Merge pull request #1429 from Codium-ai/tr/sub_bullet_points

fix: improve PR description and title field descriptions for clarity

pr_agent/algo/pr_processing.py

@@ -205,10 +205,11 @@ def pr_generate_extended_diff(pr_languages: list,
             if not extended_patch:
                 get_logger().warning(f"Failed to extend patch for file: {file.filename}")
                 continue
-            full_extended_patch = f"\n\n## {file.filename}\n{extended_patch.rstrip()}\n"
 
             if add_line_numbers_to_hunks:
                 full_extended_patch = convert_to_hunks_with_lines_numbers(extended_patch, file)
+            else:
+                full_extended_patch = f"\n\n## File: '{file.filename.strip()}'\n{extended_patch.rstrip()}\n"
 
             # add AI-summary metadata to the patch
             if file.ai_file_summary and  get_settings().get("config.enable_ai_metadata", False):

pr_agent/settings/pr_description_prompts.toml

@@ -52,8 +52,8 @@ class PRDescription(BaseModel):
 {%- if enable_semantic_files_types %}
     pr_files: List[FileDescription] = Field(max_items=20, description="a list of all the files that were changed in the PR, and summary of their changes. Each file must be analyzed regardless of change size.")
 {%- endif %}
-    description: str = Field(description="an informative and concise description of the PR. Use bullet points. Display first the most significant changes.")
-    title: str = Field(description="an informative title for the PR, describing its main theme")
+    description: str = Field(description="summarize the PR changes in up to four bullet points, each up to 8 words. Add sub-bullets if needed. Order bullets by importance, with each bullet highlighting a key change group.")
+    title: str = Field(description="a concise and descriptive title that captures the PR's main theme")
 =====
 
 

pr_agent/tools/pr_description.py

@@ -556,6 +556,11 @@ def _prepare_pr_answer(self) -> Tuple[str, str, str, List[dict]]:
             elif 'pr_files' in key.lower() and get_settings().pr_description.enable_semantic_files_types:
                 changes_walkthrough, pr_file_changes = self.process_pr_files_prediction(changes_walkthrough, value)
                 changes_walkthrough = f"{PRDescriptionHeader.CHANGES_WALKTHROUGH.value}\n{changes_walkthrough}"
+            elif key.lower().strip() == 'description':
+                if isinstance(value, list):
+                    value = ', '.join(v.rstrip() for v in value)
+                value = value.replace('\n-', '\n\n-').strip() # makes the bullet points more readable by adding double space
+                pr_body += f"{value}\n"
             else:
                 # if the value is a list, join its items by comma
                 if isinstance(value, list):

tests/unittest/test_extend_patch.py

@@ -145,14 +145,15 @@ def test_extend_patches_with_extra_lines(self, token_handler, pr_languages):
         # Check that with no extra lines, the patches are the same as the original patches
         p0 = patches_extended_no_extra_lines[0].strip()
         p1 = patches_extended_no_extra_lines[1].strip()
-        assert p0 == '## file1\n' + pr_languages[0]['files'][0].patch.strip()
-        assert p1 == '## file2\n' + pr_languages[0]['files'][1].patch.strip()
+        assert p0 == "## File: 'file1'\n" + pr_languages[0]['files'][0].patch.strip()
+        assert p1 == "## File: 'file2'\n" + pr_languages[0]['files'][1].patch.strip()
 
         patches_extended_with_extra_lines, total_tokens, patches_extended_tokens = pr_generate_extended_diff(
             pr_languages, token_handler, add_line_numbers_to_hunks=False,
             patch_extra_lines_before=2,
             patch_extra_lines_after=1
         )
 
+
         p0_extended = patches_extended_with_extra_lines[0].strip()
-        assert p0_extended == '## file1\n\n@@ -3,8 +3,8 @@ \n line0\n line1\n-original content\n+modified content\n line2\n line3\n line4\n line5\n line6'
+        assert p0_extended == "## File: 'file1'\n\n@@ -3,8 +3,8 @@ \n line0\n line1\n-original content\n+modified content\n line2\n line3\n line4\n line5\n line6"