from lxml.html.clean import clean_html, Cleaner
from bs4 import BeautifulSoup


def PostKeyError(KeyError):
    pass


def post_to_solr(p):
    # if a list is given call for each element
    if type(p) is list:
        return [post_to_solr(o) for o in p]
    if not type(p) is dict:
        raise TypeError("Argument for post_to_solr needs to be a dict or list of dicts")
    for k in ["url", "body"]:
        if not k in p:
            raise PostKeyError("Post needs to have key '%s'" % k)

    def get_text2(l):
        if not l:
            return ""
        soup = BeautifulSoup(l, features="lxml")
        return soup.get_text().replace("\n", " ").replace("\r", " ").replace("\t", " ")

    # Return the solr structure
    return {
        "id": p["url"],
        "date_dt": p["public_date"],
        "tags_tkn": p["tag_string"],
        "title_txt": p["title"],
        "text_txt": (get_text2(p.get("body", "")) or "")
        + " "
        + get_text2(p.get("agenda_html", "") or ""),
    }


def member_to_solr(m):
    if type(m) is list:
        return [member_to_solr(o) for o in m]

    return {
        "id": "/member/" + str(m["id"]),
        "date_dt": None,
        "tags_tkn": m["nickname"],
        "title_txt": m["firstname"] + " " + m["surname"] + "(" + m["nickname"] + ")",
        "text_txt": m["description"],
    }