view mercurial/revlogutils/sidedata.py @ 46472:98e39f04d60e

upgrade: implement partial upgrade for upgrading persistent-nodemap Upgrading repositories to use persistent nodemap should be fast and easy as it requires only two things: 1) Updating the requirements 2) Writing a persistent-nodemap on disk For both of the steps above, we don't need to edit existing revlogs. This patch makes upgrade only do the above mentioned two steps if we are only upgarding to use persistent-nodemap feature. Since `nodemap.persist_nodemap()` assumes that there exists a nodemap file for the given revlog if we are trying to call it, this patch adds `force` argument to create a file if does not exist which is true in our upgrade case. The test changes demonstrate that we no longer write nodemap files for manifest after upgrade which I think is desirable. Differential Revision: https://phab.mercurial-scm.org/D9936
author Pulkit Goyal <7895pulkit@gmail.com>
date Mon, 01 Feb 2021 00:02:00 +0530
parents d6a9e690d620
children 3d740058b467
line wrap: on
line source

# sidedata.py - Logic around store extra data alongside revlog revisions
#
# Copyright 2019 Pierre-Yves David <pierre-yves.david@octobus.net)
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
"""core code for "sidedata" support

The "sidedata" are stored alongside the revision without actually being part of
its content and not affecting its hash. It's main use cases is to cache
important information related to a changesets.

The current implementation is experimental and subject to changes. Do not rely
on it in production.

Sidedata are stored in the revlog itself, within the revision rawtext. They
are inserted and removed from it using the flagprocessors mechanism. The following
format is currently used::

    initial header:
        <number of sidedata; 2 bytes>
    sidedata (repeated N times):
        <sidedata-key; 2 bytes>
        <sidedata-entry-length: 4 bytes>
        <sidedata-content-sha1-digest: 20 bytes>
        <sidedata-content; X bytes>
    normal raw text:
        <all bytes remaining in the rawtext>

This is a simple and effective format. It should be enough to experiment with
the concept.
"""

from __future__ import absolute_import

import struct

from .. import error
from ..utils import hashutil

## sidedata type constant
# reserve a block for testing purposes.
SD_TEST1 = 1
SD_TEST2 = 2
SD_TEST3 = 3
SD_TEST4 = 4
SD_TEST5 = 5
SD_TEST6 = 6
SD_TEST7 = 7

# key to store copies related information
SD_P1COPIES = 8
SD_P2COPIES = 9
SD_FILESADDED = 10
SD_FILESREMOVED = 11
SD_FILES = 12

# internal format constant
SIDEDATA_HEADER = struct.Struct('>H')
SIDEDATA_ENTRY = struct.Struct('>HL20s')


def sidedatawriteprocessor(rl, text, sidedata):
    sidedata = list(sidedata.items())
    sidedata.sort()
    rawtext = [SIDEDATA_HEADER.pack(len(sidedata))]
    for key, value in sidedata:
        digest = hashutil.sha1(value).digest()
        rawtext.append(SIDEDATA_ENTRY.pack(key, len(value), digest))
    for key, value in sidedata:
        rawtext.append(value)
    rawtext.append(bytes(text))
    return b''.join(rawtext), False


def sidedatareadprocessor(rl, text):
    sidedata = {}
    offset = 0
    (nbentry,) = SIDEDATA_HEADER.unpack(text[: SIDEDATA_HEADER.size])
    offset += SIDEDATA_HEADER.size
    dataoffset = SIDEDATA_HEADER.size + (SIDEDATA_ENTRY.size * nbentry)
    for i in range(nbentry):
        nextoffset = offset + SIDEDATA_ENTRY.size
        key, size, storeddigest = SIDEDATA_ENTRY.unpack(text[offset:nextoffset])
        offset = nextoffset
        # read the data associated with that entry
        nextdataoffset = dataoffset + size
        entrytext = text[dataoffset:nextdataoffset]
        readdigest = hashutil.sha1(entrytext).digest()
        if storeddigest != readdigest:
            raise error.SidedataHashError(key, storeddigest, readdigest)
        sidedata[key] = entrytext
        dataoffset = nextdataoffset
    text = text[dataoffset:]
    return text, True, sidedata


def sidedatarawprocessor(rl, text):
    # side data modifies rawtext and prevent rawtext hash validation
    return False


processors = (
    sidedatareadprocessor,
    sidedatawriteprocessor,
    sidedatarawprocessor,
)