few more changes suggested by AI implemented

otherwise gitlab web doesn't able show diff on page and return 500 or 400 errors based on different scenarios

.github/workflows/build-and-test.yaml

@@ -0,0 +1,36 @@
+name: Build-and-test
+
+on:
+  push:
+
+jobs:
+  build-and-test:
+    runs-on: ubuntu-latest
+
+    steps:
+      - id: checkout
+        uses: actions/checkout@v2
+
+      - id: dockerx
+        name: Setup Docker Buildx
+        uses: docker/setup-buildx-action@v2
+
+      - id: build
+        name: Build dev docker
+        uses: docker/build-push-action@v2
+        with:
+          context: .
+          file: ./docker/Dockerfile
+          push: false
+          load: true
+          tags: codiumai/pr-agent:test
+          cache-from: type=gha,scope=dev
+          cache-to: type=gha,mode=max,scope=dev
+          target: test
+
+      - id: test
+        name: Test dev docker
+        run: |
+          docker run --rm codiumai/pr-agent:test pytest -v
+        
+          

.github/workflows/pr-agent-review.yaml

@@ -1,6 +1,17 @@
+# This workflow enables developers to call PR-Agents `/[actions]` in PR's comments and upon PR creation. 
+# Learn more at https://www.codium.ai/pr-agent/
+# This is v0.2 of this workflow file
+
+name: PR-Agent
+
 on:
   pull_request:
   issue_comment:
+
+permissions:
+  issues: write
+  pull-requests: write
+
 jobs:
   pr_agent_job:
     runs-on: ubuntu-latest

INSTALL.md

@@ -92,6 +92,7 @@ pip install -r requirements.txt
 
 ```
 cp pr_agent/settings/.secrets_template.toml pr_agent/settings/.secrets.toml
+chmod 600 pr_agent/settings/.secrets.toml
 # Edit .secrets.toml file
 ```
 
@@ -128,6 +129,7 @@ Allowing you to automate the review process on your private or public repositori
      - Pull requests: Read & write
      - Issue comment: Read & write
      - Metadata: Read-only
+     - Contents: Read-only
    - Set the following events:
      - Issue comment
      - Pull request

README.md

@@ -79,7 +79,7 @@ CodiumAI `PR-Agent` is an open-source tool aiming to help developers review pull
 |-------|---------------------------------------------|:------:|:------:|:---------:|
 | TOOLS | Review                                      |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:       |
 |       | ⮑ Inline review                             |   :white_check_mark:    |   :white_check_mark:    |           |
-|       | Ask                                         |   :white_check_mark:    |   :white_check_mark:    |           |
+|       | Ask                                         |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:
 |       | Auto-Description                            |   :white_check_mark:    |  :white_check_mark:      |           |
 |       | Improve Code                                |   :white_check_mark:    |   :white_check_mark:    |           |
 |       | Reflect and Review                          |   :white_check_mark:    |                         |           |
@@ -97,12 +97,12 @@ CodiumAI `PR-Agent` is an open-source tool aiming to help developers review pull
 |       | Incremental PR Review |   :white_check_mark:    |      |         |
 
 Examples for invoking the different tools via the CLI:
-- **Review**:       python cli.py --pr-url=<pr_url>  review
-- **Describe**:     python cli.py --pr-url=<pr_url>  describe
-- **Improve**:      python cli.py --pr-url=<pr_url>  improve
-- **Ask**:          python cli.py --pr-url=<pr_url>  ask "Write me a poem about this PR"
-- **Reflect**:      python cli.py --pr-url=<pr_url>  reflect
-- **Update Changelog**:      python cli.py --pr-url=<pr_url>  update_changelog
+- **Review**:       python cli.py --pr_url=<pr_url>  review
+- **Describe**:     python cli.py --pr_url=<pr_url>  describe
+- **Improve**:      python cli.py --pr_url=<pr_url>  improve
+- **Ask**:          python cli.py --pr_url=<pr_url>  ask "Write me a poem about this PR"
+- **Reflect**:      python cli.py --pr_url=<pr_url>  reflect
+- **Update Changelog**:      python cli.py --pr_url=<pr_url>  update_changelog
 
 "<pr_url>" is the url of the relevant PR (for example: https://github.com/Codium-ai/pr-agent/pull/50).
 

docker/Dockerfile

@@ -4,17 +4,21 @@ WORKDIR /app
 ADD pyproject.toml .
 RUN pip install . && rm pyproject.toml
 ENV PYTHONPATH=/app
-ADD pr_agent pr_agent
 
 FROM base as github_app
+ADD pr_agent pr_agent
 CMD ["python", "pr_agent/servers/github_app.py"]
 
 FROM base as github_polling
+ADD pr_agent pr_agent
 CMD ["python", "pr_agent/servers/github_polling.py"]
 
 FROM base as test
 ADD requirements-dev.txt .
 RUN pip install -r requirements-dev.txt && rm requirements-dev.txt
+ADD pr_agent pr_agent
+ADD tests tests
 
 FROM base as cli
+ADD pr_agent pr_agent
 ENTRYPOINT ["python", "pr_agent/cli.py"]

pr_agent/algo/ai_handler.py

@@ -29,7 +29,6 @@ class AiHandler:
             self.azure = False
             if get_settings().get("OPENAI.ORG", None):
                 litellm.organization = get_settings().openai.org
-            self.deployment_id = get_settings().get("OPENAI.DEPLOYMENT_ID", None)
             if get_settings().get("OPENAI.API_TYPE", None):
                 if get_settings().openai.api_type == "azure":
                     self.azure = True
@@ -47,6 +46,13 @@ class AiHandler:
         except AttributeError as e:
             raise ValueError("OpenAI key is required") from e
 
+    @property
+    def deployment_id(self):
+        """
+        Returns the deployment ID for the OpenAI API.
+        """
+        return get_settings().get("OPENAI.DEPLOYMENT_ID", None)
+
     @retry(exceptions=(APIError, Timeout, TryAgain, AttributeError, RateLimitError),
            tries=OPENAI_RETRIES, delay=2, backoff=2, jitter=(1, 3))
     async def chat_completion(self, model: str, temperature: float, system: str, user: str):
@@ -70,9 +76,15 @@ class AiHandler:
             TryAgain: If there is an attribute error during OpenAI inference.
         """
         try:
+            deployment_id = self.deployment_id
+            if get_settings().config.verbosity_level >= 2:
+                logging.debug(
+                    f"Generating completion with {model}"
+                    f"{(' from deployment ' + deployment_id) if deployment_id else ''}"
+                )
             response = await acompletion(
                 model=model,
-                deployment_id=self.deployment_id,
+                deployment_id=deployment_id,
                 messages=[
                     {"role": "system", "content": system},
                     {"role": "user", "content": user}

pr_agent/algo/pr_processing.py

@@ -208,18 +208,45 @@ def pr_generate_compressed_diff(top_langs: list, token_handler: TokenHandler, mo
 
 
 async def retry_with_fallback_models(f: Callable):
+    all_models = _get_all_models()
+    all_deployments = _get_all_deployments(all_models)
+    # try each (model, deployment_id) pair until one is successful, otherwise raise exception
+    for i, (model, deployment_id) in enumerate(zip(all_models, all_deployments)):
+        try:
+            get_settings().set("openai.deployment_id", deployment_id)
+            return await f(model)
+        except Exception as e:
+            logging.warning(
+                f"Failed to generate prediction with {model}"
+                f"{(' from deployment ' + deployment_id) if deployment_id else ''}: "
+                f"{traceback.format_exc()}"
+            )
+            if i == len(all_models) - 1:  # If it's the last iteration
+                raise  # Re-raise the last exception
+
+
+def _get_all_models() -> List[str]:
     model = get_settings().config.model
     fallback_models = get_settings().config.fallback_models
     if not isinstance(fallback_models, list):
-        fallback_models = [fallback_models]
+        fallback_models = [m.strip() for m in fallback_models.split(",")]
     all_models = [model] + fallback_models
-    for i, model in enumerate(all_models):
-        try:
-            return await f(model)
-        except Exception as e:
-            logging.warning(f"Failed to generate prediction with {model}: {traceback.format_exc()}")
-            if i == len(all_models) - 1:  # If it's the last iteration
-                raise  # Re-raise the last exception
+    return all_models
+
+
+def _get_all_deployments(all_models: List[str]) -> List[str]:
+    deployment_id = get_settings().get("openai.deployment_id", None)
+    fallback_deployments = get_settings().get("openai.fallback_deployments", [])
+    if not isinstance(fallback_deployments, list) and fallback_deployments:
+        fallback_deployments = [d.strip() for d in fallback_deployments.split(",")]
+    if fallback_deployments:
+        all_deployments = [deployment_id] + fallback_deployments
+        if len(all_deployments) < len(all_models):
+            raise ValueError(f"The number of deployments ({len(all_deployments)}) "
+                             f"is less than the number of models ({len(all_models)})")
+    else:
+        all_deployments = [deployment_id] * len(all_models)
+    return all_deployments
 
 
 def find_line_number_of_relevant_line_in_file(diff_files: List[FilePatchInfo],
@@ -298,12 +325,16 @@ def clip_tokens(text: str, max_tokens: int) -> str:
         str: The clipped string.
     """
     # We'll estimate the number of tokens by hueristically assuming 2.5 tokens per word
-    encoder = get_token_encoder()
-    num_input_tokens = len(encoder.encode(text))
-    if num_input_tokens <= max_tokens:
-        return text
-    num_chars = len(text)
-    chars_per_token = num_chars / num_input_tokens
-    num_output_chars = int(chars_per_token * max_tokens)
-    clipped_text = text[:num_output_chars]
-    return clipped_text
+    try:
+        encoder = get_token_encoder()
+        num_input_tokens = len(encoder.encode(text))
+        if num_input_tokens <= max_tokens:
+            return text
+        num_chars = len(text)
+        chars_per_token = num_chars / num_input_tokens
+        num_output_chars = int(chars_per_token * max_tokens)
+        clipped_text = text[:num_output_chars]
+        return clipped_text
+    except Exception as e:
+        logging.warning(f"Failed to clip tokens: {e}")
+        return text

pr_agent/algo/utils.py

@@ -8,8 +8,8 @@ import textwrap
 from datetime import datetime
 from typing import Any, List
 
+import yaml
 from starlette_context import context
-
 from pr_agent.config_loader import get_settings, global_settings
 
 
@@ -258,3 +258,26 @@ def update_settings_from_args(args: List[str]) -> List[str]:
             else:
                 other_args.append(arg)
     return other_args
+
+
+def load_yaml(review_text: str) -> dict:
+    review_text = review_text.removeprefix('```yaml').rstrip('`')
+    try:
+        data = yaml.load(review_text, Loader=yaml.SafeLoader)
+    except Exception as e:
+        logging.error(f"Failed to parse AI prediction: {e}")
+        data = try_fix_yaml(review_text)
+    return data
+
+def try_fix_yaml(review_text: str) -> dict:
+    review_text_lines = review_text.split('\n')
+    data = {}
+    for i in range(1, len(review_text_lines)):
+        review_text_lines_tmp = '\n'.join(review_text_lines[:-i])
+        try:
+            data = yaml.load(review_text_lines_tmp, Loader=yaml.SafeLoader)
+            logging.info(f"Successfully parsed AI prediction after removing {i} lines")
+            break
+        except:
+            pass
+    return data

pr_agent/cli.py

@@ -10,13 +10,13 @@ from pr_agent.config_loader import get_settings
 def run(inargs=None):
     parser = argparse.ArgumentParser(description='AI based pull request analyzer', usage=
 """\
-Usage: cli.py --pr-url <URL on supported git hosting service> <command> [<args>].
+Usage: cli.py --pr-url=<URL on supported git hosting service> <command> [<args>].
 For example:
-- cli.py --pr-url=... review
-- cli.py --pr-url=... describe
-- cli.py --pr-url=... improve
-- cli.py --pr-url=... ask "write me a poem about this PR"
-- cli.py --pr-url=... reflect
+- cli.py --pr_url=... review
+- cli.py --pr_url=... describe
+- cli.py --pr_url=... improve
+- cli.py --pr_url=... ask "write me a poem about this PR"
+- cli.py --pr_url=... reflect
 
 Supported commands:
 review / review_pr - Add a review that includes a summary of the PR and specific suggestions for improvement.
@@ -27,7 +27,7 @@ reflect - Ask the PR author questions about the PR.
 update_changelog - Update the changelog based on the PR's contents.
 
 To edit any configuration parameter from 'configuration.toml', just add -config_path=<value>.
-For example: '- cli.py --pr-url=... review --pr_reviewer.extra_instructions="focus on the file: ..."'
+For example: 'python cli.py --pr_url=... review --pr_reviewer.extra_instructions="focus on the file: ..."'
 """)
     parser.add_argument('--pr_url', type=str, help='The URL of the PR to review', required=True)
     parser.add_argument('command', type=str, help='The', choices=commands, default='review')

pr_agent/git_providers/bitbucket_provider.py

@@ -26,6 +26,13 @@ class BitbucketProvider:
         if pr_url:
             self.set_pr(pr_url)
 
+    def get_repo_settings(self):
+        try:
+            contents = self.repo_obj.get_contents(".pr_agent.toml", ref=self.pr.head.sha).decoded_content
+            return contents
+        except Exception:
+            return ""
+
     def is_supported(self, capability: str) -> bool:
         if capability in ['get_issue_comments', 'create_inline_comment', 'publish_inline_comments', 'get_labels']:
             return False
@@ -93,6 +100,13 @@ class BitbucketProvider:
     def get_issue_comments(self):
         raise NotImplementedError("Bitbucket provider does not support issue comments yet")
 
+    def get_repo_settings(self):
+        try:
+            contents = self.repo_obj.get_contents(".pr_agent.toml", ref=self.pr.head.sha).decoded_content
+            return contents
+        except Exception:
+            return ""
+
     def add_eyes_reaction(self, issue_comment_id: int) -> Optional[int]:
         return True
 
@@ -104,7 +118,7 @@ class BitbucketProvider:
         parsed_url = urlparse(pr_url)
         
         if 'bitbucket.org' not in parsed_url.netloc:
-            raise ValueError("The provided URL is not a valid GitHub URL")
+            raise ValueError("The provided URL is not a valid Bitbucket URL")
 
         path_parts = parsed_url.path.strip('/').split('/')
         

pr_agent/git_providers/git_provider.py

@@ -89,6 +89,10 @@ class GitProvider(ABC):
     def get_issue_comments(self):
         pass
 
+    @abstractmethod
+    def get_repo_settings(self):
+        pass
+
     @abstractmethod
     def add_eyes_reaction(self, issue_comment_id: int) -> Optional[int]:
         pass

pr_agent/git_providers/github_provider.py

@@ -153,7 +153,7 @@ class GithubProvider(GitProvider):
 
 
     def create_inline_comment(self, body: str, relevant_file: str, relevant_line_in_file: str):
-        position = find_line_number_of_relevant_line_in_file(self.diff_files, relevant_file.strip('`'), relevant_line_in_file)
+        position, absolute_position = find_line_number_of_relevant_line_in_file(self.diff_files, relevant_file.strip('`'), relevant_line_in_file)
         if position == -1:
             if get_settings().config.verbosity_level >= 2:
                 logging.info(f"Could not find position for {relevant_file} {relevant_line_in_file}")
@@ -398,10 +398,13 @@ class GithubProvider(GitProvider):
 
     def generate_link_to_relevant_line_number(self, suggestion) -> str:
         try:
-            relevant_file = suggestion['relevant file']
+            relevant_file = suggestion['relevant file'].strip('`').strip("'")
             relevant_line_str = suggestion['relevant line']
+            if not relevant_line_str:
+                return ""
+
             position, absolute_position = find_line_number_of_relevant_line_in_file \
-                (self.diff_files, relevant_file.strip('`'), relevant_line_str)
+                (self.diff_files, relevant_file, relevant_line_str)
 
             if absolute_position != -1:
                 # # link to right file only
@@ -416,4 +419,4 @@ class GithubProvider(GitProvider):
             if get_settings().config.verbosity_level >= 2:
                 logging.info(f"Failed adding line link, error: {e}")
 
-        return ""
+        return ""

pr_agent/git_providers/gitlab_provider.py

@@ -14,6 +14,9 @@ from .git_provider import EDIT_TYPE, FilePatchInfo, GitProvider
 
 logger = logging.getLogger()
 
+class DiffNotFoundError(Exception):
+    """Raised when the diff for a merge request cannot be found."""
+    pass
 
 class GitLabProvider(GitProvider):
 
@@ -56,7 +59,7 @@ class GitLabProvider(GitProvider):
             self.last_diff = self.mr.diffs.list(get_all=True)[-1]
         except IndexError as e:
             logger.error(f"Could not get diff for merge request {self.id_mr}")
-            raise ValueError(f"Could not get diff for merge request {self.id_mr}") from e
+            raise DiffNotFoundError(f"Could not get diff for merge request {self.id_mr}") from e
 
 
     def _get_pr_file_content(self, file_path: str, branch: str) -> str:
@@ -150,16 +153,20 @@ class GitLabProvider(GitProvider):
     def create_inline_comments(self, comments: list[dict]):
         raise NotImplementedError("Gitlab provider does not support publishing inline comments yet")
 
-    def send_inline_comment(self, body, edit_type, found, relevant_file, relevant_line_in_file, source_line_no,
-                            target_file, target_line_no):
+    def send_inline_comment(self,body: str,edit_type: str,found: bool,relevant_file: str,relevant_line_in_file: int,
+                            source_line_no: int, target_file: str,target_line_no: int) -> None:
         if not found:
             logging.info(f"Could not find position for {relevant_file} {relevant_line_in_file}")
         else:
-            d = self.last_diff
+            # in order to have exact sha's we have to find correct diff for this change
+            diff = self.get_relevant_diff(relevant_file, relevant_line_in_file)
+            if diff is None:
+                logger.error(f"Could not get diff for merge request {self.id_mr}")
+                raise DiffNotFoundError(f"Could not get diff for merge request {self.id_mr}")
             pos_obj = {'position_type': 'text',
                        'new_path': target_file.filename,
                        'old_path': target_file.old_filename if target_file.old_filename else target_file.filename,
-                       'base_sha': d.base_commit_sha, 'start_sha': d.start_commit_sha, 'head_sha': d.head_commit_sha}
+                       'base_sha': diff.base_commit_sha, 'start_sha': diff.start_commit_sha, 'head_sha': diff.head_commit_sha}
             if edit_type == 'deletion':
                 pos_obj['old_line'] = source_line_no - 1
             elif edit_type == 'addition':
@@ -171,6 +178,23 @@ class GitLabProvider(GitProvider):
             self.mr.discussions.create({'body': body,
                                         'position': pos_obj})
 
+    def get_relevant_diff(self, relevant_file: str, relevant_line_in_file: int) -> Optional[dict]:
+        changes = self.mr.changes()  # Retrieve the changes for the merge request once
+        if not changes:
+            logging.error('No changes found for the merge request.')
+            return None
+        all_diffs = self.mr.diffs.list(get_all=True)
+        if not all_diffs:
+            logging.error('No diffs found for the merge request.')
+            return None
+        for diff in all_diffs:
+            for change in changes['changes']:
+                if change['new_path'] == relevant_file and relevant_line_in_file in change['diff']:
+                    return diff
+            logging.debug(
+                f'No relevant diff found for {relevant_file} {relevant_line_in_file}. Falling back to last diff.')
+        return self.last_diff  # fallback to last_diff if no relevant diff is found
+
     def publish_code_suggestions(self, code_suggestions: list):
         for suggestion in code_suggestions:
             try:

pr_agent/servers/help.py

@@ -2,8 +2,9 @@ commands_text = "> **/review [-i]**: Request a review of your Pull Request. For
                 "considers changes since the last review, include the '-i' option.\n" \
                 "> **/describe**: Modify the PR title and description based on the contents of the PR.\n" \
                 "> **/improve**: Suggest improvements to the code in the PR. \n" \
-                "> **/ask \\<QUESTION\\>**: Pose a question about the PR.\n\n" \
-                ">To edit any configuration parameter from 'configuration.toml', add --config_path=new_value\n" \
+                "> **/ask \\<QUESTION\\>**: Pose a question about the PR.\n" \
+                "> **/update_changelog**: Update the changelog based on the PR's contents.\n\n" \
+                ">To edit any configuration parameter from **configuration.toml**, add --config_path=new_value\n" \
                 ">For example: /review --pr_reviewer.extra_instructions=\"focus on the file: ...\" \n" \
                 ">To list the possible configuration parameters, use the **/config** command.\n" \
 

pr_agent/settings/.secrets_template.toml

@@ -14,6 +14,7 @@ key = ""  # Acquire through https://platform.openai.com
 #api_version = '2023-05-15'  # Check Azure documentation for the current API version
 #api_base = ""  # The base URL for your Azure OpenAI resource. e.g. "https://<your resource name>.openai.azure.com"
 #deployment_id = ""  # The deployment name you chose when you deployed the engine
+#fallback_deployments = []  # For each fallback model specified in configuration.toml in the [config] section, specify the appropriate deployment_id
 
 [anthropic]
 key = "" # Optional, uncomment if you want to use Anthropic. Acquire through https://www.anthropic.com/

pr_agent/settings/pr_description_prompts.toml

@@ -2,38 +2,67 @@
 system="""You are CodiumAI-PR-Reviewer, a language model designed to review git pull requests.
 Your task is to provide full description of the PR content.
 - Make sure not to focus the new PR code (the '+' lines).
-
+- Notice that the 'Previous title', 'Previous description' and 'Commit messages' sections may be partial, simplistic, non-informative or not up-to-date. Hence, compare them to the PR diff code, and use them only as a reference.
+-  If needed, each YAML output should be in block scalar format ('|-')
 {%- if extra_instructions %}
 
 Extra instructions from the user:
 {{ extra_instructions }}
 {% endif %}
 
-You must use the following JSON schema to format your answer:
-```json
-{
-  "PR Title": {
-      "type": "string",
-      "description": "an informative title for the PR, describing its main theme"
-  },
-  "PR Type": {
-      "type": "string",
-      "description": possible values are: ["Bug fix", "Tests", "Bug fix with tests", "Refactoring", "Enhancement", "Documentation", "Other"]
-    },
-  "PR Description": {
-      "type": "string",
-      "description": "an informative and concise description of the PR"
-  },
-  "PR Main Files Walkthrough": {
-      "type": "string",
-      "description": "a walkthrough of the PR changes. Review main files, in bullet points, and shortly describe the changes in each file (up to 10 most important files). Format: -`filename`: description of changes\n..."
-  }
-}
+You must use the following YAML schema to format your answer:
+```yaml
+PR Title:
+  type: string
+  description: an informative title for the PR, describing its main theme
+PR Type:
+  type: array
+  items:
+    type: string
+    enum:
+      - Bug fix
+      - Tests
+      - Bug fix with tests
+      - Refactoring
+      - Enhancement
+      - Documentation
+      - Other
+PR Description:
+  type: string
+  description: an informative and concise description of the PR
+PR Main Files Walkthrough:
+  type: array
+  maxItems: 10
+  description: |-
+    a walkthrough of the PR changes. Review main files, and shortly describe the changes in each file (up to 10 most important files).
+  items:
+    filename:
+      type: string
+      description: the relevant file full path
+    changes in file:
+      type: string
+      description: minimal and concise description of the changes in the relevant file
 
-Don't repeat the prompt in the answer, and avoid outputting the 'type' and 'description' fields.
+
+Example output:
+```yaml
+PR Title: |-
+  ...
+PR Type:
+  - Bug fix
+PR Description: |-
+  ...
+PR Main Files Walkthrough:
+  - ...
+  - ...
+```
+
+Make sure to output a valid YAML. Don't repeat the prompt in the answer, and avoid outputting the 'type' and 'description' fields.
 """
 
 user="""PR Info:
+Previous title: '{{title}}'
+Previous description: '{{description}}'
 Branch: '{{branch}}'
 {%- if language %}
 
@@ -52,6 +81,6 @@ The PR Git Diff:
 ```
 Note that lines in the diff body are prefixed with a symbol that represents the type of change: '-' for deletions, '+' for additions, and ' ' (a space) for unchanged lines.
 
-Response (should be a valid JSON, and nothing else):
-```json
+Response (should be a valid YAML, and nothing else):
+```yaml
 """

pr_agent/settings/pr_reviewer_prompts.toml

@@ -7,6 +7,7 @@ Your task is to provide constructive and concise feedback for the PR, and also p
 - Suggestions should focus on improving the new added code lines.
 - Make sure not to provide suggestions repeating modifications already implemented in the new PR code (the '+' lines).
 {%- endif %}
+- If needed, each YAML output should be in block scalar format ('|-')
 
 {%- if extra_instructions %}
 
@@ -14,117 +15,121 @@ Extra instructions from the user:
 {{ extra_instructions }}
 {% endif %}
 
-You must use the following JSON schema to format your answer:
-```json
-{
-  "PR Analysis": {
-    "Main theme": {
-      "type": "string",
-      "description": "a short explanation of the PR"
-    },
-    "Type of PR": {
-      "type": "string",
-      "enum": ["Bug fix", "Tests", "Refactoring", "Enhancement", "Documentation", "Other"]
-    },
+You must use the following YAML schema to format your answer:
+```yaml
+PR Analysis:
+  Main theme:
+    type: string
+    description: a short explanation of the PR
+  Type of PR:
+    type: string
+    enum:
+      - Bug fix
+      - Tests
+      - Refactoring
+      - Enhancement
+      - Documentation
+      - Other
 {%- if require_score %}
-    "Score": {
-      "type": "int",
-      "description": "Rate this PR on a scale of 0-100 (inclusive), where 0 means the worst possible PR code, and 100 means PR code of the highest quality, without any bugs or performance issues, that is ready to be merged immediately and run in production at scale."
-    },
+  Score:
+    type: int
+    description: >-
+      Rate this PR on a scale of 0-100 (inclusive), where 0 means the worst
+      possible PR code, and 100 means PR code of the highest quality, without
+      any bugs or performance issues, that is ready to be merged immediately and
+      run in production at scale.
 {%- endif %}
 {%- if require_tests %}
-    "Relevant tests added": {
-      "type": "string",
-      "description": "yes\\no question: does this PR have relevant tests ?"
-    },
+  Relevant tests added:
+    type: string
+    description: yes\\no question: does this PR have relevant tests ?
 {%- endif %}
 {%- if question_str %}
-    "Insights from user's answer": {
-      "type": "string",
-      "description": "shortly summarize the insights you gained from the user's answers to the questions"
-    },
+  Insights from user's answer:
+    type: string
+    description: >-
+      shortly summarize the insights you gained from the user's answers to the questions
 {%- endif %}
 {%- if require_focused %}
-    "Focused PR": {
-      "type": "string",
-      "description": "Is this a focused PR, in the sense that all the PR code diff changes are united under a single focused theme ? If the theme is too broad, or the PR code diff changes are too scattered, then the PR is not focused. Explain your answer shortly."
-    }
-  },
+  Focused PR:
+    type: string
+    description: >-
+      Is this a focused PR, in the sense that all the PR code diff changes are
+      united under a single focused theme ? If the theme is too broad, or the PR
+      code diff changes are too scattered, then the PR is not focused. Explain
+      your answer shortly.
 {%- endif %}
-  "PR Feedback": {
-    "General suggestions": {
-      "type": "string",
-      "description": "General suggestions and feedback for the contributors and maintainers of this PR. May include important suggestions for the overall structure, primary purpose, best practices, critical bugs, and other aspects of the PR. Don't address PR title and description, or lack of tests. Explain your suggestions."
-    },
+PR Feedback:
+  General suggestions:
+    type: string
+    description: >-
+      General suggestions and feedback for the contributors and maintainers of
+      this PR. May include important suggestions for the overall structure,
+      primary purpose, best practices, critical bugs, and other aspects of the
+      PR. Don't address PR title and description, or lack of tests. Explain your
+      suggestions.
 {%- if num_code_suggestions > 0 %}
-    "Code feedback": {
-      "type": "array",
-      "maxItems": {{ num_code_suggestions }},
-      "uniqueItems": true,
-      "items": {
-        "relevant file": {
-          "type": "string",
-          "description": "the relevant file full path"
-        },
-        "suggestion": {
-          "type": "string",
-          "description": "a concrete suggestion for meaningfully improving the new PR code. Also describe how, specifically, the suggestion can be applied to new PR code. Add tags with importance measure that matches each suggestion ('important' or 'medium'). Do not make suggestions for updating or adding docstrings, renaming PR title and description, or linter like.
-        },
-        "relevant line": {
-          "type": "string",
-          "description": "a single code line taken from the relevant file, to which the suggestion applies. The line should be a '+' line. Make sure to output the line exactly as it appears in the relevant file"
-        }
-      }
-    },
+  Code feedback:
+    type: array
+    maxItems: {{ num_code_suggestions }}
+    uniqueItems: true
+    items:
+      relevant file:
+        type: string
+        description: the relevant file full path
+      suggestion:
+        type: string
+        description: |
+          a concrete suggestion for meaningfully improving the new PR code. Also
+          describe how, specifically, the suggestion can be applied to new PR
+          code. Add tags with importance measure that matches each suggestion
+          ('important' or 'medium'). Do not make suggestions for updating or
+          adding docstrings, renaming PR title and description, or linter like.
+      relevant line:
+        type: string
+        description: |
+          a single code line taken from the relevant file, to which the suggestion applies.
+          The line should be a '+' line.
+          Make sure to output the line exactly as it appears in the relevant file
 {%- endif %}
 {%- if require_security %}
-    "Security concerns": {
-      "type": "string",
-      "description": "yes\\no question: does this PR code introduce possible security concerns or issues, like SQL injection, XSS, CSRF, and others ? If answered 'yes', explain your answer shortly"
-       ? explain your answer shortly"
-    }
+  Security concerns:
+    type: string
+    description: >-
+      yes\\no question: does this PR code introduce possible security concerns or
+      issues, like SQL injection, XSS, CSRF, and others ? If answered 'yes',explain your answer shortly
 {%- endif %}
-  }
-}
 ```
 
 Example output:
-'
-{
-    "PR Analysis":
-    {
-        "Main theme": "xxx",
-        "Type of PR": "Bug fix",
+```yaml
+PR Analysis:
+  Main theme: xxx
+  Type of PR: Bug fix
 {%- if require_score %}
-        "Score": 89,
-{%- endif %}
-{%- if require_tests %}
-        "Relevant tests added": "No",
+  Score: 89
 {%- endif %}
+  Relevant tests added: No
 {%- if require_focused %}
-        "Focused PR": "yes\\no, because ..."
+  Focused PR: no, because ...
 {%- endif %}
-    },
-    "PR Feedback":
-    {
-        "General PR suggestions": "..., `xxx`...",
+PR Feedback:
+  General PR suggestions: ...
 {%- if num_code_suggestions > 0 %}
-        "Code feedback": [
-            {
-                "relevant file": "directory/xxx.py",
-                "suggestion": "xxx [important]",
-                "relevant line": "xxx",
-            },
-            ...
-        ]
+  Code feedback:
+    - relevant file: |-
+        directory/xxx.py
+      suggestion: xxx [important]
+      relevant line: |-
+        xxx
+    ...
 {%- endif %}
 {%- if require_security %}
-        "Security concerns": "No, because ..."
+  Security concerns: No
 {%- endif %}
-    }
-}
-'
+```
 
+Make sure to output a valid YAML. Use multi-line block scalar ('|') if needed.
 Don't repeat the prompt in the answer, and avoid outputting the 'type' and 'description' fields.
 """
 
@@ -158,6 +163,6 @@ The PR Git Diff:
 ```
 Note that lines in the diff body are prefixed with a symbol that represents the type of change: '-' for deletions, '+' for additions, and ' ' (a space) for unchanged lines.
 
-Response (should be a valid JSON, and nothing else):
-```json
+Response (should be a valid YAML, and nothing else):
+```yaml
 """

pr_agent/tools/pr_code_suggestions.py

@@ -93,6 +93,10 @@ class PRCodeSuggestions:
 
     def push_inline_code_suggestions(self, data):
         code_suggestions = []
+
+        if not data['Code suggestions']:
+            return self.git_provider.publish_comment('No suggestions found to improve this PR.')
+
         for d in data['Code suggestions']:
             try:
                 if get_settings().config.verbosity_level >= 2:

pr_agent/tools/pr_description.py

@@ -8,6 +8,7 @@ from jinja2 import Environment, StrictUndefined
 from pr_agent.algo.ai_handler import AiHandler
 from pr_agent.algo.pr_processing import get_pr_diff, retry_with_fallback_models
 from pr_agent.algo.token_handler import TokenHandler
+from pr_agent.algo.utils import load_yaml
 from pr_agent.config_loader import get_settings
 from pr_agent.git_providers import get_git_provider
 from pr_agent.git_providers.git_provider import get_main_pr_language
@@ -139,34 +140,45 @@ class PRDescription:
         - title: a string containing the PR title.
         - pr_body: a string containing the PR body in a markdown format.
         - pr_types: a list of strings containing the PR types.
-        - markdown_text: a string containing the AI prediction data in a markdown format.
+        - markdown_text: a string containing the AI prediction data in a markdown format. used for publishing a comment
         """
         # Load the AI prediction data into a dictionary
-        data = json.loads(self.prediction)
+        data = load_yaml(self.prediction.strip())
 
         # Initialization
-        markdown_text = pr_body = ""
         pr_types = []
 
         # Iterate over the dictionary items and append the key and value to 'markdown_text' in a markdown format
+        markdown_text = ""
         for key, value in data.items():
             markdown_text += f"## {key}\n\n"
             markdown_text += f"{value}\n\n"
 
         # If the 'PR Type' key is present in the dictionary, split its value by comma and assign it to 'pr_types'
         if 'PR Type' in data:
-            pr_types = data['PR Type'].split(',')
+            if type(data['PR Type']) == list:
+                pr_types = data['PR Type']
+            elif type(data['PR Type']) == str:
+                pr_types = data['PR Type'].split(',')
 
         # Assign the value of the 'PR Title' key to 'title' variable and remove it from the dictionary
         title = data.pop('PR Title')
 
         # Iterate over the remaining dictionary items and append the key and value to 'pr_body' in a markdown format,
         # except for the items containing the word 'walkthrough'
+        pr_body = ""
         for key, value in data.items():
             pr_body += f"## {key}:\n"
             if 'walkthrough' in key.lower():
-                pr_body += f"{value}\n"
+                # for filename, description in value.items():
+                for file in value:
+                    filename = file['filename'].replace("'", "`")
+                    description = file['changes in file']
+                    pr_body += f'`{filename}`: {description}\n'
             else:
+                # if the value is a list, join its items by comma
+                if type(value) == list:
+                    value = ', '.join(v for v in value)
                 pr_body += f"{value}\n\n___\n"
 
         if get_settings().config.verbosity_level >= 2:

pr_agent/tools/pr_reviewer.py

@@ -4,13 +4,15 @@ import logging
 from collections import OrderedDict
 from typing import List, Tuple
 
+import yaml
 from jinja2 import Environment, StrictUndefined
+from yaml import SafeLoader
 
 from pr_agent.algo.ai_handler import AiHandler
 from pr_agent.algo.pr_processing import get_pr_diff, retry_with_fallback_models, \
     find_line_number_of_relevant_line_in_file, clip_tokens
 from pr_agent.algo.token_handler import TokenHandler
-from pr_agent.algo.utils import convert_to_markdown, try_fix_json
+from pr_agent.algo.utils import convert_to_markdown, try_fix_json, try_fix_yaml, load_yaml
 from pr_agent.config_loader import get_settings
 from pr_agent.git_providers import get_git_provider
 from pr_agent.git_providers.git_provider import IncrementalPR, get_main_pr_language
@@ -160,19 +162,17 @@ class PRReviewer:
         Prepare the PR review by processing the AI prediction and generating a markdown-formatted text that summarizes
         the feedback.
         """
-        review = self.prediction.strip()
-
-        try:
-            data = json.loads(review)
-        except json.decoder.JSONDecodeError:
-            data = try_fix_json(review)
+        data = load_yaml(self.prediction.strip())
 
         # Move 'Security concerns' key to 'PR Analysis' section for better display
         pr_feedback = data.get('PR Feedback', {})
         security_concerns = pr_feedback.get('Security concerns')
-        if security_concerns:
+        if security_concerns is not None:
             del pr_feedback['Security concerns']
-            data.setdefault('PR Analysis', {})['Security concerns'] = security_concerns
+            if type(security_concerns) == bool and security_concerns == False:
+                data.setdefault('PR Analysis', {})['Security concerns'] = 'No security concerns found'
+            else:
+                data.setdefault('PR Analysis', {})['Security concerns'] = security_concerns
 
         #
         if 'Code feedback' in pr_feedback:
@@ -183,6 +183,12 @@ class PRReviewer:
                 del pr_feedback['Code feedback']
             else:
                 for suggestion in code_feedback:
+                    if ('relevant file' in suggestion) and (not suggestion['relevant file'].startswith('``')):
+                        suggestion['relevant file'] = f"``{suggestion['relevant file']}``"
+
+                    if 'relevant line' not in suggestion:
+                        suggestion['relevant line'] = ''
+
                     relevant_line_str = suggestion['relevant line'].split('\n')[0]
 
                     # removing '+'
@@ -219,7 +225,7 @@ class PRReviewer:
             logging.info(f"Markdown response:\n{markdown_text}")
 
         if markdown_text == None or len(markdown_text) == 0:
-            markdown_text = review
+            markdown_text = ""
 
         return markdown_text
 
@@ -230,11 +236,13 @@ class PRReviewer:
         if get_settings().pr_reviewer.num_code_suggestions == 0:
             return
 
-        review = self.prediction.strip()
+        review_text = self.prediction.strip()
+        review_text = review_text.removeprefix('```yaml').rstrip('`')
         try:
-            data = json.loads(review)
-        except json.decoder.JSONDecodeError:
-            data = try_fix_json(review)
+            data = yaml.load(review_text, Loader=SafeLoader)
+        except Exception as e:
+            logging.error(f"Failed to parse AI prediction: {e}")
+            data = try_fix_yaml(review_text)
 
         comments: List[str] = []
         for suggestion in data.get('PR Feedback', {}).get('Code feedback', []):

pyproject.toml

@@ -42,7 +42,8 @@ dependencies = [
   "atlassian-python-api==3.39.0",
   "GitPython~=3.1.32",
   "starlette-context==0.3.6",
-  "litellm~=0.1.351"
+  "litellm~=0.1.351",
+  "PyYAML==6.0"
 ]
 
 [project.urls]

requirements.txt

@@ -11,4 +11,7 @@ pytest~=7.4.0
 aiohttp~=3.8.4
 atlassian-python-api==3.39.0
 GitPython~=3.1.32
+litellm~=0.1.351
+PyYAML==6.0
+starlette-context==0.3.6
 litellm~=0.1.351

tests/unittest/test_bitbucket_provider.py

@@ -0,0 +1,10 @@
+from pr_agent.git_providers.bitbucket_provider import BitbucketProvider
+
+
+class TestBitbucketProvider:
+    def test_parse_pr_url(self):
+        url = "https://bitbucket.org/WORKSPACE_XYZ/MY_TEST_REPO/pull-requests/321"
+        workspace_slug, repo_slug, pr_number = BitbucketProvider._parse_pr_url(url)
+        assert workspace_slug == "WORKSPACE_XYZ"
+        assert repo_slug == "MY_TEST_REPO"
+        assert pr_number == 321

tests/unittest/test_load_yaml.py

@@ -0,0 +1,32 @@
+
+# Generated by CodiumAI
+
+import pytest
+from pr_agent.algo.utils import load_yaml
+
+
+class TestLoadYaml:
+    #  Tests that load_yaml loads a valid YAML string
+    def test_load_valid_yaml(self):
+        yaml_str = 'name: John Smith\nage: 35'
+        expected_output = {'name': 'John Smith', 'age': 35}
+        assert load_yaml(yaml_str) == expected_output
+
+    def test_load_complicated_yaml(self):
+        yaml_str = \
+'''\
+PR Analysis:
+  Main theme: Enhancing the `/describe` command prompt by adding title and description
+  Type of PR: Enhancement
+  Relevant tests added: No
+  Focused PR: Yes, the PR is focused on enhancing the `/describe` command prompt.
+
+PR Feedback:
+  General suggestions: The PR seems to be well-structured and focused on a specific enhancement. However, it would be beneficial to add tests to ensure the new feature works as expected.
+  Code feedback:
+    - relevant file: pr_agent/settings/pr_description_prompts.toml
+      suggestion: Consider using a more descriptive variable name than 'user' for the command prompt. A more descriptive name would make the code more readable and maintainable. [medium]
+      relevant line: 'user="""PR Info:'
+  Security concerns: No'''
+        expected_output = {'PR Analysis': {'Main theme': 'Enhancing the `/describe` command prompt by adding title and description', 'Type of PR': 'Enhancement', 'Relevant tests added': False, 'Focused PR': 'Yes, the PR is focused on enhancing the `/describe` command prompt.'}, 'PR Feedback': {'General suggestions': 'The PR seems to be well-structured and focused on a specific enhancement. However, it would be beneficial to add tests to ensure the new feature works as expected.', 'Code feedback': [{'relevant file': 'pr_agent/settings/pr_description_prompts.toml', 'suggestion': "Consider using a more descriptive variable name than 'user' for the command prompt. A more descriptive name would make the code more readable and maintainable. [medium]", 'relevant line': 'user="""PR Info:'}], 'Security concerns': False}}
+        assert load_yaml(yaml_str) == expected_output