Update app.py

app.py

@@ -2,16 +2,23 @@ import gradio as gr
 import google.generativeai as genai
 import paramiko
 import io
+import re
+
+dependency_file_patterns = "|".join([
+    "package\.json", "requirements\.txt", "Gemfile", "pom\.xml", "build\.gradle",
+    "composer\.json", "Cargo\.toml", "go\.mod", "project\.clj", "mix\.exs",
+    "build\.sbt", "Package\.swift", ".*\.csproj", "packages\.config", "yarn\.lock",
+    "package-lock\.json", "Pipfile", "poetry\.lock", "environment\.yml",
+    "pubspec\.yaml", "deps\.edn", "rebar\.config", "bower\.json", "Podfile", "Cartfile"
+])
 
 def fetch_github_file(github_url, ssh_private_key):
     try:
-        # Parse the GitHub URL
-        parts = github_url.split('/')
-        owner = parts[3]
-        repo = parts[4]
-        branch = parts[6]
-        file_path = '/'.join(parts[7:])
-
+        # Parse the GitHub SSH URL
+        parts = github_url.split(':')
+        owner_repo = parts[1].split('.git')[0]
+        owner, repo = owner_repo.split('/')
+        
         # Set up SSH client
         ssh = paramiko.SSHClient()
         ssh.set_missing_host_key_policy(paramiko.AutoAddPolicy())
@@ -23,21 +30,28 @@ def fetch_github_file(github_url, ssh_private_key):
         ssh.connect('github.com', username='git', pkey=private_key)
 
         # Construct the git command to fetch the file content
-        git_command = f"git archive --[email protected]:{owner}/{repo}.git {branch} {file_path} | tar -xO"
+        git_command = f"git ls-tree -r --name-only HEAD | grep -E '({dependency_file_patterns})'"
 
         # Execute the command
         stdin, stdout, stderr = ssh.exec_command(git_command)
         
-        # Read the file content
-        file_content = stdout.read().decode('utf-8')
+        # Read the file list
+        file_list = stdout.read().decode('utf-8').splitlines()
+
+        file_contents = {}
+        for file_path in file_list:
+            cat_command = f"git show HEAD:{file_path}"
+            stdin, stdout, stderr = ssh.exec_command(cat_command)
+            file_contents[file_path] = stdout.read().decode('utf-8')
 
         # Close the SSH connection
         ssh.close()
 
-        if not file_content:
-            return "Error: File not found or empty."
+        if not file_contents:
+            return "Error: No dependency files found."
+
+        return file_contents
 
-        return file_content
     except paramiko.AuthenticationException:
         return "Error: Authentication failed. Please check your SSH key."
     except paramiko.SSHException as ssh_err:
@@ -45,17 +59,19 @@ def fetch_github_file(github_url, ssh_private_key):
     except Exception as e:
         return f"Error accessing GitHub: {str(e)}"
 
-def process_with_gemini(file_content, gemini_api_key):
+def process_with_gemini(file_contents, gemini_api_key):
     genai.configure(api_key=gemini_api_key)
-    model = genai.GenerativeModel('gemini-2.5-pro-preview-03-25')
-    
+    model = genai.GenerativeModel('gemini-pro')
+
+    combined_content = "\n\n".join([f"File: {name}\n{content}" for name, content in file_contents.items()])
+
     prompt = f"""
-    Analyze the following file content for open-source license information:
+    Analyze the following dependency files:
 
-    {file_content}
+    {combined_content}
 
     Please provide:
-    1. A list of dependencies and their versions
+    1. A list of dependencies and their versions for each file
     2. The licenses associated with each dependency (if available)
     3. A summary of the project based on these dependencies
     4. Any potential license conflicts or considerations
@@ -70,14 +86,14 @@ def process_input(file, github_url, ssh_private_key, gemini_api_key):
         return "Error: Please either upload a file OR provide a GitHub URL, not both."
     
     if file is not None:
-        file_content = file.decode('utf-8')
+        file_content = {file.name: file.read().decode('utf-8')}
     elif github_url and ssh_private_key:
-        if not github_url.startswith("https://github.com/"):
-            return "Error: Invalid GitHub URL. Please use the format: https://github.com/username/repository/blob/branch/path/to/file"
+        if not re.match(r'^git@github\.com:.+/.+\.git$', github_url):
+            return "Error: Invalid GitHub SSH URL. Please use the format: git@github.com:username/repository.git"
         if not ssh_private_key.strip():
             return "Error: SSH Private Key is empty. Please provide a valid key."
         file_content = fetch_github_file(github_url, ssh_private_key)
-        if file_content.startswith("Error:"):
+        if isinstance(file_content, str) and file_content.startswith("Error:"):
             return file_content
     else:
         return "Error: Please either upload a file OR provide both GitHub URL and SSH Private Key."
@@ -93,13 +109,13 @@ iface = gr.Interface(
     fn=process_input,
     inputs=[
         gr.File(label="Upload dependency file (e.g., requirements.txt, package.json, Gemfile)"),
-        gr.Textbox(label="GitHub File URL (optional)"),
+        gr.Textbox(label="GitHub Repository URL (optional, use SSH format)"),
         gr.Textbox(label="SSH Private Key (required if using GitHub URL)", type="password"),
         gr.Textbox(label="Gemini API Key", type="password"),
     ],
     outputs=gr.Textbox(label="License Information and Analysis"),
     title="Open Source License Extractor",
-    description="Upload a dependency file OR provide a GitHub file URL to extract and analyze open-source license information.",
+    description="Upload a dependency file or provide a GitHub repository URL to analyze open-source licenses.",
 )
 
 if __name__ == "__main__":