from sys import argv
from os import environ
from dotenv import load_dotenv
from cloudscraper import CloudScraper, create_scraper
from re import findall, sub

class Scraper:
    def __init__(self, pseudo: str, password: str, app: str, debug: bool = False):
        self.debug = debug
        self.url = "https://forum.mobilism.org"
        self.requested_app = app
        self.loginData = {
            "username": pseudo,
            "password": password,
            "login": "Login"
        }

    def errorFormat(self, code: int = None, message: str = "") -> str:
        """Pretty error message."""
        return f"{f'[{code}]' if code else ''}{' ' if len(message) > 0 and code else ''}{message}."

    def connect(self) -> CloudScraper:
        """Login to the forum using credentials."""
        session = create_scraper(browser = {"browser": "chrome", "platform": "windows"}) # connect with cloudflare bypasser with a chrome browser on windows
        if not session:
            raise SystemError(self.errorFormat(message = "The creation of the session failed")) # called only if failed at creating the session

        if self.debug: print("Connection attempt...")
        reponse = session.post(f"{self.url}/ucp.php", data = self.loginData, params = {"mode": "login"}) # connect to the forum using credentials - params are set by default but its in case forum changing that
        if reponse.status_code != 200:
            raise ConnectionRefusedError(self.errorFormat(code = reponse.status_code, message = "Unable to connect")) # called only status code isn't 200

        return session

    def search(self, session: CloudScraper) -> tuple[list[dict], list[dict]]:
        """Do the research."""
        if self.debug: print("Going to search page and check connection...", end = " ")
        reponse = session.get(f"{self.url}/search.php", params = {"keywords": self.requested_app, "sr": "topics", "sf": "titleonly"}) # fetch results page
        if "Sorry but you are not permitted to use the search system. If you're not logged in please" in reponse.text:
            raise ConnectionError(self.errorFormat(message = "Connection failed, check credentials")) # called only if login failed
        if reponse.status_code != 200:
            raise ConnectionError(self.errorFormat(code = reponse.status_code, message = "Impossible to make the search")) # called only status code isn't 200
        if self.debug: print(f"Connected.")

        if self.debug: print(f"Fetching results for {self.requested_app}...", end = " ")

        topics = self.parse(reponse.text)

        self.save(topics)

        return topics, self.getInfos(session, topics)

    def parse(self, htmlPage: str) -> list[dict]:
        """Parse HTML reponse to a clean list"""
        if "No suitable matches were found." in htmlPage:
            return []
        elements = htmlPage.split("<tr>\n<td>")[1:]
        elements[-1] = elements[-1].split("</td>\n</tr>")[0]
        for i in range(0, len(elements)):
            try:
                _title = findall(r"class=\"topictitle\">(.*)<\/a>", elements[i])[0]
                _title = sub(r" ?&amp; ?", " ", _title)
            except:
                _title = None
            try:
                _author = findall(r"(<br />|</strong>)\n\n?<i class=\"icon-user\"></i> by <a href=\"\./memberlist\.php\?mode=viewprofile&amp;u=\d+\"( style=\"color: #.*;\" class=\"username-coloured\")?>(.*)</a>", elements[i])[0][-1]
            except:
                _author = None
            try:
                _link = findall(r"\./viewtopic\.php\?f=(\d*)&amp;t=(\d*)&amp", elements[i])[0]
                _link = {"f": _link[0], "t": _link[1]}
            except:
                _link = None
            try:
                _date = findall(r"</a> <i class=\"icon-time\"></i> <small>(.*)</small>", elements[i])[0]
            except:
                _date = None
                print("\n" + elements[i] + "\n")
            elements[i] = {"title": _title, "author": _author, "date": _date, "link": f"{self.url}/viewtopic.php?f={_link['f']}&t={_link['t']}", "linkParams": _link}

        return elements

    def getInfos(self, session: CloudScraper, elements: list) -> list:
        """Go to the first n pages and get a lot of infos"""
        page = 3
        if self.debug: print(f"Going to the {page} first pages...", end = " ")
        results = []
        for i in range(0, len(elements)):
            if i < page:
                reponse = session.get(f"{self.url}/viewtopic.php", params = elements[i]["linkParams"]) # fetch results page
                results.append(reponse)
                if reponse.status_code != 200:
                    raise ConnectionError(self.errorFormat(code = reponse.status_code, message = f"Error while doing the search n°{i}")) # called only status code isn't 200
        if self.debug: print(f"Done.")

        if self.debug: print(f"Parsing results page...", end = " ")
        results = self.parsingInfos(results)
        if self.debug: print(f"Done.")

        return results

    def parsingInfos(self, elements: list) -> list[dict]:
        """Parse infos from the page of the app"""
        for i in range(0, len(elements)):
            elements[i] = elements[i].text
            if "Download Instructions" not in elements[i]:
                elements[i] = {"changelogs": None, "downloadLinks": None}
                continue
            try:
                _changelogs = findall(r"What's New:</span> ?<br />(.*)<br /><br /><span style=\"c|font-weight: bold\">T", elements[i])[0]
                if len(_changelogs) < 2: # if result none, trying other method
                    _changelogs = findall(r"What's New:</span> ?<br />(.*)<br /><br /><span style=\"font-weight: bold\">T", elements[i])[0]
            except:
                _changelogs = "No changelog found."
            try:
                elements[i] = sub(r"Download Instructions:</span>(.*)?<br /><s", "Download Instructions:</span><br /><s", elements[i])
                _downloadLinks = findall(r"Download Instructions:</span> ?<br />(.*|[\s\S]*)<br /><br />Trouble downloading|</a></div>", elements[i])[0]
                if len(_downloadLinks) < 2:# if result none, trying other method
                    _downloadLinks = findall(r"Download Instructions:</span> ?<br />(.*|[\s\S]*)</a></div>", elements[i])[0]
            except:
                _downloadLinks = None
            _downloadLinks = sub(r"\n|<a class=\"postlink\" href=\"|\(Closed Filehost\) ?|<span style=\"font-weight: bold\">|</span>|\">(\S*)</a>", "", _downloadLinks) # remove html garbage
            _downloadLinks = sub(r"<br />\n?", "\n", _downloadLinks) # convert newline html to \n
            _downloadLinks = sub(r"Mirrors(?!:)|Mirror(?!s)(?!:)", "Mirror:", _downloadLinks) # add ":"
            elements[i] = {"changelogs": _changelogs, "downloadLinks": _downloadLinks}

        return elements

    def prettyPrint(self, topics: tuple[list[dict], list[dict]]) -> list:
        """Show a pretty message with all the specialized infos"""
        topics, topicsInfos = topics
        print("\n")
        result = []
        for i in range(0, len(topicsInfos)):
            result.append({
                "title": topics[i]["title"],
                "author": topics[i]["author"],
                "date": topics[i]["date"],
                "changelogs": str(topicsInfos[i]["changelogs"]).replace("<br />", "\n"),
                "downloadLinks": topicsInfos[i]["downloadLinks"]
            })
            print(f"Title: {result[i]['title']}\n")
            print(f"Author: {result[i]['author']}\n")
            print(f"Date of release: {result[i]['date']}\n")
            print(f"Changelogs: \n{result[i]['changelogs']}\n")
            print(f"Download links: \n{result[i]['downloadLinks']}")
            print("\n\n---\n")

        return result

    def work(self) -> list:
        """Call all the others methods."""
        return self.prettyPrint(self.search(self.connect()))

    def save(self, elements: list) -> None:
        """Save all the results parsed to a CSV file."""
        taille = len(elements)
        if taille == 0:
            print("No elements were found with the search.")
            return
        filename = "results.csv"
        with open(filename, "w") as f:
            topCSV = list(elements[0].keys()) # create a copy of the first element keys
            topCSV.remove("linkParams") # remove linkParams
            f.write(";".join(topCSV))
            f.write("\n")
            for element in elements:
                if element != "linkParams":
                    f.write(";".join(str(e) for e in list(element.values())[:-1]))
                f.write("\n")
        print(f"{taille} elements have been registered in the {filename} file.")

if __name__ == "__main__":
    argv = argv[1:]
    if len(argv) < 1: # no args
        print("No App to retrieve.")
        exit(1)
    load_dotenv() # load .env file
    try:
        try: # for logs
            debug = environ["DEBUG_MOBILISM"].lower() in ("yes", "true", "1")
        except:
            debug = False
        try: # try to fetch credentials from de .env first
            pseudoMobilism = environ["PSEUDO_MOBILISM"]
            passwordMobilism = environ["PASSWORD_MOBILISM"]
        except: # if it failed try to get from the cli
            if len(argv) >= 3:
                pseudoMobilism = argv[0]
                passwordMobilism = argv[1]
                argv = argv[-2:]
            else: # if it failed again there is a problem
                raise KeyError
        Scraper(pseudoMobilism, passwordMobilism, " ".join([n for n in argv]), debug).work() # call the work() function
    except KeyError:
        print('Please fill in the username and password (with quotes) by args or with .env file and give an app to retrieve.')