hgext/convert/cvs.py
author Gregory Szorc <gregory.szorc@gmail.com>
Mon, 30 Apr 2018 15:32:11 -0700
branchstable
changeset 37832 6169d95dce3b
parent 37658 34758397ad1b
child 39823 24e493ec2229
permissions -rw-r--r--
httppeer: detect redirect to URL without query string (issue5860) 197d10e157ce subtly changed the HTTP peer's handling of HTTP redirects. Before that changeset, we instantiated an HTTP peer instance and performed the capabilities lookup with that instance. The old code had the following relevant properties: 1) The HTTP request layer would automatically follow HTTP redirects. 2) An encountered HTTP redirect would update a peer instance variable pointing to the repo URL. 3) The peer would automagically perform a "capabilities" command request if a caller requested capabilities but capabilities were not yet defined. The first HTTP request issued by a peer is for ?cmd=capabilities. If the server responds with an HTTP redirect to a ?cmd=capabilities URL, the HTTP request layer automatically followed it, retrieved a valid capabilities response, and the peer's base URL was updated automatically so subsequent requests used the proper URL. In other words, things "just worked." In the case where the server redirected to a URL without the ?cmd=capabilities query string, the HTTP request layer would follow the redirect and likely encounter HTML. The peer's base URL would be updated and the unexpected Content-Type would raise a RepoError. We would catch RepoError and immediately call between() (testing the case for pre 0.9.1 servers not supporting the "capabilities" command). e.g. try: inst._fetchcaps() except error.RepoError: inst.between([(nullid, nullid)]) between() would eventually call into _callstream(). And _callstream() made a call to self.capable('httpheader'). capable() would call self.capabilities(), which would see that no capabilities were set (because HTML was returned for that request) and call the "capabilities" command to fetch capabilities. Because the base URL had been updated from the redirect, this 2nd "capabilities" command would succeed and the client would immediately call "between," which would also succeed. The legacy handshake succeeded. Only because "capabilities" was successfully executed as a side effect did the peer recognize that it was talking to a modern server. In other words, this all appeared to work accidentally. After 197d10e157ce, we stopped calling the "capabilities" command on the peer instance. Instead, we made the request via a low-level opener, detected the redirect as part of response handling code, and passed the redirected URL into the constructed peer instance. For cases where the redirected URL included the query string, this "just worked." But for cases where the redirected URL stripped the query string, we threw RepoError and because we removed the "between" handshake fallback, we fell through to the "is a static HTTP repo" check and performed an HTTP request for .hg/requires. While 197d10e157ce was marked as backwards incompatible, the only intended backwards incompatible behavior was not performing the "between" fallback. It was not realized that the "between" command had the side-effect of recovering from an errant redirect that dropped the query string. This commit restores the previous behavior and allows clients to handle a redirect that drops the query string. In the case where the request is redirected and the query string is dropped, we raise a special case of RepoError. We then catch this special exception in the handshake code and perform another "capabilities" request against the redirected URL. If that works, all is well. Otherwise, we fall back to the "is a static HTTP repo" check. The new code is arguably better than before 197d10e157ce, as it is explicit about the expected behavior and we avoid performing a "between" request, saving a server round trip. Differential Revision: https://phab.mercurial-scm.org/D3433

# cvs.py: CVS conversion code inspired by hg-cvs-import and git-cvsimport
#
#  Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others
#
# 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 errno
import os
import re
import socket

from mercurial.i18n import _
from mercurial import (
    encoding,
    error,
    pycompat,
    util,
)
from mercurial.utils import (
    dateutil,
    procutil,
)

from . import (
    common,
    cvsps,
)

stringio = util.stringio
checktool = common.checktool
commit = common.commit
converter_source = common.converter_source
makedatetimestamp = common.makedatetimestamp
NoRepo = common.NoRepo

class convert_cvs(converter_source):
    def __init__(self, ui, repotype, path, revs=None):
        super(convert_cvs, self).__init__(ui, repotype, path, revs=revs)

        cvs = os.path.join(path, "CVS")
        if not os.path.exists(cvs):
            raise NoRepo(_("%s does not look like a CVS checkout") % path)

        checktool('cvs')

        self.changeset = None
        self.files = {}
        self.tags = {}
        self.lastbranch = {}
        self.socket = None
        self.cvsroot = open(os.path.join(cvs, "Root"), 'rb').read()[:-1]
        self.cvsrepo = open(os.path.join(cvs, "Repository"), 'rb').read()[:-1]
        self.encoding = encoding.encoding

        self._connect()

    def _parse(self):
        if self.changeset is not None:
            return
        self.changeset = {}

        maxrev = 0
        if self.revs:
            if len(self.revs) > 1:
                raise error.Abort(_('cvs source does not support specifying '
                                   'multiple revs'))
            # TODO: handle tags
            try:
                # patchset number?
                maxrev = int(self.revs[0])
            except ValueError:
                raise error.Abort(_('revision %s is not a patchset number')
                                 % self.revs[0])

        d = pycompat.getcwd()
        try:
            os.chdir(self.path)
            id = None

            cache = 'update'
            if not self.ui.configbool('convert', 'cvsps.cache'):
                cache = None
            db = cvsps.createlog(self.ui, cache=cache)
            db = cvsps.createchangeset(self.ui, db,
                fuzz=int(self.ui.config('convert', 'cvsps.fuzz')),
                mergeto=self.ui.config('convert', 'cvsps.mergeto'),
                mergefrom=self.ui.config('convert', 'cvsps.mergefrom'))

            for cs in db:
                if maxrev and cs.id > maxrev:
                    break
                id = (b"%d" % cs.id)
                cs.author = self.recode(cs.author)
                self.lastbranch[cs.branch] = id
                cs.comment = self.recode(cs.comment)
                if self.ui.configbool('convert', 'localtimezone'):
                    cs.date = makedatetimestamp(cs.date[0])
                date = dateutil.datestr(cs.date, '%Y-%m-%d %H:%M:%S %1%2')
                self.tags.update(dict.fromkeys(cs.tags, id))

                files = {}
                for f in cs.entries:
                    files[f.file] = "%s%s" % ('.'.join([(b"%d" % x)
                                                        for x in f.revision]),
                                              ['', '(DEAD)'][f.dead])

                # add current commit to set
                c = commit(author=cs.author, date=date,
                           parents=[(b"%d" % p.id) for p in cs.parents],
                           desc=cs.comment, branch=cs.branch or '')
                self.changeset[id] = c
                self.files[id] = files

            self.heads = self.lastbranch.values()
        finally:
            os.chdir(d)

    def _connect(self):
        root = self.cvsroot
        conntype = None
        user, host = None, None
        cmd = ['cvs', 'server']

        self.ui.status(_("connecting to %s\n") % root)

        if root.startswith(":pserver:"):
            root = root[9:]
            m = re.match(r'(?:(.*?)(?::(.*?))?@)?([^:\/]*)(?::(\d*))?(.*)',
                         root)
            if m:
                conntype = "pserver"
                user, passw, serv, port, root = m.groups()
                if not user:
                    user = "anonymous"
                if not port:
                    port = 2401
                else:
                    port = int(port)
                format0 = ":pserver:%s@%s:%s" % (user, serv, root)
                format1 = ":pserver:%s@%s:%d%s" % (user, serv, port, root)

                if not passw:
                    passw = "A"
                    cvspass = os.path.expanduser("~/.cvspass")
                    try:
                        pf = open(cvspass, 'rb')
                        for line in pf.read().splitlines():
                            part1, part2 = line.split(' ', 1)
                            # /1 :pserver:user@example.com:2401/cvsroot/foo
                            # Ah<Z
                            if part1 == '/1':
                                part1, part2 = part2.split(' ', 1)
                                format = format1
                            # :pserver:user@example.com:/cvsroot/foo Ah<Z
                            else:
                                format = format0
                            if part1 == format:
                                passw = part2
                                break
                        pf.close()
                    except IOError as inst:
                        if inst.errno != errno.ENOENT:
                            if not getattr(inst, 'filename', None):
                                inst.filename = cvspass
                            raise

                sck = socket.socket()
                sck.connect((serv, port))
                sck.send("\n".join(["BEGIN AUTH REQUEST", root, user, passw,
                                    "END AUTH REQUEST", ""]))
                if sck.recv(128) != "I LOVE YOU\n":
                    raise error.Abort(_("CVS pserver authentication failed"))

                self.writep = self.readp = sck.makefile('r+')

        if not conntype and root.startswith(":local:"):
            conntype = "local"
            root = root[7:]

        if not conntype:
            # :ext:user@host/home/user/path/to/cvsroot
            if root.startswith(":ext:"):
                root = root[5:]
            m = re.match(br'(?:([^@:/]+)@)?([^:/]+):?(.*)', root)
            # Do not take Windows path "c:\foo\bar" for a connection strings
            if os.path.isdir(root) or not m:
                conntype = "local"
            else:
                conntype = "rsh"
                user, host, root = m.group(1), m.group(2), m.group(3)

        if conntype != "pserver":
            if conntype == "rsh":
                rsh = encoding.environ.get("CVS_RSH") or "ssh"
                if user:
                    cmd = [rsh, '-l', user, host] + cmd
                else:
                    cmd = [rsh, host] + cmd

            # popen2 does not support argument lists under Windows
            cmd = [procutil.shellquote(arg) for arg in cmd]
            cmd = procutil.quotecommand(' '.join(cmd))
            self.writep, self.readp = procutil.popen2(cmd)

        self.realroot = root

        self.writep.write("Root %s\n" % root)
        self.writep.write("Valid-responses ok error Valid-requests Mode"
                          " M Mbinary E Checked-in Created Updated"
                          " Merged Removed\n")
        self.writep.write("valid-requests\n")
        self.writep.flush()
        r = self.readp.readline()
        if not r.startswith("Valid-requests"):
            raise error.Abort(_('unexpected response from CVS server '
                               '(expected "Valid-requests", but got %r)')
                             % r)
        if "UseUnchanged" in r:
            self.writep.write("UseUnchanged\n")
            self.writep.flush()
            r = self.readp.readline()

    def getheads(self):
        self._parse()
        return self.heads

    def getfile(self, name, rev):

        def chunkedread(fp, count):
            # file-objects returned by socket.makefile() do not handle
            # large read() requests very well.
            chunksize = 65536
            output = stringio()
            while count > 0:
                data = fp.read(min(count, chunksize))
                if not data:
                    raise error.Abort(_("%d bytes missing from remote file")
                                     % count)
                count -= len(data)
                output.write(data)
            return output.getvalue()

        self._parse()
        if rev.endswith("(DEAD)"):
            return None, None

        args = ("-N -P -kk -r %s --" % rev).split()
        args.append(self.cvsrepo + '/' + name)
        for x in args:
            self.writep.write("Argument %s\n" % x)
        self.writep.write("Directory .\n%s\nco\n" % self.realroot)
        self.writep.flush()

        data = ""
        mode = None
        while True:
            line = self.readp.readline()
            if line.startswith("Created ") or line.startswith("Updated "):
                self.readp.readline() # path
                self.readp.readline() # entries
                mode = self.readp.readline()[:-1]
                count = int(self.readp.readline()[:-1])
                data = chunkedread(self.readp, count)
            elif line.startswith(" "):
                data += line[1:]
            elif line.startswith("M "):
                pass
            elif line.startswith("Mbinary "):
                count = int(self.readp.readline()[:-1])
                data = chunkedread(self.readp, count)
            else:
                if line == "ok\n":
                    if mode is None:
                        raise error.Abort(_('malformed response from CVS'))
                    return (data, "x" in mode and "x" or "")
                elif line.startswith("E "):
                    self.ui.warn(_("cvs server: %s\n") % line[2:])
                elif line.startswith("Remove"):
                    self.readp.readline()
                else:
                    raise error.Abort(_("unknown CVS response: %s") % line)

    def getchanges(self, rev, full):
        if full:
            raise error.Abort(_("convert from cvs does not support --full"))
        self._parse()
        return sorted(self.files[rev].iteritems()), {}, set()

    def getcommit(self, rev):
        self._parse()
        return self.changeset[rev]

    def gettags(self):
        self._parse()
        return self.tags

    def getchangedfiles(self, rev, i):
        self._parse()
        return sorted(self.files[rev])