# url.py - HTTP handling for mercurial

#

# Copyright 2005, 2006, 2007, 2008 Matt Mackall <mpm@selenic.com>

# Copyright 2006, 2007 Alexis S. L. Carvalho <alexis@cecm.usp.br>

# Copyright 2006 Vadim Gelfer <vadim.gelfer@gmail.com>

#

# This software may be used and distributed according to the terms of the

# GNU General Public License version 2 or any later version.

from __future__ import absolute_import

import base64

import os

import socket

import sys

from .i18n import _

from .pycompat import getattr

from . import (

    encoding,

    error,

    httpconnection as httpconnectionmod,

    keepalive,

    pycompat,

    sslutil,

    urllibcompat,

    util,

)

from .utils import stringutil

httplib = util.httplib

stringio = util.stringio

urlerr = util.urlerr

urlreq = util.urlreq

def escape(s, quote=None):

    """Replace special characters "&", "<" and ">" to HTML-safe sequences.

    If the optional flag quote is true, the quotation mark character (")

    is also translated.

    This is the same as cgi.escape in Python, but always operates on

    bytes, whereas cgi.escape in Python 3 only works on unicodes.

    """

    s = s.replace(b"&", b"&")

    s = s.replace(b"<", b"&lt;")

    s = s.replace(b">", b">")

    if quote:

        s = s.replace(b'"', b""")

    return s

class passwordmgr(object):

    def __init__(self, ui, passwddb):

        self.ui = ui

        self.passwddb = passwddb

    def add_password(self, realm, uri, user, passwd):

        return self.passwddb.add_password(realm, uri, user, passwd)

    def find_user_password(self, realm, authuri):

        assert isinstance(realm, (type(None), str))

        assert isinstance(authuri, str)

        authinfo = self.passwddb.find_user_password(realm, authuri)

        user, passwd = authinfo

        user, passwd = pycompat.bytesurl(user), pycompat.bytesurl(passwd)

        if user and passwd:

            self._writedebug(user, passwd)

            return (pycompat.strurl(user), pycompat.strurl(passwd))

        if not user or not passwd:

            res = httpconnectionmod.readauthforuri(self.ui, authuri, user)

            if res:

                group, auth = res

                user, passwd = auth.get(b'username'), auth.get(b'password')

                self.ui.debug(b"using auth.%s.* for authentication\n" % group)

        if not user or not passwd:

            u = util.url(pycompat.bytesurl(authuri))

            u.query = None

            if not self.ui.interactive():

                raise error.Abort(

                    _(b'http authorization required for %s')

                    % util.hidepassword(bytes(u))

                )

            self.ui.write(

                _(b"http authorization required for %s\n")

                % util.hidepassword(bytes(u))

            )

            self.ui.write(_(b"realm: %s\n") % pycompat.bytesurl(realm))

            if user:

                self.ui.write(_(b"user: %s\n") % user)

            else:

                user = self.ui.prompt(_(b"user:"), default=None)

            if not passwd:

                passwd = self.ui.getpass()

        # As of Python 3.8, the default implementation of

        # AbstractBasicAuthHandler.retry_http_basic_auth() assumes the user

        # is set if pw is not None. This means (None, str) is not a valid

        # return type of find_user_password().

        if user is None:

            return None, None

        self.passwddb.add_password(realm, authuri, user, passwd)

        self._writedebug(user, passwd)

        return (pycompat.strurl(user), pycompat.strurl(passwd))

    def _writedebug(self, user, passwd):

        msg = _(b'http auth: user %s, password %s\n')

        self.ui.debug(msg % (user, passwd and b'*' * len(passwd) or b'not set'))

    def find_stored_password(self, authuri):

        return self.passwddb.find_user_password(None, authuri)

class proxyhandler(urlreq.proxyhandler):

    def __init__(self, ui):

        proxyurl = ui.config(b"http_proxy", b"host") or encoding.environ.get(

            b'http_proxy'

        )

        # XXX proxyauthinfo = None

        if proxyurl:

            # proxy can be proper url or host[:port]

            if not (

                proxyurl.startswith(b'http:') or proxyurl.startswith(b'https:')

            ):

                proxyurl = b'http://' + proxyurl + b'/'

            proxy = util.url(proxyurl)

            if not proxy.user:

                proxy.user = ui.config(b"http_proxy", b"user")

                proxy.passwd = ui.config(b"http_proxy", b"passwd")

            # see if we should use a proxy for this url

            no_list = [b"localhost", b"127.0.0.1"]

            no_list.extend(

                [p.lower() for p in ui.configlist(b"http_proxy", b"no")]

            )

            no_list.extend(

                [

                    p.strip().lower()

                    for p in encoding.environ.get(b"no_proxy", b'').split(b',')

                    if p.strip()

                ]

            )

            # "http_proxy.always" config is for running tests on localhost

            if ui.configbool(b"http_proxy", b"always"):

                self.no_list = []

            else:

                self.no_list = no_list

            # Keys and values need to be str because the standard library

            # expects them to be.

            proxyurl = str(proxy)

            proxies = {'http': proxyurl, 'https': proxyurl}

            ui.debug(b'proxying through %s\n' % util.hidepassword(bytes(proxy)))

        else:

            proxies = {}

        urlreq.proxyhandler.__init__(self, proxies)

        self.ui = ui

    def proxy_open(self, req, proxy, type_):

        host = pycompat.bytesurl(urllibcompat.gethost(req)).split(b':')[0]

        for e in self.no_list:

            if host == e:

                return None

            if e.startswith(b'*.') and host.endswith(e[2:]):

                return None

            if e.startswith(b'.') and host.endswith(e[1:]):

                return None

        return urlreq.proxyhandler.proxy_open(self, req, proxy, type_)

def _gen_sendfile(orgsend):

    def _sendfile(self, data):

        # send a file

        if isinstance(data, httpconnectionmod.httpsendfile):

            # if auth required, some data sent twice, so rewind here

            data.seek(0)

            for chunk in util.filechunkiter(data):

                orgsend(self, chunk)

        else:

            orgsend(self, data)

    return _sendfile

has_https = util.safehasattr(urlreq, b'httpshandler')

class httpconnection(keepalive.HTTPConnection):

    # must be able to send big bundle as stream.

    send = _gen_sendfile(keepalive.HTTPConnection.send)

    def getresponse(self):

        proxyres = getattr(self, 'proxyres', None)

        if proxyres:

            if proxyres.will_close:

                self.close()

            self.proxyres = None

            return proxyres

        return keepalive.HTTPConnection.getresponse(self)

# Large parts of this function have their origin from before Python 2.6

# and could potentially be removed.

def _generic_start_transaction(handler, h, req):

    tunnel_host = req._tunnel_host

    if tunnel_host:

        if tunnel_host[:7] not in ['http://', 'https:/']:

            tunnel_host = 'https://' + tunnel_host

        new_tunnel = True

    else:

        tunnel_host = urllibcompat.getselector(req)

        new_tunnel = False

    if new_tunnel or tunnel_host == urllibcompat.getfullurl(req):  # has proxy

        u = util.url(pycompat.bytesurl(tunnel_host))

        if new_tunnel or u.scheme == b'https':  # only use CONNECT for HTTPS

            h.realhostport = b':'.join([u.host, (u.port or b'443')])

            h.headers = req.headers.copy()

            h.headers.update(handler.parent.addheaders)

            return

    h.realhostport = None

    h.headers = None

def _generic_proxytunnel(self):

    proxyheaders = {

        pycompat.bytestr(x): pycompat.bytestr(self.headers[x])

        for x in self.headers

        if x.lower().startswith('proxy-')

    }

    self.send(b'CONNECT %s HTTP/1.0\r\n' % self.realhostport)

    for header in pycompat.iteritems(proxyheaders):

        self.send(b'%s: %s\r\n' % header)

    self.send(b'\r\n')

    # majority of the following code is duplicated from

    # httplib.HTTPConnection as there are no adequate places to

    # override functions to provide the needed functionality

    # strict was removed in Python 3.4.

    kwargs = {}

    if not pycompat.ispy3:

        kwargs[b'strict'] = self.strict

    res = self.response_class(self.sock, method=self._method, **kwargs)

    while True:

        version, status, reason = res._read_status()

        if status != httplib.CONTINUE:

            break

        # skip lines that are all whitespace

        list(iter(lambda: res.fp.readline().strip(), b''))

    res.status = status

    res.reason = reason.strip()

    if res.status == 200:

        # skip lines until we find a blank line

        list(iter(res.fp.readline, b'\r\n'))

        return True

    if version == b'HTTP/1.0':

        res.version = 10

    elif version.startswith(b'HTTP/1.'):

        res.version = 11

    elif version == b'HTTP/0.9':

        res.version = 9

    else:

        raise httplib.UnknownProtocol(version)

    if res.version == 9:

        res.length = None

        res.chunked = 0

        res.will_close = 1

        res.msg = httplib.HTTPMessage(stringio())

        return False

    res.msg = httplib.HTTPMessage(res.fp)

    res.msg.fp = None

    # are we using the chunked-style of transfer encoding?

    trenc = res.msg.getheader(b'transfer-encoding')

    if trenc and trenc.lower() == b"chunked":

        res.chunked = 1

        res.chunk_left = None

    else:

        res.chunked = 0

    # will the connection close at the end of the response?

    res.will_close = res._check_close()

    # do we have a Content-Length?

    # NOTE: RFC 2616, section 4.4, #3 says we ignore this if

    # transfer-encoding is "chunked"

    length = res.msg.getheader(b'content-length')

    if length and not res.chunked:

        try:

            res.length = int(length)

        except ValueError:

            res.length = None

        else:

            if res.length < 0:  # ignore nonsensical negative lengths

                res.length = None

    else:

        res.length = None

    # does the body have a fixed length? (of zero)

    if (

        status == httplib.NO_CONTENT

        or status == httplib.NOT_MODIFIED

        or 100 <= status < 200

        or res._method == b'HEAD'  # 1xx codes

    ):

        res.length = 0

    # if the connection remains open, and we aren't using chunked, and

    # a content-length was not provided, then assume that the connection

    # WILL close.

    if not res.will_close and not res.chunked and res.length is None:

        res.will_close = 1

    self.proxyres = res

    return False

class httphandler(keepalive.HTTPHandler):

    def http_open(self, req):

        return self.do_open(httpconnection, req)

    def _start_transaction(self, h, req):

        _generic_start_transaction(self, h, req)

        return keepalive.HTTPHandler._start_transaction(self, h, req)

class logginghttpconnection(keepalive.HTTPConnection):

    def __init__(self, createconn, *args, **kwargs):

        keepalive.HTTPConnection.__init__(self, *args, **kwargs)

        self._create_connection = createconn

    if sys.version_info < (2, 7, 7):

        # copied from 2.7.14, since old implementations directly call

        # socket.create_connection()

        def connect(self):

            self.sock = self._create_connection(

                (self.host, self.port), self.timeout, self.source_address

            )

            if self._tunnel_host:

                self._tunnel()

class logginghttphandler(httphandler):

    """HTTP handler that logs socket I/O."""

    def __init__(self, logfh, name, observeropts, timeout=None):

        super(logginghttphandler, self).__init__(timeout=timeout)

        self._logfh = logfh

        self._logname = name

        self._observeropts = observeropts