Preliminary capability, knowledge cores with msgpack

This commit is contained in:
Cyber MacGeddon 2024-11-25 18:39:00 +00:00
parent 319f9ac04a
commit b918b4d8aa
5 changed files with 398 additions and 3 deletions

View file

@ -23,3 +23,6 @@ if "error" in resp:
print(f"Error: {resp['error']}")
sys.exit(1)
print(resp["vectors"])

View file

@ -0,0 +1,9 @@
#!/usr/bin/env python3
import msgpack
import sys
unpacker = msgpack.Unpacker(sys.stdin.buffer, raw=False)
for unpacked in unpacker:
print(unpacked)

View file

@ -0,0 +1,190 @@
#!/usr/bin/env python3
import aiohttp
import asyncio
import msgpack
import json
import sys
import argparse
import os
async def fetch_ge(queue, user, collection, url):
async with aiohttp.ClientSession() as session:
async with session.ws_connect(f"{url}stream/graph-embeddings") as ws:
async for msg in ws:
if msg.type == aiohttp.WSMsgType.TEXT:
data = msg.json()
if user:
if data["metadata"]["user"] != user:
continue
if collection:
if data["metadata"]["collection"] != collection:
continue
await queue.put([
"ge",
{
"m": {
"i": data["metadata"]["id"],
"m": data["metadata"]["metadata"],
"u": data["metadata"]["user"],
"c": data["metadata"]["collection"],
},
"v": data["vectors"],
"e": data["entity"],
}
])
if msg.type == aiohttp.WSMsgType.ERROR:
print("Error")
break
async def fetch_triples(queue, user, collection, url):
async with aiohttp.ClientSession() as session:
async with session.ws_connect(f"{url}stream/triples") as ws:
async for msg in ws:
if msg.type == aiohttp.WSMsgType.TEXT:
data = msg.json()
if user:
if data["metadata"]["user"] != user:
continue
if collection:
if data["metadata"]["collection"] != collection:
continue
await queue.put((
"t",
{
"m": {
"i": data["metadata"]["id"],
"m": data["metadata"]["metadata"],
"u": data["metadata"]["user"],
"c": data["metadata"]["collection"],
},
"t": data["triples"],
}
))
if msg.type == aiohttp.WSMsgType.ERROR:
print("Error")
break
ge_counts = 0
t_counts = 0
async def stats():
global t_counts
global ge_counts
while True:
await asyncio.sleep(5)
print(
f"Graph embeddings: {ge_counts:10d} Triples: {t_counts:10d}"
)
async def output(queue, path, format):
global t_counts
global ge_counts
with open(path, "wb") as f:
while True:
msg = await queue.get()
if format == "msgpack":
f.write(msgpack.packb(msg, use_bin_type=True))
else:
f.write(json.dumps(msg).encode("utf-8"))
if msg[0] == "t":
t_counts += 1
else:
if msg[0] == "ge":
ge_counts += 1
async def run(**args):
q = asyncio.Queue()
ge_task = asyncio.create_task(
fetch_ge(
queue=q, user=args["user"], collection=args["collection"],
url=args["url"] + "api/v1/"
)
)
triples_task = asyncio.create_task(
fetch_triples(
queue=q, user=args["user"], collection=args["collection"],
url=args["url"] + "api/v1/"
)
)
output_task = asyncio.create_task(
output(
queue=q, path=args["output_file"], format=args["format"],
)
)
stats_task = asyncio.create_task(stats())
await output_task
await triples_task
await ge_task
await stats_task
async def main():
parser = argparse.ArgumentParser(
prog='tg-load-pdf',
description=__doc__,
)
default_url = os.getenv("TRUSTGRAPH_API", "http://localhost:8088/")
default_user = "trustgraph"
collection = "default"
parser.add_argument(
'-u', '--url',
default=default_url,
help=f'TrustGraph API URL (default: {default_url})',
)
parser.add_argument(
'-o', '--output-file',
# Make it mandatory, difficult to over-write an existing file
required=True,
help=f'Output file'
)
parser.add_argument(
'--format',
default="msgpack",
choices=["msgpack", "json"],
help=f'Output format (default: msgpack)',
)
parser.add_argument(
'--user',
help=f'User ID to filter on (default: no filter)'
)
parser.add_argument(
'--collection',
help=f'Collection ID to filter on (default: no filter)'
)
args = parser.parse_args()
await run(**vars(args))
asyncio.run(main())

View file

@ -104,5 +104,6 @@ setuptools.setup(
"scripts/triples-query-neo4j",
"scripts/triples-write-cassandra",
"scripts/triples-write-neo4j",
"scripts/tg-save-kg-core",
]
)

View file

@ -14,7 +14,7 @@ module = ".".join(__name__.split(".")[1:-1])
import asyncio
import argparse
from aiohttp import web
from aiohttp import web, WSMsgType
import json
import logging
import uuid
@ -47,9 +47,13 @@ from ... schema import GraphRagQuery, GraphRagResponse
from ... schema import graph_rag_request_queue
from ... schema import graph_rag_response_queue
from ... schema import TriplesQueryRequest, TriplesQueryResponse
from ... schema import TriplesQueryRequest, TriplesQueryResponse, Triples
from ... schema import triples_request_queue
from ... schema import triples_response_queue
from ... schema import triples_store_queue
from ... schema import GraphEmbeddings
from ... schema import graph_embeddings_store_queue
from ... schema import AgentRequest, AgentResponse
from ... schema import agent_request_queue
@ -84,6 +88,11 @@ def to_subgraph(x):
for t in x
]
class Running:
def __init__(self): self.running = True
def get(self): return self.running
def stop(self): self.running = False
class Publisher:
def __init__(self, pulsar_host, topic, schema=None, max_size=10,
@ -132,6 +141,7 @@ class Subscriber:
self.consumer_name = consumer_name
self.schema = schema
self.q = {}
self.full = {}
async def run(self):
while True:
@ -145,10 +155,19 @@ class Subscriber:
) as consumer:
while True:
msg = await consumer.receive()
id = msg.properties()["id"]
try:
id = msg.properties()["id"]
except:
id = None
value = msg.value()
if id in self.q:
await self.q[id].put(value)
for q in self.full.values():
await q.put(value)
except Exception as e:
print("Exception:", e, flush=True)
@ -164,6 +183,59 @@ class Subscriber:
if id in self.q:
del self.q[id]
async def subscribe_all(self, id):
q = asyncio.Queue()
self.full[id] = q
return q
async def unsubscribe_all(self, id):
if id in self.full:
del self.full[id]
def serialize_triples(message):
return {
"metadata": {
"id": message.metadata.id,
"metadata": [
{
"s": t.s.value,
"p": t.p.value,
"o": t.o.value,
}
for t in message.metadata.metadata
],
"user": message.metadata.user,
"collection": message.metadata.collection,
},
"triples": [
{
"s": t.s.value,
"p": t.p.value,
"o": t.o.value,
}
for t in message.triples
]
}
def serialize_graph_embeddings(message):
return {
"metadata": {
"id": message.metadata.id,
"metadata": [
{
"s": t.s.value,
"p": t.p.value,
"o": t.o.value,
}
for t in message.metadata.metadata
],
"user": message.metadata.user,
"collection": message.metadata.collection,
},
"vectors": message.vectors,
"entity": message.entity.value,
}
class Api:
def __init__(self, **config):
@ -243,6 +315,18 @@ class Api:
JsonSchema(EmbeddingsResponse)
)
self.triples_tap = Subscriber(
self.pulsar_host, triples_store_queue,
"api-gateway", "api-gateway",
schema=JsonSchema(Triples)
)
self.graph_embeddings_tap = Subscriber(
self.pulsar_host, graph_embeddings_store_queue,
"api-gateway", "api-gateway",
schema=JsonSchema(GraphEmbeddings)
)
self.document_out = Publisher(
self.pulsar_host, document_ingest_queue,
schema=JsonSchema(Document),
@ -264,6 +348,12 @@ class Api:
web.post("/api/v1/embeddings", self.embeddings),
web.post("/api/v1/load/document", self.load_document),
web.post("/api/v1/load/text", self.load_text),
web.get("/api/v1/ws", self.socket),
web.get("/api/v1/stream/triples", self.stream_triples),
web.get(
"/api/v1/stream/graph-embeddings",
self.stream_graph_embeddings
),
])
async def llm(self, request):
@ -660,6 +750,100 @@ class Api:
{ "error": str(e) }
)
async def socket(self, request):
ws = web.WebSocketResponse()
await ws.prepare(request)
async for msg in ws:
if msg.type == WSMsgType.TEXT:
if msg.data == 'close':
await ws.close()
else:
await ws.send_str(msg.data + '/answer')
elif msg.type == WSMsgType.ERROR:
print('ws connection closed with exception %s' %
ws.exception())
print('websocket connection closed')
return ws
async def stream(self, q, ws, running, fn):
while running.get():
try:
resp = await asyncio.wait_for(q.get(), 0.5)
await ws.send_json(fn(resp))
except TimeoutError:
continue
except Exception as e:
print(f"Exception: {str(e)}", flush=True)
async def stream_triples(self, request):
id = str(uuid.uuid4())
q = await self.triples_tap.subscribe_all(id)
running = Running()
ws = web.WebSocketResponse()
await ws.prepare(request)
tsk = asyncio.create_task(self.stream(
q,
ws,
running,
serialize_triples,
))
async for msg in ws:
if msg.type == WSMsgType.ERROR:
break
else:
# Ignore incoming messages
pass
running.stop()
await self.triples_tap.unsubscribe_all(id)
await tsk
return ws
async def stream_graph_embeddings(self, request):
id = str(uuid.uuid4())
q = await self.graph_embeddings_tap.subscribe_all(id)
running = Running()
ws = web.WebSocketResponse()
await ws.prepare(request)
tsk = asyncio.create_task(self.stream(
q,
ws,
running,
serialize_graph_embeddings,
))
async for msg in ws:
if msg.type == WSMsgType.ERROR:
break
else:
# Ignore incoming messages
pass
running.stop()
await self.graph_embeddings_tap.unsubscribe_all(id)
await tsk
return ws
async def app_factory(self):
self.llm_pub_task = asyncio.create_task(self.llm_in.run())
@ -688,6 +872,14 @@ class Api:
self.embeddings_out.run()
)
self.triples_tap_task = asyncio.create_task(
self.triples_tap.run()
)
self.graph_embeddings_tap_task = asyncio.create_task(
self.graph_embeddings_tap.run()
)
self.doc_ingest_pub_task = asyncio.create_task(self.document_out.run())
self.text_ingest_pub_task = asyncio.create_task(self.text_out.run())