view mercurial/revlogutils/sidedata.py @ 45095:8e04607023e5

procutil: ensure that procutil.std{out,err}.write() writes all bytes

Python 3 offers different kind of streams and it’s not guaranteed for all of
them that calling write() writes all bytes.

When Python is started in unbuffered mode, sys.std{out,err}.buffer are
instances of io.FileIO, whose write() can write less bytes for
platform-specific reasons (e.g. Linux has a 0x7ffff000 bytes maximum and could
write less if interrupted by a signal; when writing to Windows consoles, it’s
limited to 32767 bytes to avoid the "not enough space" error). This can lead to
silent loss of data, both when using sys.std{out,err}.buffer (which may in fact
not be a buffered stream) and when using the text streams sys.std{out,err}
(I’ve created a CPython bug report for that:
https://bugs.python.org/issue41221).

Python may fix the problem at some point. For now, we implement our own wrapper
for procutil.std{out,err} that calls the raw stream’s write() method until all
bytes have been written. We don’t use sys.std{out,err} for larger writes, so I
think it’s not worth the effort to patch them.

# sidedata.py - Logic around store extra data alongside revlog revisions
#
# Copyright 2019 Pierre-Yves David <[e-mail])
#
# 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, withing the revision rawtext. They
are inserted, 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 enought 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

# 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,
)