Spaces:
Sleeping
Sleeping
File size: 8,208 Bytes
9002555 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 |
from script.build_vector import IndexManager
from script.document_uploader import Uploader
from db.save_data import InsertDatabase
from db.get_data import GetDatabase
from db.delete_data import DeleteDatabase
from db.update_data import UpdateDatabase
from typing import Any
from fastapi import UploadFile
from fastapi import HTTPException
from core.chat.engine import Engine
from core.parser import clean_text, update_response, renumber_sources, seperate_to_list
from llama_index.core.composability import QASummaryQueryEngineBuilder
from service.dto import BotResponseStreaming, TestStreaming
from service.aws_loader import Loader
import logging
import re
# Configure logging
logging.basicConfig(level=logging.INFO)
# async def data_ingestion(
# db_conn, reference, file: UploadFile, content_table: UploadFile
# ) -> Any:
async def data_ingestion(
db_conn, reference, file: UploadFile
) -> Any:
insert_database = InsertDatabase(db_conn)
file_name = f"{reference['title']}.pdf"
aws_loader = Loader()
file_obj = file
aws_loader.upload_to_s3(file_obj, file_name)
print("Uploaded Success")
try:
# Insert data into the database
await insert_database.insert_data(reference)
# uploader = Uploader(reference, file, content_table)
uploader = Uploader(reference, file)
print("uploader : ", uploader)
nodes_with_metadata = await uploader.process_documents()
# Build indexes using IndexManager
index = IndexManager()
response = index.build_indexes(nodes_with_metadata)
return response
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in data ingestion: {e}")
raise HTTPException(
status_code=500,
detail="An internal server error occurred in data ingestion.",
)
async def get_data(db_conn, title="", fetch_all_data=True):
get_database = GetDatabase(db_conn)
print(get_database)
try:
if fetch_all_data:
results = await get_database.get_all_data()
print(results)
logging.info("Database fetched all data")
return results
else:
results = await get_database.get_data(title)
logging.info("Database fetched one data")
return results
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in get data.: {e}")
raise HTTPException(
status_code=500, detail="An internal server error occurred in get data."
)
async def update_data(id: int, reference, db_conn):
update_database = UpdateDatabase(db_conn)
try:
reference = reference.model_dump()
print(reference)
reference.update({"id": id})
print(reference)
await update_database.update_record(reference)
response = {"status": "Update Success"}
return response
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in update data.: {e}")
raise HTTPException(
status_code=500, detail="An internal server error occurred in update data."
)
async def delete_data(id: int, db_conn):
delete_database = DeleteDatabase(db_conn)
try:
params = {"id": id}
await delete_database.delete_record(params)
response = {"status": "Delete Success"}
return response
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in get data.: {e}")
raise HTTPException(
status_code=500, detail="An internal server error occurred in delete data."
)
def generate_completion_non_streaming(user_request, chat_engine):
try:
engine = Engine()
index_manager = IndexManager()
# Load existing indexes
index = index_manager.load_existing_indexes()
# Retrieve the chat engine with the loaded index
chat_engine = engine.get_chat_engine(index)
# Generate completion response
response = chat_engine.chat(user_request)
sources = response.sources
number_reference = list(set(re.findall(r"\[(\d+)\]", str(response))))
number_reference_sorted = sorted(number_reference)
contents = []
raw_contents = []
metadata_collection = []
scores = []
if number_reference_sorted:
for number in number_reference_sorted:
# Konversi number ke integer untuk digunakan sebagai indeks
number = int(number)
# Pastikan sources tidak kosong dan memiliki elemen yang diperlukan
if sources and len(sources) > 0:
node = dict(sources[0])["raw_output"].source_nodes
# Pastikan number valid sebagai indeks
if 0 <= number - 1 < len(node):
print(node[number - 1].node.get_text())
raw_content = seperate_to_list(
node[number - 1].node.get_text()
)
raw_contents.append(raw_content)
content = clean_text(node[number - 1].node.get_text())
contents.append(content)
metadata = dict(node[number - 1].node.metadata)
metadata_collection.append(metadata)
score = node[number - 1].score
scores.append(score)
else:
print(f"Invalid reference number: {number}")
else:
print("No sources available")
else:
print("There are no references")
response = update_response(str(response))
contents = renumber_sources(contents)
# Check the lengths of content and metadata
num_content = len(contents)
num_metadata = len(metadata_collection)
# Add content to metadata
for i in range(min(num_content, num_metadata)):
metadata_collection[i]["content"] = re.sub(r"source \d+\:", "", contents[i])
return str(response), raw_contents, contents, metadata_collection, scores
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in generate text: {e}")
raise HTTPException(
status_code=500,
detail="An internal server error occurred in generate text.",
)
async def generate_streaming_completion(user_request, chat_engine):
try:
engine = Engine()
index_manager = IndexManager()
# Load existing indexes
index = index_manager.load_existing_indexes()
# Retrieve the chat engine with the loaded index
chat_engine = engine.get_chat_engine(index)
# Generate completion response
response = chat_engine.stream_chat(user_request)
completed_response = ""
for gen in response.response_gen:
completed_response += gen # Concatenate the new string
yield BotResponseStreaming(
content=gen, completed_content=completed_response
)
nodes = response.source_nodes
for node in nodes:
reference = str(clean_text(node.node.get_text()))
metadata = dict(node.node.metadata)
score = float(node.score)
yield BotResponseStreaming(
completed_content=completed_response,
reference=reference,
metadata=metadata,
score=score,
)
except Exception as e:
yield {"error": str(e)}
except Exception as e:
# Log the error and raise HTTPException for FastAPI
logging.error(f"An error occurred in generate text: {e}")
raise HTTPException(
status_code=500,
detail="An internal server error occurred in generate text.",
) |