import requests_html as rh
import os
# import pypandoc
import subprocess

def generate_directory_pdf(url, name, s=None):
    s = rh.HTMLSession() if not s else s
    r1 = s.get(url)

    html = ""
    anchors = r1.html.find('.td-sidebar-link')
    links = [a.absolute_links.pop() for a in anchors if a.element.tag == 'a']
    # dedup
    unique_links = []
    for i in links:
        if i not in unique_links:
            unique_links.append(i)
    links = unique_links
    links = filter(lambda href: href.startswith(url), links) # filter out links

    print("downloading...")
    cwd = os.getcwd()
    for l1 in links:
        r2 = s.get(l1)
        # r2.html.render()
        div = r2.html.find('.td-content', first=True, clean=True)
        if div:
            # try:
                # if name in ["Setup", "Tutorials", "Reference"]:  # will give duplicate id error, go through pages one by one to skip error page
                    # print("testing " + l1, end='')
                    # output = pypandoc.convert_text(div.html, "pdf", format="html", outputfile="/tmp/kubernetes_pdf_doc_tmp.pdf".format(name), extra_args=['--pdf-engine=weasyprint'])
                    # print(" works")
            # except Exception as e:
            #     print(" failed!")
            #     print(e)
            # else:
            html += div.html
        with open("{}/{}.html".format(cwd, name), "wt") as f:
            f.write(html)

    print("generating pdf...")
    subprocess.run(["{}/weasy_print.sh".format(cwd), name])
    # try:
    #     output = pypandoc.convert_text(html, "pdf", format="html", outputfile="./{}.pdf".format(name), extra_args=['--pdf-engine=weasyprint', '--css=codeblock_wrap.css'])
    # except Exception as e:
    #     print(e)

if __name__ == '__main__':
    s = rh.HTMLSession()
    directories = [\
                   "Setup",
                   "Concepts",
                   "Tasks",
                   "Tutorials",
                   "Reference",
                   # "Getting-started-guides",  # same as setup
                   # "Admin",  # this direct to concepts
                   # "Imported",  # deprecated
                   ]
    directories_pairs = [("https://kubernetes.io/docs/{}/".format(n.lower()), n) for n in directories]
    for url, name in directories_pairs:
        print(name)
        generate_directory_pdf(url, name)