Fix bug for PR Content code diff calculation logic

Author: Jet Xu

CHANGELOG.md

@@ -5,6 +5,11 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.1.8] - 2024-11-03
+
+### Optimized
+- fix bugs for get pr content file diff calculate logic
+
 ## [0.1.7] - 2024-10-31
 
 ### Optimized

llama_github/data_retrieval/github_entities.py

@@ -123,7 +123,7 @@ def get_repo(self):
                         return None
         self.update_last_read_time()
         return self._repo
-
+    
     def get_structure(self, path="/") -> dict:
         """
         Retrieves the structure of the repository using a singleton design pattern.
@@ -183,6 +183,8 @@ def get_file_content(self, file_path: str, sha: Optional[str] = None) -> Optiona
                             decoded_content = base64.b64decode(file_content.content).decode('utf-8')
                         elif (file_content.encoding is None or file_content.encoding == 'none') and hasattr(file_content, 'download_url') and file_content.download_url:
                             try:
+                                logger.debug(f"Downloading file {file_path} from {file_content.download_url}")
+                                # Use requests to download the file content
                                 response = requests.get(
                                     file_content.download_url,
                                     timeout=30,
@@ -526,6 +528,7 @@ def get_pr_content(self, number, pr=None, context_lines=10) -> Dict[str, Any]:
                             pr_data["commit_stats"] = {}
 
                         # Process file changes
+                        comparison = None
                         dependency_files = ['requirements.txt', 'Pipfile', 'Pipfile.lock', 'setup.py']
                         config_files = ['.env', 'settings.py', 'config.yaml', 'config.yml', 'config.json']
                         for file in pr_files:
@@ -552,6 +555,16 @@ def get_pr_content(self, number, pr=None, context_lines=10) -> Dict[str, Any]:
                                 custom_diff = ''
                             else:
                                 custom_diff = DiffGenerator.generate_custom_diff(base_content, head_content, context_lines)
+                                if not custom_diff or custom_diff.strip() == '':
+                                    try:
+                                        #use compare API if custom diff is empty
+                                        if not comparison:
+                                            comparison = self.repo.compare(pr.base.sha, pr.head.sha)
+                                        patches = [f.patch for f in comparison.files if f.filename == file_path]
+                                        custom_diff = patches[0] if patches else None
+                                    except Exception as e:
+                                        logger.exception(f"Error fetching file diff for PR #{number}")
+                                        custom_diff = ''
 
                             # Categorize code changes
                             change_categories = CodeAnalyzer.categorize_change(custom_diff)

llama_github/version.py

@@ -1 +1 @@
-__version__ = '0.1.7'
+__version__ = '0.1.8'

setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = llama-github
-version = 0.1.7
+version = 0.1.8
 author = Jet Xu
 author_email = [email protected]
 description = Llama-github is an open-source Python library that empowers LLM Chatbots, AI Agents, and Auto-dev Agents to conduct Retrieval from actively selected GitHub public projects. It Augments through LLMs and Generates context for any coding question, in order to streamline the development of sophisticated AI-driven applications.