errors commitw

app.py

@@ -1,25 +1,53 @@
+import streamlit as st
 from resource import *
 from proper_main import *
-import streamlit as st
-import subprocess as sp
-
+from resource import llm_chain
+import time
 
 def main():
-    st.title("GPT4All Chatbot")
+    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"):
-        web_scrape(user_url)
-        curr_path = data_cloning()
-        query = analyse()
-        llm_chain=langu()
-        #response_gpt = llm_chain([str(query)])
-        # Display the response
-        #st.text_area("Bot Response:", value=response_gpt, height=100)
-        st.text_area("Bot Response:", value=sp.check_output(['ls','-la','/tmp']), height=100)
+
+        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")
+            	st.experimental_rerun()
+            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)
+            	st.experimental_rerun()
+            	
+            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("Error occurred. Please contact the admin {e}.")
+            time.sleep(5)
+            st.experimental_rerun()
 
 if __name__ == "__main__":
     main()
+

proper_main.py

@@ -20,40 +20,39 @@ except Exception as e:
         os.system("pip install -r " + curr_dir + "/requirements.txt")
 
 
-repos = []
 
 
-def web_scrape(user_url):
-    base_url = "https://www.github.com"
-    user_url = user_url
-
-    if user_url.endswith("/"):
-        user_url = user_url[:-1]
-
-    try:
-        response = requests.get(user_url + "?tab=repositories")
-    except Exception as e:
-        print("Please provide a valid link:", e)
-        web_scrape()
 
+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:
-        print("Please provide a valid link.")
-        web_scrape()
-
-    make_soup = BeautifulSoup(response.text, 'html.parser')
-    li = make_soup.findAll('div', class_='d-inline-block mb-1')
-    if len(li) == 0:
-    	print("Please Provide the Valid Link")
-    	web_scrape()
-
-    for _, i in enumerate(li):
-        for a in i.findAll('a'):
-            new_url = base_url + a['href']
-            repos.append(new_url)
-
-
-def data_cloning():
+        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(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)
@@ -61,18 +60,23 @@ def data_cloning():
     return os.getcwd()
 
 
-def data_cleaning(directory):
+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")
+    	st.experimental_rerun()
 
     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 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)
@@ -83,19 +87,24 @@ def data_cleaning(directory):
                 os.rmdir(dir_path)
 
 
-def analyse():
+def analyse(st):
     project_and_grades = {}
+    report_analysis = {}
+    st.text("Analysing...")
+    if len(os.listdir(os.getcwd())) ==0:
+    	st.text("Not a Valid Repo")
+    	st.experimental_rerun()
 
     for file in os.listdir(os.getcwd()):
         print(file)
         path = os.getcwd() + "/" + file
         
-        cmd = ["radon", "cc", "--total-average", 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 = 1
+            score = 0.5
         else:
             grade = res.decode()[index:]
             score = grade[23:-1]
@@ -104,9 +113,19 @@ def analyse():
             
 
         project_and_grades["Repo " + file] = "Grade " + grade + " Score " + str(score)
+        report_analysis["Repo " + file] = [float(score)]
     shutil.rmtree('/tmp/repos')
 
-    return project_and_grades
+    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
 
 
 

resource.py

@@ -19,24 +19,21 @@ score       Rank      Risk
 31 - 40      E        High - Complex block, alarming
 41+          F        Very high - Error-prone, unstable block
 
-based on the score and rank you have to return only one most complex repo from report 
-
-if the report Rank has similar scores and rank act accordingly
+based on the score and rank you have to return only one most complex repo which has more score from report 
+if all the repos are at the same rank and score choose randomly one
 
 Report: {question}
 
 Answer:"""
 
-def langu():
 
-	import os
-	prompt = PromptTemplate(template=template, input_variables=["question"])
+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="/tmp")
+local_path= os.getcwd() + "/ggml-gpt4all-j-v1.3-groovy.bin"
+llm = GPT4All(model_name="ggml-gpt4all-j-v1.3-groovy",model_path="/tmp",callbacks=[StreamingStdOutCallbackHandler()] )
+llm_chain = LLMChain(prompt=prompt, llm=llm)
 
-	hf_hub_download(repo_id="dnato/ggml-gpt4all-j-v1.3-groovy.bin", filename="ggml-gpt4all-j-v1.3-groovy.bin", local_dir="/tmp")
-	local_path= "/home/user/.cache/huggingface/hub/models--dnato--ggml-gpt4all-j-v1.3-groovy.bin/blobs/3b2fd7cca97284467de0be8f638925f40cbff4c70a2e10ba1094f83bfa24d86b"
-	llm = GPT4All(model=local_path,callbacks=[StreamingStdOutCallbackHandler()] )
-	llm_chain = LLMChain(prompt=prompt, llm=llm)
-	return llm_chain