view doc/gendoc.py @ 44363:f7459da77f23

nodemap: introduce an option to use mmap to read the nodemap mapping

The performance and memory benefit is much greater if we don't have to copy all
the data in memory for each information. So we introduce an option (on by
default) to read the data using mmap.

This changeset is the last one definition the API for index support nodemap
data. (they have to be able to use the mmaping).

Below are some benchmark comparing the best we currently have in 5.3 with the
final step of this series (using the persistent nodemap implementation in
Rust). The benchmark run `hg perfindex` with various revset and the following
variants:

Before:

 * do not use the persistent nodemap
 * use the CPython implementation of the index for nodemap
 * use mmapping of the changelog index

After:

 * use the MixedIndex Rust code, with the NodeTree object for nodemap access
   (still in review)
 * use the persistent nodemap data from disk
 * access the persistent nodemap data through mmap
 * use mmapping of the changelog index

The persistent nodemap greatly speed up  most operation on very large
repositories. Some of the previously very fast lookup end up a bit slower because
the persistent nodemap has to be setup. However the absolute slowdown is very
small and won't matters in the big picture.

Here are some numbers (in seconds) for the reference copy of mozilla-try:

Revset            Before   After     abs-change speedup

-10000:           0.004622 0.005532   0.000910  ×   0.83
-10:              0.000050 0.000132   0.000082  ×   0.37
tip               0.000052 0.000085   0.000033  ×   0.61
0 + (-10000:)     0.028222 0.005337  -0.022885  ×   5.29
0                 0.023521 0.000084  -0.023437  × 280.01
(-10000:) + 0     0.235539 0.005308  -0.230231  ×  44.37
(-10:) + :9       0.232883 0.000180  -0.232703  ×1293.79
(-10000:) + (:99) 0.238735 0.005358  -0.233377  ×  44.55
:99 + (-10000:)   0.317942 0.005593  -0.312349  ×  56.84
:9 + (-10:)       0.313372 0.000179  -0.313193  ×1750.68
:9                0.316450 0.000143  -0.316307  ×2212.93


On smaller repositories, the cost of nodemap related operation is not as big, so
the win is much more modest. Yet it helps shaving a handful of millisecond here
and there.

Here are some numbers (in seconds) for the reference copy of mercurial:

Revset            Before   After     abs-change speedup
-10:              0.000065 0.000097   0.000032  × 0.67
tip               0.000063 0.000078   0.000015  × 0.80
0                 0.000561 0.000079  -0.000482  × 7.10
-10000:           0.004609 0.003648  -0.000961  × 1.26
0 + (-10000:)     0.005023 0.003715  -0.001307  × 1.35
(-10:) + :9       0.002187 0.000108  -0.002079  ×20.25
(-10000:) + 0     0.006252 0.003716  -0.002536  × 1.68
(-10000:) + (:99) 0.006367 0.003707  -0.002660  × 1.71
:9 + (-10:)       0.003846 0.000110  -0.003736  ×34.96
:9                0.003854 0.000099  -0.003755  ×38.92
:99 + (-10000:)   0.007644 0.003778  -0.003866  × 2.02

Differential Revision: https://phab.mercurial-scm.org/D7894

#!/usr/bin/env python
"""usage: %s DOC ...

where DOC is the name of a document
"""

from __future__ import absolute_import

import os
import sys
import textwrap

try:
    import msvcrt

    msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
    msvcrt.setmode(sys.stderr.fileno(), os.O_BINARY)
except ImportError:
    pass

# This script is executed during installs and may not have C extensions
# available. Relax C module requirements.
os.environ['HGMODULEPOLICY'] = 'allow'
# import from the live mercurial repo
sys.path.insert(0, "..")
from mercurial import demandimport

demandimport.enable()

from mercurial import (
    commands,
    encoding,
    extensions,
    help,
    minirst,
    pycompat,
    ui as uimod,
)
from mercurial.i18n import (
    gettext,
    _,
)

table = commands.table
globalopts = commands.globalopts
helptable = help.helptable
loaddoc = help.loaddoc


def get_desc(docstr):
    if not docstr:
        return b"", b""
    # sanitize
    docstr = docstr.strip(b"\n")
    docstr = docstr.rstrip()
    shortdesc = docstr.splitlines()[0].strip()

    i = docstr.find(b"\n")
    if i != -1:
        desc = docstr[i + 2 :]
    else:
        desc = shortdesc

    desc = textwrap.dedent(desc.decode('latin1')).encode('latin1')

    return (shortdesc, desc)


def get_opts(opts):
    for opt in opts:
        if len(opt) == 5:
            shortopt, longopt, default, desc, optlabel = opt
        else:
            shortopt, longopt, default, desc = opt
            optlabel = _(b"VALUE")
        allopts = []
        if shortopt:
            allopts.append(b"-%s" % shortopt)
        if longopt:
            allopts.append(b"--%s" % longopt)
        if isinstance(default, list):
            allopts[-1] += b" <%s[+]>" % optlabel
        elif (default is not None) and not isinstance(default, bool):
            allopts[-1] += b" <%s>" % optlabel
        if b'\n' in desc:
            # only remove line breaks and indentation
            desc = b' '.join(l.lstrip() for l in desc.split(b'\n'))
        desc += default and _(b" (default: %s)") % bytes(default) or b""
        yield (b", ".join(allopts), desc)


def get_cmd(cmd, cmdtable):
    d = {}
    attr = cmdtable[cmd]
    cmds = cmd.lstrip(b"^").split(b"|")

    d[b'cmd'] = cmds[0]
    d[b'aliases'] = cmd.split(b"|")[1:]
    d[b'desc'] = get_desc(gettext(pycompat.getdoc(attr[0])))
    d[b'opts'] = list(get_opts(attr[1]))

    s = b'hg ' + cmds[0]
    if len(attr) > 2:
        if not attr[2].startswith(b'hg'):
            s += b' ' + attr[2]
        else:
            s = attr[2]
    d[b'synopsis'] = s.strip()

    return d


def showdoc(ui):
    # print options
    ui.write(minirst.section(_(b"Options")))
    multioccur = False
    for optstr, desc in get_opts(globalopts):
        ui.write(b"%s\n    %s\n\n" % (optstr, desc))
        if optstr.endswith(b"[+]>"):
            multioccur = True
    if multioccur:
        ui.write(_(b"\n[+] marked option can be specified multiple times\n"))
        ui.write(b"\n")

    # print cmds
    ui.write(minirst.section(_(b"Commands")))
    commandprinter(ui, table, minirst.subsection, minirst.subsubsection)

    # print help topics
    # The config help topic is included in the hgrc.5 man page.
    helpprinter(ui, helptable, minirst.section, exclude=[b'config'])

    ui.write(minirst.section(_(b"Extensions")))
    ui.write(
        _(
            b"This section contains help for extensions that are "
            b"distributed together with Mercurial. Help for other "
            b"extensions is available in the help system."
        )
    )
    ui.write(
        (
            b"\n\n"
            b".. contents::\n"
            b"   :class: htmlonly\n"
            b"   :local:\n"
            b"   :depth: 1\n\n"
        )
    )

    for extensionname in sorted(allextensionnames()):
        mod = extensions.load(ui, extensionname, None)
        ui.write(minirst.subsection(extensionname))
        ui.write(b"%s\n\n" % gettext(pycompat.getdoc(mod)))
        cmdtable = getattr(mod, 'cmdtable', None)
        if cmdtable:
            ui.write(minirst.subsubsection(_(b'Commands')))
            commandprinter(
                ui,
                cmdtable,
                minirst.subsubsubsection,
                minirst.subsubsubsubsection,
            )


def showtopic(ui, topic):
    extrahelptable = [
        ([b"common"], b'', loaddoc(b'common'), help.TOPIC_CATEGORY_MISC),
        ([b"hg.1"], b'', loaddoc(b'hg.1'), help.TOPIC_CATEGORY_CONFIG),
        ([b"hg-ssh.8"], b'', loaddoc(b'hg-ssh.8'), help.TOPIC_CATEGORY_CONFIG),
        (
            [b"hgignore.5"],
            b'',
            loaddoc(b'hgignore.5'),
            help.TOPIC_CATEGORY_CONFIG,
        ),
        ([b"hgrc.5"], b'', loaddoc(b'hgrc.5'), help.TOPIC_CATEGORY_CONFIG),
        (
            [b"hgignore.5.gendoc"],
            b'',
            loaddoc(b'hgignore'),
            help.TOPIC_CATEGORY_CONFIG,
        ),
        (
            [b"hgrc.5.gendoc"],
            b'',
            loaddoc(b'config'),
            help.TOPIC_CATEGORY_CONFIG,
        ),
    ]
    helpprinter(ui, helptable + extrahelptable, None, include=[topic])


def helpprinter(ui, helptable, sectionfunc, include=[], exclude=[]):
    for h in helptable:
        names, sec, doc = h[0:3]
        if exclude and names[0] in exclude:
            continue
        if include and names[0] not in include:
            continue
        for name in names:
            ui.write(b".. _%s:\n" % name)
        ui.write(b"\n")
        if sectionfunc:
            ui.write(sectionfunc(sec))
        if callable(doc):
            doc = doc(ui)
        ui.write(doc)
        ui.write(b"\n")


def commandprinter(ui, cmdtable, sectionfunc, subsectionfunc):
    """Render restructuredtext describing a list of commands and their
    documentations, grouped by command category.

    Args:
      ui: UI object to write the output to
      cmdtable: a dict that maps a string of the command name plus its aliases
        (separated with pipes) to a 3-tuple of (the command's function, a list
        of its option descriptions, and a string summarizing available
        options). Example, with aliases added for demonstration purposes:

          'phase|alias1|alias2': (
             <function phase at 0x7f0816b05e60>,
             [ ('p', 'public', False, 'set changeset phase to public'),
               ...,
               ('r', 'rev', [], 'target revision', 'REV')],
             '[-p|-d|-s] [-f] [-r] [REV...]'
          )
      sectionfunc: minirst function to format command category headers
      subsectionfunc: minirst function to format command headers
    """
    h = {}
    for c, attr in cmdtable.items():
        f = c.split(b"|")[0]
        f = f.lstrip(b"^")
        h[f] = c
    cmds = h.keys()

    def helpcategory(cmd):
        """Given a canonical command name from `cmds` (above), retrieve its
        help category. If helpcategory is None, default to CATEGORY_NONE.
        """
        fullname = h[cmd]
        details = cmdtable[fullname]
        helpcategory = details[0].helpcategory
        return helpcategory or help.registrar.command.CATEGORY_NONE

    cmdsbycategory = {category: [] for category in help.CATEGORY_ORDER}
    for cmd in cmds:
        # If a command category wasn't registered, the command won't get
        # rendered below, so we raise an AssertionError.
        if helpcategory(cmd) not in cmdsbycategory:
            raise AssertionError(
                "The following command did not register its (category) in "
                "help.CATEGORY_ORDER: %s (%s)" % (cmd, helpcategory(cmd))
            )
        cmdsbycategory[helpcategory(cmd)].append(cmd)

    # Print the help for each command. We present the commands grouped by
    # category, and we use help.CATEGORY_ORDER as a guide for a helpful order
    # in which to present the categories.
    for category in help.CATEGORY_ORDER:
        categorycmds = cmdsbycategory[category]
        if not categorycmds:
            # Skip empty categories
            continue
        # Print a section header for the category.
        # For now, the category header is at the same level as the headers for
        # the commands in the category; this is fixed in the next commit.
        ui.write(sectionfunc(help.CATEGORY_NAMES[category]))
        # Print each command in the category
        for f in sorted(categorycmds):
            if f.startswith(b"debug"):
                continue
            d = get_cmd(h[f], cmdtable)
            ui.write(subsectionfunc(d[b'cmd']))
            # short description
            ui.write(d[b'desc'][0])
            # synopsis
            ui.write(b"::\n\n")
            synopsislines = d[b'synopsis'].splitlines()
            for line in synopsislines:
                # some commands (such as rebase) have a multi-line
                # synopsis
                ui.write(b"   %s\n" % line)
            ui.write(b'\n')
            # description
            ui.write(b"%s\n\n" % d[b'desc'][1])
            # options
            opt_output = list(d[b'opts'])
            if opt_output:
                opts_len = max([len(line[0]) for line in opt_output])
                ui.write(_(b"Options:\n\n"))
                multioccur = False
                for optstr, desc in opt_output:
                    if desc:
                        s = b"%-*s  %s" % (opts_len, optstr, desc)
                    else:
                        s = optstr
                    ui.write(b"%s\n" % s)
                    if optstr.endswith(b"[+]>"):
                        multioccur = True
                if multioccur:
                    ui.write(
                        _(
                            b"\n[+] marked option can be specified"
                            b" multiple times\n"
                        )
                    )
                ui.write(b"\n")
            # aliases
            if d[b'aliases']:
                ui.write(_(b"    aliases: %s\n\n") % b" ".join(d[b'aliases']))


def allextensionnames():
    return set(extensions.enabled().keys()) | set(extensions.disabled().keys())


if __name__ == "__main__":
    doc = b'hg.1.gendoc'
    if len(sys.argv) > 1:
        doc = encoding.strtolocal(sys.argv[1])

    ui = uimod.ui.load()
    if doc == b'hg.1.gendoc':
        showdoc(ui)
    else:
        showtopic(ui, encoding.strtolocal(sys.argv[1]))