Merge branch 'main' into tr/agent_logic

.dockerignore

@@ -1,2 +1,3 @@
 venv/
-pr_agent/settings/.secrets.toml
+pr_agent/settings/.secrets.toml
+pics/

PR_COMPRESSION.md

@@ -39,4 +39,4 @@ We use [tiktoken](https://github.com/openai/tiktoken) to tokenize the patches af
 4. If we haven't reached the max token length, add the `deleted files` to the prompt until the prompt reaches the max token length (hard stop), skip the rest of the patches.
 
 ### Example
-![](./pics/git_patch_logic.png)
+![](https://codium.ai/images/git_patch_logic.png)

README.md

@@ -27,25 +27,25 @@ CodiumAI `PR-Agent` is an open-source tool aiming to help developers review PRs
 <h4>Describe:</h4>
 <div align="center">
 <p float="center">
-<img src="./pics/describe.gif" width="800">
+<img src="https://codium.ai/images/describe.gif" width="800">
 </p>
 </div>
 <h4>Review:</h4>
 <div align="center">
 <p float="center">
-<img src="./pics/review.gif" width="800">
+<img src="https://codium.ai/images/review.gif" width="800">
 </p>
 </div>
 <h4>Ask:</h4>
 <div align="center">
 <p float="center">
-<img src="./pics/ask.gif" width="800">
+<img src="https://codium.ai/images/ask.gif" width="800">
 </p>
 </div>
 <h4>Improve:</h4>
 <div align="center">
 <p float="center">
-<img src="./pics/improve.gif" width="800">
+<img src="https://codium.ai/images/improve.gif" width="800">
 </p>
 </div>
 <div align="left">
@@ -64,7 +64,7 @@ CodiumAI `PR-Agent` is an open-source tool aiming to help developers review PRs
 
 Experience GPT-4 powered PR review on your public GitHub repository with our hosted PR-Agent. To try it, just mention `@CodiumAI-Agent` and add the desired command in any PR comment! The agent will generate a response based on your command.
 
-![Review generation process](./pics/demo.gif)
+![Review generation process](https://codium.ai/images/demo.gif)
 
 To set up your own PR-Agent, see the [Quickstart](#Quickstart) section
 
@@ -72,20 +72,20 @@ To set up your own PR-Agent, see the [Quickstart](#Quickstart) section
 ## Overview
 `PR-Agent` offers extensive pull request functionalities across various git providers:
 |       |                                             | GitHub | Gitlab | Bitbucket |
-|-------|---------------------------------------------|--------|--------|-----------|
-| TOOLS | Review                                      | ✓      | ✓      | ✓         |
-|       | ⮑ Inline review                             | ✓     | ✓      |           |
-|       | Ask                                         | ✓      | ✓      |           |
-|       | Auto-Description                            | ✓      |        |           |
-|       | Improve Code                                | ✓      | ✓      |           |
+|-------|---------------------------------------------|:------:|:------:|:---------:|
+| 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:    |           |
+|       | Auto-Description                            |   :white_check_mark:    |        |           |
+|       | Improve Code                                |   :white_check_mark:    |   :white_check_mark:    |           |
 |       |                                             |        |        |           |
-| USAGE | CLI                                         | ✓      | ✓      | ✓         |
-|       | Tagging bot                                 | ✓      | ✓      |           |
-|       | Actions                                     | ✓      |        |           |
+| USAGE | CLI                                         |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:       |
+|       | Tagging bot                                 |   :white_check_mark:    |   :white_check_mark:    |           |
+|       | Actions                                     |   :white_check_mark:    |        |           |
 |       |                                             |        |        |           |
-| CORE  | PR compression                              | ✓      | ✓      | ✓         |
-|       | Repo language prioritization                | ✓      | ✓      | ✓         |
-|       | Adaptive and token-aware<br />file patch fitting | ✓      | ✓      | ✓         |
+| CORE  | PR compression                              |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:       |
+|       | Repo language prioritization                |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:       |
+|       | Adaptive and token-aware<br />file patch fitting |   :white_check_mark:    |   :white_check_mark:    |   :white_check_mark:       |
 
 Examples for invoking the different tools via the [CLI](#quickstart):
 - **Review**:       python cli.py --pr-url=<pr_url>  review
@@ -122,14 +122,14 @@ Here are several ways to install and run PR-Agent:
 
 **PR-Agent** provides four types of interactions ("tools"): `"PR Reviewer"`, `"PR Q&A"`, `"PR Description"` and `"PR Code Sueggestions"`.
 
-- The "PR Reviewer" tool automatically analyzes PRs, and provides different types of feedback.
-- The "PR Ask" tool answers free-text questions about the PR.
+- The "PR Reviewer" tool automatically analyzes PRs, and provides various types of feedback.
+- The "PR Q&A" tool answers free-text questions about the PR.
 - The "PR Description" tool automatically sets the PR Title and body.
 - The "PR Code Suggestion" tool provide inline code suggestions for the PR that can be applied and committed.
 
 ## How it works
 
-![PR-Agent Tools](./pics/pr_agent_overview.png)
+![PR-Agent Tools](https://codium.ai/images/pr_agent_overview.png)
 
 Check out the [PR Compression strategy](./PR_COMPRESSION.md) page for more details on how we convert a code diff to a manageable LLM prompt
 

pr_agent/cli.py

@@ -14,10 +14,10 @@ def run():
     parser = argparse.ArgumentParser(description='AI based pull request analyzer', usage="""\
 Usage: cli.py --pr-url <URL on supported git hosting service> <command> [<args>].
 For example:
-- cli.py --pr-url xxx review
-- cli.py --pr-url xxx describe
-- cli.py --pr-url xxx improve
-- cli.py --pr-url xxx ask "write me a poem about this PR"
+- 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"
 
 Supported commands:
 review / review_pr - Add a review that includes a summary of the PR and specific suggestions for improvement.

pr_agent/git_providers/gitlab_provider.py

@@ -123,26 +123,11 @@ class GitLabProvider(GitProvider):
         range = relevant_lines_end - relevant_lines_start + 1
         body = body.replace('```suggestion', f'```suggestion:-0+{range}')
 
-        d = self.last_diff
-        #
-        # 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}
         lines = target_file.head_file.splitlines()
         relevant_line_in_file = lines[relevant_lines_start - 1]
         edit_type, found, source_line_no, target_file, target_line_no = self.find_in_file(target_file, relevant_line_in_file)
         self.send_inline_comment(body, edit_type, found, relevant_file, relevant_line_in_file, source_line_no,
                                  target_file, target_line_no)
-        # if lines[relevant_lines_start][0] == '-':
-        #     pos_obj['old_line'] = relevant_lines_start
-        # elif lines[relevant_lines_start][0] == '+':
-        #     pos_obj['new_line'] = relevant_lines_start
-        # else:
-        #     pos_obj['new_line'] = relevant_lines_start
-        #     pos_obj['old_line'] = relevant_lines_start
-        # self.mr.discussions.create({'body': body,
-        #                             'position': pos_obj})
 
     def search_line(self, relevant_file, relevant_line_in_file):
         target_file = None

pr_agent/settings/configuration.toml

@@ -12,6 +12,9 @@ num_code_suggestions=3
 inline_code_comments = true
 ask_and_reflect=false
 
+[pr_description]
+publish_description_as_comment=false
+
 [pr_questions]
 
 [pr_code_suggestions]

pr_agent/settings/pr_reviewer_prompts.toml

@@ -2,8 +2,11 @@
 system="""You are CodiumAI-PR-Reviewer, a language model designed to review git pull requests.
 Your task is to provide constructive and concise feedback for the PR, and also provide meaningfull code suggestions to improve the new PR code (the '+' lines).
 - Provide up to {{ num_code_suggestions }} code suggestions.
+{%- if num_code_suggestions > 0 %}
 - Try to focus on important suggestions like fixing code problems, issues and bugs. As a second priority, provide suggestions for meaningfull code improvements, like performance, vulnerability, modularity, and best practices.
+- 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 %}
 
 You must use the following JSON schema to format your answer:
 ```json
@@ -41,6 +44,7 @@ You must use the following JSON schema to format your answer:
       "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. Explain your suggestions."
     },
+{%- if num_code_suggestions > 0 %}
     "Code suggestions": {
       "type": "array",
       "maxItems": {{ num_code_suggestions }},
@@ -60,6 +64,7 @@ You must use the following JSON schema to format your answer:
         }
       }
     },
+{%- endif %}
 {%- if require_security %}
     "Security concerns": {
       "type": "string",
@@ -88,6 +93,7 @@ Example output:
     "PR Feedback":
     {
         "General PR suggestions": "..., `xxx`...",
+{%- if num_code_suggestions > 0 %}
         "Code suggestions": [
             {
                 "relevant file": "directory/xxx.py",
@@ -96,6 +102,7 @@ Example output:
             },
             ...
         ]
+{%- endif %}
 {%- if require_security %},
        "Security concerns": "No, because ..."
 {%- endif %}

pr_agent/tools/pr_code_suggestions.py

@@ -42,7 +42,7 @@ class PRCodeSuggestions:
         assert type(self.git_provider) != BitbucketProvider, "Bitbucket is not supported for now"
 
         logging.info('Generating code suggestions for PR...')
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             self.git_provider.publish_comment("Preparing review...", is_temporary=True)
         logging.info('Getting PR diff...')
 
@@ -56,7 +56,7 @@ class PRCodeSuggestions:
         self.prediction = await self._get_prediction()
         logging.info('Preparing PR review...')
         data = self._prepare_pr_code_suggestions()
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             logging.info('Pushing PR review...')
             self.git_provider.remove_initial_comment()
             logging.info('Pushing inline code comments...')

pr_agent/tools/pr_description.py

@@ -36,17 +36,20 @@ class PRDescription:
 
     async def describe(self):
         logging.info('Generating a PR description...')
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             self.git_provider.publish_comment("Preparing pr description...", is_temporary=True)
         logging.info('Getting PR diff...')
         self.patches_diff = get_pr_diff(self.git_provider, self.token_handler)
         logging.info('Getting AI prediction...')
         self.prediction = await self._get_prediction()
         logging.info('Preparing answer...')
-        pr_title, pr_body = self._prepare_pr_answer()
-        if settings.config.publish_review:
+        pr_title, pr_body, markdown_text = self._prepare_pr_answer()
+        if settings.config.publish_output:
             logging.info('Pushing answer...')
-            self.git_provider.publish_description(pr_title, pr_body)
+            if settings.pr_description.publish_description_as_comment:
+                self.git_provider.publish_comment(markdown_text)
+            else:
+                self.git_provider.publish_description(pr_title, pr_body)
             self.git_provider.remove_initial_comment()
         return ""
 
@@ -66,10 +69,11 @@ class PRDescription:
 
     def _prepare_pr_answer(self):
         data = json.loads(self.prediction)
+        markdown_text = ""
+        for key, value in data.items():
+            markdown_text += f"## {key}\n\n"
+            markdown_text += f"{value}\n\n"
         pr_body = ""
-        # for key, value in data.items():
-        #     markdown_text += f"## {key}\n\n"
-        #     markdown_text += f"{value}\n\n"
         title = data['PR Title']
         del data['PR Title']
         for key, value in data.items():
@@ -80,4 +84,4 @@ class PRDescription:
                 pr_body += f"**{value}**\n\n___\n"
         if settings.config.verbosity_level >= 2:
             logging.info(f"title:\n{title}\n{pr_body}")
-        return title, pr_body
+        return title, pr_body, markdown_text

pr_agent/tools/pr_information_from_user.py

@@ -34,7 +34,7 @@ class PRInformationFromUser:
 
     async def generate_questions(self):
         logging.info('Generating question to the user...')
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             self.git_provider.publish_comment("Preparing answer...", is_temporary=True)
         logging.info('Getting PR diff...')
         self.patches_diff = get_pr_diff(self.git_provider, self.token_handler)
@@ -42,7 +42,7 @@ class PRInformationFromUser:
         self.prediction = await self._get_prediction()
         logging.info('Preparing questions...')
         pr_comment = self._prepare_pr_answer()
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             logging.info('Pushing questions...')
             self.git_provider.publish_comment(pr_comment)
             self.git_provider.remove_initial_comment()

pr_agent/tools/pr_questions.py

@@ -36,7 +36,7 @@ class PRQuestions:
 
     async def answer(self):
         logging.info('Answering a PR question...')
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             self.git_provider.publish_comment("Preparing answer...", is_temporary=True)
         logging.info('Getting PR diff...')
         self.patches_diff = get_pr_diff(self.git_provider, self.token_handler)
@@ -44,7 +44,7 @@ class PRQuestions:
         self.prediction = await self._get_prediction()
         logging.info('Preparing answer...')
         pr_comment = self._prepare_pr_answer()
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             logging.info('Pushing answer...')
             self.git_provider.publish_comment(pr_comment)
             self.git_provider.remove_initial_comment()

pr_agent/tools/pr_reviewer.py

@@ -48,7 +48,7 @@ class PRReviewer:
 
     async def review(self):
         logging.info('Reviewing PR...')
-        if settings.config.publish_review:
+        if settings.config.publish_output:
                 self.git_provider.publish_comment("Preparing review...", is_temporary=True)
         logging.info('Getting PR diff...')
         self.patches_diff = get_pr_diff(self.git_provider, self.token_handler)
@@ -56,7 +56,7 @@ class PRReviewer:
         self.prediction = await self._get_prediction()
         logging.info('Preparing PR review...')
         pr_comment = self._prepare_pr_review()
-        if settings.config.publish_review:
+        if settings.config.publish_output:
             logging.info('Pushing PR review...')
             self.git_provider.publish_comment(pr_comment)
             self.git_provider.remove_initial_comment()
@@ -94,7 +94,9 @@ class PRReviewer:
                 del data['PR Feedback']['Security concerns']
                 data['PR Analysis']['Security concerns'] = val
 
-        if settings.config.git_provider == 'github' and settings.pr_reviewer.inline_code_comments:
+        if settings.config.git_provider == 'github' and \
+                settings.pr_reviewer.inline_code_comments and \
+                'Code suggestions' in data['PR Feedback']:
             del data['PR Feedback']['Code suggestions']
 
         markdown_text = convert_to_markdown(data)
@@ -112,6 +114,9 @@ class PRReviewer:
         return markdown_text
 
     def _publish_inline_code_comments(self):
+        if settings.pr_reviewer.num_code_suggestions == 0:
+            return
+
         review = self.prediction.strip()
         try:
             data = json.loads(review)