o
    viG7                     @   s|  d Z ddlZddlZddlmZ ddlmZ ddlmZ ddl	m
Z
mZmZmZmZ ddlmZ ddlmZ d	d
lmZmZmZmZmZmZ d	dlmZ d	dlmZmZ e
r\d	dlm Z  e!e"Z#e$dZ%dZ&dddZ'eG dd dZ(e				dBdee( dee) de)de)dee) dee) dee*e)e)f  dee+e)  de,e+e* e+e* ee) f fddZ-G dd  d eZ.G d!d" d"eZ/			dCd#d$d%e*dee) dee*e)e)f  dee) ddfd&d'Z0d(e*fd)d*Z1d+e*fd,d-Z2d.e*fd/d0Z3d#d$d1e)ddfd2d3Z4d#d$d4e*d5e5d1e)ddf
d6d7Z6d4e*d8e(d5e5de+e) fd9d:Z7d;e+e* d<e)de/fd=d>Z8d#d$d?e+e) d5e5de+e* fd@dAZ9dS )Dz.Git LFS related type definitions and utilities    N)	dataclass)ceil)getsize)TYPE_CHECKINGBinaryIOIterableOptional	TypedDict)unquote)	constants   )build_hf_headersfix_hf_endpoint_in_urlhf_raise_for_statushttp_backoffloggingvalidate_hf_hub_args)SliceFileObj)sha256sha_fileobj)CommitOperationAddz^[0-9a-f]{40}$zlfs-multipart-uploadzapplication/vnd.git-lfs+json)AcceptzContent-Typec                   @   s`   e Zd ZU dZeed< eed< eed< edefddZ	edefd	d
Z
edefddZdS )
UploadInfoad  
    Dataclass holding required information to determine whether a blob
    should be uploaded to the hub using the LFS protocol or the regular protocol

    Args:
        sha256 (`bytes`):
            SHA256 hash of the blob
        size (`int`):
            Size in bytes of the blob
        sample (`bytes`):
            First 512 bytes of the blob
    r   sizesamplepathc                 C   s\   t |}t|d}|dd d }t|}W d    n1 s"w   Y  | |||dS )Nrb   r   r   r   )r   ioopenpeekr   )clsr   r   filer   sha r%   R/home/air/biblejyuku/back/venv/lib/python3.10/site-packages/huggingface_hub/lfs.py	from_pathI   s   
zUploadInfo.from_pathdatac                 C   s&   t | }| t||d d |dS )Nr   )r   r   r   )r   digestlen)r"   r(   r$   r%   r%   r&   
from_bytesQ   s   zUploadInfo.from_bytesfileobjc                 C   sD   | d}|dtj t|}| }|dtj | |||dS )Nr   r   r   )readseekr   SEEK_SETr   tell)r"   r,   r   r$   r   r%   r%   r&   from_fileobjV   s   
zUploadInfo.from_fileobjN)__name__
__module____qualname____doc__bytes__annotations__intclassmethodstrr'   r+   r   r1   r%   r%   r%   r&   r   6   s   
 r   upload_infostoken	repo_typerepo_idrevisionendpointheaders	transfersreturnc                 C   s  |dur|nt j}d}|t jv rt j| }| d| | d}	d|dur&|nddgdd	 | D d
d}
|dur?dt|i|
d< i tt|d|pJi }td|	||
d}t| | }|	dd}t
|tsktd|	d}t
|trw|nd}dd	 |D dd	 |D |fS )a  
    Requests the LFS batch endpoint to retrieve upload instructions

    Learn more: https://github.com/git-lfs/git-lfs/blob/main/docs/api/batch.md

    Args:
        upload_infos (`Iterable` of `UploadInfo`):
            `UploadInfo` for the files that are being uploaded, typically obtained
            from `CommitOperationAdd.upload_info`
        repo_type (`str`):
            Type of the repo to upload to: `"model"`, `"dataset"` or `"space"`.
        repo_id (`str`):
            A namespace (user or an organization) and a repo name separated
            by a `/`.
        revision (`str`, *optional*):
            The git revision to upload to.
        headers (`dict`, *optional*):
            Additional headers to include in the request
        transfers (`list`, *optional*):
            List of transfer methods to use. Defaults to ["basic", "multipart"].

    Returns:
        `LfsBatchInfo`: 3-tuple:
            - First element is the list of upload instructions from the server
            - Second element is a list of errors, if any
            - Third element is the chosen transfer adapter if provided by the server (e.g. "basic", "multipart", "xet")

    Raises:
        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
            If an argument is invalid or the server response is malformed.
        [`HfHubHTTPError`]
            If the server returned an error.
    N /z.git/info/lfs/objects/batchuploadbasic	multipartc                 S   s   g | ]}|j  |jd qS )oidr   )r   hexr   ).0rF   r%   r%   r&   
<listcomp>   s    z'post_lfs_batch_info.<locals>.<listcomp>r   )	operationrB   objects	hash_algonameref)r<   POSTrA   jsonrO   zMalformed response from servertransferc                 S   s   g | ]
}d |vrt |qS error)_validate_batch_actionsrL   objr%   r%   r&   rM          c                 S   s   g | ]
}d |v rt |qS rW   )_validate_batch_errorrZ   r%   r%   r&   rM      r\   )r   ENDPOINTREPO_TYPES_URL_PREFIXESr
   LFS_HEADERSr   r   r   rU   get
isinstancelist
ValueErrorr:   )r;   r<   r=   r>   r?   r@   rA   rB   
url_prefix	batch_urlpayloadresp
batch_inforO   chosen_transferr%   r%   r&   post_lfs_batch_info`   sB   ,



rk   c                   @   s   e Zd ZU eed< eed< dS )PayloadPartT
partNumberetagN)r2   r3   r4   r8   r7   r:   r%   r%   r%   r&   rl      s   
 rl   c                   @   s&   e Zd ZU dZeed< ee ed< dS )CompletionPayloadTz?Payload that will be sent to the Hub when uploading multi-part.rJ   partsN)r2   r3   r4   r5   r:   r7   rc   rl   r%   r%   r%   r&   ro      s   
 ro   rN   r   lfs_batch_actionc              	   C   sF  t | |d}|du rtd| j d dS |d d }t| |d d}|dur2t| |di }|d}	t|d	 |d
}
|	durjzt|	}	W n tt	fy`   td|	 dw t
| ||	|
d nt| |
d |durt| t|d	 |}td|t||d| jj | jjdd}t| t| j d dS )a9  
    Handles uploading a given object to the Hub with the LFS protocol.

    Can be a No-op if the content of the file is already present on the hub large file storage.

    Args:
        operation (`CommitOperationAdd`):
            The add operation triggering this upload.
        lfs_batch_action (`dict`):
            Upload instructions from the LFS batch endpoint for this object. See [`~utils.lfs.post_lfs_batch_info`] for
            more details.
        headers (`dict`, *optional*):
            Headers to include in the request, including authentication and user agent headers.

    Raises:
        [`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
            If `lfs_batch_action` is improperly formatted
        [`HfHubHTTPError`]
            If the upload resulted in an error
    actionsNzContent of file z. is already present upstream - skipping uploadrF   verifyheader
chunk_sizehref)r@   zTMalformed response from LFS batch endpoint: `chunk_size` should be an integer. Got 'z'.)rN   rt   ru   
upload_url)rN   rw   rS   )r<   rA   rI   rT   z: Upload successful)rY   ra   loggerdebugpath_in_repo_validate_lfs_actionr   r8   rd   	TypeError_upload_multi_part_upload_single_partr   r   upload_infor   rK   r   r   )rN   rq   r<   rA   r@   rr   upload_actionverify_actionrt   ru   rw   
verify_urlverify_respr%   r%   r&   
lfs_upload   sD   



r   
lfs_actionc                 C   s:   t | dtr| ddu st | dtstd| S ).validates response from the LFS batch endpointrv   rt   Nz"lfs_action is improperly formatted)rb   ra   r:   dictrd   )r   r%   r%   r&   r{   
  s
   r{   lfs_batch_actionsc                 C   sp   t | dtrt | dtstd| di d}| di d}|dur.t| |dur6t| | S )r   rJ   r   z)lfs_batch_actions is improperly formattedrr   rF   rs   N)rb   ra   r:   r8   rd   r{   )r   r   r   r%   r%   r&   rY     s    rY   lfs_batch_errorc                 C   sh   t | dtrt | dtstd| d}t |tr.t |dtr.t |dts2td| S )r   rJ   r   z'lfs_batch_error is improperly formattedrX   messagecode)rb   ra   r:   r8   rd   r   )r   
error_infor%   r%   r&   r]   "  s    
r]   rw   c                 C   sH   | j dd}td||d}t| W d   dS 1 sw   Y  dS )aK  
    Uploads `fileobj` as a single PUT HTTP request (basic LFS transfer protocol)

    Args:
        upload_url (`str`):
            The URL to PUT the file to.
        fileobj:
            The file-like object holding the data to upload.

    Raises:
        [`HfHubHTTPError`]
            If the upload resulted in an error.
    T	with_tqdmPUTr(   N)as_filer   r   )rN   rw   r,   responser%   r%   r&   r~   0  s   
"r~   rt   ru   c                 C   sH   t || j|d}t| ||d}td|t|| jj td}t| dS )z@
    Uploads file using HF multipart LFS transfer protocol.
    )rt   r   ru   )rN   sorted_parts_urlsru   rS   )rU   rA   N)	_get_sorted_parts_urlsr   _upload_parts_iterativelyr   _get_completion_payloadr   rK   r`   r   )rN   rt   ru   rw   r   response_headerscompletion_resr%   r%   r&   r}   D  s   r}   r   c                 C   sN   dd t dd |  D dd dD }t|}|t|j| kr%td|S )Nc                 S   s   g | ]\}}|qS r%   r%   )rL   _rw   r%   r%   r&   rM   \  s    z*_get_sorted_parts_urls.<locals>.<listcomp>c                 S   s2   g | ]\}}|  rt|d krt|d|fqS )r   
   )isdigitr*   r8   )rL   part_numrw   r%   r%   r&   rM   _  s    c                 S   s   | d S )Nr   r%   )tr%   r%   r&   <lambda>d  s    z(_get_sorted_parts_urls.<locals>.<lambda>)keyz0Invalid server response to upload large LFS file)sorteditemsr*   r   r   rd   )rt   r   ru   sorted_part_upload_urls	num_partsr%   r%   r&   r   [  s   r   r   rJ   c                 C   sf   g }t | D ]'\}}|d}|d u s|dkr#td| d|d  ||d |d q||dS )Nrn   rD   zInvalid etag (`z`) returned for part r   )rm   rn   )rJ   rp   )	enumeratera   rd   append)r   rJ   rp   part_numberrt   rn   r%   r%   r&   r   m  s   

r   r   c           	   
   C   s   g }| j dd<}t|D ].\}}t||| |d}td||d}t| ||j W d    n1 s6w   Y  qW d    |S 1 sGw   Y  |S )NTr   )	seek_from
read_limitr   r   )r   r   r   r   r   r   rA   )	rN   r   ru   rA   r,   part_idxpart_upload_urlfileobj_slicepart_upload_resr%   r%   r&   r   |  s&   
r   )NNNN)NNN):r5   r   redataclassesr   mathr   os.pathr   typingr   r   r   r   r	   urllib.parser
   huggingface_hubr   utilsr   r   r   r   r   r   
utils._lfsr   	utils.shar   r   _commit_apir   
get_loggerr2   rx   compile	OID_REGEXLFS_MULTIPART_UPLOAD_COMMANDr`   r   r:   r   rc   tuplerk   rl   ro   r   r{   rY   r]   r~   r8   r}   r   r   r   r%   r%   r%   r&   <module>   s    

)
	V

G
