in devai-cli/src/devai/commands/review.py [0:0]
def testcoverage(context):
"""
This function performs a test coverage review using the Generative Model API.
Args:
context (str): The code to be reviewed.
"""
source='''
### Context (code) ###:
{}
'''
qry = get_prompt('review_query')
if qry is None:
qry='''
### Instruction ###
You are an experienced software engineer specializing in test coverage analysis and best practices. Given a code snippet (in any programming language) and its associated test suite (if available), your task is to perform a thorough assessment and provide actionable recommendations.
### Output Format ###
Provide your findings in a structured format, listing:
Files/Methods with Test Coverage: Clearly indicate which files and methods within the code snippet have corresponding unit tests.
If possible, specify the names of the test classes and methods that provide coverage.
Files/Methods Lacking Test Coverage: Clearly identify which files and methods within the code snippet do not have associated unit tests.
Prioritize these based on their complexity, criticality, or potential risk of containing bugs.
Overall Test Coverage Summary: Percentage of lines covered. Percentage of branches/conditions covered (if applicable to the language). Any notable coverage gaps at a high level.
Detailed Coverage Breakdown:
Files/Functions/Methods with Test Coverage:
File name and function/method name.
Corresponding test file and function/method name (if available).
Coverage type (e.g., line, branch, condition, etc.).
Files/Functions/Methods Lacking Test Coverage:
File name and function/method name.
Reason for prioritizing (complexity, criticality, risk).
Specific test scenarios to consider.
Recommendations for Improvement:
Prioritized list of functions/methods/areas where new unit tests should be added.
Guidance on test types to use (e.g., positive, negative, edge case, etc.).
Tips on improving existing tests (if applicable).
Additional Insights (Optional):
Suggestions for refactoring code to make it more testable.
Identification of potential code smells or areas prone to errors.
Language-specific best practices for testing (if applicable).
If no unit tests are present, clearly state this and emphasize the importance of adding them.
If coverage is already comprehensive, acknowledge this and suggest ways to maintain or enhance the test suite.
Tailor recommendations to the specific codebase, its context, and the programming language used.
Use clear, concise language and avoid technical jargon where possible.
Include code examples relevant to the programming language to illustrate recommendations.
### Example Dialogue ###
User: (Provides Java code snippet and test suite)
AI: (Provides output following the structured format, including coverage metrics, detailed breakdown, prioritized recommendations, and additional insights, with examples relevant to Java)
Key Changes
Language Agnostic: Prompt is now open to any programming language, with the user specifying the language upfront.
Flexible Output: Coverage metrics and test types are adjusted to be applicable to various languages (e.g., conditions instead of branches for languages that don't have explicit branching).
Language-Specific Insights: Encourages the AI to offer best practices or insights specific to the language being analyzed.
'''
# Load files as text into source variable
source=source.format(format_files_as_string(context))
code_chat_model = GenerativeModel(MODEL_NAME)
with telemetry.tool_context_manager(USER_AGENT):
code_chat = code_chat_model.start_chat(response_validation=False)
code_chat.send_message(qry)
response = code_chat.send_message(source)
click.echo(f"{response.text}")