# blobstore.py - local and remote (speaking Git-LFS protocol) blob storages

#

# Copyright 2017 Facebook, Inc.

#

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

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

import contextlib

import errno

import hashlib

import json

import os

import re

import socket

from mercurial.i18n import _

from mercurial.node import hex

from mercurial import (

    encoding,

    error,

    httpconnection as httpconnectionmod,

    pathutil,

    pycompat,

    url as urlmod,

    util,

    vfs as vfsmod,

    worker,

)

from mercurial.utils import (

    stringutil,

    urlutil,

)

from ..largefiles import lfutil

# 64 bytes for SHA256

_lfsre = re.compile(br'\A[a-f0-9]{64}\Z')

class lfsvfs(vfsmod.vfs):

    def join(self, path):

        """split the path at first two characters, like: XX/XXXXX..."""

        if not _lfsre.match(path):

            raise error.ProgrammingError(b'unexpected lfs path: %s' % path)

        return super(lfsvfs, self).join(path[0:2], path[2:])

    def walk(self, path=None, onerror=None):

        """Yield (dirpath, [], oids) tuple for blobs under path

        Oids only exist in the root of this vfs, so dirpath is always ''.

        """

        root = os.path.normpath(self.base)

        # when dirpath == root, dirpath[prefixlen:] becomes empty

        # because len(dirpath) < prefixlen.

        prefixlen = len(pathutil.normasprefix(root))

        oids = []

        for dirpath, dirs, files in os.walk(

            self.reljoin(self.base, path or b''), onerror=onerror

        ):

            dirpath = dirpath[prefixlen:]

            # Silently skip unexpected files and directories

            if len(dirpath) == 2:

                oids.extend(

                    [dirpath + f for f in files if _lfsre.match(dirpath + f)]

                )

        yield (b'', [], oids)

class nullvfs(lfsvfs):

    def __init__(self):

        pass

    def exists(self, oid):

        return False

    def read(self, oid):

        # store.read() calls into here if the blob doesn't exist in its

        # self.vfs.  Raise the same error as a normal vfs when asked to read a

        # file that doesn't exist.  The only difference is the full file path

        # isn't available in the error.

        raise IOError(

            errno.ENOENT,

            pycompat.sysstr(b'%s: No such file or directory' % oid),

        )

    def walk(self, path=None, onerror=None):

        return (b'', [], [])

    def write(self, oid, data):

        pass

class lfsuploadfile(httpconnectionmod.httpsendfile):

    """a file-like object that supports keepalive."""

    def __init__(self, ui, filename):

        super(lfsuploadfile, self).__init__(ui, filename, b'rb')

        self.read = self._data.read

    def _makeprogress(self):

        return None  # progress is handled by the worker client

class local:

    """Local blobstore for large file contents.

    This blobstore is used both as a cache and as a staging area for large blobs

    to be uploaded to the remote blobstore.

    """

    def __init__(self, repo):

        fullpath = repo.svfs.join(b'lfs/objects')

        self.vfs = lfsvfs(fullpath)

        if repo.ui.configbool(b'experimental', b'lfs.disableusercache'):

            self.cachevfs = nullvfs()

        else:

            usercache = lfutil._usercachedir(repo.ui, b'lfs')

            self.cachevfs = lfsvfs(usercache)

        self.ui = repo.ui

    def open(self, oid):

        """Open a read-only file descriptor to the named blob, in either the

        usercache or the local store."""

        return open(self.path(oid), 'rb')

    def path(self, oid):

        """Build the path for the given blob ``oid``.

        If the blob exists locally, the path may point to either the usercache

        or the local store.  If it doesn't, it will point to the local store.

        This is meant for situations where existing code that isn't LFS aware

        needs to open a blob.  Generally, prefer the ``open`` method on this

        class.

        """

        # The usercache is the most likely place to hold the file.  Commit will

        # write to both it and the local store, as will anything that downloads

        # the blobs.  However, things like clone without an update won't

        # populate the local store.  For an init + push of a local clone,

        # the usercache is the only place it _could_ be.  If not present, the

        # missing file msg here will indicate the local repo, not the usercache.

        if self.cachevfs.exists(oid):

            return self.cachevfs.join(oid)

        return self.vfs.join(oid)

    def download(self, oid, src, content_length):

        """Read the blob from the remote source in chunks, verify the content,

        and write to this local blobstore."""

        sha256 = hashlib.sha256()

        size = 0

        with self.vfs(oid, b'wb', atomictemp=True) as fp:

            for chunk in util.filechunkiter(src, size=1048576):

                fp.write(chunk)

                sha256.update(chunk)

                size += len(chunk)

            # If the server advertised a length longer than what we actually

            # received, then we should expect that the server crashed while

            # producing the response (but the server has no way of telling us

            # that), and we really don't need to try to write the response to

            # the localstore, because it's not going to match the expected.

            # The server also uses this method to store data uploaded by the

            # client, so if this happens on the server side, it's possible

            # that the client crashed or an antivirus interfered with the

            # upload.

            if content_length is not None and int(content_length) != size:

                msg = (

                    b"Response length (%d) does not match Content-Length "

                    b"header (%d) for %s"

                )

                raise LfsRemoteError(_(msg) % (size, int(content_length), oid))

            realoid = hex(sha256.digest())

            if realoid != oid:

                raise LfsCorruptionError(

                    _(b'corrupt remote lfs object: %s') % oid

                )

        self._linktousercache(oid)

    def write(self, oid, data):

        """Write blob to local blobstore.

        This should only be called from the filelog during a commit or similar.

        As such, there is no need to verify the data.  Imports from a remote

        store must use ``download()`` instead."""

        with self.vfs(oid, b'wb', atomictemp=True) as fp:

            fp.write(data)

        self._linktousercache(oid)

    def linkfromusercache(self, oid):

        """Link blobs found in the user cache into this store.

        The server module needs to do this when it lets the client know not to

        upload the blob, to ensure it is always available in this store.

        Normally this is done implicitly when the client reads or writes the

        blob, but that doesn't happen when the server tells the client that it

        already has the blob.

        """

        if not isinstance(self.cachevfs, nullvfs) and not self.vfs.exists(oid):

            self.ui.note(_(b'lfs: found %s in the usercache\n') % oid)

            lfutil.link(self.cachevfs.join(oid), self.vfs.join(oid))

    def _linktousercache(self, oid):

        # XXX: should we verify the content of the cache, and hardlink back to

        # the local store on success, but truncate, write and link on failure?

        if not self.cachevfs.exists(oid) and not isinstance(

            self.cachevfs, nullvfs

        ):

            self.ui.note(_(b'lfs: adding %s to the usercache\n') % oid)

            lfutil.link(self.vfs.join(oid), self.cachevfs.join(oid))

    def read(self, oid, verify=True):

        """Read blob from local blobstore."""

        if not self.vfs.exists(oid):

            blob = self._read(self.cachevfs, oid, verify)

            # Even if revlog will verify the content, it needs to be verified

            # now before making the hardlink to avoid propagating corrupt blobs.

            # Don't abort if corruption is detected, because `hg verify` will

            # give more useful info about the corruption- simply don't add the

            # hardlink.

            if verify or hex(hashlib.sha256(blob).digest()) == oid:

                self.ui.note(_(b'lfs: found %s in the usercache\n') % oid)

                lfutil.link(self.cachevfs.join(oid), self.vfs.join(oid))

        else:

            self.ui.note(_(b'lfs: found %s in the local lfs store\n') % oid)

            blob = self._read(self.vfs, oid, verify)

        return blob

    def _read(self, vfs, oid, verify):

        """Read blob (after verifying) from the given store"""

        blob = vfs.read(oid)

        if verify:

            _verify(oid, blob)

        return blob

    def verify(self, oid):

        """Indicate whether or not the hash of the underlying file matches its

        name."""

        sha256 = hashlib.sha256()

        with self.open(oid) as fp:

            for chunk in util.filechunkiter(fp, size=1048576):

                sha256.update(chunk)

        return oid == hex(sha256.digest())

    def has(self, oid):

        """Returns True if the local blobstore contains the requested blob,

        False otherwise."""

        return self.cachevfs.exists(oid) or self.vfs.exists(oid)

def _urlerrorreason(urlerror):

    """Create a friendly message for the given URLError to be used in an

    LfsRemoteError message.

    """

    inst = urlerror

    if isinstance(urlerror.reason, Exception):

        inst = urlerror.reason

    if hasattr(inst, 'reason'):

        try:  # usually it is in the form (errno, strerror)

            reason = inst.reason.args[1]

        except (AttributeError, IndexError):

            # it might be anything, for example a string

            reason = inst.reason

        if isinstance(reason, str):

            # SSLError of Python 2.7.9 contains a unicode

            reason = encoding.unitolocal(reason)

        return reason

    elif getattr(inst, "strerror", None):

        return encoding.strtolocal(inst.strerror)

    else:

        return stringutil.forcebytestr(urlerror)

class lfsauthhandler(util.urlreq.basehandler):

    handler_order = 480  # Before HTTPDigestAuthHandler (== 490)

    def http_error_401(self, req, fp, code, msg, headers):

        """Enforces that any authentication performed is HTTP Basic

        Authentication.  No authentication is also acceptable.

        """

        authreq = headers.get('www-authenticate', None)

        if authreq:

            scheme = authreq.split()[0]

            if scheme.lower() != 'basic':

                msg = _(b'the server must support Basic Authentication')

                raise util.urlerr.httperror(

                    req.get_full_url(),

                    code,

                    encoding.strfromlocal(msg),

                    headers,

                    fp,

                )

        return None

class _gitlfsremote:

    def __init__(self, repo, url):

        ui = repo.ui

        self.ui = ui

        baseurl, authinfo = url.authinfo()

        self.baseurl = baseurl.rstrip(b'/')

        useragent = repo.ui.config(b'experimental', b'lfs.user-agent')

        if not useragent:

            useragent = b'git-lfs/2.3.4 (Mercurial %s)' % util.version()

        self.urlopener = urlmod.opener(ui, authinfo, useragent)

        self.urlopener.add_handler(lfsauthhandler())

        self.retry = ui.configint(b'lfs', b'retry')

    def writebatch(self, pointers, fromstore):

        """Batch upload from local to remote blobstore."""

        self._batch(_deduplicate(pointers), fromstore, b'upload')

    def readbatch(self, pointers, tostore):

        """Batch download from remote to local blostore."""

        self._batch(_deduplicate(pointers), tostore, b'download')

    def _batchrequest(self, pointers, action):

        """Get metadata about objects pointed by pointers for given action

        Return decoded JSON object like {'objects': [{'oid': '', 'size': 1}]}

        See https://github.com/git-lfs/git-lfs/blob/master/docs/api/batch.md

        """

        objects = [

            {'oid': pycompat.strurl(p.oid()), 'size': p.size()}

            for p in pointers

        ]

        requestdata = pycompat.bytesurl(

            json.dumps(

                {

                    'objects': objects,

                    'operation': pycompat.strurl(action),

                }

            )

        )

        url = b'%s/objects/batch' % self.baseurl

        batchreq = util.urlreq.request(pycompat.strurl(url), data=requestdata)

        batchreq.add_header('Accept', 'application/vnd.git-lfs+json')

        batchreq.add_header('Content-Type', 'application/vnd.git-lfs+json')

        try:

            with contextlib.closing(self.urlopener.open(batchreq)) as rsp:

                rawjson = rsp.read()

        except util.urlerr.httperror as ex:

            hints = {

                400: _(

                    b'check that lfs serving is enabled on %s and "%s" is '

                    b'supported'

                )

                % (self.baseurl, action),

                404: _(b'the "lfs.url" config may be used to override %s')

                % self.baseurl,

            }

            hint = hints.get(ex.code, _(b'api=%s, action=%s') % (url, action))

            raise LfsRemoteError(

                _(b'LFS HTTP error: %s') % stringutil.forcebytestr(ex),

                hint=hint,

            )

        except util.urlerr.urlerror as ex:

            hint = (

                _(b'the "lfs.url" config may be used to override %s')

                % self.baseurl

            )

            raise LfsRemoteError(

                _(b'LFS error: %s') % _urlerrorreason(ex), hint=hint

            )

        try: