rag/svr/task_executor.py

#
#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
#
#  Licensed under the Apache License, Version 2.0 (the "License");
#  you may not use this file except in compliance with the License.
#  You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
#  Unless required by applicable law or agreed to in writing, software
#  distributed under the License is distributed on an "AS IS" BASIS,
#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
#  See the License for the specific language governing permissions and
#  limitations under the License.
#
import datetime
import json
import logging
import os
import hashlib
import copy
import re
import sys
import time
from concurrent.futures import ThreadPoolExecutor
from functools import partial
from io import BytesIO
from multiprocessing.context import TimeoutError
from timeit import default_timer as timer

import numpy as np
import pandas as pd

from api.db import LLMType, ParserType
from api.db.services.dialog_service import keyword_extraction, question_proposal
from api.db.services.document_service import DocumentService
from api.db.services.llm_service import LLMBundle
from api.db.services.task_service import TaskService
from api.db.services.file2document_service import File2DocumentService
from api.settings import retrievaler, docStoreConn
from api.db.db_models import close_connection
from rag.app import laws, paper, presentation, manual, qa, table, book, resume, picture, naive, one, audio, knowledge_graph, email
from rag.nlp import search, rag_tokenizer
from rag.raptor import RecursiveAbstractiveProcessing4TreeOrganizedRetrieval as Raptor
from api.utils.log_utils import logger, LOG_FILE
from rag.settings import DOC_MAXIMUM_SIZE, SVR_QUEUE_NAME
from rag.utils import rmSpace, num_tokens_from_string
from rag.utils.redis_conn import REDIS_CONN, Payload
from rag.utils.storage_factory import STORAGE_IMPL

BATCH_SIZE = 64

FACTORY = {
    "general": naive,
    ParserType.NAIVE.value: naive,
    ParserType.PAPER.value: paper,
    ParserType.BOOK.value: book,
    ParserType.PRESENTATION.value: presentation,
    ParserType.MANUAL.value: manual,
    ParserType.LAWS.value: laws,
    ParserType.QA.value: qa,
    ParserType.TABLE.value: table,
    ParserType.RESUME.value: resume,
    ParserType.PICTURE.value: picture,
    ParserType.ONE.value: one,
    ParserType.AUDIO.value: audio,
    ParserType.EMAIL.value: email,
    ParserType.KG.value: knowledge_graph
}

CONSUMER_NAME = "task_consumer_" + ("0" if len(sys.argv) < 2 else sys.argv[1])
PAYLOAD: Payload | None = None


def set_progress(task_id, from_page=0, to_page=-1, prog=None, msg="Processing..."):
    global PAYLOAD
    if prog is not None and prog < 0:
        msg = "[ERROR]" + msg
    cancel = TaskService.do_cancel(task_id)
    if cancel:
        msg += " [Canceled]"
        prog = -1

    if to_page > 0:
        if msg:
            msg = f"Page({from_page + 1}~{to_page + 1}): " + msg
    d = {"progress_msg": msg}
    if prog is not None:
        d["progress"] = prog
    try:
        TaskService.update_progress(task_id, d)
    except Exception:
        logger.exception(f"set_progress({task_id}) got exception")

    close_connection()
    if cancel:
        if PAYLOAD:
            PAYLOAD.ack()
            PAYLOAD = None
        os._exit(0)


def collect():
    global CONSUMER_NAME, PAYLOAD
    try:
        PAYLOAD = REDIS_CONN.get_unacked_for(CONSUMER_NAME, SVR_QUEUE_NAME, "rag_flow_svr_task_broker")
        if not PAYLOAD:
            PAYLOAD = REDIS_CONN.queue_consumer(SVR_QUEUE_NAME, "rag_flow_svr_task_broker", CONSUMER_NAME)
        if not PAYLOAD:
            time.sleep(1)
            return pd.DataFrame()
    except Exception:
        logger.exception("Get task event from queue exception")
        return pd.DataFrame()

    msg = PAYLOAD.get_message()
    if not msg:
        return pd.DataFrame()

    if TaskService.do_cancel(msg["id"]):
        logger.info("Task {} has been canceled.".format(msg["id"]))
        return pd.DataFrame()
    tasks = TaskService.get_tasks(msg["id"])
    if not tasks:
        logger.warning("{} empty task!".format(msg["id"]))
        return []

    tasks = pd.DataFrame(tasks)
    if msg.get("type", "") == "raptor":
        tasks["task_type"] = "raptor"
    return tasks


def get_storage_binary(bucket, name):
    return STORAGE_IMPL.get(bucket, name)


def build(row):
    if row["size"] > DOC_MAXIMUM_SIZE:
        set_progress(row["id"], prog=-1, msg="File size exceeds( <= %dMb )" %
                                             (int(DOC_MAXIMUM_SIZE / 1024 / 1024)))
        return []

    callback = partial(
        set_progress,
        row["id"],
        row["from_page"],
        row["to_page"])
    chunker = FACTORY[row["parser_id"].lower()]
    try:
        st = timer()
        bucket, name = File2DocumentService.get_storage_address(doc_id=row["doc_id"])
        binary = get_storage_binary(bucket, name)
        logger.info(
            "From minio({}) {}/{}".format(timer() - st, row["location"], row["name"]))
    except TimeoutError:
        callback(-1, "Internal server error: Fetch file from minio timeout. Could you try it again.")
        logger.exception("Minio {}/{} got timeout: Fetch file from minio timeout.".format(row["location"], row["name"]))
        return
    except Exception as e:
        if re.search("(No such file|not found)", str(e)):
            callback(-1, "Can not find file <%s> from minio. Could you try it again?" % row["name"])
        else:
            callback(-1, "Get file from minio: %s" % str(e).replace("'", ""))
        logger.exception("Chunking {}/{} got exception".format(row["location"], row["name"]))
        return

    try:
        cks = chunker.chunk(row["name"], binary=binary, from_page=row["from_page"],
                            to_page=row["to_page"], lang=row["language"], callback=callback,
                            kb_id=row["kb_id"], parser_config=row["parser_config"], tenant_id=row["tenant_id"])
        logger.info("Chunking({}) {}/{} done".format(timer() - st, row["location"], row["name"]))
    except Exception as e:
        callback(-1, "Internal server error while chunking: %s" %
                     str(e).replace("'", ""))
        logger.exception("Chunking {}/{} got exception".format(row["location"], row["name"]))
        return

    docs = []
    doc = {
        "doc_id": row["doc_id"],
        "kb_id": str(row["kb_id"])
    }
    el = 0
    for ck in cks:
        d = copy.deepcopy(doc)
        d.update(ck)
        md5 = hashlib.md5()
        md5.update((ck["content_with_weight"] +
                    str(d["doc_id"])).encode("utf-8"))
        d["id"] = md5.hexdigest()
        d["create_time"] = str(datetime.datetime.now()).replace("T", " ")[:19]
        d["create_timestamp_flt"] = datetime.datetime.now().timestamp()
        if not d.get("image"):
            d["img_id"] = ""
            d["page_num_list"] = json.dumps([])
            d["position_list"] = json.dumps([])
            d["top_list"] = json.dumps([])
            docs.append(d)
            continue

        try:
            output_buffer = BytesIO()
            if isinstance(d["image"], bytes):
                output_buffer = BytesIO(d["image"])
            else:
                d["image"].save(output_buffer, format='JPEG')

            st = timer()
            STORAGE_IMPL.put(row["kb_id"], d["id"], output_buffer.getvalue())
            el += timer() - st
        except Exception:
            logger.exception("Saving image of chunk {}/{}/{} got exception".format(row["location"], row["name"], d["_id"]))

        d["img_id"] = "{}-{}".format(row["kb_id"], d["id"])
        del d["image"]
        docs.append(d)
    logger.info("MINIO PUT({}):{}".format(row["name"], el))

    if row["parser_config"].get("auto_keywords", 0):
        callback(msg="Start to generate keywords for every chunk ...")
        chat_mdl = LLMBundle(row["tenant_id"], LLMType.CHAT, llm_name=row["llm_id"], lang=row["language"])
        for d in docs:
            d["important_kwd"] = keyword_extraction(chat_mdl, d["content_with_weight"],
                                                    row["parser_config"]["auto_keywords"]).split(",")
            d["important_tks"] = rag_tokenizer.tokenize(" ".join(d["important_kwd"]))

    if row["parser_config"].get("auto_questions", 0):
        callback(msg="Start to generate questions for every chunk ...")
        chat_mdl = LLMBundle(row["tenant_id"], LLMType.CHAT, llm_name=row["llm_id"], lang=row["language"])
        for d in docs:
            qst = question_proposal(chat_mdl, d["content_with_weight"], row["parser_config"]["auto_questions"])
            d["content_with_weight"] = f"Question: \n{qst}\n\nAnswer:\n" + d["content_with_weight"]
            qst = rag_tokenizer.tokenize(qst)
            if "content_ltks" in d:
                d["content_ltks"] += " " + qst
            if "content_sm_ltks" in d:
                d["content_sm_ltks"] += " " + rag_tokenizer.fine_grained_tokenize(qst)

    return docs


def init_kb(row, vector_size: int):
    idxnm = search.index_name(row["tenant_id"])
    return docStoreConn.createIdx(idxnm, row["kb_id"], vector_size)


def embedding(docs, mdl, parser_config=None, callback=None):
    if parser_config is None:
        parser_config = {}
    batch_size = 32
    tts, cnts = [rmSpace(d["title_tks"]) for d in docs if d.get("title_tks")], [
        re.sub(r"</?(table|td|caption|tr|th)( [^<>]{0,12})?>", " ", d["content_with_weight"]) for d in docs]
    tk_count = 0
    if len(tts) == len(cnts):
        tts_ = np.array([])
        for i in range(0, len(tts), batch_size):
            vts, c = mdl.encode(tts[i: i + batch_size])
            if len(tts_) == 0:
                tts_ = vts
            else:
                tts_ = np.concatenate((tts_, vts), axis=0)
            tk_count += c
            callback(prog=0.6 + 0.1 * (i + 1) / len(tts), msg="")
        tts = tts_

    cnts_ = np.array([])
    for i in range(0, len(cnts), batch_size):
        vts, c = mdl.encode(cnts[i: i + batch_size])
        if len(cnts_) == 0:
            cnts_ = vts
        else:
            cnts_ = np.concatenate((cnts_, vts), axis=0)
        tk_count += c
        callback(prog=0.7 + 0.2 * (i + 1) / len(cnts), msg="")
    cnts = cnts_

    title_w = float(parser_config.get("filename_embd_weight", 0.1))
    vects = (title_w * tts + (1 - title_w) *
             cnts) if len(tts) == len(cnts) else cnts

    assert len(vects) == len(docs)
    vector_size = 0
    for i, d in enumerate(docs):
        v = vects[i].tolist()
        vector_size = len(v)
        d["q_%d_vec" % len(v)] = v
    return tk_count, vector_size


def run_raptor(row, chat_mdl, embd_mdl, callback=None):
    vts, _ = embd_mdl.encode(["ok"])
    vector_size = len(vts[0])
    vctr_nm = "q_%d_vec" % vector_size
    chunks = []
    for d in retrievaler.chunk_list(row["doc_id"], row["tenant_id"], [str(row["kb_id"])], fields=["content_with_weight", vctr_nm]):
        chunks.append((d["content_with_weight"], np.array(d[vctr_nm])))

    raptor = Raptor(
        row["parser_config"]["raptor"].get("max_cluster", 64),
        chat_mdl,
        embd_mdl,
        row["parser_config"]["raptor"]["prompt"],
        row["parser_config"]["raptor"]["max_token"],
        row["parser_config"]["raptor"]["threshold"]
    )
    original_length = len(chunks)
    raptor(chunks, row["parser_config"]["raptor"]["random_seed"], callback)
    doc = {
        "doc_id": row["doc_id"],
        "kb_id": [str(row["kb_id"])],
        "docnm_kwd": row["name"],
        "title_tks": rag_tokenizer.tokenize(row["name"])
    }
    res = []
    tk_count = 0
    for content, vctr in chunks[original_length:]:
        d = copy.deepcopy(doc)
        md5 = hashlib.md5()
        md5.update((content + str(d["doc_id"])).encode("utf-8"))
        d["id"] = md5.hexdigest()
        d["create_time"] = str(datetime.datetime.now()).replace("T", " ")[:19]
        d["create_timestamp_flt"] = datetime.datetime.now().timestamp()
        d[vctr_nm] = vctr.tolist()
        d["content_with_weight"] = content
        d["content_ltks"] = rag_tokenizer.tokenize(content)
        d["content_sm_ltks"] = rag_tokenizer.fine_grained_tokenize(d["content_ltks"])
        res.append(d)
        tk_count += num_tokens_from_string(content)
    return res, tk_count, vector_size


def main():
    rows = collect()
    if len(rows) == 0:
        return

    for _, r in rows.iterrows():
        callback = partial(set_progress, r["id"], r["from_page"], r["to_page"])
        try:
            embd_mdl = LLMBundle(r["tenant_id"], LLMType.EMBEDDING, llm_name=r["embd_id"], lang=r["language"])
        except Exception as e:
            callback(-1, msg=str(e))
            logger.exception("LLMBundle got exception")
            continue

        if r.get("task_type", "") == "raptor":
            try:
                chat_mdl = LLMBundle(r["tenant_id"], LLMType.CHAT, llm_name=r["llm_id"], lang=r["language"])
                cks, tk_count, vector_size = run_raptor(r, chat_mdl, embd_mdl, callback)
            except Exception as e:
                callback(-1, msg=str(e))
                logger.exception("run_raptor got exception")
                continue
        else:
            st = timer()
            cks = build(r)
            logger.info("Build chunks({}): {}".format(r["name"], timer() - st))
            if cks is None:
                continue
            if not cks:
                callback(1., "No chunk! Done!")
                continue
            # TODO: exception handler
            ## set_progress(r["did"], -1, "ERROR: ")
            callback(
                msg="Finished slicing files(%d). Start to embedding the content." %
                    len(cks))
            st = timer()
            try:
                tk_count, vector_size = embedding(cks, embd_mdl, r["parser_config"], callback)
            except Exception as e:
                callback(-1, "Embedding error:{}".format(str(e)))
                logger.exception("run_rembedding got exception")
                tk_count = 0
            logger.info("Embedding elapsed({}): {:.2f}".format(r["name"], timer() - st))
            callback(msg="Finished embedding({:.2f})! Start to build index!".format(timer() - st))

        # logger.info(f"task_executor init_kb index {search.index_name(r["tenant_id"])} embd_mdl {embd_mdl.llm_name} vector length {vector_size}")
        init_kb(r, vector_size)
        chunk_count = len(set([c["id"] for c in cks]))
        st = timer()
        es_r = ""
        es_bulk_size = 4
        for b in range(0, len(cks), es_bulk_size):
            es_r = docStoreConn.insert(cks[b:b + es_bulk_size], search.index_name(r["tenant_id"]), r["kb_id"])
            if b % 128 == 0:
                callback(prog=0.8 + 0.1 * (b + 1) / len(cks), msg="")

        logger.info("Indexing elapsed({}): {:.2f}".format(r["name"], timer() - st))
        if es_r:
            callback(-1, f"Insert chunk error, detail info please check {LOG_FILE}. Please also check ES status!")
            docStoreConn.delete({"doc_id": r["doc_id"]}, search.index_name(r["tenant_id"]), r["kb_id"])
            logger.error('Insert chunk error: ' + str(es_r))
        else:
            if TaskService.do_cancel(r["id"]):
                docStoreConn.delete({"doc_id": r["doc_id"]}, search.index_name(r["tenant_id"]), r["kb_id"])
                continue
            callback(1., "Done!")
            DocumentService.increment_chunk_num(
                r["doc_id"], r["kb_id"], tk_count, chunk_count, 0)
            logger.info(
                "Chunk doc({}), token({}), chunks({}), elapsed:{:.2f}".format(
                    r["id"], tk_count, len(cks), timer() - st))


def report_status():
    global CONSUMER_NAME
    while True:
        try:
            obj = REDIS_CONN.get("TASKEXE")
            if not obj: obj = {}
            else: obj = json.loads(obj)
            if CONSUMER_NAME not in obj: obj[CONSUMER_NAME] = []
            obj[CONSUMER_NAME].append(timer())
            obj[CONSUMER_NAME] = obj[CONSUMER_NAME][-60:]
            REDIS_CONN.set_obj("TASKEXE", obj, 60*2)
        except Exception:
            logger.exception("report_status got exception")
        time.sleep(30)


if __name__ == "__main__":
    peewee_logger = logging.getLogger('peewee')
    peewee_logger.propagate = False
    peewee_logger.addHandler(logger.handlers[0])
    peewee_logger.setLevel(logger.handlers[0].level)

    exe = ThreadPoolExecutor(max_workers=1)
    exe.submit(report_status)

    while True:
        main()
        if PAYLOAD:
            PAYLOAD.ack()
            PAYLOAD = None