trustgraph/trustgraph-utils/scripts/load-pdf
cybermaggedon 9b91d5eee3
Feature/pkgsplit (#83)
* Starting to spawn base package
* More package hacking
* Bedrock and VertexAI
* Parquet split
* Updated templates
* Utils
2024-09-30 19:36:09 +01:00

128 lines
2.7 KiB
Python
Executable file

#!/usr/bin/env python3
"""
Loads a PDF document into TrustGraph processing.
"""
import pulsar
from pulsar.schema import JsonSchema
from trustgraph.schema import Document, Source, document_ingest_queue
import base64
import hashlib
import argparse
import os
import time
from trustgraph.log_level import LogLevel
class Loader:
def __init__(
self,
pulsar_host,
output_queue,
log_level,
file,
):
self.client = pulsar.Client(
pulsar_host,
logger=pulsar.ConsoleLogger(log_level.to_pulsar())
)
self.producer = self.client.create_producer(
topic=output_queue,
schema=JsonSchema(Document),
chunking_enabled=True,
)
self.file = file
def run(self):
try:
path = self.file
data = open(path, "rb").read()
id = hashlib.sha256(path.encode("utf-8")).hexdigest()[0:8]
r = Document(
source=Source(
source=path,
title=path,
id=id,
),
data=base64.b64encode(data),
)
self.producer.send(r)
except Exception as e:
print(e, flush=True)
def __del__(self):
self.client.close()
def main():
parser = argparse.ArgumentParser(
prog='loader',
description=__doc__,
)
default_pulsar_host = os.getenv("PULSAR_HOST", 'pulsar://localhost:6650')
default_output_queue = document_ingest_queue
parser.add_argument(
'-p', '--pulsar-host',
default=default_pulsar_host,
help=f'Pulsar host (default: {default_pulsar_host})',
)
parser.add_argument(
'-o', '--output-queue',
default=default_output_queue,
help=f'Output queue (default: {default_output_queue})'
)
parser.add_argument(
'-l', '--log-level',
type=LogLevel,
default=LogLevel.ERROR,
choices=list(LogLevel),
help=f'Output queue (default: info)'
)
parser.add_argument(
'-f', '--file',
required=True,
help=f'File to load'
)
args = parser.parse_args()
while True:
try:
p = Loader(
pulsar_host=args.pulsar_host,
output_queue=args.output_queue,
log_level=args.log_level,
file=args.file,
)
p.run()
print("File loaded.")
break
except Exception as e:
print("Exception:", e, flush=True)
print("Will retry...", flush=True)
time.sleep(10)
main()