Standard_Intelligence_Dev / scrape_3gpp.py
MaksG's picture
Update scrape_3gpp.py
f26ef33 verified
raw
history blame
24.4 kB
import os
import requests
from bs4 import BeautifulSoup
from urllib.parse import urljoin
import pandas as pd
import numpy as np
import zipfile
import textract
import gradio as gr
import shutil
def browse_folder(url):
if url.lower().endswith(('docs', 'docs/')):
return gr.update(choices=[])
response = requests.get(url)
response.raise_for_status() # This will raise an exception if there's an error
soup = BeautifulSoup(response.text, 'html.parser')
excel_links = [a['href'] + '/' for a in soup.find_all('a', href=True) if a['href'].startswith(url)]
return gr.update(choices=excel_links)
def extract_statuses(url):
# Send a GET request to the webpage
response = requests.get(url)
# Parse the webpage content
soup = BeautifulSoup(response.content, 'html.parser')
# Find all links in the webpage
links = soup.find_all('a')
# Identify and download the Excel file
for link in links:
href = link.get('href')
if href and (href.endswith('.xls') or href.endswith('.xlsx')):
excel_url = href if href.startswith('http') else url + href
excel_response = requests.get(excel_url)
file_name = 'guide_status.xlsx' #excel_url.split('/')[-1]
# Save the file
with open(file_name, 'wb') as f:
f.write(excel_response.content)
# Read the Excel file
df = pd.read_excel(file_name)
# Check if 'TDoc Status' column exists and extract unique statuses
if 'TDoc Status' in df.columns:
unique_statuses = df['TDoc Status'].unique().tolist()
print(f'Downloaded {file_name} and extracted statuses: {unique_statuses}')
if 'withdrawn' in unique_statuses:
unique_statuses.remove('withdrawn')
return gr.update(choices=unique_statuses, value=unique_statuses)
else:
print(f"'TDoc Status' column not found in {file_name}")
return []
def scrape(url, excel_file, folder_name, status_list, progress=gr.Progress()):
filenames = []
status_filenames = []
# Check if the excel_file argument is provided and if the file exists.
excel_file_path = "guide_status.xlsx" # Hardcoded path to the Excel file
if os.path.exists(excel_file_path):
try:
df = pd.read_excel(excel_file_path)
print(f"Initial DataFrame size: {len(df)}")
if 'TDoc Status' in df.columns:
df = df[df['TDoc Status'].isin(status_list)]
print(f"Filtered DataFrame size: {len(df)}")
if df.empty:
print("No files match the specified 'TDoc Status'.")
else:
if 'TDoc' in df.columns and not df['TDoc'].isnull().all():
status_filenames = [f"{url}{row['TDoc']}.zip" for index, row in df.iterrows()]
elif 'URL' in df.columns and not df['URL'].isnull().all():
status_filenames = df['URL'].tolist()
else:
print("No valid 'File' or 'URL' entries found for the filtered statuses.")
print(f"Filenames: {status_filenames}")
else:
print("'TDoc Status' column not found in the Excel file.")
except Exception as e:
print(f"Error reading Excel file: {e}")
if excel_file and os.path.exists(excel_file):
try:
df = pd.read_excel(excel_file)
# If 'Actions' in df.columns and filter based on it, and construct URLs from 'TDoc' or 'URL' columns
if 'Actions' in df.columns:
df = df[df['Actions'] == 'x']
elif 'File' in df.columns:
filenames = [f"{url}{row['File']}.zip" for index, row in df.iterrows()]
elif 'URL' in df.columns:
filenames = df['URL'].tolist()
except Exception as e:
print(f"Error reading Excel file: {e}")
# Optionally, handle the error or return a message if needed
# If no Excel file is provided or found, or if it lacks 'TDoc'/'URL', the function can still continue with predefined URLs or other logic
download_directory = folder_name
if not os.path.exists(download_directory):
os.makedirs(download_directory)
pourcentss = 0.05
print(f'filenames: {status_filenames}')
if not filenames and not status_filenames:
print("No Excel file provided, or no valid URLs found in the file.")
# You can either return here or continue with other predefined logic
response = requests.get(url)
# Analyser le contenu HTML de la page
soup = BeautifulSoup(response.content, "html.parser")
# Trouver tous les balises <a> avec des attributs href (liens)
links = soup.find_all("a", href=True)
# Filtrer les liens se terminant par ".zip"
zip_links = [link['href'] for link in links if link['href'].endswith('.zip')]
# Télécharger chaque fichier zip
for zip_link in zip_links:
progress(pourcentss,desc='Downloading')
pourcentss+=0.4/len(df)
# Construire l'URL absolue du fichier zip
absolute_url = urljoin(url, zip_link)
# Extraire le nom de fichier de l'URL
filename = os.path.basename(absolute_url)
# Chemin où le fichier sera enregistré
save_path = os.path.join(download_directory, filename)
# Envoyer une requête GET pour télécharger le fichier
with requests.get(absolute_url, stream=True) as r:
r.raise_for_status()
with open(save_path, 'wb') as f:
for chunk in r.iter_content(chunk_size=8192):
f.write(chunk)
elif not filenames:
# Proceed with downloading files using the filenames list
for file_url in status_filenames:
filename = os.path.basename(file_url)
save_path = os.path.join(download_directory, filename)
progress(pourcentss,desc='Downloading')
pourcentss+=0.4/len(df)
try:
with requests.get(file_url, stream=True) as r:
r.raise_for_status()
with open(save_path, 'wb') as f:
for chunk in r.iter_content(chunk_size=8192):
f.write(chunk)
except requests.exceptions.HTTPError as e:
print(f"skipped file: {file_url}: {e}")
else:
# Proceed with downloading files using the filenames list
for file_url in filenames:
filename = os.path.basename(file_url)
save_path = os.path.join(download_directory, filename)
try:
with requests.get(file_url, stream=True) as r:
r.raise_for_status()
with open(save_path, 'wb') as f:
for chunk in r.iter_content(chunk_size=8192):
f.write(chunk)
except requests.exceptions.HTTPError as e:
print(f"HTTP error occurred: {file_url}: {e}")
return False, "Il n'y a pas de colonne action ou alors celle ci n'est pas bien écrite, format attendu: 'Actions'"
return True, len(df)
def extractZip(url):
# Répertoire où les fichiers zip sont déjà téléchargés
nom_extract = url.split("/")[-3] + "_extraction"
if os.path.exists(nom_extract):
shutil.rmtree(nom_extract)
extract_directory = nom_extract
download_directory = url.split("/")[-3] + "_downloads"
# Répertoire où le contenu des fichiers zip sera extrait
# Extraire le contenu de tous les fichiers zip dans le répertoire de téléchargement
for zip_file in os.listdir(download_directory):
zip_path = os.path.join(download_directory, zip_file)
# Vérifier si le fichier est un fichier zip
if zip_file.endswith(".zip"):
extract_dir = os.path.join(extract_directory, os.path.splitext(zip_file)[0]) # Supprimer l'extension .zip
# Vérifier si le fichier zip existe
if os.path.exists(zip_path):
# Créer un répertoire pour extraire le contenu s'il n'existe pas
if not os.path.exists(extract_dir):
os.makedirs(extract_dir)
# Extraire le contenu du fichier zip
with zipfile.ZipFile(zip_path, 'r') as zip_ref:
zip_ref.extractall(extract_dir)
print(f"Extraction terminée pour {zip_file}")
else:
print(f"Fichier zip {zip_file} introuvable")
print("Toutes les extractions sont terminées !")
def excel3gpp(url):
response = requests.get(url)
response.raise_for_status() # This will raise an exception if there's an error
# Use BeautifulSoup to parse the HTML content
soup = BeautifulSoup(response.text, 'html.parser')
# Look for Excel file links; assuming they have .xlsx or .xls extensions
excel_links = [a['href'] for a in soup.find_all('a', href=True) if a['href'].endswith(('.xlsx', '.xls'))]
# Download the first Excel file found (if any)
if excel_links:
excel_url = excel_links[0] # Assuming you want the first Excel file
if not excel_url.startswith('http'):
excel_url = os.path.join(url, excel_url) # Handle relative URLs
# Download the Excel file
excel_response = requests.get(excel_url)
excel_response.raise_for_status()
# Define the path where you want to save the file
# Replace 'path_to_save_directory' with your desired path
# Write the content of the Excel file to a local file
# Write the content of the Excel file to a local file named 'guide.xlsx'
nom_guide = 'guide.xlsx' # Directly specify the filename
if os.path.exists(nom_guide):
os.remove(nom_guide)
filepath = nom_guide
with open(filepath, 'wb') as f:
f.write(excel_response.content)
print(f'Excel file downloaded and saved as: {filepath}')
def replace_line_breaks(text):
return text.replace("\n", "/n")
def remod_text(text):
return text.replace("/n", "\n")
def update_excel(data, excel_file, url):
new_df_columns = ["URL", "File", "Type", "Title", "Source", "Status", "Content"]
temp_df = pd.DataFrame(data, columns=new_df_columns)
try:
# Check if the Excel file already exists and append data to it
if os.path.exists(excel_file):
old_df = pd.read_excel(excel_file)
df = pd.concat([old_df, temp_df], axis=0, ignore_index=True)
else:
df = temp_df
# Save the updated data back to the Excel file
df.to_excel(excel_file, index=False)
except Exception as e:
print(f"Error updating Excel file: {e}")
def extractionPrincipale(url, excel_file=None, status_list=None, progress=gr.Progress()):
nom_download = url.split("/")[-3] + "_downloads"
if os.path.exists(nom_download):
shutil.rmtree(nom_download)
folder_name = nom_download
nom_status = url.split("/")[-3] + "_status.xlsx"
if os.path.exists(nom_status):
os.remove(nom_status)
temp_excel = nom_status
progress(0.0,desc='Downloading')
result, count = scrape(url, excel_file, folder_name, status_list)
if result:
print("Success")
else:
return(None)
progress(0.4,desc='Extraction')
extractZip(url)
progress(0.5,desc='Extraction 2')
excel3gpp(url)
progress(0.6,desc='Creating Excel File')
extract_directory = url.split("/")[-3] + "_extraction"
categories = {
"Other": ["URL", "File", "Type", "Title", "Source", "Content"],
"CR": ["URL", "File", "Type", "Title", "Source", "Content"],
"pCR":["URL", "File", "Type", "Title", "Source", "Content"],
"LS": ["URL", "File", "Type", "Title", "Source", "Content"],
"WID": ["URL", "File", "Type", "Title", "Source", "Content"],
"SID": ["URL", "File", "Type", "Title", "Source", "Content"],
"DISCUSSION": ["URL", "File", "Type", "Title", "Source", "Content"],
"pdf": ["URL", "File", "Type", "Title", "Source", "Content"],
"ppt": ["URL", "File", "Type", "Title", "Source", "Content"],
"pptx": ["URL", "File", "Type", "Title", "Source", "Content"]
}
pourcents2=0.6
data = []
errors_count = 0
processed_count = 0 # Counter for processed files
pre_title_section = None
try:
df = pd.read_excel(temp_excel)
except Exception as e:
print(f"Initializing a new DataFrame because: {e}")
df = pd.DataFrame(columns=["URL", "File", "Type", "Title", "Source", "Status", "Content"])
for folder in os.listdir(extract_directory):
folder_path = os.path.join(extract_directory, folder)
if os.path.isdir(folder_path):
for file in os.listdir(folder_path):
progress(min(pourcents2,0.99),desc='Creating Excel File')
pourcents2+=0.4/count
if file == "__MACOSX":
continue
file_path = os.path.join(folder_path, file)
if file.endswith((".pptx", ".ppt", ".pdf", ".docx", ".doc", ".DOCX")):
try:
text = textract.process(file_path).decode('utf-8')
except Exception as e:
print(f"Error processing {file_path}: {e}")
errors_count += 1
continue
cleaned_text_lines = text.split('\n')
cleaned_text = '\n'.join([line.strip('|').strip() for line in cleaned_text_lines if line.strip()])
title = ""
debut = ""
sections = cleaned_text.split("Title:")
if len(sections) > 1:
pre_title_section = sections[0].strip().split()
title = sections[1].strip().split("\n")[0].strip()
debut = sections[0].strip()
category = "Other"
if file.endswith(".pdf"):
category = "pdf"
elif file.endswith((".ppt", ".pptx")):
category = "ppt" # assuming all ppt and pptx files go into the same category
elif "CHANGE REQUEST" in debut:
category = "CR"
elif "Discussion" in title:
category = "DISCUSSION"
elif "WID" in title:
category = "WID"
elif "SID" in title:
category = "SID"
elif "LS" in title:
category = "LS"
elif pre_title_section and pre_title_section[-1] == 'pCR':
category = "pCR"
elif "Pseudo-CR" in title:
category = "pCR"
contenu = "" # This will hold the concatenated content for 'Contenu' column
if category in categories:
columns = categories[category]
extracted_content = []
if category == "CR":
reason_for_change = ""
summary_of_change = ""
if len(sections) > 1:
reason_for_change = sections[1].split("Reason for change", 1)[-1].split("Summary of change")[0].strip()
summary_of_change = sections[1].split("Summary of change", 1)[-1].split("Consequences if not")[0].strip()
extracted_content.append(f"Reason for change: {reason_for_change}")
extracted_content.append(f"Summary of change: {summary_of_change}")
elif category == "pCR":
if len(sections) > 1:# Handle 'pCR' category-specific content extraction
pcr_specific_content = sections[1].split("Introduction", 1)[-1].split("First Change")[0].strip()
extracted_content.append(f"Introduction: {pcr_specific_content}")
elif category == "LS":
overall_review = ""
if len(sections) > 1:
overall_review = sections[1].split("Overall description", 1)[-1].strip()
extracted_content.append(f"Overall review: {overall_review}")
elif category in ["WID", "SID"]:
objective = ""
start_index = cleaned_text.find("Objective")
end_index = cleaned_text.find("Expected Output and Time scale")
if start_index != -1 and end_index != -1:
objective = cleaned_text[start_index + len("Objective"):end_index].strip()
extracted_content.append(f"Objective: {objective}")
elif category == "DISCUSSION":
Discussion = ""
extracted_text = replace_line_breaks(cleaned_text)
start_index_doc_for = extracted_text.find("Document for:")
if start_index_doc_for != -1:
start_index_word_after_doc_for = start_index_doc_for + len("Document for:")
end_index_word_after_doc_for = start_index_word_after_doc_for + extracted_text[start_index_word_after_doc_for:].find("/n")
word_after_doc_for = extracted_text[start_index_word_after_doc_for:end_index_word_after_doc_for].strip()
result_intro = ''
result_conclusion = ''
result_info = ''
if word_after_doc_for.lower() == "discussion":
start_index_intro = extracted_text.find("Introduction")
end_index_intro = extracted_text.find("Discussion", start_index_intro)
intro_text = ""
if start_index_intro != -1 and end_index_intro != -1:
intro_text = extracted_text[start_index_intro + len("Introduction"):end_index_intro].strip()
result_intro = remod_text(intro_text) # Convert back line breaks
else:
result_intro = "Introduction section not found."
# Attempt to find "Conclusion"
start_index_conclusion = extracted_text.find("Conclusion", end_index_intro)
end_index_conclusion = extracted_text.find("Proposal", start_index_conclusion if start_index_conclusion != -1 else end_index_intro)
conclusion_text = ""
if start_index_conclusion != -1 and end_index_conclusion != -1:
conclusion_text = extracted_text[start_index_conclusion + len("Conclusion"):end_index_conclusion].strip()
result_conclusion = remod_text(conclusion_text)
elif start_index_conclusion == -1: # Conclusion not found, look for Proposal directly
start_index_proposal = extracted_text.find("Proposal", end_index_intro)
if start_index_proposal != -1:
end_index_proposal = len(extracted_text) # Assuming "Proposal" section goes till the end if present
proposal_text = extracted_text[start_index_proposal + len("Proposal"):end_index_proposal].strip()
result_conclusion = remod_text(proposal_text) # Using "Proposal" content as "Conclusion"
else:
result_conclusion = "Conclusion/Proposal section not found."
else:
# Handle case where "Conclusion" exists but no "Proposal" to mark its end
conclusion_text = extracted_text[start_index_conclusion + len("Conclusion"):].strip()
result_conclusion = remod_text(conclusion_text)
Discussion=f"Introduction: {result_intro}\nConclusion/Proposal: {result_conclusion}"
elif word_after_doc_for.lower() == "information":
start_index_info = extracted_text.find(word_after_doc_for)
if start_index_info != -1:
info_to_end = extracted_text[start_index_info + len("Information"):].strip()
result_info = remod_text(info_to_end)
Discussion = f"Discussion:{result_info}"
else:
Discussion = "The word after 'Document for:' is not 'Discussion', 'DISCUSSION', 'Information', or 'INFORMATION'."
else:
Discussion = "The phrase 'Document for:' was not found."
# Since DISCUSSION category handling requires more specific processing, adapt as necessary
# Here's a simplified example
discussion_details = Discussion
extracted_content.append(discussion_details)
# Add more categories as needed
contenu = "\n".join(extracted_content)
# Assuming 'source' needs to be filled from the guide.xlsx mapping
# Placeholder for source value calculation
source = "" # Update this with actual source determination logic
status = ""
data.append([url+ "/" + folder + '.zip', folder , category, title, source,status, contenu])
# After processing all files and directories
# Read the guide.xlsx file into a DataFrame to map 'TDoc' to 'Source'
guide_df = pd.read_excel('guide.xlsx', usecols=['Source', 'TDoc','TDoc Status'])
tdoc_source_map = {row['TDoc']: row['Source'] for index, row in guide_df.iterrows()}
tdoc_status_map = {row['TDoc']: row['TDoc Status'] for index, row in guide_df.iterrows()}
# Update the 'Source' in your data based on matching 'Nom du fichier' with 'TDoc'
for item in data:
nom_du_fichier = item[1] # Assuming 'Nom du fichier' is the first item in your data list
if nom_du_fichier in tdoc_source_map:
item[4] = tdoc_source_map[nom_du_fichier] # Update the 'Source' field, assuming it's the fourth item
item[5] = tdoc_status_map[nom_du_fichier]
processed_count += 1
# Check if it's time to update the Excel file
if processed_count % 20 == 0:
update_excel(data, temp_excel, url)
print(f"Updated after processing {processed_count} files.")
data = [] # Clear the data list after updating
if data:
# This final call ensures that any remaining data is processed and saved.
update_excel(data, temp_excel, url)
print(f"Final update after processing all files.")
file_name = temp_excel
# Save the updated DataFrame to Excel
return file_name