ScriptDemon/GitHub_Summarize
2
1import streamlit as st2import google.generativeai as genai3import os4import requests5import docx6import PyPDF27 8# Configure Generative AI9genai.configure(api_key="AIzaSyDOQi9d3dxzNuH9RXGdtizXeQT7MD_cdSA")10model = genai.GenerativeModel("gemini-1.5-flash-8b")11 12# Helper Functions13def api_get(url, headers={'Accept': 'application/vnd.github.v3+json'}):14 response = requests.get(url, headers=headers)15 response.raise_for_status()16 return response.json()17 18def fetch_readme_from_github(repo_url):19 try:20 readme_data = api_get(repo_url.replace("https://github.com/", "https://api.github.com/repos/") + "/readme")21 return requests.get(readme_data['download_url']).text22 except Exception as e:23 st.error(f"Error fetching README: {e}")24 return None25 26def read_docx(file):27 return "\n".join([para.text for para in docx.Document(file).paragraphs])28 29def read_pdf(file):30 return "\n".join([page.extract_text() for page in PyPDF2.PdfReader(file).pages])31 32def fetch_repo_contents(repo_url):33 try:34 return api_get(repo_url.replace("https://github.com/", "https://api.github.com/repos/") + "/contents")35 except Exception as e:36 st.error(f"Error fetching repository contents: {e}")37 return None38 39def traverse_files(contents, file_list):40 """41 Recursively traverse through the repository and collect all files.42 """43 for item in contents:44 if item['type'] == 'dir':45 sub_dir_contents = api_get(item['url'])46 traverse_files(sub_dir_contents, file_list)47 elif item['type'] == 'file':48 file_list.append(item)49 50def fetch_additional_repo_data(repo_url):51 """Fetch additional metadata for repository comparison."""52 try:53 metadata = api_get(repo_url.replace("https://github.com/", "https://api.github.com/repos/"))54 contributors = api_get(metadata['contributors_url']) if 'contributors_url' in metadata else []55 contents = fetch_repo_contents(repo_url)56 languages = api_get(metadata['languages_url']) if 'languages_url' in metadata else {}57 return {58 "metadata": metadata,59 "contributors": contributors,60 "contents": contents,61 "languages": languages,62 }63 except Exception as e:64 st.error(f"Error fetching additional repo data: {e}")65 return None66 67def compare_repositories(repo1_data, repo2_data):68 """Compare repositories on various parameters."""69 comparison = {}70 try:71 comparison["Stars"] = (repo1_data["metadata"]["stargazers_count"], repo2_data["metadata"]["stargazers_count"])72 comparison["Forks"] = (repo1_data["metadata"]["forks_count"], repo2_data["metadata"]["forks_count"])73 comparison["Contributors"] = (len(repo1_data["contributors"]), len(repo2_data["contributors"]))74 comparison["Languages"] = list(set(repo1_data["languages"].keys()).intersection(repo2_data["languages"].keys()))75 comparison["Last Updated"] = (repo1_data["metadata"]["updated_at"], repo2_data["metadata"]["updated_at"])76 repo1_files = {item['name'] for item in repo1_data["contents"] if item['type'] == 'file'}77 repo2_files = {item['name'] for item in repo2_data["contents"] if item['type'] == 'file'}78 comparison["Common Files"] = list(repo1_files.intersection(repo2_files))79 except Exception as e:80 st.error(f"Error comparing repositories: {e}")81 return comparison82 83def suggest_preferred_repo(comparison):84 """Suggest the preferred repository based on comparison."""85 reasons = []86 suggestion = "Repo 1" if comparison["Stars"][0] >= comparison["Stars"][1] else "Repo 2"87 if comparison["Stars"][0] > comparison["Stars"][1]:88 reasons.append("Repo 1 has more stars and is likely more popular.")89 elif comparison["Stars"][1] > comparison["Stars"][0]:90 reasons.append("Repo 2 has more stars and is likely more popular.")91 if comparison["Forks"][0] > comparison["Forks"][1]:92 reasons.append("Repo 1 has more forks, indicating active usage.")93 elif comparison["Forks"][1] > comparison["Forks"][0]:94 reasons.append("Repo 2 has more forks, indicating active usage.")95 if comparison["Contributors"][0] > comparison["Contributors"][1]:96 reasons.append("Repo 1 has more contributors, indicating stronger community support.")97 elif comparison["Contributors"][1] > comparison["Contributors"][0]:98 reasons.append("Repo 2 has more contributors, indicating stronger community support.")99 return suggestion, reasons100 101def detect_language(file_extension):102 file_extensions_map = {103 ".py": "Python", ".java": "Java", ".js": "JavaScript", ".cpp": "C++", ".cs": "C#",104 ".rb": "Ruby", ".php": "PHP", ".go": "Go", ".ts": "TypeScript", ".swift": "Swift",105 ".kt": "Kotlin", ".rs": "Rust", ".dart": "Dart", ".scala": "Scala", ".r": "R",106 ".pl": "Perl", ".sh": "Shell", ".html": "HTML", ".css": "CSS", ".json": "JSON",107 ".ipynb": "Jupyter Notebook"108 }109 return file_extensions_map.get(file_extension, "Unknown")110 111def categorize_file(file_name, file_extension):112 """Categorize files based on their names or extensions."""113 # Mapping file extensions to categories114 frontend_extensions = ['.html', '.css', '.js', '.jsx', '.ts', '.tsx', '.vue']115 backend_extensions = ['.py', '.java', '.go', '.js']116 model_extensions = ['.h5', '.pkl', '.pt', '.model']117 documentation_extensions = ['.md', '.txt', '.rst']118 configuration_extensions = ['.json', '.yml', '.env', '.ini', '.xml']119 misc_extensions = ['.sh', '.bash', '.sql', '.csv', '.log', '.xml']120 121 # Match the file extension to categories122 if file_extension in frontend_extensions:123 return "Frontend"124 elif file_extension in backend_extensions:125 return "Backend"126 elif file_extension in model_extensions:127 return "Model"128 elif file_extension in documentation_extensions:129 return "Documentation"130 elif file_extension in configuration_extensions:131 return "Configuration"132 elif file_extension in misc_extensions:133 return "Miscellaneous"134 else:135 return "Miscellaneous"136 137def display_file_structure(contents, level=0):138 """Recursively display the file structure of the repository, categorized."""139 if not contents:140 return141 for item in contents:142 indentation = " " * level143 file_name = item['name']144 145 # Extract the file extension or handle files without extensions146 file_extension = os.path.splitext(file_name)[1].lower() if '.' in file_name else ''147 148 # Categorize the file149 category = categorize_file(file_name, file_extension)150 151 # Debugging log for file categorization152 st.write(f"{indentation}File: {file_name} - Category: {category}")153 154 if item['type'] == 'dir':155 st.write(f"{indentation}📁 {file_name} (Directory) - Category: {category}")156 sub_contents = api_get(item['url'])157 display_file_structure(sub_contents, level + 1)158 elif item['type'] == 'file':159 st.write(f"{indentation}📄 {file_name} (File) - Category: {category}")160 161 162def explain_code(file_data, explanation_type, display_code):163 """Explain the code using AI."""164 code_content = requests.get(file_data["download_url"]).text165 file_extension = os.path.splitext(file_data["name"])[1]166 language = detect_language(file_extension)167 prompt = (168 f"Give a {explanation_type.lower()} explanation of the following {language} code:\n\n{code_content}"169 )170 if display_code:171 st.subheader(f"Code from {file_data['name']}")172 st.code(code_content, language=language.lower())173 st.subheader(f"{explanation_type} Explanation")174 st.write(model.generate_content(prompt).text)175 176def refactor_code(file_data):177 """Provide refactoring suggestions for the code."""178 code_content = requests.get(file_data["download_url"]).text179 prompt = f"Provide refactoring suggestions for the following code:\n\n{code_content}"180 st.subheader(f"Refactoring Suggestions for {file_data['name']}")181 st.write(model.generate_content(prompt).text)182 183# Main Application184st.set_page_config(page_title="GitSummarize", page_icon="📄", layout="centered")185st.title("GitSummarize")186 187option = st.sidebar.radio("Choose an option:", [188 "Summarize File",189 "Summarize GitHub README",190 "Explain GitHub Code",191 "Explain File Structure",192 "Compare GitHub Repos"193])194 195if option == "Summarize File":196 uploaded_file = st.file_uploader("Choose a file", type=["txt", "md", "docx", "pdf"])197 if uploaded_file:198 st.write("File content read. Summarizing...")199 custom_prompt = st.text_area("Enter your custom prompt for summarizing the file (optional):")200 word_count = st.number_input("Enter the number of words for the summary:", min_value=10, max_value=1000, value=100)201 if st.button("Confirm Summarization"):202 file_extension = os.path.splitext(uploaded_file.name)[1]203 file_content = {204 ".txt": lambda: uploaded_file.read().decode("utf-8"),205 ".md": lambda: uploaded_file.read().decode("utf-8"),206 ".docx": lambda: read_docx(uploaded_file),207 ".pdf": lambda: read_pdf(uploaded_file),208 }.get(file_extension, lambda: None)()209 if file_content:210 prompt = f"{custom_prompt or 'Summarize the following content'} in {word_count} words:\n\n{file_content}"211 st.write("Summary", model.generate_content(prompt).text)212 213elif option == "Summarize GitHub README":214 repo_url = st.text_input("Enter the GitHub repository URL")215 if st.button("Confirm Fetch and Summarize") and repo_url:216 readme_content = fetch_readme_from_github(repo_url)217 if readme_content:218 prompt = f"Summarize this README in detail with skills required:\n\n{readme_content}"219 st.write("Detailed Summary", model.generate_content(prompt).text)220 221elif option == "Explain GitHub Code":222 repo_url = st.text_input("Enter the GitHub repository URL:")223 explanation_type = st.radio("Select the type of explanation:", ["Gist", "Detailed", "Line by Line"])224 display_code = st.checkbox("Display the code before explanation", value=True)225 226 if "all_files" not in st.session_state:227 st.session_state.all_files = []228 229 if repo_url and st.button("Fetch All Repository Files"):230 contents = fetch_repo_contents(repo_url)231 if contents:232 file_list = []233 traverse_files(contents, file_list)234 st.session_state.all_files = file_list235 st.success(f"Found {len(file_list)} files in the repository.")236 237 if st.session_state.all_files:238 file_names = [file["name"] for file in st.session_state.all_files]239 selected_file = st.selectbox("Select a file to explain or refactor:", file_names)240 241 if st.button("Explain Selected File"):242 selected_file_data = next(243 (file for file in st.session_state.all_files if file["name"] == selected_file), None244 )245 if selected_file_data:246 explain_code(selected_file_data, explanation_type, display_code)247 248 if st.button("Refactor Selected File"):249 selected_file_data = next(250 (file for file in st.session_state.all_files if file["name"] == selected_file), None251 )252 if selected_file_data:253 refactor_code(selected_file_data)254 255elif option == "Explain File Structure":256 repo_url = st.text_input("Enter the GitHub repository URL")257 if st.button("Confirm Fetch and Explain Structure") and repo_url:258 explanation = fetch_repo_contents(repo_url)259 if explanation:260 st.write("Repository Structure Explanation")261 display_file_structure(explanation)262 263elif option == "Compare GitHub Repos":264 repo1_url = st.text_input("Enter the first GitHub repository URL")265 repo2_url = st.text_input("Enter the second GitHub repository URL")266 if repo1_url and repo2_url:267 repo1_data = fetch_additional_repo_data(repo1_url)268 repo2_data = fetch_additional_repo_data(repo2_url)269 270 if repo1_data and repo2_data:271 comparison = compare_repositories(repo1_data, repo2_data)272 st.subheader("Repository Comparison")273 st.json(comparison)274 preferred_repo, reasons = suggest_preferred_repo(comparison)275 st.subheader(f"Preferred Repository: {preferred_repo}")276 for reason in reasons:277 st.write(reason)278