git-repo/progress.py
Mike Frysinger b2fa30a2b8 sync: switch network fetch to multiprocessing
This avoids GIL limitations with using threads for parallel processing.

This reworks the fetch logic to return results for processing in the
main thread instead of leaving every thread to do its own processing.

We have to tweak the chunking logic a little here because multiprocessing
favors batching over returning immediate results when using a larger value
for chunksize.  When a single job can be quite slow, this tradeoff is not
good UX.

Bug: https://crbug.com/gerrit/12389
Change-Id: I0f0512d15ad7332d1eb28aff52c29d378acc9e1d
Reviewed-on: https://gerrit-review.googlesource.com/c/git-repo/+/298642
Reviewed-by: Chris Mcdonald <cjmcdonald@google.com>
Tested-by: Mike Frysinger <vapier@google.com>
2021-04-01 14:52:57 +00:00

124 lines
3.4 KiB
Python

# Copyright (C) 2009 The Android Open Source Project
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import sys
from time import time
from repo_trace import IsTrace
_NOT_TTY = not os.isatty(2)
# This will erase all content in the current line (wherever the cursor is).
# It does not move the cursor, so this is usually followed by \r to move to
# column 0.
CSI_ERASE_LINE = '\x1b[2K'
def duration_str(total):
"""A less noisy timedelta.__str__.
The default timedelta stringification contains a lot of leading zeros and
uses microsecond resolution. This makes for noisy output.
"""
hours, rem = divmod(total, 3600)
mins, secs = divmod(rem, 60)
ret = '%.3fs' % (secs,)
if mins:
ret = '%im%s' % (mins, ret)
if hours:
ret = '%ih%s' % (hours, ret)
return ret
class Progress(object):
def __init__(self, title, total=0, units='', print_newline=False, delay=True):
self._title = title
self._total = total
self._done = 0
self._start = time()
self._show = not delay
self._units = units
self._print_newline = print_newline
# Only show the active jobs section if we run more than one in parallel.
self._show_jobs = False
self._active = 0
def start(self, name):
self._active += 1
if not self._show_jobs:
self._show_jobs = self._active > 1
self.update(inc=0, msg='started ' + name)
def finish(self, name):
self.update(msg='finished ' + name)
self._active -= 1
def update(self, inc=1, msg=''):
self._done += inc
if _NOT_TTY or IsTrace():
return
if not self._show:
if 0.5 <= time() - self._start:
self._show = True
else:
return
if self._total <= 0:
sys.stderr.write('%s\r%s: %d,' % (
CSI_ERASE_LINE,
self._title,
self._done))
sys.stderr.flush()
else:
p = (100 * self._done) / self._total
if self._show_jobs:
jobs = '[%d job%s] ' % (self._active, 's' if self._active > 1 else '')
else:
jobs = ''
sys.stderr.write('%s\r%s: %2d%% %s(%d%s/%d%s)%s%s%s' % (
CSI_ERASE_LINE,
self._title,
p,
jobs,
self._done, self._units,
self._total, self._units,
' ' if msg else '', msg,
'\n' if self._print_newline else ''))
sys.stderr.flush()
def end(self):
if _NOT_TTY or IsTrace() or not self._show:
return
duration = duration_str(time() - self._start)
if self._total <= 0:
sys.stderr.write('%s\r%s: %d, done in %s\n' % (
CSI_ERASE_LINE,
self._title,
self._done,
duration))
sys.stderr.flush()
else:
p = (100 * self._done) / self._total
sys.stderr.write('%s\r%s: %3d%% (%d%s/%d%s), done in %s\n' % (
CSI_ERASE_LINE,
self._title,
p,
self._done, self._units,
self._total, self._units,
duration))
sys.stderr.flush()