X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/0e0c1400b57d5de8aa8c18dd4897527f905a4b42..d047c1cb9ceecb6e324adf102e5e38e11fe698e1:/sdk/python/arvados/keep.py diff --git a/sdk/python/arvados/keep.py b/sdk/python/arvados/keep.py index ee91491efa..e6e93f0806 100644 --- a/sdk/python/arvados/keep.py +++ b/sdk/python/arvados/keep.py @@ -1,14 +1,20 @@ +# Copyright (C) The Arvados Authors. All rights reserved. +# +# SPDX-License-Identifier: Apache-2.0 + from __future__ import absolute_import from __future__ import division from future import standard_library +from future.utils import native_str standard_library.install_aliases() from builtins import next from builtins import str from builtins import range from builtins import object -import io +import collections import datetime import hashlib +import io import logging import math import os @@ -72,8 +78,9 @@ class KeepLocator(object): def __str__(self): return '+'.join( - str(s) for s in [self.md5sum, self.size, - self.permission_hint()] + self.hints + native_str(s) + for s in [self.md5sum, self.size, + self.permission_hint()] + self.hints if s is not None) def stripped(self): @@ -90,7 +97,7 @@ class KeepLocator(object): return getattr(self, data_name) def setter(self, hex_str): if not arvados.util.is_hex(hex_str, length): - raise ValueError("{} is not a {}-digit hex string: {}". + raise ValueError("{} is not a {}-digit hex string: {!r}". format(name, length, hex_str)) setattr(self, data_name, hex_str) return property(getter, setter) @@ -290,6 +297,7 @@ class KeepClient(object): self._result = {'error': None} self._usable = True self._session = None + self._socket = None self.get_headers = {'Accept': 'application/octet-stream'} self.get_headers.update(headers) self.put_headers = headers @@ -320,15 +328,28 @@ class KeepClient(object): except: ua.close() - @staticmethod - def _socket_open(family, socktype, protocol, address=None): + def _socket_open(self, *args, **kwargs): + if len(args) + len(kwargs) == 2: + return self._socket_open_pycurl_7_21_5(*args, **kwargs) + else: + return self._socket_open_pycurl_7_19_3(*args, **kwargs) + + def _socket_open_pycurl_7_19_3(self, family, socktype, protocol, address=None): + return self._socket_open_pycurl_7_21_5( + purpose=None, + address=collections.namedtuple( + 'Address', ['family', 'socktype', 'protocol', 'addr'], + )(family, socktype, protocol, address)) + + def _socket_open_pycurl_7_21_5(self, purpose, address): """Because pycurl doesn't have CURLOPT_TCP_KEEPALIVE""" - s = socket.socket(family, socktype, protocol) + s = socket.socket(address.family, address.socktype, address.protocol) s.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1) # Will throw invalid protocol error on mac. This test prevents that. if hasattr(socket, 'TCP_KEEPIDLE'): s.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPIDLE, 75) s.setsockopt(socket.IPPROTO_TCP, socket.TCP_KEEPINTVL, 75) + self._socket = s return s def get(self, locator, method="GET", timeout=None): @@ -342,7 +363,8 @@ class KeepClient(object): self._headers = {} response_body = BytesIO() curl.setopt(pycurl.NOSIGNAL, 1) - curl.setopt(pycurl.OPENSOCKETFUNCTION, self._socket_open) + curl.setopt(pycurl.OPENSOCKETFUNCTION, + lambda *args, **kwargs: self._socket_open(*args, **kwargs)) curl.setopt(pycurl.URL, url.encode('utf-8')) curl.setopt(pycurl.HTTPHEADER, [ '{}: {}'.format(k,v) for k,v in self.get_headers.items()]) @@ -356,6 +378,10 @@ class KeepClient(object): curl.perform() except Exception as e: raise arvados.errors.HttpError(0, str(e)) + finally: + if self._socket: + self._socket.close() + self._socket = None self._result = { 'status_code': curl.getinfo(pycurl.RESPONSE_CODE), 'body': response_body.getvalue(), @@ -420,7 +446,8 @@ class KeepClient(object): body_reader = BytesIO(body) response_body = BytesIO() curl.setopt(pycurl.NOSIGNAL, 1) - curl.setopt(pycurl.OPENSOCKETFUNCTION, self._socket_open) + curl.setopt(pycurl.OPENSOCKETFUNCTION, + lambda *args, **kwargs: self._socket_open(*args, **kwargs)) curl.setopt(pycurl.URL, url.encode('utf-8')) # Using UPLOAD tells cURL to wait for a "go ahead" from the # Keep server (in the form of a HTTP/1.1 "100 Continue" @@ -440,9 +467,13 @@ class KeepClient(object): curl.perform() except Exception as e: raise arvados.errors.HttpError(0, str(e)) + finally: + if self._socket: + self._socket.close() + self._socket = None self._result = { 'status_code': curl.getinfo(pycurl.RESPONSE_CODE), - 'body': response_body.getvalue(), + 'body': response_body.getvalue().decode('utf-8'), 'headers': self._headers, 'error': False, } @@ -851,7 +882,7 @@ class KeepClient(object): The weight is md5(h + u) where u is the last 15 characters of the service endpoint's UUID. """ - return hashlib.md5(data_hash + service_uuid[-15:]).hexdigest() + return hashlib.md5((data_hash + service_uuid[-15:]).encode()).hexdigest() def weighted_service_roots(self, locator, force_rebuild=False, need_writable=False): """Return an array of Keep service endpoints, in the order in @@ -1141,7 +1172,7 @@ class KeepClient(object): """ md5 = hashlib.md5(data).hexdigest() locator = '%s+%d' % (md5, len(data)) - with open(os.path.join(self.local_store, md5 + '.tmp'), 'w') as f: + with open(os.path.join(self.local_store, md5 + '.tmp'), 'wb') as f: f.write(data) os.rename(os.path.join(self.local_store, md5 + '.tmp'), os.path.join(self.local_store, md5)) @@ -1155,8 +1186,8 @@ class KeepClient(object): raise arvados.errors.NotFoundError( "Invalid data locator: '%s'" % loc_s) if locator.md5sum == config.EMPTY_BLOCK_LOCATOR.split('+')[0]: - return '' - with open(os.path.join(self.local_store, locator.md5sum), 'r') as f: + return b'' + with open(os.path.join(self.local_store, locator.md5sum), 'rb') as f: return f.read() def is_cached(self, locator):