oss2/resumable.py

# -*- coding: utf-8 -*-

"""
oss2.resumable
~~~~~~~~~~~~~~

该模块包含了断点续传相关的函数和类。
"""

import os

from . import utils
from .utils import b64encode_as_string, b64decode_from_string
from . import iterators
from . import exceptions
from . import defaults
from . import http
from . import models
from .crypto_bucket import CryptoBucket
from . import Bucket
from .iterators import PartIterator

from .models import PartInfo
from .compat import json, stringify, to_unicode, to_string
from .task_queue import TaskQueue
from .headers import *

import functools
import threading
import random
import string

import logging

logger = logging.getLogger(__name__)


def resumable_upload(bucket, key, filename,
                     store=None,
                     headers=None,
                     multipart_threshold=None,
                     part_size=None,
                     progress_callback=None,
                     num_threads=None,
                     params=None):
    """断点上传本地文件。

    实现中采用分片上传方式上传本地文件，缺省的并发数是 `oss2.defaults.multipart_num_threads` ，并且在
    本地磁盘保存已经上传的分片信息。如果因为某种原因上传被中断，下次上传同样的文件，即源文件和目标文件路径都
    一样，就只会上传缺失的分片。

    缺省条件下，该函数会在用户 `HOME` 目录下保存断点续传的信息。当待上传的本地文件没有发生变化，
    且目标文件名没有变化时，会根据本地保存的信息，从断点开始上传。

    使用该函数应注意如下细节：
        #. 如果使用CryptoBucket，函数会退化为普通上传

    :param bucket: :class:`Bucket <oss2.Bucket>` 或者 ：:class:`CryptoBucket <oss2.CryptoBucket>` 对象
    :param key: 上传到用户空间的文件名
    :param filename: 待上传本地文件名
    :param store: 用来保存断点信息的持久存储，参见 :class:`ResumableStore` 的接口。如不指定，则使用 `ResumableStore` 。

    :param headers: HTTP头部
        # 调用外部函数put_object 或 init_multipart_upload传递完整headers
        # 调用外部函数uplpad_part目前只传递OSS_REQUEST_PAYER, OSS_TRAFFIC_LIMIT
        # 调用外部函数complete_multipart_upload目前只传递OSS_REQUEST_PAYER, OSS_OBJECT_ACL
    :type headers: 可以是dict，建议是oss2.CaseInsensitiveDict

    :param multipart_threshold: 文件长度大于该值时，则用分片上传。
    :param part_size: 指定分片上传的每个分片的大小。如不指定，则自动计算。
    :param progress_callback: 上传进度回调函数。参见 :ref:`progress_callback` 。
    :param num_threads: 并发上传的线程数，如不指定则使用 `oss2.defaults.multipart_num_threads` 。

    :param params: HTTP请求参数
        # 只有'sequential'这个参数才会被传递到外部函数init_multipart_upload中。
        # 其他参数视为无效参数不会往外部函数传递。
    :type params: dict
    """
    logger.debug("Start to resumable upload, bucket: {0}, key: {1}, filename: {2}, headers: {3}, "
                "multipart_threshold: {4}, part_size: {5}, num_threads: {6}".format(bucket.bucket_name, to_string(key),
                                                                                    filename, headers, multipart_threshold,
                                                                                    part_size, num_threads))
    size = os.path.getsize(filename)
    multipart_threshold = defaults.get(multipart_threshold, defaults.multipart_threshold)

    logger.debug("The size of file to upload is: {0}, multipart_threshold: {1}".format(size, multipart_threshold))
    if size >= multipart_threshold:
        uploader = _ResumableUploader(bucket, key, filename, size, store,
                                      part_size=part_size,
                                      headers=headers,
                                      progress_callback=progress_callback,
                                      num_threads=num_threads,
                                      params=params)
        result = uploader.upload()
    else:
        with open(to_unicode(filename), 'rb') as f:
            result = bucket.put_object(key, f, headers=headers, progress_callback=progress_callback)

    return result


def resumable_download(bucket, key, filename,
                       multiget_threshold=None,
                       part_size=None,
                       progress_callback=None,
                       num_threads=None,
                       store=None,
                       params=None,
                       headers=None):
    """断点下载。

    实现的方法是：
        #. 在本地创建一个临时文件，文件名由原始文件名加上一个随机的后缀组成；
        #. 通过指定请求的 `Range` 头按照范围并发读取OSS文件，并写入到临时文件里对应的位置；
        #. 全部完成之后，把临时文件重命名为目标文件 （即 `filename` ）

    在上述过程中，断点信息，即已经完成的范围，会保存在磁盘上。因为某种原因下载中断，后续如果下载
    同样的文件，也就是源文件和目标文件一样，就会先读取断点信息，然后只下载缺失的部分。

    缺省设置下，断点信息保存在 `HOME` 目录的一个子目录下。可以通过 `store` 参数更改保存位置。

    使用该函数应注意如下细节：
        #. 对同样的源文件、目标文件，避免多个程序（线程）同时调用该函数。因为断点信息会在磁盘上互相覆盖，或临时文件名会冲突。
        #. 避免使用太小的范围（分片），即 `part_size` 不宜过小，建议大于或等于 `oss2.defaults.multiget_part_size` 。
        #. 如果目标文件已经存在，那么该函数会覆盖此文件。
        #. 如果使用CryptoBucket，函数会退化为普通下载


    :param bucket: :class:`Bucket <oss2.Bucket>` 或者 ：:class:`CryptoBucket <oss2.CryptoBucket>` 对象
    :param str key: 待下载的远程文件名。
    :param str filename: 本地的目标文件名。
    :param int multiget_threshold: 文件长度大于该值时，则使用断点下载。
    :param int part_size: 指定期望的分片大小，即每个请求获得的字节数，实际的分片大小可能有所不同。
    :param progress_callback: 下载进度回调函数。参见 :ref:`progress_callback` 。
    :param num_threads: 并发下载的线程数，如不指定则使用 `oss2.defaults.multiget_num_threads` 。

    :param store: 用来保存断点信息的持久存储，可以指定断点信息所在的目录。
    :type store: `ResumableDownloadStore`

    :param dict params: 指定下载参数，可以传入versionId下载指定版本文件

    :param headers: HTTP头部,
        # 调用外部函数head_object目前只传递OSS_REQUEST_PAYER
        # 调用外部函数get_object_to_file, get_object目前需要向下传递的值有OSS_REQUEST_PAYER, OSS_TRAFFIC_LIMIT
    :type headers: 可以是dict，建议是oss2.CaseInsensitiveDict

    :raises: 如果OSS文件不存在，则抛出 :class:`NotFound <oss2.exceptions.NotFound>` ；也有可能抛出其他因下载文件而产生的异常。
    """
    logger.debug("Start to resumable download, bucket: {0}, key: {1}, filename: {2}, multiget_threshold: {3}, "
                "part_size: {4}, num_threads: {5}".format(bucket.bucket_name, to_string(key), filename,
                                                          multiget_threshold, part_size, num_threads))
    multiget_threshold = defaults.get(multiget_threshold, defaults.multiget_threshold)

    valid_headers = _populate_valid_headers(headers, [OSS_REQUEST_PAYER, OSS_TRAFFIC_LIMIT])
    result = bucket.head_object(key, params=params, headers=valid_headers)
    logger.debug("The size of object to download is: {0}, multiget_threshold: {1}".format(result.content_length,
                                                                                          multiget_threshold))
    if result.content_length >= multiget_threshold:
        downloader = _ResumableDownloader(bucket, key, filename, _ObjectInfo.make(result), part_size=part_size,
                                          progress_callback=progress_callback, num_threads=num_threads, store=store,
                                          params=params, headers=valid_headers)
        downloader.download(result.server_crc)
    else:
        bucket.get_object_to_file(key, filename, progress_callback=progress_callback, params=params,
                                  headers=valid_headers)


_MAX_MULTIGET_PART_COUNT = 100000


def determine_part_size(total_size,
                        preferred_size=None):
    """确定分片上传是分片的大小。

    :param int total_size: 总共需要上传的长度
    :param int preferred_size: 用户期望的分片大小。如果不指定则采用defaults.part_size

    :return: 分片大小
    """
    if not preferred_size:
        preferred_size = defaults.part_size

    return _determine_part_size_internal(total_size, preferred_size, defaults.max_part_count)


def _determine_part_size_internal(total_size, preferred_size, max_count):
    if total_size < preferred_size:
        return total_size

    while preferred_size * max_count < total_size or preferred_size < defaults.min_part_size:
        preferred_size = preferred_size * 2

    return preferred_size


def _split_to_parts(total_size, part_size):
    parts = []
    num_parts = utils.how_many(total_size, part_size)

    for i in range(num_parts):
        if i == num_parts - 1:
            start = i * part_size
            end = total_size
        else:
            start = i * part_size
            end = part_size + start

        parts.append(_PartToProcess(i + 1, start, end))

    return parts

def _populate_valid_headers(headers=None, valid_keys=None):
    """构建只包含有效keys的http header

    :param headers: 需要过滤的header
    :type headers: 可以是dict，建议是oss2.CaseInsensitiveDict

    :param valid_keys: 有效的关键key列表
    :type valid_keys: list

    :return: 只包含有效keys的http header, type: oss2.CaseInsensitiveDict
    """
    if headers is None or valid_keys is None:
        return None

    headers = http.CaseInsensitiveDict(headers)
    valid_headers = http.CaseInsensitiveDict()

    for key in valid_keys:
        if headers.get(key) is not None:
            valid_headers[key] = headers[key]

    if len(valid_headers) == 0:
        valid_headers = None

    return valid_headers

def _filter_invalid_headers(headers=None, invalid_keys=None):
    """过滤无效keys的http header

    :param headers: 需要过滤的header
    :type headers: 可以是dict，建议是oss2.CaseInsensitiveDict

    :param invalid_keys: 无效的关键key列表
    :type invalid_keys: list

    :return: 过滤无效header之后的http headers, type: oss2.CaseInsensitiveDict
    """
    if headers is None or invalid_keys is None:
        return None

    headers = http.CaseInsensitiveDict(headers)
    valid_headers = headers.copy()

    for key in invalid_keys:
        if valid_headers.get(key) is not None:
            valid_headers.pop(key)

    if len(valid_headers) == 0:
        valid_headers = None

    return valid_headers


def _populate_valid_params(params=None, valid_keys=None):
    """构建只包含有效keys的params

    :param params: 需要过滤的params
    :type params: dict

    :param valid_keys: 有效的关键key列表
    :type valid_keys: list

    :return: 只包含有效keys的params
    """
    if params is None or valid_keys is None:
        return None

    valid_params = dict()

    for key in valid_keys:
        if params.get(key) is not None:
            valid_params[key] = params[key]

    if len(valid_params) == 0:
        valid_params = None

    return valid_params

class _ResumableOperation(object):
    def __init__(self, bucket, key, filename, size, store,
                 progress_callback=None, versionid=None):
        self.bucket = bucket
        self.key = to_string(key)
        self.filename = filename
        self.size = size

        self._abspath = os.path.abspath(filename)

        self.__store = store

        if versionid is None:
            self.__record_key = self.__store.make_store_key(bucket.bucket_name, self.key, self._abspath)
        else:
            self.__record_key = self.__store.make_store_key(bucket.bucket_name, self.key, self._abspath, versionid)

        logger.debug("Init _ResumableOperation, record_key: {0}".format(self.__record_key))

        # protect self.__progress_callback
        self.__plock = threading.Lock()
        self.__progress_callback = progress_callback

    def _del_record(self):
        self.__store.delete(self.__record_key)

    def _put_record(self, record):
        self.__store.put(self.__record_key, record)

    def _get_record(self):
        return self.__store.get(self.__record_key)

    def _report_progress(self, consumed_size):
        if self.__progress_callback:
            with self.__plock:
                self.__progress_callback(consumed_size, self.size)


class _ObjectInfo(object):
    def __init__(self):
        self.size = None
        self.etag = None
        self.mtime = None

    @staticmethod
    def make(head_object_result):
        objectInfo = _ObjectInfo()
        objectInfo.size = head_object_result.content_length
        objectInfo.etag = head_object_result.etag
        objectInfo.mtime = head_object_result.last_modified

        return objectInfo


class _ResumableDownloader(_ResumableOperation):
    def __init__(self, bucket, key, filename, objectInfo,
                 part_size=None,
                 store=None,
                 progress_callback=None,
                 num_threads=None,
                 params=None,
                 headers=None):
        versionid = None
        if params is not None and params.get('versionId') is not None:
            versionid = params.get('versionId')
        super(_ResumableDownloader, self).__init__(bucket, key, filename, objectInfo.size,
                                                   store or ResumableDownloadStore(),
                                                   progress_callback=progress_callback, 
                                                   versionid=versionid)
        self.objectInfo = objectInfo
        self.__op = 'ResumableDownload'
        self.__part_size = defaults.get(part_size, defaults.multiget_part_size)
        self.__part_size = _determine_part_size_internal(self.size, self.__part_size, _MAX_MULTIGET_PART_COUNT)

        self.__tmp_file = None
        self.__num_threads = defaults.get(num_threads, defaults.multiget_num_threads)
        self.__finished_parts = None
        self.__finished_size = None
        self.__params = params
        self.__headers = headers

        # protect record
        self.__lock = threading.Lock()
        self.__record = None
        logger.debug("Init _ResumableDownloader, bucket: {0}, key: {1}, part_size: {2}, num_thread: {3}".format(
            bucket.bucket_name, to_string(key), self.__part_size, self.__num_threads))

    def download(self, server_crc = None):
        self.__load_record()

        parts_to_download = self.__get_parts_to_download()
        logger.debug("Parts need to download: {0}".format(parts_to_download))

        # create tmp file if it is does not exist
        open(self.__tmp_file, 'a').close()

        q = TaskQueue(functools.partial(self.__producer, parts_to_download=parts_to_download),
                      [self.__consumer] * self.__num_threads)
        q.run()

        if self.bucket.enable_crc:
            parts = sorted(self.__finished_parts, key=lambda p: p.part_number)
            object_crc = utils.calc_obj_crc_from_parts(parts)
            utils.check_crc('resume download', object_crc, server_crc, None)

        utils.force_rename(self.__tmp_file, self.filename)

        self._report_progress(self.size)
        self._del_record()

    def __producer(self, q, parts_to_download=None):
        for part in parts_to_download:
            q.put(part)

    def __consumer(self, q):
        while q.ok():
            part = q.get()
            if part is None:
                break

            self.__download_part(part)

    def __download_part(self, part):
        self._report_progress(self.__finished_size)

        with open(self.__tmp_file, 'rb+') as f:
            f.seek(part.start, os.SEEK_SET)

            headers = _populate_valid_headers(self.__headers, [OSS_REQUEST_PAYER, OSS_TRAFFIC_LIMIT])
            if headers is None:
                headers = http.CaseInsensitiveDict()
            headers[IF_MATCH] = self.objectInfo.etag
            headers[IF_UNMODIFIED_SINCE] = utils.http_date(self.objectInfo.mtime)

            result = self.bucket.get_object(self.key, byte_range=(part.start, part.end - 1), headers=headers, params=self.__params)
            utils.copyfileobj_and_verify(result, f, part.end - part.start, request_id=result.request_id)

        part.part_crc = result.client_crc
        logger.debug("down part success, add part info to record, part_number: {0}, start: {1}, end: {2}".format(
            part.part_number, part.start, part.end))

        self.__finish_part(part)

    def __load_record(self):
        record = self._get_record()
        logger.debug("Load record return {0}".format(record))

        if record and not self.__is_record_sane(record):
            logger.warn("The content of record is invalid, delete the record")
            self._del_record()
            record = None

        if record and not os.path.exists(self.filename + record['tmp_suffix']):
            logger.warn("Temp file: {0} does not exist, delete the record".format(
                self.filename + record['tmp_suffix']))
            self._del_record()
            record = None

        if record and self.__is_remote_changed(record):
            logger.warn("Object: {0} has been overwritten，delete the record and tmp file".format(self.key))
            utils.silently_remove(self.filename + record['tmp_suffix'])
            self._del_record()
            record = None

        if not record:
            record = {'op_type': self.__op, 'bucket': self.bucket.bucket_name, 'key': self.key,
                      'size': self.objectInfo.size, 'mtime': self.objectInfo.mtime, 'etag': self.objectInfo.etag,
                      'part_size': self.__part_size, 'file_path': self._abspath, 'tmp_suffix': self.__gen_tmp_suffix(),
                      'parts': []}
            logger.debug('Add new record, bucket: {0}, key: {1}, part_size: {2}'.format(
                self.bucket.bucket_name, self.key, self.__part_size))
            self._put_record(record)

        self.__tmp_file = self.filename + record['tmp_suffix']
        self.__part_size = record['part_size']
        self.__finished_parts = list(
            _PartToProcess(p['part_number'], p['start'], p['end'], p['part_crc']) for p in record['parts'])
        self.__finished_size = sum(p.size for p in self.__finished_parts)
        self.__record = record

    def __get_parts_to_download(self):
        assert self.__record

        all_set = set(_split_to_parts(self.size, self.__part_size))
        finished_set = set(self.__finished_parts)

        return sorted(list(all_set - finished_set), key=lambda p: p.part_number)

    def __is_record_sane(self, record):
        try:
            if record['op_type'] != self.__op:
                logger.error('op_type invalid, op_type in record:{0} is invalid'.format(record['op_type']))
                return False

            for key in ('etag', 'tmp_suffix', 'file_path', 'bucket', 'key'):
                if not isinstance(record[key], str):
                    logger.error('{0} is not a string: {1}'.format(key, record[key]))
                    return False

            for key in ('part_size', 'size', 'mtime'):
                if not isinstance(record[key], int):
                    logger.error('{0} is not an integer: {1}'.format(key, record[key]))
                    return False

            if not isinstance(record['parts'], list):
                logger.error('{0} is not a list: {1}'.format(key, record[key]))
                return False
        except KeyError as e:
            logger.error('Key not found: {0}'.format(e.args))
            return False

        return True

    def __is_remote_changed(self, record):
        return (record['mtime'] != self.objectInfo.mtime or
            record['size'] != self.objectInfo.size or
            record['etag'] != self.objectInfo.etag)

    def __finish_part(self, part):
        with self.__lock:
            self.__finished_parts.append(part)
            self.__finished_size += part.size

            self.__record['parts'].append({'part_number': part.part_number,
                                           'start': part.start,
                                           'end': part.end,
                                           'part_crc': part.part_crc})
            self._put_record(self.__record)

    def __gen_tmp_suffix(self):
        return '.tmp-' + ''.join(random.choice(string.ascii_lowercase) for i in range(12))


class _ResumableUploader(_ResumableOperation):
    """以断点续传方式上传文件。

    :param bucket: :class:`Bucket <oss2.Bucket>` 对象
    :param key: 文件名
    :param filename: 待上传的文件名
    :param size: 文件总长度
    :param store: 用来保存进度的持久化存储
    :param headers: 传给 `init_multipart_upload` 的HTTP头部
    :param part_size: 分片大小。优先使用用户提供的值。如果用户没有指定，那么对于新上传，计算出一个合理值；对于老的上传，采用第一个
        分片的大小。
    :param progress_callback: 上传进度回调函数。参见 :ref:`progress_callback` 。
    """

    def __init__(self, bucket, key, filename, size,
                 store=None,
                 headers=None,
                 part_size=None,
                 progress_callback=None,
                 num_threads=None,
                 params=None):
        super(_ResumableUploader, self).__init__(bucket, key, filename, size,
                                                 store or ResumableStore(),
                                                 progress_callback=progress_callback)

        self.__op = 'ResumableUpload'
        self.__headers = headers

        self.__part_size = defaults.get(part_size, defaults.part_size)

        self.__mtime = os.path.getmtime(filename)

        self.__num_threads = defaults.get(num_threads, defaults.multipart_num_threads)

        self.__upload_id = None

        self.__params = params

        # protect below fields
        self.__lock = threading.Lock()
        self.__record = None
        self.__finished_size = 0
        self.__finished_parts = None
        self.__encryption = False
        self.__record_upload_context = False
        self.__upload_context = None

        if isinstance(self.bucket, CryptoBucket):
            self.__encryption = True
            self.__record_upload_context = True

        logger.debug("Init _ResumableUploader, bucket: {0}, key: {1}, part_size: {2}, num_thread: {3}".format(
            bucket.bucket_name, to_string(key), self.__part_size, self.__num_threads))

    def upload(self):
        self.__load_record()

        parts_to_upload = self.__get_parts_to_upload(self.__finished_parts)
        parts_to_upload = sorted(parts_to_upload, key=lambda p: p.part_number)
        logger.debug("Parts need to upload: {0}".format(parts_to_upload))

        q = TaskQueue(functools.partial(self.__producer, parts_to_upload=parts_to_upload),
                      [self.__consumer] * self.__num_threads)
        q.run()

        self._report_progress(self.size)

        headers = _populate_valid_headers(self.__headers, [OSS_REQUEST_PAYER, OSS_OBJECT_ACL])
        result = self.bucket.complete_multipart_upload(self.key, self.__upload_id, self.__finished_parts, headers=headers)
        self._del_record()

        return result

    def __producer(self, q, parts_to_upload=None):
        for part in parts_to_upload:
            q.put(part)

    def __consumer(self, q):
        while True:
            part = q.get()
            if part is None:
                break

            self.__upload_part(part)

    def __upload_part(self, part):
        with open(to_unicode(self.filename), 'rb') as f:
            self._report_progress(self.__finished_size)

            f.seek(part.start, os.SEEK_SET)
            headers = _populate_valid_headers(self.__headers, [OSS_REQUEST_PAYER, OSS_TRAFFIC_LIMIT])
            if self.__encryption:
                result = self.bucket.upload_part(self.key, self.__upload_id, part.part_number,
                                                 utils.SizedFileAdapter(f, part.size), headers=headers,
                                                 upload_context=self.__upload_context)
            else:
                result = self.bucket.upload_part(self.key, self.__upload_id, part.part_number,
                                                 utils.SizedFileAdapter(f, part.size), headers=headers)

            logger.debug("Upload part success, add part info to record, part_number: {0}, etag: {1}, size: {2}".format(
                part.part_number, result.etag, part.size))
            self.__finish_part(PartInfo(part.part_number, result.etag, size=part.size, part_crc=result.crc))

    def __finish_part(self, part_info):
        with self.__lock:
            self.__finished_parts.append(part_info)
            self.__finished_size += part_info.size

    def __load_record(self):
        record = self._get_record()
        logger.debug("Load record return {0}".format(record))

        if record and not self.__is_record_sane(record):
            logger.warn("The content of record is invalid, delete the record")
            self._del_record()
            record = None

        if record and self.__file_changed(record):
            logger.warn("File: {0} has been changed, delete the record".format(self.filename))
            self._del_record()
            record = None

        if record and not self.__upload_exists(record['upload_id']):
            logger.warn('Multipart upload: {0} does not exist, delete the record'.format(record['upload_id']))
            self._del_record()
            record = None

        if not record:
            params = _populate_valid_params(self.__params, [Bucket.SEQUENTIAL])
            part_size = determine_part_size(self.size, self.__part_size)
            logger.debug("Upload File size: {0}, User-specify part_size: {1}, Calculated part_size: {2}".format(
                self.size, self.__part_size, part_size))
            if self.__encryption:
                upload_context = models.MultipartUploadCryptoContext(self.size, part_size)
                upload_id = self.bucket.init_multipart_upload(self.key, self.__headers, params,
                                                              upload_context).upload_id
                if self.__record_upload_context:
                    material = upload_context.content_crypto_material
                    material_record = {'wrap_alg': material.wrap_alg, 'cek_alg': material.cek_alg,
                                       'encrypted_key': b64encode_as_string(material.encrypted_key),
                                       'encrypted_iv': b64encode_as_string(material.encrypted_iv),
                                       'mat_desc': material.mat_desc}
            else:
                upload_id = self.bucket.init_multipart_upload(self.key, self.__headers, params).upload_id

            record = {'op_type': self.__op, 'upload_id': upload_id, 'file_path': self._abspath, 'size': self.size,
                      'mtime': self.__mtime, 'bucket': self.bucket.bucket_name, 'key': self.key, 'part_size': part_size}

            if self.__record_upload_context:
                record['content_crypto_material'] = material_record

            logger.debug('Add new record, bucket: {0}, key: {1}, upload_id: {2}, part_size: {3}'.format(
                self.bucket.bucket_name, self.key, upload_id, part_size))

            self._put_record(record)

        self.__record = record
        self.__part_size = self.__record['part_size']
        self.__upload_id = self.__record['upload_id']
        if self.__record_upload_context:
            if 'content_crypto_material' in self.__record:
                material_record = self.__record['content_crypto_material']
                wrap_alg = material_record['wrap_alg']
                cek_alg = material_record['cek_alg']
                if cek_alg != self.bucket.crypto_provider.cipher.alg or wrap_alg != self.bucket.crypto_provider.wrap_alg:
                    err_msg = 'Envelope or data encryption/decryption algorithm is inconsistent'
                    raise exceptions.InconsistentError(err_msg, self)
                content_crypto_material = models.ContentCryptoMaterial(self.bucket.crypto_provider.cipher,
                                                                       material_record['wrap_alg'],
                                                                       b64decode_from_string(
                                                                           material_record['encrypted_key']),
                                                                       b64decode_from_string(
                                                                           material_record['encrypted_iv']),
                                                                       material_record['mat_desc'])
                self.__upload_context = models.MultipartUploadCryptoContext(self.size, self.__part_size,
                                                                            content_crypto_material)

            else:
                err_msg = 'If record_upload_context flag is true, content_crypto_material must in the the record'
                raise exceptions.InconsistentError(err_msg, self)

        else:
            if 'content_crypto_material' in self.__record:
                err_msg = 'content_crypto_material must in the the record, but record_upload_context flat is false'
                raise exceptions.InvalidEncryptionRequest(err_msg, self)

        self.__finished_parts = self.__get_finished_parts()
        self.__finished_size = sum(p.size for p in self.__finished_parts)

    def __get_finished_parts(self):
        parts = []

        valid_headers = _filter_invalid_headers(self.__headers,
                [OSS_SERVER_SIDE_ENCRYPTION, OSS_SERVER_SIDE_DATA_ENCRYPTION])

        for part in PartIterator(self.bucket, self.key, self.__upload_id, headers=valid_headers):
            parts.append(part)

        return parts

    def __upload_exists(self, upload_id):
        try:
            valid_headers = _filter_invalid_headers(self.__headers,
                      [OSS_SERVER_SIDE_ENCRYPTION, OSS_SERVER_SIDE_DATA_ENCRYPTION])
            list(iterators.PartIterator(self.bucket, self.key, upload_id, '0', max_parts=1, headers=valid_headers))
        except exceptions.NoSuchUpload:
            return False
        else:
            return True

    def __file_changed(self, record):
        return record['mtime'] != self.__mtime or record['size'] != self.size

    def __get_parts_to_upload(self, parts_uploaded):
        all_parts = _split_to_parts(self.size, self.__part_size)
        if not parts_uploaded:
            return all_parts

        all_parts_map = dict((p.part_number, p) for p in all_parts)

        for uploaded in parts_uploaded:
            if uploaded.part_number in all_parts_map:
                del all_parts_map[uploaded.part_number]

        return all_parts_map.values()

    def __is_record_sane(self, record):
        try:
            if record['op_type'] != self.__op:
                logger.error('op_type invalid, op_type in record:{0} is invalid'.format(record['op_type']))
                return False

            for key in ('upload_id', 'file_path', 'bucket', 'key'):
                if not isinstance(record[key], str):
                    logger.error('Type Error, {0} in record is not a string type: {1}'.format(key, record[key]))
                    return False

            for key in ('size', 'part_size'):
                if not isinstance(record[key], int):
                    logger.error('Type Error, {0} in record is not an integer type: {1}'.format(key, record[key]))
                    return False

            if not isinstance(record['mtime'], int) and not isinstance(record['mtime'], float):
                logger.error(
                    'Type Error, mtime in record is not a float or an integer type: {0}'.format(record['mtime']))
                return False

        except KeyError as e:
            logger.error('Key not found: {0}'.format(e.args))
            return False

        return True


_UPLOAD_TEMP_DIR = '.py-oss-upload'
_DOWNLOAD_TEMP_DIR = '.py-oss-download'


class _ResumableStoreBase(object):
    def __init__(self, root, dir):
        logger.debug("Init ResumableStoreBase, root path: {0}, temp dir: {1}".format(root, dir))
        self.dir = os.path.join(root, dir)

        if os.path.isdir(self.dir):
            return

        utils.makedir_p(self.dir)

    def get(self, key):
        pathname = self.__path(key)

        logger.debug('ResumableStoreBase: get key: {0} from file path: {1}'.format(key, pathname))

        if not os.path.exists(pathname):
            logger.debug("file {0} is not exist".format(pathname))
            return None

        # json.load()返回的总是unicode，对于Python2，我们将其转换
        # 为str。

        try:
            with open(to_unicode(pathname), 'r') as f:
                content = json.load(f)
        except ValueError:
            os.remove(pathname)
            return None
        else:
            return stringify(content)

    def put(self, key, value):
        pathname = self.__path(key)

        with open(to_unicode(pathname), 'w') as f:
            json.dump(value, f)

        logger.debug('ResumableStoreBase: put key: {0} to file path: {1}, value: {2}'.format(key, pathname, value))

    def delete(self, key):
        pathname = self.__path(key)
        os.remove(pathname)

        logger.debug('ResumableStoreBase: delete key: {0}, file path: {1}'.format(key, pathname))

    def __path(self, key):
        return os.path.join(self.dir, key)


def _normalize_path(path):
    return os.path.normpath(os.path.normcase(path))


class ResumableStore(_ResumableStoreBase):
    """保存断点上传断点信息的类。

    每次上传的信息会保存在 `root/dir/` 下面的某个文件里。

    :param str root: 父目录，缺省为HOME
    :param str dir: 子目录，缺省为 `_UPLOAD_TEMP_DIR`
    """

    def __init__(self, root=None, dir=None):
        super(ResumableStore, self).__init__(root or os.path.expanduser('~'), dir or _UPLOAD_TEMP_DIR)

    @staticmethod
    def make_store_key(bucket_name, key, filename):
        filepath = _normalize_path(filename)

        oss_pathname = 'oss://{0}/{1}'.format(bucket_name, key)
        return utils.md5_string(oss_pathname) + '--' + utils.md5_string(filepath)


class ResumableDownloadStore(_ResumableStoreBase):
    """保存断点下载断点信息的类。

    每次下载的断点信息会保存在 `root/dir/` 下面的某个文件里。

    :param str root: 父目录，缺省为HOME
    :param str dir: 子目录，缺省为 `_DOWNLOAD_TEMP_DIR`
    """

    def __init__(self, root=None, dir=None):
        super(ResumableDownloadStore, self).__init__(root or os.path.expanduser('~'), dir or _DOWNLOAD_TEMP_DIR)

    @staticmethod
    def make_store_key(bucket_name, key, filename, version_id=None):
        filepath = _normalize_path(filename)

        if version_id is None:
            oss_pathname = 'oss://{0}/{1}'.format(bucket_name, key)
        else:
            oss_pathname = 'oss://{0}/{1}?versionid={2}'.format(bucket_name, key, version_id)
        return utils.md5_string(oss_pathname) + '--' + utils.md5_string(filepath)


def make_upload_store(root=None, dir=None):
    return ResumableStore(root=root, dir=dir)


def make_download_store(root=None, dir=None):
    return ResumableDownloadStore(root=root, dir=dir)


class _PartToProcess(object):
    def __init__(self, part_number, start, end, part_crc=None):
        self.part_number = part_number
        self.start = start
        self.end = end
        self.part_crc = part_crc

    @property
    def size(self):
        return self.end - self.start

    def __hash__(self):
        return hash(self.__key)

    def __eq__(self, other):
        return self.__key == other.__key

    @property
    def __key(self):
        return self.part_number, self.start, self.end