252 lines
8.7 KiB
Python
252 lines
8.7 KiB
Python
# Copyright (c) 2014 Dark Secret Software Inc.
|
|
#
|
|
# 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 datetime
|
|
import fnmatch
|
|
import hashlib
|
|
import os
|
|
import os.path
|
|
import tarfile
|
|
|
|
import notification_utils
|
|
|
|
import archive
|
|
import disk_storage
|
|
|
|
|
|
class NoMoreFiles(Exception):
|
|
pass
|
|
|
|
|
|
class NoValidFile(Exception):
|
|
pass
|
|
|
|
|
|
class BadWorkingDirectory(Exception):
|
|
pass
|
|
|
|
|
|
class RollManager(object):
|
|
def __init__(self, filename_template, directory=".",
|
|
archive_class=None, archive_callback=None):
|
|
self.filename_template = filename_template
|
|
self.directory = directory
|
|
self.active_archive = None
|
|
self.archive_class = archive_class
|
|
self.active_filename = None
|
|
self.archive_callback = archive_callback
|
|
|
|
def _roll_archive(self):
|
|
self.close()
|
|
self.get_active_archive()
|
|
|
|
def close(self):
|
|
if self.active_archive:
|
|
self.active_archive.close()
|
|
if self.archive_callback:
|
|
self.archive_callback.on_close(self.active_filename)
|
|
self.active_archive = None
|
|
self.active_filename = None
|
|
|
|
|
|
class ReadingRollManager(RollManager):
|
|
def __init__(self, filename_template, directory=".",
|
|
destination_directory=".",
|
|
archive_class = archive.ArchiveReader,
|
|
archive_callback=None, roll_size_mb=1000):
|
|
super(ReadingRollManager, self).__init__(
|
|
filename_template,
|
|
directory=directory,
|
|
archive_callback=archive_callback,
|
|
archive_class=archive_class)
|
|
self.files_to_read = self._get_matching_files(directory,
|
|
filename_template)
|
|
|
|
def _get_matching_files(self, directory, filename_template):
|
|
files = [os.path.join(directory, f)
|
|
for f in os.listdir(self.directory)
|
|
if os.path.isfile(os.path.join(directory, f))]
|
|
return sorted(fnmatch.filter(files, filename_template))
|
|
|
|
def read(self):
|
|
# (metadata, payload)
|
|
for x in range(3): # 3 attempts to file a valid file.
|
|
a = self.get_active_archive()
|
|
try:
|
|
return a.read()
|
|
except disk_storage.EndOfFile:
|
|
self._roll_archive()
|
|
raise NoValidFile("Unable to find a valid file after 3 attempts")
|
|
|
|
def get_active_archive(self):
|
|
if not self.active_archive:
|
|
if not self.files_to_read:
|
|
raise NoMoreFiles()
|
|
self.active_filename = self.files_to_read.pop(0)
|
|
self.active_archive = self.archive_class(self.active_filename)
|
|
if self.archive_callback:
|
|
self.archive_callback.on_open(self.active_filename)
|
|
return self.active_archive
|
|
|
|
|
|
class WritingRollManager(RollManager):
|
|
def __init__(self, filename_template, roll_checker, directory=".",
|
|
destination_directory=".",
|
|
archive_class=archive.ArchiveWriter,
|
|
archive_callback=None, roll_size_mb=1000):
|
|
super(WritingRollManager, self).__init__(
|
|
filename_template,
|
|
directory=directory,
|
|
archive_callback=archive_callback,
|
|
archive_class=archive_class)
|
|
self.roll_checker = roll_checker
|
|
|
|
def write(self, metadata, payload):
|
|
"""metadata is string:string dict.
|
|
payload must be encoded as string.
|
|
"""
|
|
a = self.get_active_archive()
|
|
a.write(metadata, payload)
|
|
if self._should_roll_archive():
|
|
self._roll_archive()
|
|
|
|
def _make_filename(self):
|
|
f = notification_utils.now().strftime(self.filename_template)
|
|
f = f.replace(" ", "_")
|
|
f = f.replace("/", "_")
|
|
f = f.replace(":", "_")
|
|
return os.path.join(self.directory, f)
|
|
|
|
def get_active_archive(self):
|
|
if not self.active_archive:
|
|
self.active_filename = self._make_filename()
|
|
self.active_archive = self.archive_class(self.active_filename)
|
|
if self.archive_callback:
|
|
self.archive_callback.on_open(self.active_filename)
|
|
self.roll_checker.start(self.active_archive)
|
|
return self.active_archive
|
|
|
|
def _should_roll_archive(self):
|
|
return self.roll_checker.check(self.active_archive)
|
|
|
|
|
|
class WritingJSONRollManager(object):
|
|
"""No archiver. No roll checker. Just write 1 file line per json payload.
|
|
Once the file gets big enough, .tar.gz the file and move
|
|
into the destination_directory.
|
|
Expects an external tool like rsync to move the file.
|
|
A SHA-256 of the payload may be included in the tarball filename."""
|
|
def __init__(self, *args, **kwargs):
|
|
self.filename_template = args[0]
|
|
self.directory = kwargs.get('directory', '.')
|
|
self.destination_directory = kwargs.get('destination_directory', '.')
|
|
self.roll_size_mb = int(kwargs.get('roll_size_mb', 1000))
|
|
minutes = kwargs.get('roll_minutes', 15)
|
|
self.roll_after = datetime.timedelta(minutes=minutes)
|
|
|
|
# Look in the working directory for any files. Move them to the
|
|
# destination directory before we start. This implies we
|
|
# have to make sure multiple workers don't point at the same
|
|
# working directory.
|
|
self.handle = None
|
|
self.size = 0
|
|
self.start_time = self._get_time()
|
|
|
|
self._archive_working_files()
|
|
|
|
def _get_time(self):
|
|
# Broken out for testing ...
|
|
return datetime.datetime.utcnow()
|
|
|
|
def _archive_working_files(self):
|
|
for f in os.listdir(self.directory):
|
|
full = os.path.join(self.directory, f)
|
|
if os.path.isfile(full):
|
|
self._do_roll(full)
|
|
|
|
def _make_filename(self, crc, prefix):
|
|
now = notification_utils.now()
|
|
dt = str(notification_utils.dt_to_decimal(now))
|
|
f = now.strftime(self.filename_template)
|
|
f = f.replace(" ", "_")
|
|
f = f.replace("/", "_")
|
|
f = f.replace(":", "_")
|
|
f = f.replace("[[CRC]]", crc)
|
|
f = f.replace("[[TIMESTAMP]]", dt)
|
|
return os.path.join(prefix, f)
|
|
|
|
def _should_roll(self, size):
|
|
return ((size / 1048576) >= self.roll_size_mb or
|
|
(size > 0 and
|
|
self._get_time() >= (self.start_time + self.roll_after)))
|
|
|
|
def _get_file_sha(self, filename):
|
|
block_size=2**20
|
|
sha256 = hashlib.sha256()
|
|
# no context manager, just to keep testing simple.
|
|
f = open(filename, "r")
|
|
while True:
|
|
data = f.read(block_size)
|
|
if not data:
|
|
break
|
|
sha256.update(data)
|
|
f.close()
|
|
return sha256.hexdigest()
|
|
|
|
def _tar_working_file(self, filename):
|
|
# tar all the files in working directory into an archive
|
|
# in destination_directory.
|
|
crc = self._get_file_sha(filename)
|
|
|
|
# No contextmgr for tarfile in 2.6 :(
|
|
fn = self._make_filename(crc, self.destination_directory) + ".tar.gz"
|
|
tar = tarfile.open(fn, "w:gz")
|
|
just_name = os.path.basename(filename)
|
|
tar.add(filename, arcname=just_name)
|
|
tar.close()
|
|
|
|
def _clean_working_directory(self, filename):
|
|
os.remove(filename)
|
|
self.size = 0
|
|
|
|
def _do_roll(self, filename):
|
|
self.close()
|
|
self._tar_working_file(filename)
|
|
self._clean_working_directory(filename)
|
|
|
|
def write(self, metadata, json_payload):
|
|
# Metadata is ignored.
|
|
handle = self._get_handle()
|
|
handle.write("%s\n" % json_payload)
|
|
handle.flush()
|
|
|
|
self.size += len(json_payload)
|
|
|
|
if self._should_roll(self.size):
|
|
self._do_roll(self.filename)
|
|
|
|
def _get_handle(self):
|
|
if not self.handle:
|
|
self.filename = self._make_filename('[[CRC]]', self.directory)
|
|
self.handle = open(self.filename, "w")
|
|
self.start_time = self._get_time()
|
|
|
|
return self.handle
|
|
|
|
def close(self):
|
|
if self.handle:
|
|
self.handle.close()
|
|
self.handle = None
|