Typo

The new feature can be enabled via the new configuration `github_app.handle_push_event`. To avoid any unwanted side-effects, the current default of this configuration is set to `false`.

The high level flow (assuming the configuration is enabled):
1. receive push event from GitHub
2. extract branch and commits from event
3. find PR url for branch (currently does not support PRs from forks)
4. perform configured commands (e.g. `/describe`, `/review -i`)

The push event flow is guarded by a backlog queue so that multiple push events on the same branch won't trigger multiple duplicate runs of the PR-Agent commands.
Example timeline:
1. push 1 - start handling event
2. push 2 - waiting to be handled while push 1 event is still running
3. push 3 - event is dropped since handling it and handling push 2 is the same, so it is redundant
4. push 1 finished being handled
5. push 2 awakens from wait and continues handling (potentially reviewing the commits of both push 2 and push 3)

All of these options are configurable and can be enabled/disabled as per the user's desire.

Additional minor changes in this PR:
1. Created `DefaultDictWithTimeout` utility class to avoid too much boilerplate code in managing caches for outdated triggers.
2. Guard against running increment review when there are no new commits.
3. Minor styling changes for incremented review text.

docs/DESCRIBE.md

@@ -27,18 +27,14 @@ Under the section 'pr_description', the [configuration file](./../pr_agent/setti
 
 - `extra_instructions`: Optional extra instructions to the tool. For example: "focus on the changes in the file X. Ignore change in ...".
 
-#### Markers template
+### Markers template
 
 markers enable to easily integrate user's content and auto-generated content, with a template-like mechanism.
 
-- `use_description_markers`: if set to true, the tool will use markers template. It replaces every marker of the form `pr_agent:marker_name` with the relevant content. Default is false.
-
 For example, if the PR original description was:
 ```
 User content...
 
-## PR Type:
-pr_agent:pr_type
 
 ## PR Description:
 pr_agent:summary
@@ -46,6 +42,21 @@ pr_agent:summary
 ## PR Walkthrough:
 pr_agent:walkthrough
 ```
-The marker `pr_agent:pr_type` will be replaced with the PR type, `pr_agent:summary` will be replaced with the PR summary, and `pr_agent:walkthrough` will be replaced with the PR walkthrough.
+The marker `pr_agent:summary` will be replaced with the PR summary, and `pr_agent:walkthrough` will be replaced with the PR walkthrough.
 
+##### Example:
+```
+        env:
+          pr_description.use_description_markers: 'true'
+```          
+
+<kbd><img src=./../pics/describe_markers_before.png width="768"></kbd>
+
+==>
+
+<kbd><img src=./../pics/describe_markers_after.png width="768"></kbd>
+
+##### Configuration params:
+
+- `use_description_markers`: if set to true, the tool will use markers template. It replaces every marker of the form `pr_agent:marker_name` with the relevant content. Default is false.
 - `include_generated_by_header`: if set to true, the tool will add a dedicated header: 'Generated by PR Agent at ...' to any automatic content. Default is true.

docs/IMPROVE.md

@@ -32,3 +32,14 @@ Under the section 'pr_code_suggestions', the [configuration file](./../pr_agent/
 - `rank_extended_suggestions`: if set to true, the tool will rank the suggestions, based on importance. Default is true.
 - `max_number_of_calls`: maximum number of chunks. Default is 5.
 - `final_clip_factor`: factor to remove suggestions with low confidence. Default is 0.9.
+
+
+#### A note on code suggestions quality
+
+- With current level of AI for code (GPT-4), mistakes can happen. Not all the suggestions will be perfect, and a user should not accept all of them automatically.
+
+- Suggestions are not meant to be [simplistic](./../pr_agent/settings/pr_code_suggestions_prompts.toml#L34). Instead, they aim to give deep feedback and raise questions, ideas and thoughts to the user, who can then use his judgment, experience, and understanding of the code base.
+
+- Recommended to use the 'extra_instructions' field to guide the model to suggestions that are more relevant to the specific needs of the project.
+
+- Best quality will be obtained by using 'improve --extended' mode.

docs/REVIEW.md

@@ -44,3 +44,14 @@ The tool will first ask the author questions about the PR, and will guide the re
 <kbd><img src=./../pics/reflection_questions.png width="768"></kbd>
 <kbd><img src=./../pics/reflection_answers.png width="768"></kbd>
 <kbd><img src=./../pics/reflection_insights.png width="768"></kbd>
+
+
+#### A note on code suggestions quality
+
+- With current level of AI for code (GPT-4), mistakes can happen. Not all the suggestions will be perfect, and a user should not accept all of them automatically.
+
+- Suggestions are not meant to be [simplistic](./../pr_agent/settings/pr_reviewer_prompts.toml#L29). Instead, they aim to give deep feedback and raise questions, ideas and thoughts to the user, who can then use his judgment, experience, and understanding of the code base.
+
+- Recommended to use the 'extra_instructions' field to guide the model to suggestions that are more relevant to the specific needs of the project.
+
+- Unlike the 'review' feature, which does a lot of things, the ['improve --extended'](./IMPROVE.md) feature is dedicated only to suggestions, and usually gives better results.

pr_agent/git_providers/bitbucket_provider.py

@@ -142,10 +142,15 @@ class BitbucketProvider(GitProvider):
     def remove_initial_comment(self):
         try:
             for comment in self.temp_comments:
-                self.pr.delete(f"comments/{comment}")
+                self.remove_comment(comment)
         except Exception as e:
             get_logger().exception(f"Failed to remove temp comments, error: {e}")
 
+    def remove_comment(self, comment):
+        try:
+            self.pr.delete(f"comments/{comment}")
+        except Exception as e:
+            get_logger().exception(f"Failed to remove comment, error: {e}")
 
     # funtion to create_inline_comment
     def create_inline_comment(self, body: str, relevant_file: str, relevant_line_in_file: str):

pr_agent/git_providers/codecommit_provider.py

@@ -221,6 +221,9 @@ class CodeCommitProvider(GitProvider):
     def remove_initial_comment(self):
         return ""  # not implemented yet
 
+    def remove_comment(self, comment):
+        return ""  # not implemented yet
+
     def publish_inline_comment(self, body: str, relevant_file: str, relevant_line_in_file: str):
         # https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/codecommit/client/post_comment_for_compared_commit.html
         raise NotImplementedError("CodeCommit provider does not support publishing inline comments yet")

pr_agent/git_providers/gerrit_provider.py

@@ -396,5 +396,8 @@ class GerritProvider(GitProvider):
         # shutil.rmtree(self.repo_path)
         pass
 
+    def remove_comment(self, comment):
+        pass
+
     def get_pr_branch(self):
         return self.repo.head

pr_agent/git_providers/git_provider.py

@@ -71,6 +71,10 @@ class GitProvider(ABC):
     def remove_initial_comment(self):
         pass
 
+    @abstractmethod
+    def remove_comment(self, comment):
+        pass
+
     @abstractmethod
     def get_languages(self):
         pass

pr_agent/git_providers/github_provider.py

@@ -50,7 +50,7 @@ class GithubProvider(GitProvider):
     def get_incremental_commits(self):
         self.commits = list(self.pr.get_commits())
 
-        self.get_previous_review()
+        self.previous_review = self.get_previous_review(full=True, incremental=True)
         if self.previous_review:
             self.incremental.commits_range = self.get_commit_range()
             # Get all files changed during the commit range
@@ -63,7 +63,7 @@ class GithubProvider(GitProvider):
 
     def get_commit_range(self):
         last_review_time = self.previous_review.created_at
-        first_new_commit_index = 0
+        first_new_commit_index = None
         for index in range(len(self.commits) - 1, -1, -1):
             if self.commits[index].commit.author.date > last_review_time:
                 self.incremental.first_new_commit_sha = self.commits[index].sha
@@ -71,15 +71,21 @@ class GithubProvider(GitProvider):
             else:
                 self.incremental.last_seen_commit_sha = self.commits[index].sha
                 break
-        return self.commits[first_new_commit_index:]
+        return self.commits[first_new_commit_index:] if first_new_commit_index is not None else []
 
-    def get_previous_review(self):
+    def get_previous_review(self, *, full: bool, incremental: bool):
-        self.previous_review = None
+        if not (full or incremental):
-        self.comments = list(self.pr.get_issue_comments())
+            raise ValueError("At least one of full or incremental must be True")
+        if not getattr(self, "comments", None):
+            self.comments = list(self.pr.get_issue_comments())
+        prefixes = []
+        if full:
+            prefixes.append("## PR Analysis")
+        if incremental:
+            prefixes.append("## Incremental PR Review")
         for index in range(len(self.comments) - 1, -1, -1):
-            if self.comments[index].body.startswith("## PR Analysis") or self.comments[index].body.startswith("## Incremental PR Review"):
+            if any(self.comments[index].body.startswith(prefix) for prefix in prefixes):
-                self.previous_review = self.comments[index]
+                return self.comments[index]
-                break
 
     def get_files(self):
         if self.incremental.is_incremental and self.file_set:
@@ -218,10 +224,16 @@ class GithubProvider(GitProvider):
         try:
             for comment in getattr(self.pr, 'comments_list', []):
                 if comment.is_temporary:
-                    comment.delete()
+                    self.remove_comment(comment)
         except Exception as e:
             get_logger().exception(f"Failed to remove initial comment, error: {e}")
 
+    def remove_comment(self, comment):
+        try:
+            comment.delete()
+        except Exception as e:
+            get_logger().exception(f"Failed to remove comment, error: {e}")
+
     def get_title(self):
         return self.pr.title
 

pr_agent/git_providers/gitlab_provider.py

@@ -287,10 +287,16 @@ class GitLabProvider(GitProvider):
     def remove_initial_comment(self):
         try:
             for comment in self.temp_comments:
-                comment.delete()
+                self.remove_comment(comment)
         except Exception as e:
             get_logger().exception(f"Failed to remove temp comments, error: {e}")
 
+    def remove_comment(self, comment):
+        try:
+            comment.delete()
+        except Exception as e:
+            get_logger().exception(f"Failed to remove comment, error: {e}")
+
     def get_title(self):
         return self.mr.title
 

pr_agent/git_providers/local_git_provider.py

@@ -140,6 +140,9 @@ class LocalGitProvider(GitProvider):
     def remove_initial_comment(self):
         pass  # Not applicable to the local git provider, but required by the interface
 
+    def remove_comment(self, comment):
+        pass  # Not applicable to the local git provider, but required by the interface
+
     def get_languages(self):
         """
         Calculate percentage of languages in repository. Used for hunk prioritisation.

pr_agent/secret_providers/google_cloud_storage_secret_provider.py

@@ -1,9 +1,8 @@
 import ujson
-
 from google.cloud import storage
 
 from pr_agent.config_loader import get_settings
-from pr_agent.git_providers.gitlab_provider import logger
+from pr_agent.log import get_logger
 from pr_agent.secret_providers.secret_provider import SecretProvider
 
 
@@ -15,7 +14,7 @@ class GoogleCloudStorageSecretProvider(SecretProvider):
             self.bucket_name = get_settings().google_cloud_storage.bucket_name
             self.bucket = self.client.bucket(self.bucket_name)
         except Exception as e:
-            logger.error(f"Failed to initialize Google Cloud Storage Secret Provider: {e}")
+            get_logger().error(f"Failed to initialize Google Cloud Storage Secret Provider: {e}")
             raise e
 
     def get_secret(self, secret_name: str) -> str:
@@ -23,7 +22,7 @@ class GoogleCloudStorageSecretProvider(SecretProvider):
             blob = self.bucket.blob(secret_name)
             return blob.download_as_string()
         except Exception as e:
-            logger.error(f"Failed to get secret {secret_name} from Google Cloud Storage: {e}")
+            get_logger().error(f"Failed to get secret {secret_name} from Google Cloud Storage: {e}")
             return ""
 
     def store_secret(self, secret_name: str, secret_value: str):
@@ -31,5 +30,5 @@ class GoogleCloudStorageSecretProvider(SecretProvider):
             blob = self.bucket.blob(secret_name)
             blob.upload_from_string(secret_value)
         except Exception as e:
-            logger.error(f"Failed to store secret {secret_name} in Google Cloud Storage: {e}")
+            get_logger().error(f"Failed to store secret {secret_name} in Google Cloud Storage: {e}")
             raise e

pr_agent/servers/bitbucket_app.py

@@ -65,15 +65,17 @@ async def handle_manifest(request: Request, response: Response):
 
 @router.post("/webhook")
 async def handle_github_webhooks(background_tasks: BackgroundTasks, request: Request):
-    print(request.headers)
+    log_context = {"server_type": "bitbucket_app"}
+    get_logger().debug(request.headers)
     jwt_header = request.headers.get("authorization", None)
     if jwt_header:
         input_jwt = jwt_header.split(" ")[1]
     data = await request.json()
-    print(data)
+    get_logger().debug(data)
     async def inner():
         try:
             owner = data["data"]["repository"]["owner"]["username"]
+            log_context["sender"] = owner
             secrets = json.loads(secret_provider.get_secret(owner))
             shared_secret = secrets["shared_secret"]
             client_key = secrets["client_key"]
@@ -85,11 +87,17 @@ async def handle_github_webhooks(background_tasks: BackgroundTasks, request: Req
             agent = PRAgent()
             if event == "pullrequest:created":
                 pr_url = data["data"]["pullrequest"]["links"]["html"]["href"]
-                await agent.handle_request(pr_url, "review")
+                log_context["api_url"] = pr_url
+                log_context["event"] = "pull_request"
+                with get_logger().contextualize(**log_context):
+                    await agent.handle_request(pr_url, "review")
             elif event == "pullrequest:comment_created":
                 pr_url = data["data"]["pullrequest"]["links"]["html"]["href"]
+                log_context["api_url"] = pr_url
+                log_context["event"] = "comment"
                 comment_body = data["data"]["comment"]["content"]["raw"]
-                await agent.handle_request(pr_url, comment_body)
+                with get_logger().contextualize(**log_context):
+                    await agent.handle_request(pr_url, comment_body)
         except Exception as e:
             get_logger().error(f"Failed to handle webhook: {e}")
     background_tasks.add_task(inner)
@@ -102,9 +110,10 @@ async def handle_github_webhooks(request: Request, response: Response):
 @router.post("/installed")
 async def handle_installed_webhooks(request: Request, response: Response):
     try:
-        print(request.headers)
+        get_logger().info("handle_installed_webhooks")
+        get_logger().info(request.headers)
         data = await request.json()
-        print(data)
+        get_logger().info(data)
         shared_secret = data["sharedSecret"]
         client_key = data["clientKey"]
         username = data["principal"]["username"]
@@ -119,8 +128,10 @@ async def handle_installed_webhooks(request: Request, response: Response):
 
 @router.post("/uninstalled")
 async def handle_uninstalled_webhooks(request: Request, response: Response):
+    get_logger().info("handle_uninstalled_webhooks")
+
     data = await request.json()
-    print(data)
+    get_logger().info(data)
 
 
 def start():

pr_agent/servers/github_app.py

@@ -1,7 +1,7 @@
 import copy
 import os
-import time
+import asyncio.locks
-from typing import Any, Dict
+from typing import Any, Dict, List, Tuple
 
 import uvicorn
 from fastapi import APIRouter, FastAPI, HTTPException, Request, Response
@@ -14,8 +14,9 @@ from pr_agent.algo.utils import update_settings_from_args
 from pr_agent.config_loader import get_settings, global_settings
 from pr_agent.git_providers import get_git_provider
 from pr_agent.git_providers.utils import apply_repo_settings
+from pr_agent.git_providers.git_provider import IncrementalPR
 from pr_agent.log import LoggingFormat, get_logger, setup_logger
-from pr_agent.servers.utils import verify_signature
+from pr_agent.servers.utils import verify_signature, DefaultDictWithTimeout
 
 setup_logger(fmt=LoggingFormat.JSON)
 
@@ -47,6 +48,7 @@ async def handle_marketplace_webhooks(request: Request, response: Response):
     body = await get_body(request)
     get_logger().info(f'Request body:\n{body}')
 
+
 async def get_body(request):
     try:
         body = await request.json()
@@ -61,7 +63,9 @@ async def get_body(request):
     return body
 
 
-_duplicate_requests_cache = {}
+_duplicate_requests_cache = DefaultDictWithTimeout(ttl=get_settings().github_app.duplicate_requests_cache_ttl)
+_duplicate_push_triggers = DefaultDictWithTimeout(ttl=get_settings().github_app.push_trigger_pending_tasks_ttl)
+_pending_task_duplicate_push_conditions = DefaultDictWithTimeout(asyncio.locks.Condition, ttl=get_settings().github_app.push_trigger_pending_tasks_ttl)
 
 
 async def handle_request(body: Dict[str, Any], event: str):
@@ -109,40 +113,110 @@ async def handle_request(body: Dict[str, Any], event: str):
     # handle pull_request event:
     #   automatically review opened/reopened/ready_for_review PRs as long as they're not in draft,
     #   as well as direct review requests from the bot
-    elif event == 'pull_request':
+    elif event == 'pull_request' and action != 'synchronize':
-        pull_request = body.get("pull_request")
+        pull_request, api_url = _check_pull_request_event(action, body, log_context, bot_user)
-        if not pull_request:
+        if not (pull_request and api_url):
-            return {}
-        api_url = pull_request.get("url")
-        if not api_url:
-            return {}
-        log_context["api_url"] = api_url
-        if pull_request.get("draft", True) or pull_request.get("state") != "open" or pull_request.get("user", {}).get("login", "") == bot_user:
             return {}
         if action in get_settings().github_app.handle_pr_actions:
             if action == "review_requested":
                 if body.get("requested_reviewer", {}).get("login", "") != bot_user:
                     return {}
-                if pull_request.get("created_at") == pull_request.get("updated_at"):
+            get_logger().info(f"Performing review for {api_url=} because of {event=} and {action=}")
-                    # avoid double reviews when opening a PR for the first time
+            await _perform_commands(get_settings().github_app.pr_commands, agent, body, api_url, log_context)
-                    return {}
+
-            get_logger().info(f"Performing review because of event={event} and action={action}")
+    # handle pull_request event with synchronize action - "push trigger" for new commits
-            apply_repo_settings(api_url)
+    elif event == 'pull_request' and action == 'synchronize' and get_settings().github_app.handle_push_trigger:
-            for command in get_settings().github_app.pr_commands:
+        pull_request, api_url = _check_pull_request_event(action, body, log_context, bot_user)
-                split_command = command.split(" ")
+        if not (pull_request and api_url):
-                command = split_command[0]
+            return {}
-                args = split_command[1:]
+
-                other_args = update_settings_from_args(args)
+        # TODO: do we still want to get the list of commits to filter bot/merge commits?
-                new_command = ' '.join([command] + other_args)
+        before_sha = body.get("before")
-                get_logger().info(body)
+        after_sha = body.get("after")
-                get_logger().info(f"Performing command: {new_command}")
+        merge_commit_sha = pull_request.get("merge_commit_sha")
-                with get_logger().contextualize(**log_context):
+        if before_sha == after_sha:
-                    await agent.handle_request(api_url, new_command)
+            return {}
+        if get_settings().github_app.push_trigger_ignore_merge_commits and after_sha == merge_commit_sha:
+            return {}
+        if get_settings().github_app.push_trigger_ignore_bot_commits and body.get("sender", {}).get("login", "") == bot_user:
+            return {}
+
+        # Prevent triggering multiple times for subsequent push triggers when one is enough:
+        # The first push will trigger the processing, and if there's a second push in the meanwhile it will wait.
+        # Any more events will be discarded, because they will all trigger the exact same processing on the PR.
+        # We let the second event wait instead of discarding it because while the first event was being processed,
+        # more commits may have been pushed that led to the subsequent events,
+        # so we keep just one waiting as a delegate to trigger the processing for the new commits when done waiting.
+        current_active_tasks = _duplicate_push_triggers.setdefault(api_url, 0)
+        max_active_tasks = 2 if get_settings().github_app.push_trigger_pending_tasks_backlog else 1
+        if current_active_tasks < max_active_tasks:
+            # first task can enter, and second tasks too if backlog is enabled
+            get_logger().info(
+                f"Continue processing push trigger for {api_url=} because there are {current_active_tasks} active tasks"
+            )
+            _duplicate_push_triggers[api_url] += 1
+        else:
+            get_logger().info(
+                f"Skipping push trigger for {api_url=} because another event already triggered the same processing"
+            )
+            return {}
+        async with _pending_task_duplicate_push_conditions[api_url]:
+            if current_active_tasks == 1:
+                # second task waits
+                get_logger().info(
+                    f"Waiting to process push trigger for {api_url=} because the first task is still in progress"
+                )
+                await _pending_task_duplicate_push_conditions[api_url].wait()
+                get_logger().info(f"Finished waiting to process push trigger for {api_url=} - continue with flow")
+
+        try:
+            if get_settings().github_app.push_trigger_wait_for_initial_review and not get_git_provider()(api_url, incremental=IncrementalPR(True)).previous_review:
+                get_logger().info(f"Skipping incremental review because there was no initial review for {api_url=} yet")
+                return {}
+            get_logger().info(f"Performing incremental review for {api_url=} because of {event=} and {action=}")
+            await _perform_commands(get_settings().github_app.push_commands, agent, body, api_url, log_context)
+
+        finally:
+            # release the waiting task block
+            async with _pending_task_duplicate_push_conditions[api_url]:
+                _pending_task_duplicate_push_conditions[api_url].notify(1)
+                _duplicate_push_triggers[api_url] -= 1
 
     get_logger().info("event or action does not require handling")
     return {}
 
 
+def _check_pull_request_event(action: str, body: dict, log_context: dict, bot_user: str) -> Tuple[Dict[str, Any], str]:
+    invalid_result = {}, ""
+    pull_request = body.get("pull_request")
+    if not pull_request:
+        return invalid_result
+    api_url = pull_request.get("url")
+    if not api_url:
+        return invalid_result
+    log_context["api_url"] = api_url
+    if pull_request.get("draft", True) or pull_request.get("state") != "open" or pull_request.get("user", {}).get("login", "") == bot_user:
+        return invalid_result
+    if action in ("review_requested", "synchronize") and pull_request.get("created_at") == pull_request.get("updated_at"):
+        # avoid double reviews when opening a PR for the first time
+        return invalid_result
+    return pull_request, api_url
+
+
+async def _perform_commands(commands: List[str], agent: PRAgent, body: dict, api_url: str, log_context: dict):
+    apply_repo_settings(api_url)
+    for command in commands:
+        split_command = command.split(" ")
+        command = split_command[0]
+        args = split_command[1:]
+        other_args = update_settings_from_args(args)
+        new_command = ' '.join([command] + other_args)
+        get_logger().info(body)
+        get_logger().info(f"Performing command: {new_command}")
+        with get_logger().contextualize(**log_context):
+            await agent.handle_request(api_url, new_command)
+
+
 def _is_duplicate_request(body: Dict[str, Any]) -> bool:
     """
     In some deployments its possible to get duplicate requests if the handling is long,
@@ -150,13 +224,8 @@ def _is_duplicate_request(body: Dict[str, Any]) -> bool:
     """
     request_hash = hash(str(body))
     get_logger().info(f"request_hash: {request_hash}")
-    request_time = time.monotonic()
+    is_duplicate = _duplicate_requests_cache.get(request_hash, False)
-    ttl = get_settings().github_app.duplicate_requests_cache_ttl  # in seconds
+    _duplicate_requests_cache[request_hash] = True
-    to_delete = [key for key, key_time in _duplicate_requests_cache.items() if request_time - key_time > ttl]
-    for key in to_delete:
-        del _duplicate_requests_cache[key]
-    is_duplicate = request_hash in _duplicate_requests_cache
-    _duplicate_requests_cache[request_hash] = request_time
     if is_duplicate:
         get_logger().info(f"Ignoring duplicate request {request_hash}")
     return is_duplicate

pr_agent/servers/gitlab_webhook.py

@@ -12,23 +12,33 @@ from starlette_context.middleware import RawContextMiddleware
 
 from pr_agent.agent.pr_agent import PRAgent
 from pr_agent.config_loader import get_settings, global_settings
-from pr_agent.log import get_logger, setup_logger
+from pr_agent.log import LoggingFormat, get_logger, setup_logger
 from pr_agent.secret_providers import get_secret_provider
 
-setup_logger()
+setup_logger(fmt=LoggingFormat.JSON)
 router = APIRouter()
 
 secret_provider = get_secret_provider() if get_settings().get("CONFIG.SECRET_PROVIDER") else None
 
 
+def handle_request(background_tasks: BackgroundTasks, url: str, body: str, log_context: dict):
+    log_context["action"] = body
+    log_context["event"] = "pull_request" if body == "/review" else "comment"
+    log_context["api_url"] = url
+    with get_logger().contextualize(**log_context):
+        background_tasks.add_task(PRAgent().handle_request, url, body)
+
+
 @router.post("/webhook")
 async def gitlab_webhook(background_tasks: BackgroundTasks, request: Request):
+    log_context = {"server_type": "gitlab_app"}
     if request.headers.get("X-Gitlab-Token") and secret_provider:
         request_token = request.headers.get("X-Gitlab-Token")
         secret = secret_provider.get_secret(request_token)
         try:
             secret_dict = json.loads(secret)
             gitlab_token = secret_dict["gitlab_token"]
+            log_context["sender"] = secret_dict["id"]
             context["settings"] = copy.deepcopy(global_settings)
             context["settings"].gitlab.personal_access_token = gitlab_token
         except Exception as e:
@@ -48,13 +58,13 @@ async def gitlab_webhook(background_tasks: BackgroundTasks, request: Request):
     if data.get('object_kind') == 'merge_request' and data['object_attributes'].get('action') in ['open', 'reopen']:
         get_logger().info(f"A merge request has been opened: {data['object_attributes'].get('title')}")
         url = data['object_attributes'].get('url')
-        background_tasks.add_task(PRAgent().handle_request, url, "/review")
+        handle_request(background_tasks, url, "/review")
     elif data.get('object_kind') == 'note' and data['event_type'] == 'note':
         if 'merge_request' in data:
             mr = data['merge_request']
             url = mr.get('url')
             body = data.get('object_attributes', {}).get('note')
-            background_tasks.add_task(PRAgent().handle_request, url, body)
+            handle_request(background_tasks, url, body)
     return JSONResponse(status_code=status.HTTP_200_OK, content=jsonable_encoder({"message": "success"}))
 
 

pr_agent/servers/utils.py

@@ -1,5 +1,8 @@
 import hashlib
 import hmac
+import time
+from collections import defaultdict
+from typing import Callable, Any
 
 from fastapi import HTTPException
 
@@ -25,3 +28,59 @@ def verify_signature(payload_body, secret_token, signature_header):
 class RateLimitExceeded(Exception):
     """Raised when the git provider API rate limit has been exceeded."""
     pass
+
+
+class DefaultDictWithTimeout(defaultdict):
+    """A defaultdict with a time-to-live (TTL)."""
+
+    def __init__(
+        self,
+        default_factory: Callable[[], Any] = None,
+        ttl: int = None,
+        refresh_interval: int = 60,
+        update_key_time_on_get: bool = True,
+        *args,
+        **kwargs,
+    ):
+        """
+        Args:
+            default_factory: The default factory to use for keys that are not in the dictionary.
+            ttl: The time-to-live (TTL) in seconds.
+            refresh_interval: How often to refresh the dict and delete items older than the TTL.
+            update_key_time_on_get: Whether to update the access time of a key also on get (or only when set).
+        """
+        super().__init__(default_factory, *args, **kwargs)
+        self.__key_times = dict()
+        self.__ttl = ttl
+        self.__refresh_interval = refresh_interval
+        self.__update_key_time_on_get = update_key_time_on_get
+        self.__last_refresh = self.__time() - self.__refresh_interval
+
+    @staticmethod
+    def __time():
+        return time.monotonic()
+
+    def __refresh(self):
+        if self.__ttl is None:
+            return
+        request_time = self.__time()
+        if request_time - self.__last_refresh > self.__refresh_interval:
+            return
+        to_delete = [key for key, key_time in self.__key_times.items() if request_time - key_time > self.__ttl]
+        for key in to_delete:
+            del self[key]
+        self.__last_refresh = request_time
+
+    def __getitem__(self, __key):
+        if self.__update_key_time_on_get:
+            self.__key_times[__key] = self.__time()
+        self.__refresh()
+        return super().__getitem__(__key)
+
+    def __setitem__(self, __key, __value):
+        self.__key_times[__key] = self.__time()
+        return super().__setitem__(__key, __value)
+
+    def __delitem__(self, __key):
+        del self.__key_times[__key]
+        return super().__delitem__(__key)

pr_agent/settings/configuration.toml

@@ -24,6 +24,7 @@ num_code_suggestions=4
 inline_code_comments = false
 ask_and_reflect=false
 automatic_review=true
+remove_previous_review_comment=false
 extra_instructions = ""
 
 [pr_description] # /describe #
@@ -31,6 +32,7 @@ publish_labels=true
 publish_description_as_comment=false
 add_original_user_description=false
 keep_original_user_title=false
+use_bullet_points=true
 extra_instructions = ""
 # markers
 use_description_markers=false
@@ -82,6 +84,27 @@ pr_commands = [
     "/describe --pr_description.add_original_user_description=true --pr_description.keep_original_user_title=true",
     "/auto_review",
 ]
+# settings for "pull_request" event with "synchronize" action - used to detect and handle push triggers for new commits
+handle_push_trigger = false
+push_trigger_ignore_bot_commits = true
+push_trigger_ignore_merge_commits = true
+push_trigger_wait_for_initial_review = true
+push_trigger_pending_tasks_backlog = true
+push_trigger_pending_tasks_ttl = 300
+push_commands = [
+    "/describe --pr_description.add_original_user_description=true --pr_description.keep_original_user_title=true",
+    """/auto_review -i \
+       --pr_reviewer.require_focused_review=false \
+       --pr_reviewer.require_score_review=false \
+       --pr_reviewer.require_tests_review=false \
+       --pr_reviewer.require_security_review=false \
+       --pr_reviewer.require_estimate_effort_to_review=false \
+       --pr_reviewer.num_code_suggestions=0 \
+       --pr_reviewer.inline_code_comments=false \
+       --pr_reviewer.remove_previous_review_comment=true \
+       --pr_reviewer.extra_instructions='' \
+    """
+]
 
 [gitlab]
 # URL to the gitlab service

pr_agent/settings/language_extensions.toml

@@ -433,3 +433,6 @@ reStructuredText = [".rst", ".rest", ".rest.txt", ".rst.txt", ]
 wisp = [".wisp", ]
 xBase = [".prg", ".prw", ]
 
+[docs_blacklist_extensions]
+# Disable docs for these extensions of text files and scripts that are not programming languages of function, classes and methods
+docs_blacklist = ['sql', 'txt', 'yaml', 'json', 'xml', 'md', 'rst', 'rest', 'rest.txt', 'rst.txt', 'mdpolicy', 'mdown', 'markdown', 'mdwn', 'mkd', 'mkdn', 'mkdown', 'sh']

pr_agent/settings/pr_description_prompts.toml

@@ -31,7 +31,8 @@ PR Type:
       - Other
 PR Description:
   type: string
-  description: an informative and concise description of the PR
+  description: an informative and concise description of the PR.
+  {%- if use_bullet_points %} Use bullet points. {% endif %}
 PR Main Files Walkthrough:
   type: array
   maxItems: 10

pr_agent/settings/pr_reviewer_prompts.toml

@@ -25,7 +25,7 @@ code line that already existed in the file....
 The review should focus on new code added in the PR (lines starting with '+'), and not on code that already existed in the file (lines starting with '-', or without prefix).
 
 {%- if num_code_suggestions > 0 %}
-- Provide up to {{ num_code_suggestions }} code suggestions.
+- Provide up to {{ num_code_suggestions }} code suggestions. Try to provide diverse and insightful suggestions.
 - Focus on important suggestions like fixing code problems, issues and bugs. As a second priority, provide suggestions for meaningful code improvements, like performance, vulnerability, modularity, and best practices.
 - Avoid making suggestions that have already been implemented in the PR code. For example, if you want to add logs, or change a variable to const, or anything else, make sure it isn't already in the PR code.
 - Don't suggest to add docstring, type hints, or comments.
@@ -99,10 +99,10 @@ PR Feedback:
   General suggestions:
     type: string
     description: |-
-      General suggestions and feedback for the contributors and maintainers of
+      General suggestions and feedback for the contributors and maintainers of this PR.
-      this PR. May include important suggestions for the overall structure,
+      May include important suggestions for the overall structure,
-      primary purpose, best practices, critical bugs, and other aspects of the
+      primary purpose, best practices, critical bugs, and other aspects of the PR.
-      PR. Don't address PR title and description, or lack of tests. Explain your suggestions.
+      Don't address PR title and description, or lack of tests. Explain your suggestions.
 {%- if num_code_suggestions > 0 %}
   Code feedback:
     type: array
@@ -115,11 +115,10 @@ PR Feedback:
       suggestion:
         type: string
         description: |-
-          a concrete suggestion for meaningfully improving the new PR code. Also
+          a concrete suggestion for meaningfully improving the new PR code.
-          describe how, specifically, the suggestion can be applied to new PR
+          Also describe how, specifically, the suggestion can be applied to new PR code.
-          code. Add tags with importance measure that matches each suggestion
+          Add tags with importance measure that matches each suggestion ('important' or 'medium').
-          ('important' or 'medium'). Do not make suggestions for updating or
+          Do not make suggestions for updating or adding docstrings, renaming PR title and description, or linter like.
-          adding docstrings, renaming PR title and description, or linter like.
       relevant line:
         type: string
         description: |-

pr_agent/tools/pr_add_docs.py

@@ -65,6 +65,11 @@ class PRAddDocs:
 
     async def _prepare_prediction(self, model: str):
         get_logger().info('Getting PR diff...')
+
+        # Disable adding docs to scripts and other non-relevant text files
+        from pr_agent.algo.language_handler import bad_extensions
+        bad_extensions += get_settings().docs_blacklist_extensions.docs_blacklist
+
         self.patches_diff = get_pr_diff(self.git_provider,
                                         self.token_handler,
                                         model,

pr_agent/tools/pr_description.py

@@ -40,6 +40,7 @@ class PRDescription:
             "description": self.git_provider.get_pr_description(full=False),
             "language": self.main_pr_language,
             "diff": "",  # empty diff for initial calculation
+            "use_bullet_points": get_settings().pr_description.use_bullet_points,
             "extra_instructions": get_settings().pr_description.extra_instructions,
             "commit_messages_str": self.git_provider.get_commit_messages()
         }
@@ -185,6 +186,11 @@ class PRDescription:
         else:
             ai_header = ""
 
+        ai_type = self.data.get('PR Type')
+        if ai_type and not re.search(r'<!--\s*pr_agent:type\s*-->', body):
+            pr_type = f"{ai_header}{ai_type}"
+            body = body.replace('pr_agent:type', pr_type)
+
         ai_summary = self.data.get('PR Description')
         if ai_summary and not re.search(r'<!--\s*pr_agent:summary\s*-->', body):
             summary = f"{ai_header}{ai_summary}"

pr_agent/tools/pr_reviewer.py

@@ -98,6 +98,9 @@ class PRReviewer:
             if self.is_auto and not get_settings().pr_reviewer.automatic_review:
                 get_logger().info(f'Automatic review is disabled {self.pr_url}')
                 return None
+            if self.is_auto and self.incremental.is_incremental and not self.incremental.first_new_commit_sha:
+                get_logger().info(f"Incremental review is enabled for {self.pr_url} but there are no new commits")
+                return None
 
             get_logger().info(f'Reviewing PR: {self.pr_url} ...')
 
@@ -111,9 +114,10 @@ class PRReviewer:
 
             if get_settings().config.publish_output:
                 get_logger().info('Pushing PR review...')
+                previous_review_comment = self._get_previous_review_comment()
                 self.git_provider.publish_comment(pr_comment)
                 self.git_provider.remove_initial_comment()
-
+                self._remove_previous_review_comment(previous_review_comment)
                 if get_settings().pr_reviewer.inline_code_comments:
                     get_logger().info('Pushing inline code comments...')
                     self._publish_inline_code_comments()
@@ -228,9 +232,13 @@ class PRReviewer:
         if self.incremental.is_incremental:
             last_commit_url = f"{self.git_provider.get_pr_url()}/commits/" \
                               f"{self.git_provider.incremental.first_new_commit_sha}"
+            last_commit_msg = self.incremental.commits_range[0].commit.message if self.incremental.commits_range else ""
+            incremental_review_markdown_text = f"Starting from commit {last_commit_url}"
+            if last_commit_msg:
+                incremental_review_markdown_text += f"  \n_({last_commit_msg.splitlines(keepends=False)[0]})_"
             data = OrderedDict(data)
             data.update({'Incremental PR Review': {
-                "⏮️ Review for commits since previous PR-Agent review": f"Starting from commit {last_commit_url}"}})
+                "⏮️ Review for commits since previous PR-Agent review": incremental_review_markdown_text}})
             data.move_to_end('Incremental PR Review', last=False)
 
         markdown_text = convert_to_markdown(data, self.git_provider.is_supported("gfm_markdown"))
@@ -311,3 +319,26 @@ class PRReviewer:
                     break
 
         return question_str, answer_str
+
+    def _get_previous_review_comment(self):
+        """
+        Get the previous review comment if it exists.
+        """
+        try:
+            if get_settings().pr_reviewer.remove_previous_review_comment and hasattr(self.git_provider, "get_previous_review"):
+                return self.git_provider.get_previous_review(
+                    full=not self.incremental.is_incremental,
+                    incremental=self.incremental.is_incremental,
+                )
+        except Exception as e:
+            get_logger().exception(f"Failed to get previous review comment, error: {e}")
+
+    def _remove_previous_review_comment(self, comment):
+        """
+        Remove the previous review comment if it exists.
+        """
+        try:
+            if get_settings().pr_reviewer.remove_previous_review_comment and comment:
+                self.git_provider.remove_comment(comment)
+        except Exception as e:
+            get_logger().exception(f"Failed to remove previous review comment, error: {e}")

pr_agent/tools/pr_similar_issue.py

@@ -48,9 +48,9 @@ class PRSimilarIssue:
         # check if index exists, and if repo is already indexed
         run_from_scratch = False
         if run_from_scratch:  # for debugging
-            if not index_name in pinecone.list_indexes():
+            pinecone.init(api_key=api_key, environment=environment)
+            if index_name in pinecone.list_indexes():
                 get_logger().info('Removing index...')
-                pinecone.init(api_key=api_key, environment=environment)
                 pinecone.delete_index(index_name)
                 get_logger().info('Done')