simbora

main.py

import requests
import io

import PyPDF2  # pip install PyPDF2
from seleniumwire import webdriver  # pip install selenium-wire


def payload_from_gene_data(dataset: list[str], signature, cutoff_1, cutoff_2):
    return {
        'methodoption': 'os',
        'dataset': '\n'.join(dataset),
        'signature': 'signature',
        'highcol': '#ff0000',
        'lowcol': '#0000ff',
        'groupcutoff1': str(cutoff_1),
        'groupcutoff2': str(cutoff_2),
        'axisunit': 'month',
        'ifhr': 'hr',
        'ifconf': 'conf',
        'signature_norm': '',
        'is_sub': 'false',
        'subtype': '',
    }

headers = {
    'Accept': 'application/json, text/javascript, */*; q=0.01',
    'Accept-Language': 'en-US,en;q=0.9',
    'Cache-Control': 'no-cache',
    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
    'Origin': 'http://gepia2.cancer-pku.cn',
    'Pragma': 'no-cache',
    'Proxy-Connection': 'keep-alive',
    'Referer': 'http://gepia2.cancer-pku.cn/',
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/113.0.0.0 Safari/537.36',
    'X-Requested-With': 'XMLHttpRequest',
}

all_genes_data = [
    payload_from_gene_data(dataset=['BRCA', 'COAD'], signature='ERBB2', cutoff_1=50, cutoff_2=50)
]

driver = webdriver.Chrome(seleniumwire_options={
    'suppress_connection_errors': False
})

driver.get('http://gepia2.cancer-pku.cn/#survival')
cookies = requests.cookies.RequestsCookieJar()

for cookie in driver.get_cookies():
    cookies.set(cookie['name'], cookie['value'], domain=cookie['domain'], path=cookie['path'])

# breakpoint()

for data in all_genes_data:
    # gera PDF no server
    response = requests.post(
        'http://gepia2.cancer-pku.cn/assets/PHP4/survival_zf.php',
        data=data,
        verify=False,
        cookies=cookies,
        headers=headers)

    # baixa PDF
    response_dict = eval(response.content.decode())
    response = requests.get(
        f'http://gepia2.cancer-pku.cn/tmp/{response_dict["outdir"]}',
        cookies=cookies,
        headers=headers)

    pdf_bytes = response.content

    # TODO: processa PDF
    pdf_bytes.append('%%EOF\r\n')
    read_pdf = PyPDF2.PdfReader(io.BytesIO(pdf_bytes))
    page = read_pdf.getPage(0)
    page_content = page.extractText()

    breakpoint()