
    YxiU                    &    d dl mZ d dlmZ dddZy)    )annotations)fuzzc                $   g }| D ]  }dj                  |j                               j                         }|s3d}|D ];  }t        j                  |j                         |j                               |k\  s9d} n |rx|j                  |        |S )zQ
    Keep first occurrence; merge near-duplicates based on fuzzy similarity.
     FT)joinsplitstripr   ratiolowerappend)items	thresholdkeptss_normis_dupks          +/home/sandhiya/dr-transcripts/src/dedupe.pydedupe_stringsr      s     D 
 !'')$**, 	Azz&,,.!'')4	A	 KK
  K    N)Z   )r   	list[str]r   intreturnr   )
__future__r   	rapidfuzzr   r    r   r   <module>r      s    " r   