o MfP@sdZddlZddlZddlmZddlmZddlmZddlm Z ddlm Z dZ d Z hd Z hd ZGd d d ejZGdddejZGdddeZdS)zFModule for file-like access of blobs, usually invoked via Blob.open().N)RequestRangeNotSatisfiable)_NUM_RETRIES_MESSAGE) DEFAULT_RETRY)%DEFAULT_RETRY_IF_GENERATION_SPECIFIED)ConditionalRetryPolicyii>retrytimeout raw_downloadif_generation_matchif_generation_not_matchif_metageneration_matchif_metageneration_not_match> rrchecksum num_retries content_typepredefined_aclr r r r c@sheZdZdZdefddZdddZddd Zdd d Zd dZ e ddZ ddZ ddZ ddZdS) BlobReaderaA file-like object that reads from a blob. :type blob: 'google.cloud.storage.blob.Blob' :param blob: The blob to download. :type chunk_size: long :param chunk_size: (Optional) The minimum number of bytes to read at a time. If fewer bytes than the chunk_size are requested, the remainder is buffered. The default is the chunk_size of the blob, or 40MiB. :type retry: google.api_core.retry.Retry or google.cloud.storage.retry.ConditionalRetryPolicy :param retry: (Optional) How to retry the RPC. A None value will disable retries. A google.api_core.retry.Retry value will enable retries, and the object will define retriable response codes and errors and configure backoff and timeout options. A google.cloud.storage.retry.ConditionalRetryPolicy value wraps a Retry object and activates it only if certain conditions are met. This class exists to provide safe defaults for RPC calls that are not technically safe to retry normally (due to potential data duplication or other side-effects) but become safe to retry if a condition such as if_metageneration_match is set. See the retry.py source code and docstrings in this package (google.cloud.storage.retry) for information on retry types and how to configure them. Media operations (downloads and uploads) do not support non-default predicates in a Retry object. The default will always be used. Other configuration changes for Retry objects such as delays and deadlines are respected. :param download_kwargs: Keyword arguments to pass to the underlying API calls. The following arguments are supported: - ``if_generation_match`` - ``if_generation_not_match`` - ``if_metageneration_match`` - ``if_metageneration_not_match`` - ``timeout`` Note that download_kwargs are also applied to blob.reload(), if a reload is needed during seek(). NcKsX|D]}|tvrtd|dq||_d|_t|_|p"|jp"t|_ ||_ ||_ dS)Nz-BlobReader does not support keyword argument .r) VALID_DOWNLOAD_KWARGS ValueError_blob_posioBytesIO_buffer chunk_sizeDEFAULT_CHUNK_SIZE _chunk_size_retry_download_kwargs)selfblobrrdownload_kwargskwargr$]/var/www/html/analyze/labelStudio/lib/python3.10/site-packages/google/cloud/storage/fileio.py__init__ms   zBlobReader.__init__cCs||j|}|t|}|dks|dkr|j|j7_t|}|jd|jd|j}|dkrB|t||j }nd}z||j j d||d|j d|j 7}Wn tyaYnw|dkrt||kr|j||d|jd|d|}|jt||7_|S)Nr)startendrrr$) _checkClosedrreadlenrtellseektruncatemaxrrdownload_as_bytesrrrwrite)r sizeresultremaining_size read_size fetch_start fetch_endr$r$r%r+{s<        zBlobReader.readcCs ||SN)r+)r r3r$r$r%read1 zBlobReader.read1rcCs||jjdur|jjdi|j|j|j}|dkr#|}n|dkr,||}n |dkr6|jj|}|dvr>td||jjkrH|jj}||jkr`|j d|j d|}||_|S||}|j|j |d}|S)aSeek within the blob. This implementation of seek() uses knowledge of the blob size to validate that the reported position does not exceed the blob last byte. If the blob size is not already known it will call blob.reload(). Nr>rr<r=zinvalid whence valuer$) r*rr3reloadrrrr-rr.r/)r poswhenceinitial_offset target_posnew_pos differencer$r$r%r.s.       zBlobReader.seekcCs|jdSr9rcloser r$r$r%rFszBlobReader.closecC|jjSr9rclosedrGr$r$r%rJzBlobReader.closedcCdSNTr$rGr$r$r%readablezBlobReader.readablecCrLNFr$rGr$r$r%writablerOzBlobReader.writablecCrLrMr$rGr$r$r%seekablerOzBlobReader.seekabler')r)__name__ __module__ __qualname____doc__rr&r+r:r.rFpropertyrJrNrQrRr$r$r$r%r;s1  - (  rc@seZdZdZdddefddZeddZejddZd d Z d d Z d dZ ddZ ddZ ddZeddZddZddZddZdS) BlobWriteram A file-like object that writes to a blob. :type blob: 'google.cloud.storage.blob.Blob' :param blob: The blob to which to write. :type chunk_size: long :param chunk_size: (Optional) The maximum number of bytes to buffer before sending data to the server, and the size of each request when data is sent. Writes are implemented as a "resumable upload", so chunk_size for writes must be exactly a multiple of 256KiB as with other resumable uploads. The default is the chunk_size of the blob, or 40 MiB. :type text_mode: bool :param text_mode: (Deprecated) A synonym for ignore_flush. For backwards-compatibility, if True, sets ignore_flush to True. Use ignore_flush instead. This parameter will be removed in a future release. :type ignore_flush: bool :param ignore_flush: Makes flush() do nothing instead of raise an error. flush() without closing is not supported by the remote service and therefore calling it on this class normally results in io.UnsupportedOperation. However, that behavior is incompatible with some consumers and wrappers of file objects in Python, such as zipfile.ZipFile or io.TextIOWrapper. Setting ignore_flush will cause flush() to successfully do nothing, for compatibility with those contexts. The correct way to actually flush data to the remote server is to close() (using this object as a context manager is recommended). :type retry: google.api_core.retry.Retry or google.cloud.storage.retry.ConditionalRetryPolicy :param retry: (Optional) How to retry the RPC. A None value will disable retries. A google.api_core.retry.Retry value will enable retries, and the object will define retriable response codes and errors and configure backoff and timeout options. A google.cloud.storage.retry.ConditionalRetryPolicy value wraps a Retry object and activates it only if certain conditions are met. This class exists to provide safe defaults for RPC calls that are not technically safe to retry normally (due to potential data duplication or other side-effects) but become safe to retry if a condition such as if_metageneration_match is set. See the retry.py source code and docstrings in this package (google.cloud.storage.retry) for information on retry types and how to configure them. Media operations (downloads and uploads) do not support non-default predicates in a Retry object. The default will always be used. Other configuration changes for Retry objects such as delays and deadlines are respected. :param upload_kwargs: Keyword arguments to pass to the underlying API calls. The following arguments are supported: - ``if_generation_match`` - ``if_generation_not_match`` - ``if_metageneration_match`` - ``if_metageneration_not_match`` - ``timeout`` - ``content_type`` - ``num_retries`` - ``predefined_acl`` - ``checksum`` NFcKs`|D]}|tvrtd|dq||_t|_d|_|p!|jp!t|_|p&||_ ||_ ||_ dS)Nz-BlobWriter does not support keyword argument r) VALID_UPLOAD_KWARGSrr SlidingBufferr_upload_and_transportrrr _ignore_flushr_upload_kwargs)r r!r text_mode ignore_flushr upload_kwargsr#r$r$r%r&+s    zBlobWriter.__init__cC|jS)zGet the blob's default chunk size. :rtype: int or ``NoneType`` :returns: The current blob's chunk size, if it is set. )_BlobWriter__chunk_sizerGr$r$r%rDszBlobWriter._chunk_sizecCs2|dur|dkr|tdkrtdt||_dS)a Set the blob's default chunk size. :type value: int :param value: (Optional) The current blob's chunk size, if it is set. :raises: :class:`ValueError` if ``value`` is not ``None`` and is not a multiple of 256 KiB. Nrz$Chunk size must be a multiple of %d.)CHUNK_SIZE_MULTIPLErrc)r valuer$r$r%rMs   cCs6||j|}t|j|j}|r|||Sr9)r*rr2r,r_upload_chunks_from_buffer)r br? num_chunksr$r$r%r2]s   zBlobWriter.writecCs|jdd}|j}|jdd}|dur#tjttdd|tur#d}t|t r;|j d|j dd}|j |d}|j j |j jj|j|d|f|j|d |j|_dS) Nrrr=) stacklevelr r )ifGenerationMatchifMetagenerationMatch) query_params)rr)r^poprwarningswarnrDeprecationWarningr isinstancerget"get_retry_policy_if_conditions_metr_initiate_resumable_uploadbucketclientrrr\)r rrrrlr$r$r%_initiate_uploadis4    zBlobWriter._initiate_uploadcCsd|js||j\}}i}d|jvrd|jdi}t|D] }|j|fi|q|jdS)z$Upload a specified number of chunks.rN)r\rwr^rrrangetransmit_next_chunkrflush)r rhupload transportkwargs_r$r$r%rfs   z%BlobWriter._upload_chunks_from_buffercCs|jt|jSr9)rr-r,rGr$r$r%r-szBlobWriter.tellcCs|jstddS)NzCannot flush without finalizing upload. Use close() instead, or set ignore_flush=True when constructing this class (see docstring).)r]rUnsupportedOperationrGr$r$r%rzs zBlobWriter.flushcCs |jjs |d|jdS)Nr<)rrJrfrFrGr$r$r%rFs zBlobWriter.closecCrHr9rIrGr$r$r%rJrKzBlobWriter.closedcCrLrPr$rGr$r$r%rNrOzBlobWriter.readablecCrLrMr$rGr$r$r%rQrOzBlobWriter.writablecCrLrPr$rGr$r$r%rRrOzBlobWriter.seekable)rTrUrVrWrr&rXrsetterr2rwrfr-rzrFrJrNrQrRr$r$r$r%rYs,I    '   rYc@sfeZdZdZddZddZdddZd d Zd d Zd dZ ddZ ddZ ddZ e ddZdS)r[aA non-rewindable buffer that frees memory of chunks already consumed. This class is necessary because `google-resumable-media-python` expects `tell()` to work relative to the start of the file, not relative to a place in an intermediate buffer. Using this class, we present an external interface with consistent seek and tell behavior without having to actually store bytes already sent. Behavior of this class differs from an ordinary BytesIO buffer. `write()` will always append to the end of the file only and not change the seek position otherwise. `flush()` will delete all data already read (data to the left of the seek position). `tell()` will report the seek position of the buffer including all deleted data. Additionally the class implements __len__() which will report the size of the actual underlying buffer. This class does not attempt to implement the entire Python I/O interface. cCst|_d|_dS)Nr)rrr_cursorrGr$r$r%r&s  zSlidingBuffer.__init__cCsD||j}|jdtj|j|}|j||j|S)z>Append to the end of the buffer without changing the position.r)r*rr-r.rSEEK_ENDr2r)r rgbookmarkr?r$r$r%r2s     zSlidingBuffer.writer'cCs*||j|}|jt|7_|S)zRead and move the cursor.)r*rr+rr,)r r3datar$r$r%r+s zSlidingBuffer.readcCsF||j}|jd|jd|j||jddS)z@Delete already-read data (all data to the left of the position).rN)r*rr+r.r/r2)r leftoverr$r$r%rzs     zSlidingBuffer.flushcCrb)z>Report how many bytes have been read from the buffer in total.)rrGr$r$r%r-szSlidingBuffer.tellcCsb||j}||j}|j|tj}|||kr!||jkr+|j|td||_|jS)a]Seek to a position (backwards only) within the internal buffer. This implementation of seek() verifies that the seek destination is contained in _buffer. It will raise ValueError if the destination byte has already been purged from the buffer. The "whence" argument is not supported in this implementation. zCannot seek() to that value.)r*rr-rr.rSEEK_CURr)r r?buffer_initial_posrDbuffer_seek_resultr$r$r%r.s     zSlidingBuffer.seekcCs*|j}|jdtj}|j||S)z7Determine the size of the buffer by seeking to the end.r)rr-r.rr)r rlengthr$r$r%__len__s  zSlidingBuffer.__len__cC |jSr9rErGr$r$r%rF r;zSlidingBuffer.closecCrr9)rr*rGr$r$r%r*#r;zSlidingBuffer._checkClosedcCrHr9rIrGr$r$r%rJ&rKzSlidingBuffer.closedNrS)rTrUrVrWr&r2r+rzr-r.rrFr*rXrJr$r$r$r%r[s  r[)rWrrngoogle.api_core.exceptionsrgoogle.cloud.storage._helpersrgoogle.cloud.storage.retryrrrrdrrrZBufferedIOBaserrYobjectr[r$r$r$r%s"      *c