###
# Copyright (c) 2005, Jeremiah Fincher
# Copyright (c) 2009, James McCoy
# Copyright (c) 2010-2024, Valentin Lorentz
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
#   * Redistributions of source code must retain the above copyright notice,
#     this list of conditions, and the following disclaimer.
#   * Redistributions in binary form must reproduce the above copyright notice,
#     this list of conditions, and the following disclaimer in the
#     documentation and/or other materials provided with the distribution.
#   * Neither the name of the author of this software nor the name of
#     contributors to this software may be used to endorse or promote products
#     derived from this software without specific prior written consent.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
# ARE DISCLAIMED.  IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
###

import re
import socket

import supybot.conf as conf
import supybot.utils as utils
from supybot.commands import *
import supybot.plugins as plugins
import supybot.commands as commands
import supybot.ircutils as ircutils
import supybot.callbacks as callbacks
from supybot.i18n import PluginInternationalization, internationalizeDocstring
_ = PluginInternationalization('Web')

from html.parser import HTMLParser
from html.entities import entitydefs
import http.client as http_client

class Title(utils.web.HtmlToText):
    entitydefs = entitydefs.copy()
    entitydefs['nbsp'] = ' '
    def __init__(self):
        self.inTitle = False
        self.inSvg = 0 # counter instead of boolean because svg can be nested
        utils.web.HtmlToText.__init__(self)

    @property
    def inHtmlTitle(self):
        return self.inTitle and self.inSvg == 0

    def handle_starttag(self, tag, attrs):
        if tag == 'title':
            self.inTitle = True
        elif tag == 'svg':
            self.inSvg += 1

    def handle_endtag(self, tag):
        if tag == 'title':
            self.inTitle = False
        elif tag == 'svg':
            self.inSvg = max(0, self.inSvg - 1)

    def append(self, data):
        if self.inHtmlTitle:
            super(Title, self).append(data)

class DelayedIrc:
    def __init__(self, irc):
        self._irc = irc
        self._replies = []
    def reply(self, *args, **kwargs):
        self._replies.append(('reply', args, kwargs))
    def error(self, *args, **kwargs):
        self._replies.append(('error', args, kwargs))
    def __getattr__(self, name):
        assert name not in ('reply', 'error', '_irc', '_msg', '_replies')
        return getattr(self._irc, name)

if hasattr(http_client, '_MAXHEADERS'):
    def fetch_sandbox(f):
        """Runs a command in a forked process with limited memory resources
        to prevent memory bomb caused by specially crafted http responses.

        On CPython versions with support for limiting the number of headers,
        this is the identity function"""
        return f
else:
    # For the following CPython versions (as well as the matching Pypy
    # versions):
    # * 2.6 before 2.6.9
    # * 2.7 before 2.7.9
    # * 3.2 before 3.2.6
    # * 3.3 before 3.3.3
    def fetch_sandbox(f):
        """Runs a command in a forked process with limited memory resources
        to prevent memory bomb caused by specially crafted http responses."""
        def process(self, irc, msg, *args, **kwargs):
            delayed_irc = DelayedIrc(irc)
            f(self, delayed_irc, msg, *args, **kwargs)
            return delayed_irc._replies
        def newf(self, irc, *args):
            try:
                replies = commands.process(process, self, irc, *args,
                        timeout=10, heap_size=10*1024*1024,
                        pn=self.name(), cn=f.__name__)
            except (commands.ProcessTimeoutError, MemoryError):
                raise utils.web.Error(_('Page is too big or the server took '
                        'too much time to answer the request.'))
            else:
                for (method, args, kwargs) in replies:
                    getattr(irc, method)(*args, **kwargs)
        newf.__doc__ = f.__doc__
        return newf

def catch_web_errors(f):
    """Display a nice error instead of "An error has occurred"."""
    def newf(self, irc, *args, **kwargs):
        try:
            f(self, irc, *args, **kwargs)
        except utils.web.Error as e:
            irc.reply(str(e))
    return utils.python.changeFunctionName(newf, f.__name__, f.__doc__)

class Web(callbacks.PluginRegexp):
    """Add the help for 'help Web' here."""
    regexps = ['titleSnarfer']
    threaded = True

    def noIgnore(self, irc, msg):
        return not self.registryValue('checkIgnored', msg.channel, irc.network)

    def getTitle(self, irc, url, raiseErrors, msg):
        size = conf.supybot.protocols.http.peekSize()

        def url_workaround(url):
            """Returns a new URL that should be the target of a new request,
            or None if the request is fine as it is.

            The returned URL may be the same as the parameter, in case
            something else was changed by this function through side-effects.
            """
            nonlocal size
            parsed_url = utils.web.urlparse(url)
            if parsed_url.netloc in ('youtube.com', 'youtu.be') \
                    or parsed_url.netloc.endswith(('.youtube.com')):
                # there is a lot of Javascript before the <title>
                if size < 819200:
                    size = max(819200, size)
                    return url
                else:
                    return None
            if parsed_url.netloc in ('reddit.com', 'www.reddit.com', 'new.reddit.com'):
                # Since 2022-03, New Reddit has 'Reddit - Dive into anything' as
                # <title> on every page.
                parsed_url = parsed_url._replace(netloc='old.reddit.com')
                url = utils.web.urlunparse(parsed_url)
                self.log.debug("Rewrite URL to %s", url)
                return url

            return None

        url = url_workaround(url) or url
        timeout = self.registryValue('timeout')
        headers = conf.defaultHttpHeaders(irc.network, msg.channel)
        try:
            fd = utils.web.getUrlFd(url, timeout=timeout, headers=headers)
            target = fd.geturl()
            fixed_target = url_workaround(target)
            if fixed_target is not None:
                # happens when using minification services linking to one of
                # the websites handled by url_workaround; eg. v.redd.it
                fd.close()
                fd = utils.web.getUrlFd(fixed_target, timeout=timeout, headers=headers)
                target = fd.geturl()
            text = fd.read(size)
            response_headers = fd.headers
            fd.close()
        except socket.timeout:
            if raiseErrors:
                irc.error(_('Connection to %s timed out') % url, Raise=True)
            else:
                self.log.info('Web plugins TitleSnarfer: URL <%s> timed out',
                              url)
                return
        except Exception as e:
            if raiseErrors:
                irc.error(_('That URL raised <' + str(e)) + '>',
                          Raise=True)
            else:
                self.log.info('Web plugin TitleSnarfer: URL <%s> raised <%s>',
                              url, str(e))
                return

        encoding = None
        if 'Content-Type' in fd.headers:
            # using p.partition('=') instead of 'p.split('=', 1)' because,
            # unlike RFC 7231, RFC 9110 allows an empty parameter list
            # after ';':
            # * https://www.rfc-editor.org/rfc/rfc9110.html#name-media-type
            # * https://www.rfc-editor.org/rfc/rfc9110.html#parameter
            mime_params = [p.partition('=')
                for p in fd.headers['Content-Type'].split(';')[1:]]
            mime_params = {k.strip(): v.strip() for (k, sep, v) in mime_params}
            if mime_params.get('charset'):
                encoding = mime_params['charset']

        encoding = encoding or utils.web.getEncoding(text) or 'utf8'

        try:
            text = text.decode(encoding, 'replace')
        except UnicodeDecodeError:
            if raiseErrors:
                irc.error(_('Could not guess the page\'s encoding. (Try '
                            'installing python-charade.)'), Raise=True)
            else:
                self.log.info('Web plugin TitleSnarfer: URL <%s> Could '
                              'not guess the page\'s encoding. (Try '
                              'installing python-charade.)', url)
                return
        try:
            parser = Title()
            parser.feed(text)
        except UnicodeDecodeError:
            # Workaround for Python 2
            # https://github.com/progval/Limnoria/issues/1359
            parser = Title()
            parser.feed(text.encode('utf8'))
        parser.close()
        title = utils.str.normalizeWhitespace(''.join(parser.data).strip())
        if title:
            return (target, title)
        elif raiseErrors:
            if len(text) < size:
                irc.error(_('That URL appears to have no HTML title.'),
                        Raise=True)
            else:
                irc.error(format(_('That URL appears to have no HTML title '
                                 'within the first %S.'), size), Raise=True)
        else:
            if len(text) < size:
                self.log.debug('Web plugin TitleSnarfer: URL <%s> appears '
                               'to have no HTML title. ', url)
            else:
                self.log.debug('Web plugin TitleSnarfer: URL <%s> appears '
                               'to have no HTML title within the first %S.',
                               url, size)

    @fetch_sandbox
    def titleSnarfer(self, irc, msg, match):
        channel = msg.channel
        network = irc.network
        if not channel:
            return
        if callbacks.addressed(irc, msg):
            return
        if self.registryValue('titleSnarfer', channel, network):
            url = match.group(0)
            if not self._checkURLWhitelist(irc, msg, url):
                return
            r = self.registryValue('nonSnarfingRegexp', channel, network)
            if r and r.search(url):
                self.log.debug('Not titleSnarfing %q.', url)
                return
            r = self.getTitle(irc, url, False, msg)
            if not r:
                return
            (target, title) = r
            if title:
                domain = utils.web.getDomain(target
                        if self.registryValue('snarferShowTargetDomain',
                                              channel, network)
                        else url)
                prefix = self.registryValue('snarferPrefix', channel, network)
                if prefix:
                    s = "%s %s" % (prefix, title)
                else:
                    s = title
                if self.registryValue('snarferShowDomain', channel, network):
                    s += format(_(' (at %s)'), domain)
                irc.reply(s, prefixNick=False)
        if self.registryValue('snarfMultipleUrls', channel, network):
            # FIXME: hack
            msg.tag('repliedTo', False)
    titleSnarfer = urlSnarfer(titleSnarfer)
    titleSnarfer.__doc__ = utils.web._httpUrlRe

    def _checkURLWhitelist(self, irc, msg, url):
        if not self.registryValue('urlWhitelist',
                                  channel=msg.channel, network=irc.network):
            return True
        passed = False
        for wu in self.registryValue('urlWhitelist',
                                     channel=msg.channel, network=irc.network):
            if wu.endswith('/') and url.find(wu) == 0:
                passed = True
                break
            if (not wu.endswith('/')) and (url.find(wu + '/') == 0 or url == wu):
                passed = True
                break
        return passed

    @wrap(['httpUrl'])
    @catch_web_errors
    @fetch_sandbox
    def headers(self, irc, msg, args, url):
        """<url>

        Returns the HTTP headers of <url>.  Only HTTP urls are valid, of
        course.
        """
        if not self._checkURLWhitelist(irc, msg, url):
            irc.error("This url is not on the whitelist.")
            return
        timeout = self.registryValue('timeout')
        fd = utils.web.getUrlFd(url, timeout=timeout)
        try:
            s = ', '.join([format(_('%s: %s'), k, v)
                           for (k, v) in fd.headers.items()])
            irc.reply(s)
        finally:
            fd.close()

    @wrap(['httpUrl'])
    @catch_web_errors
    @fetch_sandbox
    def location(self, irc, msg, args, url):
        """<url>

        If the <url> is redirected to another page, returns the URL of that
        page. This works even if there are multiple redirects.
        Only HTTP urls are valid.
        Useful to "un-tinify" URLs."""
        timeout = self.registryValue('timeout')
        (target, text) = utils.web.getUrlTargetAndContent(url, size=60,
            timeout=timeout)
        irc.reply(target)

    _doctypeRe = re.compile(r'(<!DOCTYPE[^>]+>)', re.M)
    @wrap(['httpUrl'])
    @catch_web_errors
    @fetch_sandbox
    def doctype(self, irc, msg, args, url):
        """<url>

        Returns the DOCTYPE string of <url>.  Only HTTP urls are valid, of
        course.
        """
        if not self._checkURLWhitelist(irc, msg, url):
            irc.error("This url is not on the whitelist.")
            return
        size = conf.supybot.protocols.http.peekSize()
        timeout = self.registryValue('timeout')
        s = utils.web.getUrl(url, size=size, timeout=timeout).decode('utf8')
        m = self._doctypeRe.search(s)
        if m:
            s = utils.str.normalizeWhitespace(m.group(0))
            irc.reply(s)
        else:
            irc.reply(_('That URL has no specified doctype.'))

    @wrap(['httpUrl'])
    @catch_web_errors
    @fetch_sandbox
    def size(self, irc, msg, args, url):
        """<url>

        Returns the Content-Length header of <url>.  Only HTTP urls are valid,
        of course.
        """
        if not self._checkURLWhitelist(irc, msg, url):
            irc.error("This url is not on the whitelist.")
            return
        timeout = self.registryValue('timeout')
        fd = utils.web.getUrlFd(url, timeout=timeout)
        try:
            try:
                size = fd.headers['Content-Length']
                if size is None:
                    raise KeyError('content-length')
                irc.reply(format(_('%u is %S long.'), url, int(size)))
            except KeyError:
                size = conf.supybot.protocols.http.peekSize()
                s = fd.read(size)
                if len(s) != size:
                    irc.reply(format(_('%u is %S long.'), url, len(s)))
                else:
                    irc.reply(format(_('The server didn\'t tell me how long %u '
                                     'is but it\'s longer than %S.'),
                                     url, size))
        finally:
            fd.close()

    @wrap([getopts({'no-filter': ''}), 'httpUrl'])
    @catch_web_errors
    @fetch_sandbox
    def title(self, irc, msg, args, optlist, url):
        """[--no-filter] <url>

        Returns the HTML <title>...</title> of a URL.
        If --no-filter is given, the bot won't strip special chars (action,
        DCC, ...).
        """
        if not self._checkURLWhitelist(irc, msg, url):
            irc.error("This url is not on the whitelist.")
            return
        r = self.getTitle(irc, url, True, msg)
        if not r:
            return
        (target, title) = r
        if title:
            if not [y for x,y in optlist if x == 'no-filter']:
                for i in range(1, 4):
                    title = title.replace(chr(i), '')
            irc.reply(title)

    @wrap(['text'])
    def urlquote(self, irc, msg, args, text):
        """<text>

        Returns the URL quoted form of the text.
        """
        irc.reply(utils.web.urlquote(text))

    @wrap(['text'])
    def urlunquote(self, irc, msg, args, text):
        """<text>

        Returns the text un-URL quoted.
        """
        s = utils.web.urlunquote(text)
        irc.reply(s)

    @wrap(['url'])
    @catch_web_errors
    @fetch_sandbox
    def fetch(self, irc, msg, args, url):
        """<url>

        Returns the contents of <url>, or as much as is configured in
        supybot.plugins.Web.fetch.maximum.  If that configuration variable is
        set to 0, this command will be effectively disabled.
        """
        if not self._checkURLWhitelist(irc, msg, url):
            irc.error("This url is not on the whitelist.")
            return
        max = self.registryValue('fetch.maximum')
        timeout = self.registryValue('fetch.timeout')
        if not max:
            irc.error(_('This command is disabled '
                      '(supybot.plugins.Web.fetch.maximum is set to 0).'),
                      Raise=True)
        fd = utils.web.getUrl(url, size=max, timeout=timeout).decode('utf8')
        irc.reply(fd)

Class = Web

# vim:set shiftwidth=4 softtabstop=4 expandtab textwidth=79: