
    qiG7                        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jB                  e"      Z# ejH                  d      Z%dZ&dddZ'e G d d             Z(e	 	 	 	 d8d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       Z- G d de      Z. G d  d!e      Z/	 	 	 d9d"d#d$e*dee)   dee*e)e)f      dee)   ddfd%Z0d&e*fd'Z1d(e*fd)Z2d*e*fd+Z3d"d#d,e)ddfd-Z4d"d#d.e*d/e5d,e)ddf
d0Z6d.e*d1e(d/e5de+e)   fd2Z7d3e+e*   d4e)de/fd5Z8d"d#d6e+e)   d/e5de+e*   fd7Z9y):z.Git LFS related type definitions and utilities    N)	dataclass)ceil)getsize)TYPE_CHECKINGBinaryIOIterableOptional	TypedDict)unquote)	constants   )build_hf_headersfix_hf_endpoint_in_urlhf_raise_for_statushttp_backoffloggingvalidate_hf_hub_args)SliceFileObj)sha256sha_fileobj)CommitOperationAddz^[0-9a-f]{40}$zlfs-multipart-uploadzapplication/vnd.git-lfs+json)AcceptzContent-Typec                   r    e Zd ZU dZeed<   eed<   eed<   edefd       Z	edefd       Z
ed	efd
       Zy)
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                     t        |      }t        j                  |d      5 }|j                  d      d d }t	        |      }d d d         | |      S # 1 sw Y   xY w)Nrb   r   r   r   )r   ioopenpeekr   )clsr   r   filer   shas         E/opt/pipecat/venv/lib/python3.12/site-packages/huggingface_hub/lfs.py	from_pathzUploadInfo.from_pathI   s_    t}WWT4  	$DYYs^DS)Fd#C	$ S88	$ 	$s    AAdatac                 b    t        |      j                         } | t        |      |d d |      S )Nr    )r   r   r   )r   digestlen)r%   r*   r'   s      r(   
from_byteszUploadInfo.from_bytesQ   s-    Tl!!#D	$t*SAA    fileobjc                     |j                  d      }|j                  dt        j                         t	        |      }|j                         }|j                  dt        j                          | |||      S )Nr    r   r!   )readseekr"   SEEK_SETr   tell)r%   r0   r   r'   r   s        r(   from_fileobjzUploadInfo.from_fileobjV   sX    c"Q$'"||~Q$S88r/   N)__name__
__module____qualname____doc__bytes__annotations__intclassmethodstrr)   r.   r   r6    r/   r(   r   r   6   sj     M
IM9S 9 9 Be B B 98 9 9r/   r   upload_infostoken	repo_typerepo_idrevisionendpointheaders	transfersreturnc                    ||nt         j                  }d}|t         j                  v rt         j                  |   }| d| | d}	d||nddg| D 
cg c])  }
|
j                  j	                         |
j
                  d+ c}
d	d
}|dt        |      i|d<   i t        t        |      |xs i }t        d|	||      }t        |       |j                         }|j                  dd      }t        |t              st        d      |j                  d      }t        |t               r|nd}|D cg c]  }d|vst#        |       c}|D cg c]  }d|v st%        |       c}|fS c c}
w c c}w c c}w )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	multipartoidr   r   )	operationrH   objects	hash_algonameref)rB   POSTrG   jsonrS   zMalformed response from servertransfererror)r   ENDPOINTREPO_TYPES_URL_PREFIXESr   hexr   r   LFS_HEADERSr   r   r   rY   get
isinstancelist
ValueErrorr?   _validate_batch_actions_validate_batch_error)rA   rB   rC   rD   rE   rF   rG   rH   
url_prefix	batch_urlrM   payloadresp
batch_inforS   chosen_transferobjs                    r(   post_lfs_batch_inform   `   s   X $/xY5G5GHJI55566yA
*Aj\'2MNI"+"7Yg{=S '

  }}((*
 G  '("34


' =bG
 	7IDJnnY-Ggt$9:: nnZ0O)3OS)IotO 29O#G3<N	 	%O/6I'S.	s	#I 9
: 	PIs   .E+2	E0<E0	E5E5c                   "    e Zd ZU eed<   eed<   y)PayloadPartT
partNumberetagN)r7   r8   r9   r=   r<   r?   r@   r/   r(   ro   ro      s    O
Ir/   ro   c                   ,    e Zd ZU dZeed<   ee   ed<   y)CompletionPayloadTz?Payload that will be sent to the Hub when uploading multi-part.rQ   partsN)r7   r8   r9   r:   r?   r<   rb   ro   r@   r/   r(   rs   rs      s    I	Hr/   rs   rR   r   lfs_batch_actionc                 (   t        |       |j                  d      }|$t        j                  d| j                   d       y|d   d   }t        |       |d   j                  d      }|t        |       |j                  di       }|j                  d      }	t        |d	   |
      }
|		 t        |	      }	t        | ||	|
       nt        | |
       |xt        |       t        |d	   |      }t        d|t        ||      | j                  j                  j!                         | j                  j"                  d      }t%        |       t        j                  | j                   d       y# t        t        f$ r t        d|	 d      w xY w)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 uploadrM   verifyheader
chunk_sizehref)rF   zTMalformed response from LFS batch endpoint: `chunk_size` should be an integer. Got 'z'.)rR   ry   rz   
upload_url)rR   r|   rW   )rB   rG   rP   rX   z: Upload successful)rd   r`   loggerdebugpath_in_repo_validate_lfs_actionr   r=   rc   	TypeError_upload_multi_part_upload_single_partr   r   upload_infor   r^   r   r   )rR   ru   rB   rG   rF   rw   upload_actionverify_actionry   rz   r|   
verify_urlverify_resps                r(   
lfs_uploadr      s   8 ,-""9-G'	(>(>'??mno %Y/9M'$Y/33H=M ]+ x,FL)J'f(=QJ	ZJ
 	Yv*akliJG  ]++M&,A8L
"$5'B"..5599;YEZEZE_E_`	
 	K(
LLI**++>?@' I& 	fgqfrrtu 	s   -E2 2F
lfs_actionc                     t        | j                  d      t              r0| j                  d      *t        | j                  d      t              st	        d      | S ).validates response from the LFS batch endpointr{   ry   z"lfs_action is improperly formatted)ra   r`   r?   dictrc   )r   s    r(   r   r   
  sK     	:>>&)3/^^H%-JNN8<TVZ1[=>>r/   lfs_batch_actionsc                 P   t        | j                  d      t              rt        | j                  d      t              st	        d      | j                  di       j                  d      }| j                  di       j                  d      }|t        |       |t        |       | S )r   rQ   r   z)lfs_batch_actions is improperly formattedrw   rM   rx   )ra   r`   r?   r=   rc   r   )r   r   r   s      r(   rd   rd     s    (,,U3S9jIZI^I^_eIfhk>lDEE%)))R8<<XFM%)))R8<<XFM ]+ ]+r/   lfs_batch_errorc                 l   t        | j                  d      t              rt        | j                  d      t              st	        d      | j                  d      }t        |t
              r>t        |j                  d      t              rt        |j                  d      t              st	        d      | S )r   rQ   r   z'lfs_batch_error is improperly formattedr[   messagecode)ra   r`   r?   r=   rc   r   )r   
error_infos     r(   re   re   "  s    **5137JGZGZ[aGbdg<hBCC $$W-J:t$z~~i0#6z~~f-s3BCCr/   r|   c                     | j                  d      5 }t        d||      }t        |       ddd       y# 1 sw Y   yxY w)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   )rR   r|   r0   responses       r(   r   r   0  sB     
		T		* &gz@H%& & &s   6?ry   rz   c           	          t        || j                  |      }t        | ||      }t        d|t	        || j                  j
                  j                               t              }t        |       y)z@
    Uploads file using HF multipart LFS transfer protocol.
    )ry   r   rz   )rR   sorted_parts_urlsrz   rW   )rY   rG   N)	_get_sorted_parts_urlsr   _upload_parts_iterativelyr   _get_completion_payloadr   r^   r_   r   )rR   ry   rz   r|   r   response_headerscompletion_ress          r(   r   r   D  sn    
 /f)J_J_lvw 1/@Z "$%5y7L7L7S7S7W7W7YZ	N 'r/   r   c           
      V   t        | j                         D cg c]1  \  }}|j                         rt        |      dkD  rt	        |d      |f3 c}}d       D cg c]  \  }}|	 }}}t        |      }|t        |j                  |z        k7  rt        d      |S c c}}w c c}}w )Nr   
   c                     | d   S )Nr   r@   )ts    r(   <lambda>z(_get_sorted_parts_urls.<locals>.<lambda>d  s
    !A$ r/   )keyz0Invalid server response to upload large LFS file)sorteditemsisdigitr-   r=   r   r   rc   )ry   r   rz   part_numr|   _sorted_part_upload_urls	num_partss           r(   r   r   [  s     $ -3LLN(Hj##%#h-!*; Xr"J/
 

Az 	
 
 +,ID))J677KLL""
s   6B
B%r   rQ   c                     g }t        |       D ]H  \  }}|j                  d      }||dk(  rt        d| d|dz          |j                  |dz   |d       J ||dS )Nrq   rK   zInvalid etag (`z`) returned for part r   )rp   rq   )rQ   rt   )	enumerater`   rc   append)r   rQ   rt   part_numberry   rq   s         r(   r   r   m  s     "E()9: 	
Vzz&!<42:tf4I+XY/IZ[\\)Ao	
		
 ''r/   r   c                 6   g }| j                  d      5 }t        |      D ]S  \  }}t        |||z  |      5 }t        d||      }t	        |       |j                  |j                         d d d        U 	 d d d        |S # 1 sw Y   jxY w# 1 sw Y   |S xY w)NTr   )	seek_from
read_limitr   r   )r   r   r   r   r   r   rG   )	rR   r   rz   rG   r0   part_idxpart_upload_urlfileobj_slicepart_upload_ress	            r(   r   r   |  s     G			T		* 
8g)23D)E 		8%Ho$x/% 8 ".uoM"Z#O46678 8		8
8 N8 8
8 Ns"   #B5B-
BBBB)NNNN)NNN):r:   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_loggerr7   r}   compile	OID_REGEXLFS_MULTIPART_UPLOAD_COMMANDr_   r   r?   r   rb   tuplerm   ro   rs   r   r   rd   re   r   r=   r   r   r   r   r@   r/   r(   <module>r      s   5 	 	 !   I I   %  % * /			H	%BJJ()	5  -2 &9 &9 &9R  #"(,%)S:&SC=S S 	S
 smS smS d38n%S S	"S 4:tDz8C=01S Sl9 
   (,"DA#DADA C=DA d38n%	DA
 smDA 
DANT t 4 &#7 &S &T &(("6 ( (RU (cf (ko (.#4 #j #c #VZ[^V_ #$(d4j (s (GY (#8<S	OR	$Zr/   