wd_importPEP/crawlers/members_parliament_crawler.py

import os

import yaml
import json

import urllib.request, urllib.error, urllib.parse

from lxml import etree
import lxml.html
import lxml.html.soupparser


class members_parliament_crawler(object):
    def __init__(self, config_file):
        with open(config_file, "r") as stream:
            try:
                self.config = yaml.safe_load(stream)
            except yaml.YAMLError as exc:
                print(exc)

    # input list of countries in form of ['nicaragua', 'honduras', .. , 'mexico']

    def download_member_list_pages_of_countries(self, list_of_countries):
        # download only html pages of the countries specified in input

        for country in list_of_countries:
            for key in self.config:
                if key in list_of_countries:
                    try:
                        member_list = self.config.get(key).get("memberList")
                    except Exception as e:
                        print(
                            "There is a problem with the entry memberList in the config.yaml - the original error message is:",
                            e,
                        )
                    try:
                        member_list_link = member_list.get("link")
                    except Exception as e:
                        print(
                            "No memberListLink defined in config.yaml - the original error message is:",
                            e,
                        )

                    # download the html page of the List of Members

                    response = urllib.request.urlopen(member_list_link)
                    web_content = response.read().decode("UTF-8")

                    # save interim results to files

                    f = open("crawlers/pages/" + key + "MemberList.html", "w+")
                    f.write(webContent)
                    f.close

    def parse_member_list_data2dictionary(self, list_of_countries):
        for country in list_of_countries:
            try:
                # use soupparser to handle broken html

                tree = lxml.html.soupparser.parse(
                    "crawlers/pages/" + country + "MemberList.html"
                )

                #                for e in tree.iter():
                #
                #                    print(e.tag)
                #
                #                 for e in tree.xpath('//html//body//form//table//tr//td//table//tr'):
                #
                #                     #print(etree.tostring(e).decode())

                dictionary_member_list = {}

                country_conf = self.config.get(country)
                country_domain = country_conf.get("domain")
                country_conf_member_list = country_conf.get("memberList")
                country_conf_member_list_parent = country_conf_member_list.get("parent")
                country_conf_member_list_child_name = country_conf_member_list.get("child-name")
                country_conf_member_list_child_link = country_conf_member_list.get("child-link")

                for n in range(len(tree.xpath(country_conf_member_list_parent))):
                    name = tree.xpath(
                        country_conf_member_list_parent
                        + "["
                        + str(n)
                        + "]"
                        + country_conf_member_list_child_name
                    )
                    link = tree.xpath(
                        country_conf_member_list_parent
                        + "["
                        + str(n)
                        + "]"
                        + country_conf_member_list_child_link
                    )

                    if len(name) > 0:
                        dictionary_member_list[n] = {}
                        dictionary_member_list[n]["name"] = name[0]

                        if country_domain in link[0]:
                            dictionary_member_list[n]["link"] = link[0]

                        if country_domain not in link[0]:
                            dictionary_member_list[n]["link"] = country_domain + link[0]

            except Exception as e:
                print(
                    "parsing the html did not work. Possibly you first have to downloadMemberListPagesOfCountries(). The original error message is:",
                    e,
                )

            # save interim results to files

            f = open("crawlers/output/" + country + "MemberList.txt", "w+")
            f.write(str(dictionary_member_list))
            f.close

    def download_member_data_htmls(self, list_of_countries):
        for country in list_of_countries:
            f = open("crawlers/output/" + country + "MemberList.txt")
            text = f.read()

            dictionary_member_list = eval(text)

            for member_id in dictionary_member_list:
                member_link = dictionary_member_list[member_id]["link"]

                # download the html page of the Member

                response = urllib.request.urlopen(member_link)
                web_content = response.read().decode("UTF-8")

                # save interim results to files

                file_name = "crawlers/pages/" + country + "/" + str(member_id) + ".html"

                os.makedirs(os.path.dirname(file_name), exist_ok=True)
                f = open(file_name, "w+")
                f.write(web_content)
                f.close

    def parse_member_data2dictionary(self, list_of_countries):
        for country in list_of_countries:
            print("started to parse data of member of " + country + " ..")

            f = open("crawlers/output/" + country + "MemberList.txt")
            text = f.read()

            dictionary_member_list = eval(text)

            country_conf = self.config.get(country)
            country_domain = country_conf.get("domain")
            country_conf_member = country_conf.get("member")
            country_conf_member_info1 = country_conf_member.get("info-1")
            country_conf_member_info1_parent = country_conf_member_info1.get("parent")
            country_conf_member_info1_child_political_party = country_conf_member_info1.get(
                "child-politicalParty"
            )

            for member_id in dictionary_member_list:
                print(
                    "started to parse data of member with name "
                    + dictionary_member_list[member_id]["name"]
                    + " .."
                )

                file_name = "crawlers/pages/" + country + "/" + str(member_id) + ".html"

                tree = lxml.html.soupparser.parse(file_name)

                political_party = tree.xpath(
                    country_conf_member_info1_parent
                    + country_conf_member_info1_child_political_party
                )

                print("oi", political_party)

                if len(political_party) > 0:
                    dictionary_member_list[member_id]["political party"] = political_party[
                        0
                    ]

            f = open("crawlers/output/" + country + "MemberList.txt", "w+")
            f.write(str(dictionary_member_list))
            f.close