initial Commit

Dockerfile

@@ -0,0 +1,22 @@
+# Use an official Ubuntu base image
+FROM kalilinux/kali-rolling
+
+WORKDIR /code
+
+COPY . /code
+# Update the package lists
+RUN apt-get update
+
+# Install system packages
+RUN apt-get install -y grep git python3 python3-pip coreutils curl
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# Copy the rest of your application files
+
+RUN curl -O https://gpt4all.io/models/ggml-gpt4all-j-v1.3-groovy.bin
+
+EXPOSE 7860
+
+CMD ["streamlit", "run", "app.py","--server.port", "7860"]
+

app.py

@@ -0,0 +1,54 @@
+import streamlit as st
+from resource import *
+from proper_main import *
+from resource import llm_chain
+import time
+
+def main():
+    while True:
+        st.title("Github Automated Repo Analysis")
+
+        # User input
+        user_url = st.text_input("Enter the Github URL")
+        
+        option = st.radio("What you want me to do",["Python Analysis","GPT Evaluation"])
+
+        # Generate response
+        if st.button("Submit"):
+
+            st.text("Please wait Automation is Processing")
+            strttime=time.time()
+            repos, status = web_scrape(user_url,st)
+            
+            #task_progress = st.progress(0)
+            #task_progress.progress("Tools is taking action please wait")
+            if status == 0:
+                repo_path = data_cloning(repos,st)
+                data_cleaning(repo_path,st)
+                query,report_analysis = analyse(st)
+                if len(query) == 0:
+                    st.write("The given User's URL doesnt Contain Python Repository")
+                    break
+                if option == "Python Analysis":
+                    repo_name,score=self_analysis(report_analysis)
+                    output="The Complex Repo is "+ str(repo_name)+" Because the Complexity Score is "+str(score)
+                    #st.write("The Complex Repo is",repo_name," Because the Complexity Score is",score)
+                    st.text_area("Bot Response:", value=output, height=100)
+                    time.sleep(15)
+                    break
+                    
+                elif option == "GPT Evaluation":
+                    response_gpt = llm_chain([str(query)])
+                    # Display the response
+                    st.text_area("Bot Response:", value=response_gpt['text'], height=100)
+                    elapsed_time = time.time() - strttime
+                    st.text(f"Execution time: {elapsed_time:.2f} seconds")
+            else:
+                output = st.empty()
+                output.error(f"Error occurred. Please contact the admin {repos}.")
+                time.sleep(5)
+                break
+
+if __name__ == "__main__":
+    main()
+

proper_main.py

@@ -0,0 +1,141 @@
+try:
+    import requests
+    import os
+    import subprocess as sp
+    from bs4 import BeautifulSoup
+    from nbconvert import PythonExporter
+    import shutil
+
+except Exception as e:
+    print("Some modules are missing:", e)
+    print("Do you want to install them via this Python program?")
+    option = input("Y or N: ")
+    if option.lower() not in ["y", "n"]:
+        exit()
+    elif option.lower() == "n":
+        exit()
+    elif option.lower() == "y":
+        print("Make sure your internet connection is active; otherwise, it may throw an error. Press 'N' to exit.")
+        curr_dir = os.getcwd()
+        os.system("pip install -r " + curr_dir + "/requirements.txt")
+
+
+
+
+
+def web_scrape(user_url,st):
+    username=user_url[19:]
+    if username.endswith("/"):
+    	username=username[:-1]
+    print(username)
+    base_url = f"https://api.github.com/users/{username}/repos"
+    
+    response = requests.get(base_url)
+    if response.status_code != 200:
+        return ("Please provide a valid link.",1)
+    st.text("Extracting the Repos")
+    repos = []
+    repositories = response.json()
+    for repo in repositories:
+    	
+    	repo_name = repo["name"]
+    	
+    	repos.append("https://github.com/"+username + "/" +repo_name)
+    return repos,0
+
+
+def data_cloning(repos,st):
+    
+    if os.path.isdir("/tmp/repos"):
+    	shutil.rmtree("/tmp/repos")
+    
+    os.mkdir("/tmp/repos")
+    
+    
+    st.text("Cloning the Repos")
+    os.chdir("/tmp/repos")
+    for i in repos:
+        sp.run(["git", "clone", i], stdout=sp.DEVNULL, stderr=sp.DEVNULL)
+
+    return os.getcwd()
+
+
+def data_cleaning(directory,st):
+    exporter = PythonExporter()
+    st.text("Cleaning the Repos")
+    
+    if len(os.listdir(os.getcwd())) ==0:
+    	st.text("Not a Valid Repo")
+    	return
+
+    for root, dirs, files in os.walk(directory, topdown=False):
+        for filename in files:
+            file_path = os.path.join(root, filename)
+
+            #if filename.endswith(".ipynb"):
+                #output, _ = exporter.from_filename(file_path)
+                #with open(os.path.join(root, filename[:-6] + ".py"), "w") as script_file:
+                #    script_file.write(output)
+                #os.remove(file_path)
+
+            if not (filename.endswith(".py") or filename.endswith(".ipynb")):
+                os.remove(file_path)
+
+        for dir_name in dirs:
+            dir_path = os.path.join(root, dir_name)
+            if not os.listdir(dir_path):
+                os.rmdir(dir_path)
+
+
+def analyse(st):
+    project_and_grades = {}
+    report_analysis = {}
+    st.text("Analysing...")
+    if len(os.listdir(os.getcwd())) ==0:
+    	st.text("Not a Valid Repo")
+    	return
+
+    for file in os.listdir(os.getcwd()):
+        print(file)
+        path = os.getcwd() + "/" + file
+        
+        cmd = ["radon", "cc", "--total-average","--include-ipynb", file]
+        res = sp.check_output(cmd)
+        index = res.decode().find("Average")
+        if index <= 0:
+            grade = "A"
+            score = 0.5
+        else:
+            grade = res.decode()[index:]
+            score = grade[23:-1]
+            score = score[:3]
+            grade=grade[20]
+            
+
+        project_and_grades["Repo " + file] = " Score " + str(score)
+        report_analysis["Repo " + file] = [float(score)]
+    shutil.rmtree('/tmp/repos')
+
+    return project_and_grades,report_analysis
+    
+    
+    
+def self_analysis(report_analysis):
+	score= max(report_analysis.values())
+	for keyss in report_analysis:
+		if report_analysis[keyss]==score:
+			repo = keyss
+	return repo,score
+
+
+
+"""def main():
+	web_scrape()
+	curr_path=data_cloning()
+	data_cleaning(curr_path)
+	report=analyse()
+	print(report)
+	
+if __name__ == main():
+	main()
+"""

requirements.txt

@@ -0,0 +1,9 @@
+streamlit
+gpt4all
+langchain
+huggingface
+huggingface_hub
+radon
+requests
+bs4
+nbconvert

resource.py

@@ -0,0 +1,40 @@
+from langchain import PromptTemplate, LLMChain
+from huggingface_hub import hf_hub_download
+from langchain.llms import GPT4All
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
+import os
+
+
+template ="""
+You are a friendly chatbot assistant you are going to evaluate a Github Repo i analysed all the repos and made a report
+i'm gonna give a report you just refer the below table
+
+Table:
+
+score                 Risk
+1 - 5                 Low - Simple block
+6 - 10                Low - Well structured and stable block
+11 - 20               Moderate - Slightly complex block
+21 - 30               More than moderate - More complex block
+31 - 40               High - Complex block, alarming
+41+                   Very high - Error-prone, unstable block
+
+based on the score and you have to return only one most complex repo which has more score from report 
+when the score is high thats the complex repo and
+if all the repos are at the same score choose randomly one
+
+Report: {question}
+
+Answer:"""
+
+
+prompt = PromptTemplate(template=template, input_variables=["question"])
+
+#hf_hub_download(repo_id="dnato/ggml-gpt4all-j-v1.3-groovy.bin", filename="ggml-gpt4all-j-v1.3-groovy.bin", local_dir="/code")
+local_path= os.getcwd() + "/ggml-gpt4all-j-v1.3-groovy.bin"
+llm = GPT4All(model=local_path,callbacks=[StreamingStdOutCallbackHandler()] )
+llm_chain = LLMChain(prompt=prompt, llm=llm)
+
+
+
+