lambda_sample

from textractor import Textractor
import re
import boto3
from textractcaller.t_call import call_textract, Textract_Features, Query, QueriesConfig, Adapter, AdaptersConfig
import trp.trp2 as t2
import pandas as pd
import json

textract_client = boto3.client('textract')

def tabulate_query_answers(textract_json):
    d = t2.TDocumentSchema().load(textract_json)
    for page in d.pages:
        query_answers = d.get_query_answers(page=page)
        display(pd.DataFrame(query_answers))

session = boto3.Session( aws_access_key_id=['ACCESS_KEY'], aws_secret_access_key=['SECRET_ACCESS_KEY'])
extractor = Textractor(profile_name="default")


queries = []
queries.append(Query(text=["TEXT1"], alias=["TEXT1"], pages=["*"]))
queries.append(Query(text=["TEXT2"], alias=["TEXT2"], pages=["*"]))

queries_config = QueriesConfig(queries=queries)

s3 = session.resource('s3')

my_bucket = s3.Bucket(['SOURCE_DATA_BUCKET'])
upload_bucket =['CHUNKED_DATA_BUCKET']
substring =  "\d"
adapter_id=['ADAPTER_ID']
adapter1 = Adapter(adapter_id=adapter_id, version="7", pages=["*"])
adapters_config = AdaptersConfig(adapters=[adapter1])
print(f"Calling Custom Queries with Adapter:{adapter_id}")

for obj in my_bucket.objects.all():
    original_file = 's3://'+ obj.bucket_name+'/'+ obj.key
    # print (original_file)
    if re.search(substring,  obj.key):

        adapter1 = Adapter(adapter_id=adapter_id, version="7", pages=["*"])
        adapters_config = AdaptersConfig(adapters=[adapter1])
        print(f"Calling Custom Queries with Adapter:{adapter_id}")
        textract__with_adapter = call_textract(input_document=original_file,
                  boto3_textract_client=textract_client,
                  features=[Textract_Features.QUERIES],
                  queries_config=queries_config,
                  adapters_config=adapters_config)
        tabulate_query_answers(textract__with_adapter)
        # Extract data from JSON dict and upload to S3
        Count = 0
        content = []
        while Count in range(len(textract_json_with_adapter.get('Blocks'))):

          content_kv = list(textract_json_with_adapter.get('Blocks')[Count].items())
          content_string = content_kv[2][1]
          # print ('content',content)
          Count += 1
          content.append(content_string)

        download_file = obj.key + '.txt'
        s3.Object(upload_bucket, download_file).put(Body=str(content))