mirror of
https://github.com/infiniflow/ragflow.git
synced 2025-12-08 20:42:30 +08:00
Use consistent log file names, introduced initLogger (#3403)
### What problem does this PR solve? Use consistent log file names, introduced initLogger ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [ ] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [x] Refactoring - [ ] Performance Improvement - [ ] Other (please describe):
This commit is contained in:
Zhichang Yu
@ -13,9 +13,20 @@
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
#
|
||||
import logging
|
||||
import inspect
|
||||
from api.utils.log_utils import initRootLogger
|
||||
initRootLogger(inspect.getfile(inspect.currentframe()))
|
||||
for module in ["pdfminer"]:
|
||||
module_logger = logging.getLogger(module)
|
||||
module_logger.setLevel(logging.WARNING)
|
||||
for module in ["peewee"]:
|
||||
module_logger = logging.getLogger(module)
|
||||
module_logger.handlers.clear()
|
||||
module_logger.propagate = True
|
||||
|
||||
import datetime
|
||||
import json
|
||||
import logging
|
||||
import os
|
||||
import hashlib
|
||||
import copy
|
||||
@ -42,7 +53,6 @@ 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
|
||||
@ -90,7 +100,7 @@ def set_progress(task_id, from_page=0, to_page=-1, prog=None, msg="Processing...
|
||||
try:
|
||||
TaskService.update_progress(task_id, d)
|
||||
except Exception:
|
||||
logger.exception(f"set_progress({task_id}) got exception")
|
||||
logging.exception(f"set_progress({task_id}) got exception")
|
||||
|
||||
close_connection()
|
||||
if cancel:
|
||||
@ -110,7 +120,7 @@ def collect():
|
||||
time.sleep(1)
|
||||
return pd.DataFrame()
|
||||
except Exception:
|
||||
logger.exception("Get task event from queue exception")
|
||||
logging.exception("Get task event from queue exception")
|
||||
return pd.DataFrame()
|
||||
|
||||
msg = PAYLOAD.get_message()
|
||||
@ -118,11 +128,11 @@ def collect():
|
||||
return pd.DataFrame()
|
||||
|
||||
if TaskService.do_cancel(msg["id"]):
|
||||
logger.info("Task {} has been canceled.".format(msg["id"]))
|
||||
logging.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"]))
|
||||
logging.warning("{} empty task!".format(msg["id"]))
|
||||
return []
|
||||
|
||||
tasks = pd.DataFrame(tasks)
|
||||
@ -151,29 +161,29 @@ def build(row):
|
||||
st = timer()
|
||||
bucket, name = File2DocumentService.get_storage_address(doc_id=row["doc_id"])
|
||||
binary = get_storage_binary(bucket, name)
|
||||
logger.info(
|
||||
logging.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"]))
|
||||
logging.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"]))
|
||||
logging.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"]))
|
||||
logging.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"]))
|
||||
logging.exception("Chunking {}/{} got exception".format(row["location"], row["name"]))
|
||||
return
|
||||
|
||||
docs = []
|
||||
@ -210,12 +220,12 @@ def build(row):
|
||||
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"]))
|
||||
logging.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))
|
||||
logging.info("MINIO PUT({}):{}".format(row["name"], el))
|
||||
|
||||
if row["parser_config"].get("auto_keywords", 0):
|
||||
st = timer()
|
||||
@ -345,7 +355,7 @@ def main():
|
||||
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")
|
||||
logging.exception("LLMBundle got exception")
|
||||
continue
|
||||
|
||||
if r.get("task_type", "") == "raptor":
|
||||
@ -354,12 +364,12 @@ def main():
|
||||
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")
|
||||
logging.exception("run_raptor got exception")
|
||||
continue
|
||||
else:
|
||||
st = timer()
|
||||
cks = build(r)
|
||||
logger.info("Build chunks({}): {}".format(r["name"], timer() - st))
|
||||
logging.info("Build chunks({}): {}".format(r["name"], timer() - st))
|
||||
if cks is None:
|
||||
continue
|
||||
if not cks:
|
||||
@ -375,12 +385,12 @@ def main():
|
||||
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")
|
||||
logging.exception("run_rembedding got exception")
|
||||
tk_count = 0
|
||||
logger.info("Embedding elapsed({}): {:.2f}".format(r["name"], timer() - st))
|
||||
logging.info("Embedding elapsed({}): {:.2f}".format(r["name"], timer() - st))
|
||||
callback(msg="Finished embedding (in {:.2f}s)! 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}")
|
||||
# logging.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()
|
||||
@ -391,11 +401,11 @@ def main():
|
||||
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))
|
||||
logging.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))
|
||||
logging.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"])
|
||||
@ -404,7 +414,7 @@ def main():
|
||||
callback(1., "Done!")
|
||||
DocumentService.increment_chunk_num(
|
||||
r["doc_id"], r["kb_id"], tk_count, chunk_count, 0)
|
||||
logger.info(
|
||||
logging.info(
|
||||
"Chunk doc({}), token({}), chunks({}), elapsed:{:.2f}".format(
|
||||
r["id"], tk_count, len(cks), timer() - st))
|
||||
|
||||
@ -421,16 +431,11 @@ def report_status():
|
||||
obj[CONSUMER_NAME] = obj[CONSUMER_NAME][-60:]
|
||||
REDIS_CONN.set_obj("TASKEXE", obj, 60*2)
|
||||
except Exception:
|
||||
logger.exception("report_status got exception")
|
||||
logging.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)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user