import boto3
s3 = boto3.client('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3.download_file(
Bucket = 'my_bucket',
Key = 'my_container/my_file.csv'
Filename = 'downloaded_file.csv')

## [a.csv, b.csv, c.csv] -> [abc.zip]으로 압축되었다고 가정
## a.csv를 다운로드 받으려는 코드

import boto3
s3 = boto3.client('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3.download_file(Bucket = 'my_bucket',
                 Key = 'my_container/abc.zip',
                 Filename = 'downloaded_abc.zip')

from zipfile import ZipFile
z = ZipFile('downloaded_abc.zip')
z.extract('a.csv')
z.close()

s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
s3_object
# s3.Object(bucket_name = 'my_bucket', key='my_container/abc.zip')
# A resource representing an Amazon Simple Storage Service (S3) Object 
print(s3_object.content_length) # 전체 파일 길이를 반환(byte)
# 123
print(s3_object.get(Range='bytes=1-3'))
# s3_object의 1바이트 ~ 3바이트에 해당하는 부분을 반환합니다
# 1바이트~3바이트에 해당하는 데이터는 s3_object.get(Range='bytes=1-3')['Body']로 전달됩니다.
# 그럼 어디부터 어디까지 읽어야하는 걸까요? <- 우리가 구현하고자 하는 핵심입니다.

class botocore.response.StreamingBody(raw_stream, content_length):
'''Wrapper class for an http response body.
This provides a few additional conveniences that do not exist in the urllib3 model:
Set the timeout on the socket (i.e read() timeouts)
Auto validation of content length, if the amount of bytes we read does not match the content length, an exception is raised.'''
close()
'''Close the underlying http response stream.'''
iter_chunks(chunk_size=1024)
'''Return an iterator to yield chunks of chunk_size bytes from the raw stream.'''
iter_lines(chunk_size=1024, keepends=False)
'''Return an iterator to yield lines from the raw stream.
This is achieved by reading chunk of bytes (of size chunk_size) at a time from the raw stream, and then yielding lines from there.'''
next()
'''Return the next 1k chunk from the raw stream.'''
read(amt=None)
'''Read at most amt bytes from the stream.
If the amt argument is omitted, read all data.'''
set_socket_timeout(timeout)
'''Set the timeout seconds on the socket.'''

class zipfile.ZipFile(file, mode='r', compression=ZIP_STORED, allowZip64=True, compresslevel=None, *, strict_timestamps=True):
'''ZIP 파일을 엽니다, 여기서 file은 파일에 대한 경로 (문자열), 파일류 객체 또는 경로류 객체일 수 있습니다.'''

import boto3
from zipfile import ZipFile
s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
streaming_body = s3_object.get()['Body'] # Range를 지정하지 않으면 전체 바이트를 읽어온다
with ZipFile(streaming_body) as z:
  print(z.namelist())

import io

# io.RawIOBase는 io.IOBase를 상속하는 클래스
class S3File(io.RawIOBase):
    def __init__(self, s3_object):
        self.s3_object = s3_object 
        self.position = 0

    def seek(self, offset, whence=io.SEEK_SET):
        if whence == io.SEEK_SET: # 시작 위치로부터 position설정
            self.position = offset
        elif whence == io.SEEK_CUR: # 현재 위치로부터 position 설정
            self.position += offset
        elif whence == io.SEEK_END: # 끝 위치로부터 offset(음수)만큼 돌아가서 설정
            self.position = self.s3_object.content_length + offset
        else:
            raise ValueError("invalid whence (%r, should be %d, %d, %d)" % (
                whence, io.SEEK_SET, io.SEEK_CUR, io.SEEK_END
            ))

        return self.position

    def seekable(self):
        return True

import boto3
s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
s3_file = S3File(s3_object)
with ZipFile(s3_file) as z:
  print(z.namelist())

class S3File(io.RawIOBase):
    ...위와 동일...
		# 추가 부분
    @property
    def size(self):
        return self.s3_object.content_length

    def read(self, size=-1):
        if size == -1:
            # 함수 정의에 따라, EOF까지의 바이트를 반환
            range_header = "bytes=%d-" % self.position
            self.seek(offset=0, whence=io.SEEK_END)
        else:
            # new_position 설정
            new_position = self.position + size
            # content_length(size)를 넘긴 경우, EOF까지의 바이트를 반환
            if new_position >= self.size:
                return self.read()
						
            range_header = "bytes=%d-%d" % (self.position, new_position - 1)
            self.seek(offset=size, whence=io.SEEK_CUR)
				# rage_header에서 지정한 만큼만, StreamingBody로부터 읽어들여옵니다
        return self.s3_object.get(Range=range_header)["Body"].read()

    def readable(self):
        return True

import boto3
s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
s3_file = S3File(s3_object)
with ZipFile(s3_file) as z: # 이제 ZipFile은 s3_object의 내용을 읽을 수 있습니다!!
  print(z.namelist())
# [a.csv, b.csv, c.csv]

import boto3
s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
s3_file = S3File(s3_object)
with ZipFile(s3_file) as z:
  z.extract('a.csv')

# abc.zip에 400개의 파일을 압축했습니다.
# 방법 1 : zip 다운로드 이후 압축 해제
import boto3
from zipfile import ZipFile

s3 = boto3.client('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3.download_file(Bucket = 'my_bucket',
                 Key = 'my_container/abc.zip',
                 Filename = 'downloaded_abc.zip')

z = ZipFile('downloaded_abc.zip')
z.extract('a.csv')
z.close()
# 9.74s 소요

# 방법 2 : 정의한 S3File을 이용하여 zip 압축 내 필요 파일만 다운로드
s3 = boto3.resource('s3',
                 region_name = 'us-east-1',
                 aws_acess_key_id=AWS_KEY_ID,
                 aws_secret_access_key=AWS_SECRET)
s3_object = s3.Object('my_bucket',
                      'my_container/abc.zip')
s3_file = S3File(s3_object)
z = ZipFile(s3_file)
z.extract('a.csv')
z.close()
# 475ms 소요 (약 0.5초)