app.py

import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
import os
import  tempfile
import pandas as pd
########################################################################################
def find_most_similar_paragraph(input_sentence, text):
    # Split the text into paragraphs
    paragraphs = text.split("\n\n")  # Assuming paragraphs are separated by at least one empty line

    # Initialize variables to keep track of similarity
    max_similarity = 0
    most_similar_paragraph = ""

    # Compare input sentence with each paragraph
    for paragraph in paragraphs:
        # Preprocess the paragraph (convert to lowercase and remove extra spaces)
        cleaned_paragraph = paragraph.lower().strip()

        # Calculate similarity (you can use other similarity metrics as well)
        similarity = len(set(input_sentence.lower().split()) & set(cleaned_paragraph.split())) / len(set(input_sentence.lower().split()))

        # Update max similarity if needed
        if similarity > max_similarity:
            max_similarity = similarity
            most_similar_paragraph = paragraph
    
    return most_similar_paragraph
#######################################################################################
def copy_file_content(src_file, dest_file):
    try:
        shutil.copyfile(src_file, dest_file)
        print(f"محتوای فایل {src_file} با موفقیت به {dest_file} کپی شد.")
    except FileNotFoundError:
        print(f"فایل {src_file} یافت نشد.")
    except Exception as e:
        print(f"خطا در کپی فایل: {e}")
########################################################################################

def show_path():
    uploaded_file = st.file_uploader("فایل را آپلود کنید", type=["csv", "txt"])
    if uploaded_file:
        # ایجاد یک پوشه موقت
        temp_dir = tempfile.mkdtemp()
        
        # ایجاد مسیر فایل آپلود شده در پوشه موقت
        path = os.path.join(temp_dir,uploaded_file.name)
        
        # ذخیره محتوای فایل آپلود شده در مسیر
        with open(path, "wb") as f:
            f.write(uploaded_file.getvalue())
        
        # نمایش مسیر فایل آپلود شده
        #st.success(f"مسیر فایل آپلود شده: {path}")
    return path

#########################################################    
    
def get_pdf_text(pdf_docs):
    text = ""
    for pdf in pdf_docs:
        pdf_reader = PdfReader(pdf)
        for page in pdf_reader.pages:
            text += page.extract_text()
    return text
#################################################################
def wrap_text_preserve_newlines(text, width=300):
    # Split the input text into lines based on newline characters
    lines = text.split('\n')
    # Wrap each line individually
    wrapped_lines = [textwrap.fill(line, width=width) for line in lines]
    # Join the wrapped lines back together using newline characters
    wrapped_text = '\n'.join(wrapped_lines)
    return wrapped_text
########################################################################
# Text Splitter
def text_split(ducuments):
    text_splitter = CharacterTextSplitter(chunk_size=300, chunk_overlap=20)
    docs = text_splitter.split_documents(documents)
    return docs
######################################################################
# Embeddings
def embeddings_f():
    embeddings = HuggingFaceEmbeddings()
    return embeddings
##########################################################################
#upload pdf    
def pdf_upload():
    pdf_docs=st.file_uploader("Upload your PDF Files",accept_multiple_files=True)
    return pdf_docs
##############################################################################
def show(text):
    text_show=st.write(text)
    return text_show
###########################################################################
def upload_txt_file():
  # ایجاد یک ورودی برای آپلود فایل txt
  uploaded_file = st.file_uploader("لطفا یک فایل txt را انتخاب کنید", type="txt")
  # بررسی اینکه آیا فایلی آپلود شده است یا خیر
  if uploaded_file is not None:
    # باز کردن فایل با انکد UTF-8 و قرار دادن محتوای آن در یک متغیر به نام text
    with open(uploaded_file) as f:
      text = f.read()
    # برگرداندن متغیر text
    return text
  else:
    # برگرداندن یک پیام خطا اگر فایلی آپلود نشده باشد
    return "لطفا یک فایل txt را آپلود کنید"
###############################################################################
def find_most_similar_paragraph(input_sentence):
    try:
        # Open the text file for reading
        with open("d1.txt", 'r', encoding='utf-8') as file:
            text = file.read()

        # Split the text into paragraphs
        paragraphs = text.split("\n\n")  # Assuming paragraphs are separated by at least one empty line

        # Initialize variables to keep track of similarity
        max_similarity = 0
        most_similar_paragraph = ""

        # Compare input sentence with each paragraph
        for paragraph in paragraphs:
            # Preprocess the paragraph (convert to lowercase and remove extra spaces)
            cleaned_paragraph = paragraph.lower().strip()

            # Calculate similarity (you can use other similarity metrics as well)
            similarity = len(set(input_sentence.lower().split()) & set(cleaned_paragraph.split())) / len(set(input_sentence.lower().split()))

            # Update max similarity if needed
            if similarity > max_similarity:
                max_similarity = similarity
                most_similar_paragraph = paragraph

        return most_similar_paragraph
    except FileNotFoundError:
        return "فایل مورد نظر یافت نشد."




# Call the function
########################################################################################################


def main():
    st.set_page_config(page_title="Multiple pdf chat", page_icon=":books:")
    #chunk_size=st.text_input("enter a chunk size:")
    #chunk_overlap=st.text_input("enter a chunk overlap:")
    
    if st.button("build model"):
        with st.spinner("waiting"):
            import app2
            
            
            st.write("compelete build model") 

    query=st.text_input("enter a question:")
    if st.button("answer your question"):
      with st.spinner("waiting"):
          from app2 import db
          from app2 import load_model
          from app2 import find_help
          out=load_model(query,db)
          st.write(out)
          #st.write("-------------------------------------------------")
          st.write(query)
          st.write("************************************************")
          with open('d1.txt', 'w', encoding='utf-8') as file:
            file.write("")
            file.write(out)
          result = find_most_similar_paragraph(query)
          #st.write("************************************************")
          st.write(result)
          #find_help(out)
        ############### 
    with st.sidebar:
        uploaded_file = st.file_uploader("فایل را آپلود کنید", type=["csv", "txt"])
        #st.subheader("your document")
        #source_path=show_path()
        #if st.button("copy process"):
            #with st.spinner("processing"):
                #app1.copy_file(source_path)
    
      
        
                   
# Press the green button in the gutter to run the script.
if __name__ == '__main__':
    main()