def lambda_handler(event, context):
if (event['status'] <= 0):
return {**event, "errorMessage": "Status isnt positive"}
aws_env = {
**event,
"bucketName": os.environ.get('DOCUMENTS_BUCKET'),
"awsRegion": 'eu-west-1',
"tmpJsonOutput": "/tmp/tmp_result.json",
"tmpTxtOutput": "/tmp/tmp_result.txt",
"outputBucket": os.environ.get('DOCUMENTS_BUCKET'),
"outputNameJson": get_bbox_filename(event['objectName'], ".json"),
"outputNameTxt": get_bbox_filename(event['objectName'], ".txt"),
"textractOnly": os.environ.get('TEXTRACT_ONLY'),
"minCharNeeded": int(os.environ.get('MIN_CHAR_NEEDED')),
"extract_pdf_lines": os.environ.get('EXTRACT_PDF_LINES'),
}
status = {"statusCode": 200, "body": "All right"}
extract_pdf_lines = aws_env['extract_pdf_lines']
textract_only = aws_env['textractOnly']
tmp_folder = "/tmp/pdfToBbox"
pdf_tmp_path = copy_pdf_to_tmp(tmp_folder, aws_env)
print("==> aws_env: ", aws_env)
if textract_only == "false" and is_pdf_has_enough_characters(
pdf_tmp_path, aws_env['minCharNeeded']) is True:
print("=> Extracting bounding box with pdfplumber")
if extract_pdf_lines == "true":
print("=> Extracting pdf lines bbox")
pdf = Pdf(pdf_tmp_path, aws_env['tmpJsonOutput'],
aws_env['tmpTxtOutput'])
pdf.parse_pdf()
pdf.save_in_json()
pdf.save_in_txt()
write_bbox_to_s3(aws_env)
else:
print("=> Extracting pdf words bbox")
if execute_pdf_to_bbox(pdf_tmp_path, aws_env['tmpJsonOutput']):
print("=> Error while trying to get pdf information")
aws_env["status"] = -1
aws_env["errorMessage"] = "PDF format not supported."
else:
write_bbox_to_s3(aws_env)
else:
print("Extracting bounding box with textract")
#send_to_textract(aws_env)
aws_env['size'] = S3Helper.getS3FileSize(aws_env['bucketName'],
aws_env['outputNameTxt'],
aws_env['awsRegion'])
aws_env["s3Url"] = get_new_s3_url(aws_env['s3Url'], "txt")
aws_env["status"] = status
aws_env["status"] = 1
aws_env["errorMessage"] = None
aws_env["contentType"] = "text/txt"
aws_env['objectName'] = aws_env['outputNameTxt']
aws_env["sourceUrl"] = aws_env["s3Url"]
AwsHelper.refreshTmpFolder(tmp_folder)
return update_event(aws_env, event)
def copy_pdf_to_tmp(tmp_folder: str, aws_env: dict) -> str:
pdf_content = S3Helper.readBytesFromS3(aws_env['bucketName'],
aws_env['objectName'],
aws_env['awsRegion'])
pdf_tmp = "tmp_0.pdf"
index = 0
if os.path.isdir(tmp_folder) is True:
AwsHelper.refreshTmpFolder(tmp_folder)
os.makedirs(tmp_folder)
for _ in os.listdir(tmp_folder):
if os.path.isfile(os.path.join(tmp_folder, pdf_tmp)) is False:
break
pdf_tmp = "tmp_{0}.pdf".format(index)
index += 1
pdf_tmp = os.path.join(tmp_folder, pdf_tmp)
with open(pdf_tmp, "wb") as tmp_file:
tmp_file.write(pdf_content)
print("Copy {0} to {1}".format(aws_env["objectName"], pdf_tmp))
return pdf_tmp
def lambda_handler(event, context):
print("=> Event: {0}".format(json.dumps(event)))
aws_env = {
**event, "bucketName": os.environ['DOCUMENTS_BUCKET'],
"outputBucket": os.environ['DOCUMENTS_BUCKET'],
"awsRegion": "eu-west-1",
"outputName": get_zip_output(event['objectName'])
}
print("=> AWS env: {0}".format(json.dumps(aws_env)))
tmp_folder = "/tmp/zip_extraction"
if os.path.isdir(tmp_folder) is True:
AwsHelper.refreshTmpFolder(tmp_folder)
extraction_output = os.path.join(tmp_folder, "extractions")
prepare_output_zip(extraction_output)
zip_tmp = copy_zip_to_tmp(tmp_folder, aws_env)
print("[DEBUG]: Extracting {0} into tmp file: {1}".format(
zip_tmp, extraction_output))
extract_nested_zip(zip_tmp, extraction_output)
write_extracted_zip(aws_env, extraction_output)
aws_env["status"] = 0
aws_env["errorMessage"] = None
AwsHelper.refreshTmpFolder(tmp_folder)
return update_event(aws_env, event)