|
import os |
|
import re |
|
import shutil |
|
from functools import partial |
|
from multiprocessing import Pool |
|
from urllib.parse import urlparse |
|
|
|
import jinja2 |
|
import progressbar |
|
import requests |
|
from InquirerPy import prompt |
|
from playwright.sync_api import sync_playwright |
|
from pypdf import PdfMerger |
|
|
|
TEMP_FOLDER = 'temp' |
|
PRINT_TEMPLATE = """<html><head><meta charset="UTF-8"><style>{{custom_css}}</style><style>{{base_css}}</style><style>{{page_css}}</style></head><body><a name="{{page}}"></a><div class="viewer-page"><div class="page-{{page}} pf w0 h0">{{content}}</div></div></body></html>""" |
|
|
|
|
|
def create_folder_if_not_exists(folder: str) -> None: |
|
"""Create folder if not exists""" |
|
if not os.path.exists(folder): |
|
os.makedirs(folder) |
|
|
|
|
|
def sanitize(input_string: str) -> str: |
|
""" |
|
Sanitize the input string by replacing '/' with '_' |
|
and removing any unwanted characters. |
|
|
|
Args: |
|
input_string (str): The string to be sanitized. |
|
|
|
Returns: |
|
str: The sanitized string. |
|
""" |
|
# Replace '/' with '_' |
|
sanitized_string = input_string.replace('/', '_') |
|
|
|
# Remove any characters that are not alphanumeric or underscores |
|
sanitized_string = re.sub(r'[^\w_]', '', sanitized_string) |
|
|
|
return sanitized_string |
|
|
|
|
|
def get_domain(url: str) -> str: |
|
"""Extracts the base domain from a URL.""" |
|
parsed_url = urlparse(url) |
|
base_domain = f"{parsed_url.scheme}://{parsed_url.netloc}" |
|
return base_domain |
|
|
|
|
|
def get_manual_url() -> str: |
|
"""Prompt input for Manual PDF url""" |
|
url_question = [{ |
|
'type': 'input', |
|
'name': 'url', |
|
'message': 'Enter Manual PDF url:', |
|
}] |
|
url_answer = prompt(url_question) |
|
|
|
return url_answer.get('url').split('#')[0].split('?')[0] |
|
|
|
|
|
def get_data(url: str) -> dict: |
|
"""Process url and return a dictionary with the data""" |
|
html = requests.get(url).text |
|
file_id = re.search(r'viewer/([\d/]+)/1/bg1', html).group(1) |
|
pages = re.search(r'<title>(.*)\(.*?(\d+).*?\)</title>', html) |
|
title = pages.group(1).strip() |
|
total_pages = int(pages.group(2)) |
|
|
|
with sync_playwright() as p: |
|
browser = p.chromium.launch() |
|
page = browser.new_page() |
|
page.goto(url) |
|
|
|
page.wait_for_load_state("networkidle") |
|
css_url = page.locator("link[rel='stylesheet'][href*='_nuxt/manual']" |
|
).get_attribute("href") |
|
|
|
custom_css = requests.get(css_url).text |
|
|
|
return dict(file_id=file_id, |
|
title=title, |
|
total_pages=total_pages, |
|
custom_css=custom_css) |
|
|
|
|
|
def replace_urls_to_absolute(url_viewer: str, content: str) -> str: |
|
"""Get html content and replace url relatives for absolutes""" |
|
|
|
content = content.replace('src="', f'src="{url_viewer}') |
|
content = content.replace('src:url(', f'src:url({url_viewer}') |
|
return content |
|
|
|
|
|
def get_html_page(domain: str, file_id: str, page: int) -> str: |
|
"""Get html page from manualpdf.es""" |
|
url_viewer = f"{domain}/viewer/{file_id}/{page}/" |
|
|
|
# url return file, download it and read it |
|
content = requests.get(f"{url_viewer}page-{page}.page").text |
|
|
|
# replace relative links to absolute links |
|
content = replace_urls_to_absolute(url_viewer=url_viewer, content=content) |
|
|
|
return content |
|
|
|
|
|
def generate_page(domain: str, file_id: str, page: int, content: str, |
|
path: str, landscape: bool, custom_css: str): |
|
"""Generate html page with jinja2 template""" |
|
url_viewer = f"{domain}/viewer/{file_id}/{page}/" |
|
template = jinja2.Template(PRINT_TEMPLATE) |
|
|
|
base_url = "https://www.manualpdf.es/css/base.css" |
|
base_css = requests.get(base_url).text |
|
base_css = replace_urls_to_absolute(url_viewer=url_viewer, |
|
content=base_css) |
|
|
|
page_url = f"https://www.manualpdf.es/viewer/{file_id}/{page}/page.css" |
|
page_css = requests.get(page_url).text |
|
page_css = replace_urls_to_absolute(url_viewer=url_viewer, |
|
content=page_css) |
|
|
|
html = template.render(file_id=file_id, |
|
page=page, |
|
content=content, |
|
custom_css=custom_css, |
|
base_css=base_css, |
|
page_css=page_css) |
|
|
|
# Save html page |
|
file_name = f'{sanitize(file_id)}_{page:04}.html' |
|
with open(path + '/' + file_name, 'w', encoding='utf-8') as f: |
|
f.write(html) |
|
generate_pdf(path, file_name, landscape) |
|
|
|
|
|
def generate_pdf(path: str, file_name: str, landscape: bool = False): |
|
"""Generate PDF from html""" |
|
apath = os.path.abspath(path + '/' + file_name) |
|
out_name = file_name.split('.')[0] + '.pdf' |
|
|
|
with sync_playwright() as p: |
|
browser = p.chromium.launch() |
|
context = browser.new_context() |
|
page = context.new_page() |
|
|
|
# Load local HTML |
|
file_url = f"file://{apath}" |
|
page.goto(file_url) |
|
|
|
# Generate PDF file |
|
page.pdf(path=f'{path}/{out_name}', format="A4", landscape=landscape) |
|
|
|
browser.close() |
|
|
|
|
|
def join_pdf_pages(path: str, file_id: str, title: str, out_path: str): |
|
"""Join all pdf pages in a single pdf file""" |
|
pdfs = [path + '/' + f for f in os.listdir(path) if f.endswith('.pdf')] |
|
pdfs.sort() |
|
|
|
merger = PdfMerger() |
|
for pdf in pdfs: |
|
merger.append(pdf) |
|
|
|
title = re.sub(r'[^\w\s]', '', title).replace(' ', '_') |
|
|
|
out_file_path = out_path + '/' + f'{sanitize(file_id)}_{title}.pdf' |
|
merger.write(out_file_path) |
|
merger.close() |
|
return out_file_path |
|
|
|
|
|
def delete_temp_folder(): |
|
"""Delete temp folder if all ok""" |
|
ok_question = [{ |
|
'type': 'confirm', |
|
'name': 'ok', |
|
'message': '¿All ok? Delete temp folder?', |
|
'default': True |
|
}] |
|
ok_answer = prompt(ok_question) |
|
if ok_answer.get('ok'): |
|
shutil.rmtree(TEMP_FOLDER) |
|
|
|
|
|
def process_page(domain: str, file_id: str, page: int, wpath: str, |
|
landscape: bool, custom_css: str): |
|
"""Download and process a single page""" |
|
content = get_html_page(domain, file_id, page) |
|
generate_page(domain, file_id, page, content, wpath, landscape, custom_css) |
|
return page |
|
|
|
|
|
if __name__ == '__main__': |
|
# Create temp folder if not exists |
|
wpath = os.path.abspath(TEMP_FOLDER) |
|
create_folder_if_not_exists(wpath) |
|
|
|
# Enter url |
|
url = get_manual_url() |
|
|
|
# Get data from url |
|
try: |
|
domain = get_domain(url) |
|
print("Loading manual data...") |
|
pdf_data = get_data(url) |
|
file_id = pdf_data['file_id'] |
|
except Exception as e: |
|
print(e) |
|
print('Error: pdf data not found') |
|
exit() |
|
|
|
# Ask continue downloading file |
|
print(f'{pdf_data["title"]} with {pdf_data["total_pages"]} pages') |
|
continue_question = [{ |
|
'type': 'confirm', |
|
'name': 'continue', |
|
'message': f'Continue downloading file?', |
|
'default': True |
|
}] |
|
continue_answer = prompt(continue_question) |
|
if not continue_answer.get('continue'): |
|
exit() |
|
|
|
# Create file_id folder |
|
wpath = wpath + f'/{sanitize(file_id)}' |
|
create_folder_if_not_exists(wpath) |
|
|
|
# Files in temp folder for skip already downloaded pages |
|
generated_files = [f for f in os.listdir(wpath) if f.endswith('.pdf')] |
|
|
|
# Ask for landscape |
|
landscape_question = [{ |
|
'type': 'confirm', |
|
'name': 'landscape', |
|
'message': '¿Landscape?', |
|
'default': False |
|
}] |
|
landscape_answer = prompt(landscape_question) |
|
landscape = landscape_answer.get('landscape') |
|
|
|
# Ask for multiprocessing |
|
multiprocessing_question = [{ |
|
'type': 'confirm', |
|
'name': 'multiprocessing', |
|
'message': '¿Multiprocessing?', |
|
'default': True |
|
}] |
|
multiprocessing_answer = prompt(multiprocessing_question) |
|
|
|
if multiprocessing_answer.get('multiprocessing'): |
|
# Use multiprocessing to download and process pages in parallel |
|
total_pages = pdf_data['total_pages'] |
|
pages_to_process = [ |
|
page for page in range(1, total_pages + 1) |
|
if f'{sanitize(file_id)}_{page:04}.pdf' not in generated_files |
|
] |
|
|
|
with progressbar.ProgressBar(max_value=len(pages_to_process)) as bar: |
|
bar.update(0) |
|
with Pool() as pool: |
|
for i, _ in enumerate( |
|
pool.imap_unordered( |
|
partial(process_page, |
|
domain, |
|
file_id, |
|
wpath=wpath, |
|
landscape=landscape, |
|
custom_css=pdf_data["custom_css"]), |
|
pages_to_process), 1): |
|
bar.update(i) |
|
|
|
else: |
|
with progressbar.ProgressBar(max_value=pdf_data['total_pages']) as bar: |
|
bar.update(0) |
|
for page in range(1, pdf_data['total_pages'] + 1): |
|
|
|
# If pdf page already exists, skip it |
|
if f'{sanitize(file_id)}_{page:04}.pdf' in generated_files: |
|
bar.update(page - 1) |
|
continue |
|
|
|
# Generate html page |
|
generate_page(domain, |
|
file_id, |
|
page, |
|
get_html_page(domain, file_id, page), |
|
wpath, |
|
landscape, |
|
custom_css=pdf_data["custom_css"]) |
|
bar.update(page - 1) |
|
|
|
# Join all pdf pages in a single pdf file |
|
out_path = os.path.abspath('output') |
|
create_folder_if_not_exists(out_path) |
|
out_file = join_pdf_pages(wpath, file_id, pdf_data['title'], out_path) |
|
|
|
# Open pdf file |
|
os.system(f'open {out_file}') |
|
|
|
# Delete temp folder if ok |
|
delete_temp_folder() |
@Fabienjulio @asntcrz @Acide-Burn
It seems there was an issue with the "pyhtml2pdf" library when using the webdrivers. I have switched to a different library. It now uses "playwright."
Please make sure to reinstall the script and then install playwright properly:
I hope it works for you now!