fb_scraper_request 0.2.0__py3-none-any.whl

fb_graphql_scraper/__init__.py

@@ -0,0 +1,14 @@
+"""Facebook GraphQL Scraper - Simple API to scrape public Facebook posts without login.
+
+Example:
+    from facebook_graphql_scraper import FacebookGraphqlScraper
+
+    fb = FacebookGraphqlScraper()
+    result = fb.get_user_posts("vietgiaitri", days_limit=3)
+    print(result["data"])
+"""
+
+from .facebook_graphql_scraper import FacebookGraphqlScraper
+
+__version__ = "0.2.0"
+__all__ = ["FacebookGraphqlScraper"]

fb_graphql_scraper/base/base_page.py

@@ -0,0 +1,90 @@
+# -*- coding:utf-8 -*-
+from selenium.webdriver.chrome.options import Options as ChromeOptions
+from selenium.webdriver.chrome.service import Service
+
+
+class BasePage:
+    def __init__(self, driver_path: str, open_browser: bool = False):
+        chrome_options = self._build_options(open_browser)
+        normalized_driver_path = self._normalize_path(driver_path)
+
+        if self._looks_like_chrome_binary(normalized_driver_path):
+            chrome_options.binary_location = normalized_driver_path
+            service = Service()
+        elif normalized_driver_path:
+            service = Service(normalized_driver_path)
+        else:
+            service = Service()
+
+        self.driver = self._build_driver(service=service, chrome_options=chrome_options)
+        self.driver.maximize_window()
+
+    @staticmethod
+    def _build_options(open_browser: bool) -> ChromeOptions:
+        options = ChromeOptions()
+        options.add_argument("--disable-blink-features")
+        options.add_argument("--disable-notifications")
+        options.add_argument("--disable-blink-features=AutomationControlled")
+        if not open_browser:
+            options.add_argument("--headless=new")
+        options.add_argument("--blink-settings=imagesEnabled=false")
+        return options
+
+    @staticmethod
+    def _normalize_path(path: str | None) -> str | None:
+        if not path:
+            return None
+        # In notebook strings users often escape spaces (e.g. "Google\ Chrome").
+        return path.replace("\\ ", " ").strip()
+
+    @staticmethod
+    def _looks_like_chrome_binary(path: str | None) -> bool:
+        if not path:
+            return False
+        normalized = path.lower()
+        return normalized.endswith("/google chrome") or normalized.endswith("/chrome")
+
+    @staticmethod
+    def _build_driver(service: Service, chrome_options: ChromeOptions):
+        from seleniumwire import webdriver
+
+        try:
+            return webdriver.Chrome(service=service, options=chrome_options)
+        except AttributeError as exc:
+            if "VERSION_CHOICES" not in str(exc):
+                raise
+            # Some selenium-wire/pyOpenSSL combinations expose this at runtime.
+            BasePage._patch_seleniumwire_tls_version_choices()
+            return webdriver.Chrome(service=service, options=chrome_options)
+
+    @staticmethod
+    def _patch_seleniumwire_tls_version_choices() -> None:
+        from OpenSSL import SSL
+        from seleniumwire.thirdparty.mitmproxy.net import tls
+
+        if hasattr(tls, "VERSION_CHOICES"):
+            return
+
+        basic_options = SSL.OP_CIPHER_SERVER_PREFERENCE
+        if hasattr(SSL, "OP_NO_COMPRESSION"):
+            basic_options |= SSL.OP_NO_COMPRESSION
+
+        default_method = getattr(SSL, "SSLv23_METHOD", getattr(SSL, "TLS_METHOD", None))
+        default_options = basic_options
+        if hasattr(SSL, "OP_NO_SSLv2"):
+            default_options |= SSL.OP_NO_SSLv2
+        if hasattr(SSL, "OP_NO_SSLv3"):
+            default_options |= SSL.OP_NO_SSLv3
+
+        version_choices = {
+            "all": (default_method, basic_options),
+            "secure": (default_method, default_options),
+        }
+
+        for name in ("TLSv1", "TLSv1_1", "TLSv1_2"):
+            method_name = f"{name}_METHOD"
+            method = getattr(SSL, method_name, None)
+            if method is not None:
+                version_choices[name] = (method, basic_options)
+
+        tls.VERSION_CHOICES = version_choices

fb_graphql_scraper/example.py

@@ -0,0 +1,26 @@
+# -*- coding: utf-8 -*-
+from fb_graphql_scraper.facebook_graphql_scraper import FacebookGraphqlScraper as fb_graphql_scraper
+
+
+## Example.1 - without logging in
+if __name__ == "__main__":
+    facebook_user_name = "love.yuweishao"
+    facebook_user_id = "100044253168423"
+    days_limit = 100 # Number of days within which to scrape posts
+    driver_path = "/Users/hongshangren/Downloads/chromedriver-mac-arm64_136/chromedriver" 
+    fb_spider = fb_graphql_scraper(driver_path=driver_path, open_browser=False)
+    res = fb_spider.get_user_posts(fb_username_or_userid=facebook_user_id, days_limit=days_limit,display_progress=True)
+
+
+## Example.2 - login in your facebook account to collect data
+# if __name__ == "__main__":
+    # facebook_user_name = "love.yuweishao"
+    # facebook_user_id = "100044253168423"
+    # fb_account = "facebook_account"
+    # fb_pwd = "facebook_paswword"
+    # days_limit = 30 # Number of days within which to scrape posts
+    # driver_path = "/Users/hongshangren/Downloads/chromedriver-mac-arm64_136/chromedriver" 
+    # fb_spider = fb_graphql_scraper(fb_account=fb_account,fb_pwd=fb_pwd, driver_path=driver_path, open_browser=False)
+    # res = fb_spider.get_user_posts(fb_username_or_userid=facebook_user_name, days_limit=days_limit,display_progress=True)
+    # print(res)
+    

fb_graphql_scraper/facebook_graphql_scraper.py

@@ -0,0 +1,304 @@
+# -*- coding: utf-8 -*-
+import re
+import pandas as pd
+import requests
+from fb_graphql_scraper.utils.parser import RequestsParser
+from fb_graphql_scraper.utils.locator import *
+from fb_graphql_scraper.utils.utils import *
+
+
+class FacebookSettings:
+    """Facebook GraphQL Scraper - No login required
+
+    Example:
+        from fb_graphql_scraper.facebook_graphql_scraper import FacebookGraphqlScraper as fb_graphql_scraper
+
+        if __name__ == "__main__":
+            facebook_user_name = "love.yuweishao"
+            days_limit = 30
+            fb_spider = fb_graphql_scraper()
+            res = fb_spider.get_user_posts(fb_username_or_userid=facebook_user_name, days_limit=days_limit, display_progress=True)
+            print(res)
+    """
+
+    def __init__(self):
+        self._set_container()
+        self._set_stop_point()
+        self.doc_id = "26420831597536910"
+        self.requests_parser = RequestsParser()
+
+    def _set_container(self):
+        self.post_id_list = []
+        self.reaction_count_list = []
+        self.profile_feed = []
+        self.res = {
+            "post_caption": [],
+            "post_date": [],
+            "post_likes": [],
+            "comment_share_type": [],
+            "comment_share_value": [],
+        }
+
+    def _set_stop_point(self):
+        self.pre_diff_days = float("-inf")
+        self.counts_of_same_diff_days = 0
+
+
+class FacebookGraphqlScraper(FacebookSettings):
+    def __init__(self):
+        super().__init__()
+
+    def get_user_id_from_username(self, username: str) -> tuple:
+        """Resolve Facebook username to numeric user ID and extract profile info"""
+        # Check if already numeric ID
+        if username.isdigit():
+            return username, []
+
+        # Try to get user ID from profile page
+        url = f"https://www.facebook.com/{username}?locale=en_us"
+        headers = {
+            "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
+        }
+
+        profile_feed = []
+
+        try:
+            response = requests.get(
+                url, headers=headers, allow_redirects=True, timeout=10
+            )
+            # Look for user ID in page content
+            patterns = [
+                r'"userID":"(\d+)"',
+                r'"actorID":"(\d+)"',
+                r'"id":"(\d+)"',
+                r'"profile_owner":"(\d+)"',
+                r"entity_id=(\d+)",
+                r'"owner":{"__typename":"User","id":"(\d+)"}',
+            ]
+            user_id = username
+            for pattern in patterns:
+                match = re.search(pattern, response.text)
+                if match:
+                    user_id = match.group(1)
+                    print(f"Resolved '{username}' -> '{user_id}'")
+                    break
+
+            # Extract profile name
+            name_patterns = [
+                r'"name":"([^"]+)","__typename":"User"',
+                r'"pageName":"([^"]+)"',
+                r"<title>([^<]+)</title>",
+            ]
+            for pattern in name_patterns:
+                match = re.search(pattern, response.text)
+                if match:
+                    name = match.group(1).replace(" | Facebook", "").strip()
+                    if name:
+                        profile_feed.append(name)
+                        break
+
+            # Extract followers count if available
+            follower_patterns = [
+                r"(\d+(?:[,.]\d+)?)\s*followers",
+                r'"follower_count":(\d+)',
+                r'"followers":\{"count":(\d+)\}',
+            ]
+            for pattern in follower_patterns:
+                match = re.search(pattern, response.text, re.IGNORECASE)
+                if match:
+                    followers = match.group(1)
+                    profile_feed.append(f"{followers} followers")
+                    break
+
+            return user_id, profile_feed
+
+        except Exception as e:
+            print(f"Error resolving user ID: {e}")
+            return username, profile_feed
+
+    def format_data(self, res_in, fb_username_or_userid, new_reactions):
+        final_res = pd.json_normalize(res_in)
+        final_res["context"] = self.requests_parser.context_list
+        final_res["username_or_userid"] = fb_username_or_userid
+        final_res["owing_profile"] = self.requests_parser.owning_profile
+        final_res["sub_reactions"] = new_reactions
+        final_res["post_url"] = "https://www.facebook.com/" + final_res["post_id"]
+        final_res["time"] = self.requests_parser.creation_list
+        final_res["published_date"] = pd.to_datetime(final_res["time"], unit="s")
+        final_res["published_date2"] = final_res["published_date"].dt.strftime(
+            "%Y-%m-%d"
+        )
+        final_res = final_res[
+            [
+                "post_id",
+                "post_url",
+                "username_or_userid",
+                "owing_profile",
+                "published_date",
+                "published_date2",
+                "time",
+                "reaction_count.count",
+                "comment_rendering_instance.comments.total_count",
+                "share_count.count",
+                "sub_reactions",
+                "context",
+                "video_view_count",
+            ]
+        ].to_dict(orient="records")
+        filtered_post_id = []
+        filtered_data = []
+        for each_data in list(final_res):
+            if each_data["post_id"] not in filtered_post_id:
+                filtered_data.append(each_data)
+                filtered_post_id.append(each_data["post_id"])
+        return filtered_data
+
+    def process_reactions(self, res_in):
+        reactions_out = []
+        for each_res in res_in:
+            each_reactions = each_res["top_reactions"]["edges"]
+            processed_reactions = self.requests_parser.process_reactions(
+                reactions_in=each_reactions
+            )
+            reactions_out.append(processed_reactions)
+        return reactions_out
+
+    def get_user_posts(
+        self,
+        fb_username_or_userid: str,
+        days_limit: int = 61,
+        display_progress: bool = True,
+    ) -> dict:
+        self.requests_parser._clean_res()
+        self._set_container()
+        self._set_stop_point()
+
+        # Auto-resolve username to user ID and extract profile info
+        user_id, profile_feed = self.get_user_id_from_username(fb_username_or_userid)
+
+        print(f"Collecting posts for {user_id} (doc_id: {self.doc_id})")
+        if profile_feed:
+            print(f"Profile info: {profile_feed}")
+
+        return self.requests_flow(
+            doc_id=self.doc_id,
+            fb_username_or_userid=user_id,
+            days_limit=days_limit,
+            profile_feed=profile_feed,
+            display_progress=display_progress,
+        )
+
+    def requests_flow(
+        self,
+        doc_id: str,
+        fb_username_or_userid: str,
+        days_limit: int,
+        profile_feed: list,
+        display_progress=True,
+    ):
+        """
+        Fetch more posts from a user's Facebook profile using the requests module.
+
+        Flow:
+            1. Get the document ID of the target Facebook profile.
+            2. Use the requests module to fetch data from the profile.
+            3. Continuously fetch data by checking for new posts until the specified days limit is reached.
+
+        Args:
+            doc_id (str): The document ID of the target Facebook account.
+            fb_username_or_userid (str): The Facebook username or user ID of the target account.
+            days_limit (int): The number of days for which to fetch posts (limits the time range of retrieved posts).
+            profile_feed (list): A list containing the posts retrieved from the target profile.
+
+        Helper Functions:
+            1. get_before_time:
+                Retrieves Facebook posts from a specified time period before the current date.
+
+            2. get_payload:
+                Prepares the payload for the next round of requests to the server.
+
+            3. get_next_page_status:
+                Checks whether the target Facebook user has more posts available for retrieval.
+
+            4. compare_timestamp:
+                Verifies whether a retrieved post falls within the specified time period for collection.
+        """
+        self.requests_parser._clean_res()  # Clear all arrays used to store the results
+        self._set_container()  # 清空用於儲存貼文資訊的array
+        self._set_stop_point()  # 設置/重置停止條件 | 停止條件: 瀏覽器無法往下取得更多貼文(n次) or 已取得目標天數內貼文
+        url = "https://www.facebook.com/api/graphql/"
+        before_time = get_before_time()
+        loop_limit = 5000
+        is_first_time = True
+        # Extract data
+        for i in range(loop_limit):
+            if is_first_time:
+                payload_in = get_payload(
+                    doc_id_in=doc_id,
+                    id_in=fb_username_or_userid,
+                    before_time=before_time,  # input before_time
+                )
+                print("playload_in:", payload_in)
+                response = requests.post(
+                    url=url,
+                    data=payload_in,
+                )
+                data = response.content
+                decoded_data = data.decode("utf-8")
+                body_content = decoded_data.split("\n")
+                print(body_content[:5])
+                self.requests_parser.parse_body(body_content=body_content)
+                is_first_time = False
+
+            # if not the first tiime send request, use function 'get_next_payload' for extracting end cursor to scrape next round
+            elif not is_first_time:
+                next_cursor = get_next_cursor(body_content_in=body_content)
+                payload_in = get_next_payload(
+                    doc_id_in=doc_id,
+                    id_in=fb_username_or_userid,
+                    before_time=before_time,  # input before_time
+                    cursor_in=next_cursor,
+                )
+
+            response = requests.post(
+                url=url,
+                data=payload_in,
+            )
+            body = response.content
+            decoded_body = body.decode("utf-8")
+            body_content = decoded_body.split("\n")
+            self.requests_parser.parse_body(body_content=body_content)
+
+            # Check progress
+            next_page_status = get_next_page_status(body_content=body_content)
+
+            before_time = str(self.requests_parser.creation_list[-1])
+            if not next_page_status:
+                print("There are no more posts.")
+                break
+
+            # date_object = int(datetime.strptime(before_time, "%Y-%m-%d"))
+            if compare_timestamp(
+                timestamp=int(before_time),
+                days_limit=days_limit,
+                display_progress=display_progress,
+            ):
+                print(
+                    f"The scraper has successfully retrieved posts from the past {str(days_limit)} days."
+                )
+                break
+
+        res_out = self.requests_parser.collect_posts()
+        new_reactions = self.process_reactions(res_in=res_out)
+        # create result
+        final_res = self.format_data(
+            res_in=res_out,
+            fb_username_or_userid=fb_username_or_userid,
+            new_reactions=new_reactions,
+        )
+        return {
+            "fb_username_or_userid": fb_username_or_userid,
+            "profile": profile_feed,
+            "data": final_res,
+            "raw_data": self.requests_parser.res_new,
+        }

fb_graphql_scraper/pages/page_optional.py

@@ -0,0 +1,127 @@
+# -*- coding: utf-8 -*-
+from fb_graphql_scraper.utils.locator import *
+from selenium.webdriver.common.by import By
+from selenium.webdriver.support.ui import WebDriverWait
+from selenium.webdriver.support import expected_conditions as EC
+from selenium.webdriver.common.action_chains import ActionChains
+from selenium.webdriver.common.keys import Keys
+import time
+
+
+class PageOptional(object):
+    def __init__(self, driver=None, fb_account: str = None, fb_pwd: str = None):
+        self.locator = PageLocators
+        self.xpath_elements = PageXpath
+        self.class_elements = PageClass
+        self.page_text = PageText
+        self.driver = driver
+        self.fb_account = fb_account
+        self.fb_pwd = fb_pwd
+
+        # Loggin account
+        if self.fb_account and self.fb_pwd:
+            login_page_url = "https://www.facebook.com/login"
+            self.driver.get(url=login_page_url)
+            self.login_page()
+
+    def login_page(self):
+        try:
+            self.login_account(user=self.fb_account, 
+                               password=self.fb_pwd,
+            )
+            time.sleep(5)
+        except Exception as e:
+            print(f"Login faield, message: {e}")
+
+    def clean_requests(self):
+        print(f"Before cleaning driver requests, the number of requests are: {len(self.driver.requests)}")
+        try:
+            print("Try to clear driver requests..")
+            del self.driver.requests
+            print(f"Clear, the number of requests are: {len(self.driver.requests)}")
+        except Exception as e:
+            print(f"Clear unsuccessfully, message: {e}")
+
+    def get_in_url(self):
+        self.driver.get(url=self.url)
+
+    def login_account(self, user: str, password: str):
+        user_element = self.driver.find_element(By.NAME, "email")
+        user_element.send_keys(user)
+        password_element = self.driver.find_element(By.NAME, "pass")
+        password_element.send_keys(password)
+        password_element.send_keys(Keys.ENTER)
+
+    def scroll_window(self):
+        self.driver.execute_script(
+            "window.scrollTo(0,document.body.scrollHeight)")
+
+    def scroll_window_with_parameter(self, parameter_in: str):
+        self.driver.execute_script(f"window.scrollBy(0, {parameter_in});")
+
+    def set_browser_zoom_percent(self, zoom_percent: int):
+        zoom_percent = str(zoom_percent)
+        self.driver.execute_script(
+            f"document.body.style.zoom='{zoom_percent}%'")
+
+    def move_to_element(self, element_in):
+        ActionChains(self.driver).move_to_element(element_in).perform()
+
+    def load_next_page(self, url:str, clear_limit:int=20):
+        """>> Move on to target facebook user page,
+        before moving, clean driver's requests first,
+        or driver would store previous account's data.
+        Args: url (str): user(kol) links"""
+        i = 0
+        while i <= clear_limit:
+            self.clean_requests()
+            if len(self.driver.requests) == 0:
+                print("Clear all driver requests already!")
+                break
+            i += 1
+        self.driver.get(url=url)
+
+    def click_display_button(self):
+        elements = self.driver.find_elements(self.locator.DISPLAY_MORE)
+        for _ in range(10):
+            for each_element in elements:
+                if each_element.text == self.page_text.DISPLAY_MORE or each_element.text == self.page_text.DISPLAY_MORE2:
+                    self.move_to_element(element_in=each_element)
+                    self.scroll_window_with_parameter(parameter_in="500")
+                    try:
+                        each_element.click()
+                        elements = self.driver.find_elements(
+                            self.locator.DISPLAY_MORE)
+                    except Exception as e:
+                        print(
+                            f"Click display more unsucessfully, error message:\n{e}")
+
+    def click_display_button2(self):
+        display_more_xpath = f"//div[@class='{PageClass.DISPLAY_MORE}' and @role='{PageRoleValue.DISPLAY_MORE}' and text()='{PageText.DISPLAY_MORE}']"
+        elements = self.driver.find_elements(By.XPATH, display_more_xpath)
+        for _ in range(10):
+            for each_element in elements:
+                if each_element.text == self.page_text.DISPLAY_MORE or each_element.text == self.page_text.DISPLAY_MORE2:
+                    self.move_to_element(element_in=each_element)
+                    self.scroll_window_with_parameter(parameter_in="500")
+                    try:
+                        each_element.click()
+                        elements = self.driver.find_elements(
+                            self.locator.DISPLAY_MORE)
+                    except Exception as e:
+                        print(
+                            f"Click display more unsucessfully, error message:\n{e}")
+
+    def click_reject_login_button(self):
+        try:
+            reject_login_button = WebDriverWait(self.driver, 10).until(
+            EC.visibility_of_element_located((self.locator.CLOSELOGIN)))
+            reject_login_button.click()
+        except Exception as e:
+            print(f"Click reject button failed, message:{e}")
+
+    def quit_driver(self):
+        self.driver.quit()
+
+    def close_driver(self):
+        self.driver.close()

fb_graphql_scraper/utils/locator.py

@@ -0,0 +1,54 @@
+# -*- coding: utf-8 -*-
+from selenium.webdriver.common.by import By
+
+
+class PageXpath(object):
+    CLOSE_LOGIN_BUTTON = "/html/body/div[1]/div/div[1]/div/div[5]/div/div/div[1]/div/div[2]/div/div/div/div[1]/div/i"
+
+
+class PageClass(object):
+    DISPLAY_MORE = "x1i10hfl xjbqb8w x1ejq31n xd10rxx x1sy0etr x17r0tee x972fbf xcfux6l x1qhh985 xm0m39n x9f619 x1ypdohk xt0psk2 xe8uvvx xdj266r x11i5rnm xat24cr x1mh8g0r xexx8yu x4uap5 x18d9i69 xkhd6sd x16tdsg8 x1hl2dhg xggy1nq x1a2a7pz xt0b8zv xzsf02u x1s688f"
+    CONTENTS = "x1yztbdb x1n2onr6 xh8yej3 x1ja2u2z"
+    CAPTION = "x193iq5w xeuugli x13faqbe x1vvkbs x1xmvt09 x1lliihq x1s928wv xhkezso x1gmr53x x1cpjm7i x1fgarty x1943h6x xudqn12 x3x7a5m x6prxxf xvq8zen xo1l8bm xzsf02u x1yc453h"
+    CAPTION2 = "xdj266r x11i5rnm xat24cr x1mh8g0r x1vvkbs x126k92a"
+    POSTDATE = "x1i10hfl xjbqb8w x1ejq31n xd10rxx x1sy0etr x17r0tee x972fbf xcfux6l x1qhh985 xm0m39n x9f619 x1ypdohk xt0psk2 xe8uvvx xdj266r x11i5rnm xat24cr x1mh8g0r xexx8yu x4uap5 x18d9i69 xkhd6sd x16tdsg8 x1hl2dhg xggy1nq x1a2a7pz x1heor9g xt0b8zv xo1l8bm"
+    POSTDATE = "x1i10hfl xjbqb8w x1ejq31n xd10rxx x1sy0etr x17r0tee x972fbf xcfux6l x1qhh985 xm0m39n x9f619 x1ypdohk xt0psk2 xe8uvvx xdj266r x11i5rnm xat24cr x1mh8g0r xexx8yu x4uap5 x18d9i69 xkhd6sd x16tdsg8 x1hl2dhg xggy1nq x1a2a7pz x1heor9g xt0b8zv xo1l8bm"
+    LIKES = "x6s0dn4 x78zum5 x1iyjqo2 x6ikm8r x10wlt62"
+    COMMENT_SHARE_PARENTS = "x9f619 x1n2onr6 x1ja2u2z x78zum5 x2lah0s x1qughib x1qjc9v5 xozqiw3 x1q0g3np xykv574 xbmpl8g x4cne27 xifccgj"
+    COMMENT_SHARE_CHILD = "x193iq5w xeuugli x13faqbe x1vvkbs x1xmvt09 x1lliihq x1s928wv xhkezso x1gmr53x x1cpjm7i x1fgarty x1943h6x xudqn12 x3x7a5m x6prxxf xvq8zen xo1l8bm xi81zsa"
+
+
+class SoupElement(object):
+    pass
+
+
+class PageText(object):
+    DISPLAY_MORE = "查看更多"
+    DISPLAY_MORE2 = "顯示更多"
+
+
+class PageRoleValue(object):
+    DISPLAY_MORE = "button"
+
+
+class PageLocators(object):
+    CLOSELOGIN = (
+        By.XPATH, "//div[@aria-label='Close' or @aria-label='Dismiss' or @aria-label='關閉']|//div[@role='button' and (contains(@aria-label, 'Close') or contains(@aria-label, 'Dismiss'))]|/html/body/div[2]/div/div[1]/div/div[5]/div/div/div[1]/div/div[2]/div/div/div/div[1]/div/i")
+    DISPLAY_MORE = (
+        By.XPATH, f"//div[@class='{PageClass.DISPLAY_MORE}' and @role='{PageRoleValue.DISPLAY_MORE}' and text()='{PageText.DISPLAY_MORE}']")
+
+    LOGGINUSR1 = (
+        "/html/body/div[1]/div[1]/div[1]/div/div/div/div[2]/div/div[1]/form/div[1]/div[1]/input"
+    )
+    LOGGINPWD1 = (
+        "/html/body/div[1]/div[1]/div[1]/div/div/div/div[2]/div/div[1]/form/div[1]/div[2]/div/input"
+    )
+    
+    LOGGINUSR2 = (
+        "/html/body/div[1]/div/div[1]/div/div[5]/div/div/div[1]/div/div[2]/div/div/div/div[2]/form/div/div[3]/div/label/div/div/input")
+    LOGGINPWD2 = (
+        "/html/body/div[1]/div/div[1]/div/div[5]/div/div/div[1]/div/div[2]/div/div/div/div[2]/form/div/div[4]/div/label/div/div/input")
+
+    # version.3: facebook login page
+    LOGGINUSR3 = (By.NAME, "email")
+    LOGGINPWD3 = (By.NAME, "pass")