app.py

import gradio as gr
from pyvis.network import Network
import networkx as nx
import numpy as np
import pandas as pd
import os
from datasets import load_dataset
from datasets import Features
from datasets import Value
from datasets import Dataset
import matplotlib.pyplot as plt

import re

pattern = r'"(.*?)"'
Secret_token = os.getenv('HF_token')

dataset = load_dataset('FDSRashid/hadith_info',data_files = 'Basic_Edge_Information.csv', token = Secret_token, split = 'train')

edge_info = dataset.to_pandas()

features = Features({'Rawi ID': Value('int32'), 'Famous Name': Value('string'), 'Narrator Rank': Value('string'), 'Number of Narrations': Value('string'),  'Generation': Value('string')})
narrator_bios = load_dataset("FDSRashid/hadith_info", data_files = 'Teacher_Bios.csv', token = Secret_token,features=features )
narrator_bios = narrator_bios['train'].to_pandas()
narrator_bios.loc[49845, 'Narrator Rank'] = 'رسول الله'
narrator_bios.loc[49845, 'Number of Narrations'] = 0
narrator_bios['Number of Narrations'] = narrator_bios['Number of Narrations'].astype(int)
narrator_bios.loc[49845, 'Number of Narrations'] = 443471
narrator_bios['Generation'] = narrator_bios['Generation'].replace([None], [-1])
narrator_bios['Generation'] = narrator_bios['Generation'].astype(int)


features = Features({'matn': Value('string'), 'taraf_ID': Value('string'), 'bookid_hadithid': Value('string')})

dataset = load_dataset("FDSRashid/hadith_info", data_files = 'All_Matns.csv',token = Secret_token, features = features)
matn_info = dataset['train'].to_pandas()
matn_info = matn_info.drop(97550)
matn_info = matn_info.drop(307206)
matn_info['taraf_ID'] = matn_info['taraf_ID'].replace('KeyAbsent', -1)


matn_info['taraf_ID'] = matn_info['taraf_ID'].astype(int)
# matn_info = matn_info.sort_values('taraf_ID')
# tarafs = matn_info['taraf_ID'].unique()
# for i, taraf in enumerate(tarafs):
#     matn_info.loc[matn_info['taraf_ID'] == taraf, 'taraf_ID_New'] = i + 1 # Replace 'a' with 'e' in column 'C' where the condition is met
# matn_info['taraf_ID_New'] = matn_info['taraf_ID_New'].astype(int)



isnad_info = load_dataset('FDSRashid/hadith_info',token = Secret_token, data_files = 'isnad_info.csv', split = 'train').to_pandas()
isnad_info['Hadiths Cleaned'] = isnad_info['Hadiths'].apply(lambda x: [re.findall(pattern, string)[0].split("_") for string in x[1:-1].split(',')])

taraf_max = np.max(matn_info['taraf_ID'].unique())

cmap = plt.colormaps['cool']

books = load_dataset('FDSRashid/Hadith_info', data_files='Books.csv', token = Secret_token)['train'].to_pandas()

matn_info['Book'] = matn_info['bookid_hadithid'].apply(lambda x: books[books['Book_ID'] == int(x.split('_')[0])]['Book_Name'].to_list()[0])
matn_info['Author'] = matn_info['bookid_hadithid'].apply(lambda x: books[books['Book_ID'] == int(x.split('_')[0])]['Author'].to_list()[0])
matn_info['Hadith Number'] = matn_info['bookid_hadithid'].apply(lambda x: x.split('_')[1])

def value_to_hex(value):
    rgba_color = cmap(value)
    return "#{:02X}{:02X}{:02X}".format(int(rgba_color[0] * 255), int(rgba_color[1] * 255), int(rgba_color[2] * 255))

#edge_info, matn_info, narrator_bios, isnad_info

def visualize_isnad(taraf_num, yaxis):
    taraf_hadith = matn_info[matn_info['taraf_ID'] == taraf_num]['bookid_hadithid'].to_list()
    taraf_matns = matn_info[matn_info['taraf_ID'] == taraf_num]['matn'].to_list()
    taraf_hadith_split = [i.split('_') for i in taraf_hadith]
    taraf_book = matn_info[matn_info['taraf_ID'] == taraf_num]['Book'].to_list()
    taraf_author = matn_info[matn_info['taraf_ID'] == taraf_num]['Author'].to_list()
    taraf_hadith_number = taraf_book = matn_info[matn_info['taraf_ID'] == taraf_num]['Hadith Number'].to_list()
    lst_hadith = []
    for i in range(len(taraf_hadith_split)):
        isnad_in_hadith1 = isnad_info['Hadiths Cleaned'].apply(lambda x: taraf_hadith_split[i] in x )
        isnad_hadith1 = isnad_info[isnad_in_hadith1][['Source', 'Destination']]
        G = nx.from_pandas_edgelist(isnad_hadith1, source = 'Source', target = 'Destination', create_using = nx.DiGraph())
        node = [int(n) for n, d in G.out_degree() if d == 0][0]
        gen_node = narrator_bios[narrator_bios['Rawi ID']==node]['Generation'].iloc[0]
        name_node = narrator_bios[narrator_bios['Rawi ID']==node]['Famous Name'].iloc[0]
        lst_hadith.append([taraf_matns[i], gen_node, name_node, taraf_book[i], taraf_author[i], taraf_hadith_number[i]])
    df = pd.DataFrame(lst_hadith, columns = ['Matn', 'Generation', 'Name'])
    
    hadith_cleaned = isnad_info['Hadiths Cleaned'].apply(lambda x: any(i in x for i in taraf_hadith_split) )
    isnad_hadith = isnad_info[hadith_cleaned][['Source', 'Destination']]
    narrators = isnad_hadith.applymap(lambda x: narrator_bios[narrator_bios['Rawi ID'] == int(x)]['Famous Name'].to_list()).rename(columns={"Source": "Teacher", "Destination": "Student"})
    isnad_hadith["Student"] = narrators['Student']
    isnad_hadith["Teacher"] = narrators['Teacher']
    filtered = isnad_hadith[(isnad_hadith['Teacher'].apply(lambda x: len(x)) == 1) & (isnad_hadith['Student'].apply(lambda x: len(x)) == 1)]
    filtered['Student'] = filtered['Student'].apply(lambda x: x[0])
    filtered['Teacher'] = filtered['Teacher'].apply(lambda x: x[0])
    net = Network(directed =True)
    for _, row in filtered.iterrows():
        source = row['Teacher']
        target = row['Student']
        teacher_info = narrator_bios[narrator_bios['Rawi ID'] == int(row['Source'])]
        student_info = narrator_bios[narrator_bios['Rawi ID'] == int(row['Destination'])]
        isnad = isnad_info[(isnad_info['Source'] == row['Source']) & (isnad_info['Destination'] == row['Destination'])]
        teacher_narrations = teacher_info['Number of Narrations'].to_list()[0]
        student_narrations = student_info['Number of Narrations'].to_list()[0]
        if row['Source'] == '99999':
            net.add_node(source, font = {'size':50, 'color': 'Black'}, color = '#000000')
        else:
            net.add_node(source, font = {'size':30, 'color': 'red'}, color = value_to_hex(teacher_narrations), label = f'{source} \n {teacher_info["Narrator Rank"].to_list()[0]}')
        net.add_node(target, font = {'size': 30, 'color': 'red'}, color = value_to_hex(student_narrations), label = f'{target} \n{student_info["Narrator Rank"].to_list()[0]}')
        net.add_edge(source, target, color = value_to_hex(int(isnad['Hadith Count'].to_list()[0])), label = f"{isnad['Hadith Count'].to_list()[0]}")
    net.barnes_hut(gravity=-5000, central_gravity=0.3, spring_length=200)
    html = net.generate_html()
    html = html.replace("'", "\"")
    return f"""<iframe style="width: 100%; height: 600px;margin:0 auto" name="result" allow="midi; geolocation; microphone; camera; 
  display-capture; encrypted-media;" sandbox="allow-modals allow-forms 
  allow-scripts allow-same-origin allow-popups 
  allow-top-navigation-by-user-activation allow-downloads" allowfullscreen="" 
  allowpaymentrequest="" frameborder="0" srcdoc='{html}'></iframe>""" , df


with gr.Blocks() as demo:
  Yaxis = gr.Dropdown(choices = ['Tarafs', 'Hadiths', 'Isnads', 'Books'], value = 'Tarafs', label = 'Variable to Display', info = 'Choose the variable to visualize.')  
  taraf_number = gr.Slider(1,taraf_max , value=10000, label="Taraf", info="Choose the Taraf to Input", step = 1)
  btn = gr.Button('Submit')
  btn.click(fn = visualize_isnad, inputs = [taraf_number, Yaxis], outputs = [gr.HTML(), gr.DataFrame()])
  demo.launch()