procutil: make stream detection in make_line_buffered more correct and strict
In make_line_buffered(), we don’t want to wrap the stream if we know that lines
get flushed to the underlying raw stream already.
Previously, the heuristic was too optimistic. It assumed that any stream which
is not an instance of io.BufferedIOBase doesn’t need wrapping. However, there
are buffered streams that aren’t instances of io.BufferedIOBase, like
Mercurial’s own winstdout.
The new logic is different in two ways:
First, only for the check, if unwraps any combination of WriteAllWrapper and
winstdout.
Second, it skips wrapping the stream only if it is an instance of io.RawIOBase
(or already wrapped). If it is an instance of io.BufferedIOBase, it gets
wrapped. In any other case, the function raises an exception. This ensures
that, if an unknown stream is passed or we add another wrapper in the future,
we don’t wrap the stream if it’s already line buffered or not wrap the stream
if it’s not line buffered. In fact, this was already helpful during development
of this change. Without it, I possibly would have forgot that WriteAllWrapper
needs to be ignored for the check, leading to unnecessary wrapping if stdout is
unbuffered.
The alternative would have been to always wrap unknown streams. However, I
don’t think that anyone would benefit from being less strict. We can expect
streams from the standard library to be subclassing either io.RawIOBase or
io.BufferedIOBase, so running Mercurial in the standard way should not regress
by this change. Py2exe might replace sys.stdout and sys.stderr, but that
currently breaks Mercurial anyway and also these streams don’t claim to be
interactive, so this function is not called for them.
#!/usr/bin/env python3
# compare various algorithm variants for a given case
#
# search-discovery-case REPO LOCAL_CASE REMOTE_CASE
#
# The description for the case input uses the same format as the ouput of
# search-discovery-case
import json
import os
import subprocess
import sys
this_script = os.path.abspath(sys.argv[0])
script_name = os.path.basename(this_script)
this_dir = os.path.dirname(this_script)
hg_dir = os.path.join(this_dir, '..', '..')
HG_REPO = os.path.normpath(hg_dir)
HG_BIN = os.path.join(HG_REPO, 'hg')
SUBSET_PATH = os.path.join(HG_REPO, 'contrib', 'perf-utils', 'subsetmaker.py')
CMD_BASE = (
HG_BIN,
'debugdiscovery',
'--template',
'json',
'--config',
'extensions.subset=%s' % SUBSET_PATH,
)
# --old
# --nonheads
#
# devel.discovery.exchange-heads=True
# devel.discovery.grow-sample=True
# devel.discovery.grow-sample.dynamic=True
VARIANTS = {
'tree-discovery': ('--old',),
'set-discovery-basic': (
'--config',
'devel.discovery.exchange-heads=no',
'--config',
'devel.discovery.grow-sample=no',
'--config',
'devel.discovery.grow-sample.dynamic=no',
'--config',
'devel.discovery.randomize=yes',
),
'set-discovery-heads': (
'--config',
'devel.discovery.exchange-heads=yes',
'--config',
'devel.discovery.grow-sample=no',
'--config',
'devel.discovery.grow-sample.dynamic=no',
'--config',
'devel.discovery.randomize=yes',
),
'set-discovery-grow-sample': (
'--config',
'devel.discovery.exchange-heads=yes',
'--config',
'devel.discovery.grow-sample=yes',
'--config',
'devel.discovery.grow-sample.dynamic=no',
'--config',
'devel.discovery.randomize=yes',
),
'set-discovery-dynamic-sample': (
'--config',
'devel.discovery.exchange-heads=yes',
'--config',
'devel.discovery.grow-sample=yes',
'--config',
'devel.discovery.grow-sample.dynamic=yes',
'--config',
'devel.discovery.randomize=yes',
),
'set-discovery-default': (
'--config',
'devel.discovery.randomize=yes',
),
}
VARIANTS_KEYS = [
'tree-discovery',
'set-discovery-basic',
'set-discovery-heads',
'set-discovery-grow-sample',
'set-discovery-dynamic-sample',
'set-discovery-default',
]
assert set(VARIANTS.keys()) == set(VARIANTS_KEYS)
def format_case(case):
return '-'.join(str(s) for s in case)
def to_revsets(case):
t = case[0]
if t == 'scratch':
return 'not scratch(all(), %d, "%d")' % (case[1], case[2])
elif t == 'randomantichain':
return '::randomantichain(all(), "%d")' % case[1]
elif t == 'rev':
return '::%d' % case[1]
else:
assert False
def compare(repo, local_case, remote_case):
case = (repo, local_case, remote_case)
for variant in VARIANTS_KEYS:
res = process(case, VARIANTS[variant])
revs = res["nb-revs"]
local_heads = res["nb-head-local"]
common_heads = res["nb-common-heads"]
roundtrips = res["total-roundtrips"]
queries = res["total-queries"]
if 'tree-discovery' in variant:
print(
repo,
format_case(local_case),
format_case(remote_case),
variant,
roundtrips,
queries,
revs,
local_heads,
common_heads,
)
else:
undecided_common = res["nb-ini_und-common"]
undecided_missing = res["nb-ini_und-missing"]
undecided = undecided_common + undecided_missing
print(
repo,
format_case(local_case),
format_case(remote_case),
variant,
roundtrips,
queries,
revs,
local_heads,
common_heads,
undecided,
undecided_common,
undecided_missing,
)
return 0
def process(case, variant):
(repo, left, right) = case
cmd = list(CMD_BASE)
cmd.append('-R')
cmd.append(repo)
cmd.append('--local-as-revs')
cmd.append(to_revsets(left))
cmd.append('--remote-as-revs')
cmd.append(to_revsets(right))
cmd.extend(variant)
s = subprocess.Popen(cmd, stdout=subprocess.PIPE)
out, err = s.communicate()
return json.loads(out)[0]
if __name__ == '__main__':
if len(sys.argv) != 4:
usage = f'USAGE: {script_name} REPO LOCAL_CASE REMOTE_CASE'
print(usage, file=sys.stderr)
sys.exit(128)
repo = sys.argv[1]
local_case = sys.argv[2].split('-')
local_case = (local_case[0],) + tuple(int(x) for x in local_case[1:])
remote_case = sys.argv[3].split('-')
remote_case = (remote_case[0],) + tuple(int(x) for x in remote_case[1:])
sys.exit(compare(repo, local_case, remote_case))