#!/bin/bash
#verbose terminal output
set -xv 
start=`date +%s`

#function to loop a list of crime variables through a folder of text files
function entitySearch {

#create books array from input file
	book_list=()
	while read book
	do
	book_list+=($book)
	done < bookArray
	
#outer loop though books
	for myBook in ${book_list[@]}
	do

#send $myBook bash variable to nameStanza.py as a filename to process
	stanzaNER.py
#send output back to bash to run grep etc.on and output to file.
	
	grep "PERSON" $myBook | sort | uniq >> ${myBook}_PERSON_output
	grep "NORP" $myBook | sort | uniq >> ${myBook}_NORP_output
	grep "FAC" $myBook | sort | uniq >> ${myBook}_FAC_output
	grep "ORG" $myBook | sort | uniq >> ${myBook}_ORG_output
	grep "GPE" $myBook | sort | uniq >> ${myBook}_GPE_output
	grep "LOC" $myBook | sort | uniq >> ${myBook}_LOC_output
	grep "PRODUCT" $myBook | sort | uniq >> ${myBook}_PRODUCT_output
	grep "EVENT" $myBook | sort | uniq >> ${myBook}_EVENT_output
	grep "WORK_OF_ART" $myBook | sort | uniq >> ${myBook}_WORK_OF_ART_output
	grep "LAW" $myBook | sort | uniq >> ${myBook}_LAW_output
	grep "LANGUAGE" $myBook | sort | uniq >> ${myBook}_LANGUAGE_output
	grep "DATE" $myBook | sort | uniq >> ${myBook}_DATE_output
	grep "TIME" $myBook | sort | uniq >> ${myBook}_TIME_output
	grep "PERCENT" $myBook | sort | uniq >> ${myBook}_PERCENT_output
	grep "MONEY" $myBook | sort | uniq >> ${myBook}_MONEY_output
	grep "QUANTITY" $myBook | sort | uniq >> ${myBook}_QUANTITY_output
	grep "ORDINAL" $myBook | sort | uniq >> ${myBook}_ORDINAL_output
	grep "CARDINAL" $myBook | sort | uniq >> ${myBook}_CARDINAL_output
	
	done
}
entitySearch
end=`date +%s`
echo Execution time was `expr $end - $start` seconds.
exit 0



#stanzaNER.py
#import stanza

#nlp = stanza.Pipeline(lang='en', processors='tokenize,ner')

#with open(myBook, 'r') as f:
    #doc = nlp(f.read())
    
#print(*[f'entity: {ent.text}\ttype: {ent.type}' for ent in doc.ents], sep='\n')

#OR - but need to figure out how to create a file for the output...

#with open("output/file_output","w") as character_names:
    #for sent in ([f'entity: {ent.text}\ttype: {ent.type}' for sent in doc.sentences for ent in sent.ents]):
    	#character_names.write(sent)
    	#character_names.write("\n")