Rpatel/codebase snapshot (#698)

# Motivation

<!-- Why is this change necessary? -->

# Content

<!-- Please include a summary of the change -->

# Testing

<!-- How was the change tested? -->

# Please check the following before marking your PR as ready for review

- [ ] I have added tests for my changes
- [ ] I have updated the documentation or added new documentation as
needed

---------

Co-authored-by: rushilpatel0 <[email protected]>

Author: rushilpatel0

codegen-examples/examples/codegen_app/app.py

@@ -16,7 +16,7 @@
 ########################################################################################################################
 
 # Create the cg_app
-cg = CodegenApp(name="codegen-test", repos=["codegen-sh/Kevin-s-Adventure-Game"])
+cg = CodegenApp(name="codegen-test", repo="codegen-sh/Kevin-s-Adventure-Game")
 
 
 @cg.slack.event("app_mention")
@@ -26,7 +26,7 @@ async def handle_mention(event: SlackEvent):
 
     # Codebase
     logger.info("[CODEBASE] Initializing codebase")
-    codebase = cg.get_codebase("codegen-sh/Kevin-s-Adventure-Game")
+    codebase = cg.get_codebase()
 
     # Code Agent
     logger.info("[CODE_AGENT] Initializing code agent")
@@ -43,7 +43,7 @@ async def handle_mention(event: SlackEvent):
 def handle_pr(event: PullRequestLabeledEvent):
     logger.info("PR labeled")
     logger.info(f"PR head sha: {event.pull_request.head.sha}")
-    codebase = cg.get_codebase("codegen-sh/Kevin-s-Adventure-Game")
+    codebase = cg.get_codebase()
 
     # =====[ Check out commit ]=====
     # Might require fetch?
@@ -62,7 +62,7 @@ def handle_pr(event: PullRequestLabeledEvent):
 @cg.linear.event("Issue")
 def handle_issue(event: LinearEvent):
     logger.info(f"Issue created: {event}")
-    codebase = cg.get_codebase("codegen-sh/Kevin-s-Adventure-Game")
+    codebase = cg.get_codebase()
     return {"message": "Linear Issue event", "num_files": len(codebase.files), "num_functions": len(codebase.functions)}
 
 
@@ -74,7 +74,7 @@ def handle_issue(event: LinearEvent):
 
 # For deploying local package
 REPO_URL = "https://github.com/codegen-sh/codegen-sdk.git"
-COMMIT_ID = "26dafad2c319968e14b90806d42c6c7aaa627bb0"
+COMMIT_ID = "6a0e101718c247c01399c60b7abf301278a41786"
 
 # Create the base image with dependencies
 base_image = (
@@ -97,4 +97,5 @@ def handle_issue(event: LinearEvent):
 @app.function(image=base_image, secrets=[modal.Secret.from_dotenv()])
 @modal.asgi_app()
 def fastapi_app():
+    print("Starting codegen fastapi app")
     return cg.app

codegen-examples/examples/codegen_app/codegen_app_app.py

@@ -0,0 +1,3 @@
+from app import app
+
+app = app

codegen-examples/examples/snapshot_event_handler/.env.template

@@ -0,0 +1,3 @@
+ANTHROPIC_API_KEY="..."
+SLACK_BOT_TOKEN="..."
+GITHUB_TOKEN="..."

codegen-examples/examples/snapshot_event_handler/README.md

@@ -0,0 +1,3 @@
+from event_handlers import app
+
+app = app

codegen-examples/examples/snapshot_event_handler/codegen_app_event_handlers.py

@@ -0,0 +1,135 @@
+from codegen.agents.code_agent import CodeAgent
+from codegen.extensions.events.codegen_app import CodegenApp
+from codegen.extensions.linear.types import LinearEvent
+from codegen.extensions.slack.types import SlackEvent
+from codegen.extensions.events.modal.base import CodebaseEventsApp, EventRouterMixin
+from codegen.extensions.github.types.pull_request import PullRequestLabeledEvent
+from pr_tasks import lint_for_dev_import_violations, review_with_codegen_agent
+from typing import Literal
+from dotenv import load_dotenv
+from fastapi import FastAPI, Request
+from classy_fastapi import Routable, post
+import modal
+import logging
+
+load_dotenv(".env")
+
+
+logging.basicConfig(level=logging.INFO, force=True)
+logger = logging.getLogger(__name__)
+
+codegen_events_app = modal.App("codegen-events-router")
+
+SNAPSHOT_DICT_ID = "codegen-events-codebase-snapshots"
+
+
+REPO_URL = "https://github.com/codegen-sh/codegen-sdk.git"
+COMMIT_ID = "f398107d31bbd53fc77bc9c5f8763d2dc8fcae5b"
+
+# Create the base image with dependencies
+base_image = (
+    modal.Image.debian_slim(python_version="3.13")
+    .apt_install("git")
+    .pip_install(
+        # =====[ Codegen ]=====
+        # "codegen",
+        f"git+{REPO_URL}@{COMMIT_ID}",
+        # =====[ Rest ]=====
+        "openai>=1.1.0",
+        "fastapi[standard]",
+        "slack_sdk",
+        "classy-fastapi>=0.6.1",
+    )
+)
+
+
+event_handlers_app = modal.App("codegen-event-handlers")
+
+
+@event_handlers_app.cls(image=base_image, secrets=[modal.Secret.from_dotenv(".env")], enable_memory_snapshot=True, container_idle_timeout=300)
+class CustomEventHandlersAPI(CodebaseEventsApp):
+    commit: str = modal.parameter(default="79114f67ccfe8700416cd541d1c7c43659a95342")
+    repo_org: str = modal.parameter(default="codegen-sh")
+    repo_name: str = modal.parameter(default="Kevin-s-Adventure-Game")
+    snapshot_index_id: str = SNAPSHOT_DICT_ID
+
+    def setup_handlers(self, cg: CodegenApp):
+        @cg.slack.event("app_mention")
+        async def handle_mention(event: SlackEvent):
+            logger.info("[APP_MENTION] Received cg_app_mention event")
+
+            # Codebase
+            logger.info("[CODEBASE] Initializing codebase")
+            codebase = cg.get_codebase()
+
+            # Code Agent
+            logger.info("[CODE_AGENT] Initializing code agent")
+            agent = CodeAgent(codebase=codebase)
+
+            logger.info("[CODE_AGENT] Running code agent")
+            response = agent.run(event.text)
+
+            cg.slack.client.chat_postMessage(channel=event.channel, text=response, thread_ts=event.ts)
+
+            return {"message": "Mentioned", "received_text": event.text, "response": response}
+
+        @cg.github.event("pull_request:labeled")
+        def handle_pr(event: PullRequestLabeledEvent):
+            logger.info("PR labeled")
+            logger.info(f"PR head sha: {event.pull_request.head.sha}")
+
+            codebase = cg.get_codebase()
+            logger.info(f"Codebase: {codebase.name} codebase.repo: {codebase.repo_path}")
+
+            # =====[ Check out commit ]=====
+            # Might require fetch?
+            logger.info("> Checking out commit")
+            codebase.checkout(commit=event.pull_request.head.sha)
+
+            logger.info("> Running PR Lints")
+            # LINT CODEMOD
+            lint_for_dev_import_violations(codebase, event)
+
+            # REVIEW CODEMOD
+            review_with_codegen_agent(codebase, event)
+
+            return {"message": "PR event handled", "num_files": len(codebase.files), "num_functions": len(codebase.functions)}
+
+        @cg.linear.event("Issue")
+        def handle_issue(event: LinearEvent):
+            logger.info(f"Issue created: {event}")
+            codebase = cg.get_codebase()
+            return {"message": "Linear Issue event", "num_files": len(codebase.files), "num_functions": len(codebase.functions)}
+
+
+@codegen_events_app.cls(image=base_image, secrets=[modal.Secret.from_dotenv(".env")])
+class WebhookEventRouterAPI(EventRouterMixin, Routable):
+    snapshot_index_id: str = SNAPSHOT_DICT_ID
+
+    def get_event_handler_cls(self):
+        modal_cls = modal.Cls.from_name(app_name="Events", name="CustomEventHandlersAPI")
+        return modal_cls
+
+    @post("/{org}/{repo}/{provider}/events")
+    async def handle_event(self, org: str, repo: str, provider: Literal["slack", "github", "linear"], request: Request):
+        # Define the route for the webhook url sink, it will need to indicate the repo repo org, and the provider
+        return await super().handle_event(org, repo, provider, request)
+
+    @modal.asgi_app()
+    def api(self):
+        """Run the FastAPI app with the Router."""
+        event_api = FastAPI()
+        route_view = WebhookEventRouterAPI()
+        event_api.include_router(route_view.router)
+        return event_api
+
+
+# Setup a cron job to trigger updates to the codebase snapshots.
+@codegen_events_app.function(schedule=modal.Cron("*/10 * * * *"), image=base_image, secrets=[modal.Secret.from_dotenv(".env")])
+def refresh_repository_snapshots():
+    WebhookEventRouterAPI().refresh_repository_snapshots(snapshot_index_id=SNAPSHOT_DICT_ID)
+
+
+app = modal.App("Events", secrets=[modal.Secret.from_dotenv(".env")])
+app.include(event_handlers_app)
+app.include(codegen_events_app)

codegen-examples/examples/snapshot_event_handler/event_handlers.py

@@ -0,0 +1,78 @@
+import logging
+from codegen.agents.code_agent import CodeAgent
+from codegen.extensions.github.types.pull_request import PullRequestLabeledEvent
+
+from codegen.extensions.langchain.tools import GithubCreatePRCommentTool, GithubCreatePRReviewCommentTool, GithubViewPRTool
+from codegen.sdk.core.codebase import Codebase
+
+logging.basicConfig(level=logging.INFO, force=True)
+logger = logging.getLogger(__name__)
+
+
+def lint_for_dev_import_violations(codebase: Codebase, event: PullRequestLabeledEvent):
+    patch, commit_shas, modified_symbols = codebase.get_modified_symbols_in_pr(event.pull_request.number)
+    modified_files = set(commit_shas.keys())
+
+    DIR_NAME = "packages/next/src/client/components/react-dev-overlay"
+    directory = codebase.get_directory(DIR_NAME)
+
+    # Initialize a list to store all violations
+    violations = []
+
+    print("modified_files", modified_files)
+
+    # Check if directory exists before proceeding
+    if directory is not None and hasattr(directory, "files"):
+        for file in directory.files:
+            print("checking file", file.filepath)
+            for imp in file.inbound_imports:
+                print("file", imp.file.filepath)
+                print("checking import", imp.import_statement)
+                # Check if the import is from outside the directory and is in the modified files
+                if imp.file not in directory and imp.file.filepath in modified_files:
+                    # Skip require statements
+                    if "require" in imp.import_statement:
+                        continue
+                    violation = f"- Violation in `{file.filepath}`: Importing from `{imp.file.filepath}` ([link]({imp.github_url}))"
+                    violations.append(violation)
+                    logger.info(f"Found violation: {violation}")
+
+        # Only create a PR comment if violations are found
+        if violations:
+            review_attention_message = "## Dev Import Violations Found\n\n"
+            review_attention_message += "The following files have imports that violate development overlay rules:\n\n"
+            review_attention_message += "\n".join(violations)
+            review_attention_message += "\n\nPlease ensure that development imports are not imported in production code."
+
+            # Create PR comment with the formatted message
+            codebase._op.create_pr_comment(event.pull_request.number, review_attention_message)
+
+
+def review_with_codegen_agent(codebase: Codebase, event: PullRequestLabeledEvent):
+    review_initial_message = "CodegenBot is starting to review the PR please wait..."
+    comment = codebase._op.create_pr_comment(event.number, review_initial_message)
+    # Define tools first
+    pr_tools = [
+        GithubViewPRTool(codebase),
+        GithubCreatePRCommentTool(codebase),
+        GithubCreatePRReviewCommentTool(codebase),
+    ]
+
+    # Create agent with the defined tools
+    agent = CodeAgent(codebase=codebase, tools=pr_tools)
+
+    # Using a prompt from SWE Bench
+    prompt = f"""
+    Hey CodegenBot!
+
+    Here's a SWE task for you. Please Review this pull request!
+    {event.pull_request.url}
+    Do not terminate until have reviewed the pull request and are satisfied with your review.
+
+    Review this Pull request like the señor ingenier you are
+    be explicit about the changes, produce a short summary, and point out possible improvements where pressent dont be self congratulatory stick to the facts
+    use the tools at your disposal to create propper pr reviews include code snippets if needed, and suggest improvements if feel its necesary
+    """
+    # Run the agent
+    agent.run(prompt)
+    comment.delete()

codegen-examples/examples/snapshot_event_handler/pr_tasks.py

@@ -0,0 +1,32 @@
+[project]
+name = "codegen-examples"
+version = "0.0.0"
+readme = "README.md"
+requires-python = ">=3.12, <3.14"
+dependencies = [
+  "fastapi-utils>=0.8.0",
+  "fastapi[standard]>=0.115.9",
+  "modal>=0.73.72",
+  "fastapi-restful[all]>=0.6.0",
+  "classy-fastapi>=0.6.1",
+  "codegen@git+https://github.com/codegen-sh/codegen-sdk.git@f398107d31bbd53fc77bc9c5f8763d2dc8fcae5b",
+]
+license = { file = "LICENSE" }
+classifiers = [
+  "License :: OSI Approved :: Apache Software License",
+  "Intended Audience :: Developers",
+  "Programming Language :: Python :: 3.12",
+  "Programming Language :: Python :: 3.13",
+  "Topic :: Software Development",
+  "Development Status :: 4 - Beta",
+  "Environment :: MacOS X",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python",
+]
+
+[tool.ruff]
+line-length = 200
+exclude = ["**/input_repo/**", "**/output_repo/**", "**/repositories/**"]
+
+[tool.uv]
+cache-keys = [{ git = { commit = true, tags = true } }]