o ~i @s4dZddlZddlZddlZddlZddlZddlmZddlm Z m Z ddl Z ddl m Z ddlmZmZmZmZmZGdd d Z  dd ed e ed edede eeff ddZedkrdZzee\ZZedeedeeWdSe yZ!zede!WYdZ![!dSdZ![!wwdS)z R2 Cloud Storage module for downloading tar files containing audio segments. Handles searching, downloading, and extracting audio data from Cloudflare R2. N)Path)OptionalTuple)Config)R2_ENDPOINT_URL R2_BUCKETR2_ACCESS_KEY_IDR2_SECRET_ACCESS_KEYDEFAULT_SETTINGSc@sheZdZdZddZddededeefdd Zd ed edefd d Z dedede ee ffddZ dS)R2StorageClientz-Client for interacting with R2 cloud storage.cCs&tjdttttddd|_t|_dS)Ns3s3v4)signature_version) endpoint_urlaws_access_key_idaws_secret_access_keyconfig) boto3clientrr r r s3_clientrbucket)selfr7/home/ubuntu/maya3_transcribe/src/backend/r2_storage.py__init__s zR2StorageClient.__init__Fvideo_id use_regexreturnc Cs|jd}|d}z|jj|j|dtd||WSY|j|j|dD]!}|dgD]}|d}|drKtd||Sq3q+|rt d t |d tj }|j|jd D]!}|dgD]}|d}| |rtd ||Sqlqdtd |dS)aD Find a tar file for the given video ID in R2. Args: video_id: The YouTube video ID to search for use_regex: If True, use regex matching (fallback for non-exact matches) Returns: The S3 key of the found tar file, or None if not found list_objects_v2z.tar)BucketKeyz[R2] Found exact match: )r PrefixContentsr!z![R2] Found tar file with prefix: z.*z.*\.tar$)r z[R2] Found via regex: z%[R2] No tar file found for video_id: N) r get_paginator head_objectrprintpaginategetendswithrecompileescape IGNORECASEmatch) rrr paginator exact_keypageobjkeypatternrrr find_tar_file"s8     zR2StorageClient.find_tar_files3_key local_pathcCsvtj|ddtj|}tj||}td|d|d|j|j||tdtj |dddd |S) a Download a tar file from R2 to local storage. Args: s3_key: The S3 key of the tar file local_path: Local directory to download to Returns: Path to the downloaded tar file Texist_okz[R2] Downloading z to ...z[R2] Download complete: iz.2fz MB) osmakedirspathbasenamejoinr&r download_filergetsize)rr6r7filename local_filerrr download_tarNs "zR2StorageClient.download_tartar_path extract_toc Cstj|ddtd|dt|d }||Wdn1s%wYd}d}t|D]\}}}d|vrCtj|d}d|vrNtj|d}q3i} |r|t|d } t | } Wdn1shwYtd t | dgd ntd |rtj |rtt|d } td t | d|| fStdtt|d} | rt| dj}td||| fS)a Extract a tar file and return paths to metadata and segments. Args: tar_path: Path to the tar file extract_to: Directory to extract to Returns: Tuple of (segments_dir, metadata_dict) Tr8z[R2] Extracting r:rNz metadata.jsonsegmentsz[R2] Loaded metadata: z segments definedz%[R2] Warning: metadata.json not foundz*.flacz [R2] Found z FLAC segment filesz*[R2] Warning: segments directory not foundz **/*.flacrz[R2] Found segments at: )r;r<r&tarfileopen extractallwalkr=r?jsonloadlenr(existslistrglobstrparent) rrErFtar metadata_path segments_dirrootdirsfilesmetadataf flac_filesrrr extract_tarcs<     zR2StorageClient.extract_tarN)F) __name__ __module__ __qualname____doc__rrSboolrr5rDrdictr^rrrrr s  ,"r Trwork_diruse_regex_fallback cleanup_tarrc Cs|durtd}tj||}tj|d}tj|d}tj|r(t|t}|j||d}|s;t d|| ||} | | |\} } |rZtj| rZt | t d| | fS)a Main function to download and extract audio segments for a video. Args: video_id: YouTube video ID work_dir: Working directory for downloads (default from config) use_regex_fallback: Try regex matching if exact match fails cleanup_tar: Remove tar file after extraction Returns: Tuple of (segments_directory, metadata_dict) Nredownload extracted)rz No tar file found for video_id: z[R2] Cleaned up tar file)r r;r=r?rPshutilrmtreer r5FileNotFoundErrorrDr^remover&) rrerfrgvideo_work_dir download_dir extract_dirrr6rErWr[rrrdownload_video_segmentss"    rq__main__ pF_BQpHaIdUz Segments directory: zMetadata keys: zError: )NTT)"rbr;r*rIrMrjpathlibrtypingrrrbotocore.configrrrrr r r r rSrcrdrqr_ test_video_idrWr[r&rQkeys ExceptionerrrrsL    2