "console": "integratedTerminal",
+                "env": {
+                    "PYTHONUNBUFFERED": "1"
+                },
+            }
+        ]
+    }
+}

codeflash/api/aiservice.py

@@ -29,6 +29,7 @@ def __init__(self) -> None:
     def get_aiservice_base_url(self) -> str:
         if os.environ.get("CODEFLASH_AIS_SERVER", default="prod").lower() == "local":
             logger.info("Using local AI Service at http://localhost:8000")
+            console.rule()
             return "http://localhost:8000"
         return "https://app.codeflash.ai"
 

codeflash/optimization/function_optimizer.py

@@ -1072,7 +1072,6 @@ def run_and_parse_tests(
                     cwd=self.project_root,
                     test_env=test_env,
                     pytest_timeout=INDIVIDUAL_TESTCASE_TIMEOUT,
-                    pytest_cmd=self.test_cfg.pytest_cmd,
                     verbose=True,
                     enable_coverage=enable_coverage,
                 )
@@ -1081,8 +1080,8 @@ def run_and_parse_tests(
                     test_files,
                     cwd=self.project_root,
                     test_env=test_env,
-                    pytest_timeout=INDIVIDUAL_TESTCASE_TIMEOUT,
                     pytest_cmd=self.test_cfg.pytest_cmd,
+                    pytest_timeout=INDIVIDUAL_TESTCASE_TIMEOUT,
                     pytest_target_runtime_seconds=testing_time,
                     pytest_min_loops=pytest_min_loops,
                     pytest_max_loops=pytest_max_loops,

codeflash/verification/test_runner.py

@@ -50,9 +50,12 @@ def run_behavioral_tests(
                 )
             else:
                 test_files.append(str(file.instrumented_behavior_file_path))
+        pytest_cmd_list = (
+            shlex.split(f"{SAFE_SYS_EXECUTABLE} -m pytest", posix=IS_POSIX)
+            if pytest_cmd == "pytest"
+            else [SAFE_SYS_EXECUTABLE, "-m", *shlex.split(pytest_cmd, posix=IS_POSIX)]
+        )
         test_files = list(set(test_files))  # remove multiple calls in the same test function
-        pytest_cmd_list = shlex.split(pytest_cmd, posix=IS_POSIX)
-
         common_pytest_args = [
             "--capture=tee-sys",
             f"--timeout={pytest_timeout}",
@@ -77,17 +80,19 @@ def run_behavioral_tests(
             )  # this cleanup is necessary to avoid coverage data from previous runs, if there are any,
             # then the current run will be appended to the previous data, which skews the results
             logger.debug(cov_erase)
+            coverage_cmd = [SAFE_SYS_EXECUTABLE, "-m", "coverage", "run", f"--rcfile={coveragercfile.as_posix()}", "-m"]
+
+            if pytest_cmd == "pytest":
+                coverage_cmd.extend(["pytest"])
+            else:
+                coverage_cmd.extend(shlex.split(pytest_cmd, posix=IS_POSIX)[1:])
 
-            coverage_cmd = f"{SAFE_SYS_EXECUTABLE} -m coverage run --rcfile={coveragercfile.as_posix()} -m"
             results = execute_test_subprocess(
-                shlex.split(coverage_cmd) + pytest_cmd_list + common_pytest_args + result_args + test_files,
-                cwd=cwd,
-                env=pytest_test_env,
-                timeout=600,
+                coverage_cmd + common_pytest_args + result_args + test_files, cwd=cwd, env=pytest_test_env, timeout=600
             )
             logger.debug(
-                f"Result return code: {results.returncode}"
-                + (f", Result stderr: {results.stderr}" if results.stderr else "")
+                f"Result return code: {results.returncode}, "
+                f"{'Result stderr:' + str(results.stderr) if results.stderr else ''}"
             )
         else:
             results = execute_test_subprocess(
@@ -97,8 +102,7 @@ def run_behavioral_tests(
                 timeout=600,  # TODO: Make this dynamic
             )
             logger.debug(
-                f"Result return code: {results.returncode}"
-                + (f", Result stderr: {results.stderr}" if results.stderr else "")
+                f"""Result return code: {results.returncode}, {"Result stderr:" + str(results.stderr) if results.stderr else ""}"""
             )
     elif test_framework == "unittest":
         if enable_coverage:
@@ -110,8 +114,7 @@ def run_behavioral_tests(
             verbose=verbose, test_file_paths=test_files, test_env=test_env, cwd=cwd
         )
         logger.debug(
-            f"Result return code: {results.returncode}"
-            + (f", Result stderr: {results.stderr}" if results.stderr else "")
+            f"""Result return code: {results.returncode}, {"Result stderr:" + str(results.stderr) if results.stderr else ""}"""
         )
     else:
         msg = f"Unsupported test framework: {test_framework}"
@@ -134,7 +137,11 @@ def run_benchmarking_tests(
     pytest_max_loops: int = 100_000,
 ) -> tuple[Path, subprocess.CompletedProcess]:
     if test_framework == "pytest":
-        pytest_cmd_list = shlex.split(pytest_cmd, posix=IS_POSIX)
+        pytest_cmd_list = (
+            shlex.split(f"{SAFE_SYS_EXECUTABLE} -m pytest", posix=IS_POSIX)
+            if pytest_cmd == "pytest"
+            else shlex.split(pytest_cmd)
+        )
         test_files: list[str] = []
         for file in test_paths.test_files:
             if file.test_type in [TestType.REPLAY_TEST, TestType.EXISTING_UNIT_TEST] and file.tests_in_file:

codeflash/verification/verification_utils.py

@@ -6,6 +6,8 @@
 
 from pydantic.dataclasses import dataclass
 
+from codeflash.code_utils.compat import SAFE_SYS_EXECUTABLE
+
 
 def get_test_file_path(test_dir: Path, function_name: str, iteration: int = 0, test_type: str = "unit") -> Path:
     assert test_type in ["unit", "inspired", "replay", "perf"]
@@ -74,4 +76,4 @@ class TestConfig:
     # tests_project_rootdir corresponds to pytest rootdir,
     # or for unittest - project_root_from_module_root(args.tests_root, pyproject_file_path)
     concolic_test_root_dir: Optional[Path] = None
-    pytest_cmd: str = "pytest"
+    pytest_cmd: str = "pytest"

pyproject.toml

@@ -69,7 +69,7 @@ exclude = [
 [tool.poetry.dependencies]
 python = ">=3.9"
 unidiff = ">=0.7.4"
-pytest = ">=7.0.0"
+pytest = ">=7.0.0,<8.3.4"
 gitpython = ">=3.1.31"
 libcst = ">=1.0.1"
 jedi = ">=0.19.1"
@@ -120,6 +120,7 @@ types-greenlet = "^3.1.0.20241221"
 types-pexpect = "^4.9.0.20241208"
 types-unidiff = "^0.7.0.20240505"
 sqlalchemy = "^2.0.38"
+uv = ">=0.6.2"
 
 [tool.poetry.build]
 script = "codeflash/update_license_version.py"
@@ -221,8 +222,8 @@ module-root = "codeflash"
 tests-root = "tests"
 test-framework = "pytest"
 formatter-cmds = [
-    "poetry run ruff check --exit-zero --fix $file",
-    "poetry run ruff format $file",
+    "uvx ruff check --exit-zero --fix $file",
+    "uvx ruff format $file",
 ]