o
     
j+7                     @   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
 ddl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$dZ%dZ&dddZ'eG dd dZ(e				dCdee( de)dB de)de)de)dB de)dB de*e)e)f dB de+e) dB de,e+e* e+e* e)dB f fddZ-G d d! d!eZ.G d"d# d#eZ/			dDd$d%d&e*de)dB de*e)e)f dB de)dB ddfd'd(Z0d)e*fd*d+Z1d,e*fd-d.Z2d/e*fd0d1Z3d$d%d2e)ddfd3d4Z4d$d%d5e*d6e5d2e)ddf
d7d8Z6d5e*d9e(d6e5de+e) fd:d;Z7d<e+e* d=e)de/fd>d?Z8d$d%d@e+e) d6e5de+e* fdAdBZ9dS )Ez.Git LFS related type definitions and utilities    N)Iterable)	dataclass)ceil)getsize)TYPE_CHECKINGBinaryIO	TypedDict)unquote)	constants   )build_hf_headersfix_hf_endpoint_in_urlhf_raise_for_statushttp_backoffloggingvalidate_hf_hub_args)SliceFileObj)sha256sha_fileobj)CommitOperationAddz^[0-9a-f]{40}$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   sZ   t |}t|d}|dd d }t|}W d    n1 s!w   Y  | |||dS )Nrb   r   r   r   )r   openpeekr   )clsr   r   filer   sha r#   F/home/kuhnn/.local/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$   
from_bytesQ   s   zUploadInfo.from_bytesfileobjc                 C   sD   | d}|dtj t|}| }|dtj | |||dS )Nr   r   r   )readseekioSEEK_SETr   tell)r    r*   r   r"   r   r#   r#   r$   from_fileobjV   s   
zUploadInfo.from_fileobjN)__name__
__module____qualname____doc__bytes__annotations__intclassmethodstrr%   r)   r   r0   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   ).0rE   r#   r#   r$   
<listcomp>   s    z'post_lfs_batch_info.<locals>.<listcomp>r   )	operationrA   objects	hash_algonameref)r;   POSTr@   jsonrN   zMalformed response from servertransferc                 S   s   g | ]
}d |vrt |qS error)_validate_batch_actionsrK   objr#   r#   r$   rL          c                 S   s   g | ]
}d |v rt |qS rV   )_validate_batch_errorrY   r#   r#   r$   rL      r[   )r
   ENDPOINTREPO_TYPES_URL_PREFIXESr	   LFS_HEADERSr   r   r   rT   get
isinstancelist
ValueErrorr9   )r:   r;   r<   r=   r>   r?   r@   rA   
url_prefix	batch_urlpayloadresp
batch_inforN   chosen_transferr#   r#   r$   post_lfs_batch_info`   sB   ,



rj   c                   @   s   e Zd ZU eed< eed< dS )PayloadPartT
partNumberetagN)r1   r2   r3   r7   r6   r9   r#   r#   r#   r$   rk      s   
 rk   c                   @   s&   e Zd ZU dZeed< ee ed< dS )CompletionPayloadTz?Payload that will be sent to the Hub when uploading multi-part.rI   partsN)r1   r2   r3   r4   r9   r6   rb   rk   r#   r#   r#   r$   rn      s   
 rn   rM   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 uploadrE   verifyheader
chunk_sizehref)r?   zTMalformed response from LFS batch endpoint: `chunk_size` should be an integer. Got 'z'.)rM   rs   rt   
upload_url)rM   rv   rR   )r;   r@   rH   rS   z: Upload successful)rX   r`   loggerdebugpath_in_repo_validate_lfs_actionr   r7   rc   	TypeError_upload_multi_part_upload_single_partr   r   upload_infor   rJ   r   r   )rM   rp   r;   r@   r?   rq   upload_actionverify_actionrs   rt   rv   
verify_urlverify_respr#   r#   r$   
lfs_upload   sD   



r   
lfs_actionc                 C   s:   t | dtr| ddu st | dtstd| S ).validates response from the LFS batch endpointru   rs   Nz"lfs_action is improperly formatted)ra   r`   r9   dictrc   )r   r#   r#   r$   rz   
  s
   rz   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   rI   r   z)lfs_batch_actions is improperly formattedrq   rE   rr   N)ra   r`   r9   r7   rc   rz   )r   r   r   r#   r#   r$   rX     s    rX   lfs_batch_errorc                 C   sh   t | dtrt | dtstd| d}t |tr.t |dtr.t |dts2td| S )r   rI   r   z'lfs_batch_error is improperly formattedrW   messagecode)ra   r`   r9   r7   rc   r   )r   
error_infor#   r#   r$   r\   "  s    
r\   rv   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   )rM   rv   r*   responser#   r#   r$   r}   0  s   
"r}   rs   rt   c                 C   sH   t || j|d}t| ||d}td|t|| jj td}t| dS )z@
    Uploads file using HF multipart LFS transfer protocol.
    )rs   r~   rt   )rM   sorted_parts_urlsrt   rR   )rT   r@   N)	_get_sorted_parts_urlsr~   _upload_parts_iterativelyr   _get_completion_payloadr   rJ   r_   r   )rM   rs   rt   rv   r   response_headerscompletion_resr#   r#   r$   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#   )rK   _rv   r#   r#   r$   rL   \  s    z*_get_sorted_parts_urls.<locals>.<listcomp>c                 S   s2   g | ]\}}|  rt|d krt|d|fqS )r   
   )isdigitr(   r7   )rK   part_numrv   r#   r#   r$   rL   _  s    c                 S   s   | d S )Nr   r#   )tr#   r#   r$   <lambda>d  s    z(_get_sorted_parts_urls.<locals>.<lambda>)keyz0Invalid server response to upload large LFS file)sorteditemsr(   r   r   rc   )rs   r~   rt   sorted_part_upload_urls	num_partsr#   r#   r$   r   [  s   r   r   rI   c                 C   sf   g }t | D ]'\}}|d}|d u s|dkr#td| d|d  ||d |d q||dS )Nrm   rC   zInvalid etag (`z`) returned for part r   )rl   rm   )rI   ro   )	enumerater`   rc   append)r   rI   ro   part_numberrs   rm   r#   r#   r$   r   m  s   

r   r   c           	   
   C   s   g }| j dd<}t|D ].\}}t||| |d}td||d}t| ||j W d    n1 s6w   Y  qW d    |S 1 sGw   Y  |S )NTr   )	seek_from
read_limitr   r   )r   r   r   r   r   r   r@   )	rM   r   rt   r@   r*   part_idxpart_upload_urlfileobj_slicepart_upload_resr#   r#   r$   r   |  s&   
r   )NNNN)NNN):r4   r-   recollections.abcr   dataclassesr   mathr   os.pathr   typingr   r   r   urllib.parser	   huggingface_hubr
   utilsr   r   r   r   r   r   
utils._lfsr   	utils.shar   r   _commit_apir   
get_loggerr1   rw   compile	OID_REGEXLFS_MULTIPART_UPLOAD_COMMANDr_   r   r9   r   rb   tuplerj   rk   rn   r   rz   rX   r\   r}   r7   r|   r   r   r   r#   r#   r#   r$   <module>   s    

)
	V

G
