KevinHuSh
commited on
Commit
·
b9d91e7
1
Parent(s):
10b4a61
add redis to accelerate access of minio (#482)
Browse files### What problem does this PR solve?
### Type of change
- [x] New Feature (non-breaking change which adds functionality)
- api/apps/__init__.py +1 -1
- api/utils/file_utils.py +1 -1
- conf/mapping.json +1 -1
- conf/service_conf.yaml +5 -0
- docker/service_conf.yaml +4 -0
- rag/app/naive.py +4 -2
- rag/settings.py +6 -0
- rag/svr/task_broker.py +10 -4
- rag/svr/task_executor.py +14 -7
- rag/utils/redis_conn.py +55 -0
api/apps/__init__.py
CHANGED
@@ -54,7 +54,7 @@ app.errorhandler(Exception)(server_error_response)
|
|
54 |
#app.config["LOGIN_DISABLED"] = True
|
55 |
app.config["SESSION_PERMANENT"] = False
|
56 |
app.config["SESSION_TYPE"] = "filesystem"
|
57 |
-
app.config['MAX_CONTENT_LENGTH'] = os.environ.get("MAX_CONTENT_LENGTH", 128 * 1024 * 1024)
|
58 |
|
59 |
Session(app)
|
60 |
login_manager = LoginManager()
|
|
|
54 |
#app.config["LOGIN_DISABLED"] = True
|
55 |
app.config["SESSION_PERMANENT"] = False
|
56 |
app.config["SESSION_TYPE"] = "filesystem"
|
57 |
+
app.config['MAX_CONTENT_LENGTH'] = int(os.environ.get("MAX_CONTENT_LENGTH", 128 * 1024 * 1024))
|
58 |
|
59 |
Session(app)
|
60 |
login_manager = LoginManager()
|
api/utils/file_utils.py
CHANGED
@@ -147,7 +147,7 @@ def filename_type(filename):
|
|
147 |
return FileType.PDF.value
|
148 |
|
149 |
if re.match(
|
150 |
-
r".*\.(docx|
|
151 |
return FileType.DOC.value
|
152 |
|
153 |
if re.match(
|
|
|
147 |
return FileType.PDF.value
|
148 |
|
149 |
if re.match(
|
150 |
+
r".*\.(docx|ppt|pptx|yml|xml|htm|json|csv|txt|ini|xls|xlsx|wps|rtf|hlp|pages|numbers|key|md)$", filename):
|
151 |
return FileType.DOC.value
|
152 |
|
153 |
if re.match(
|
conf/mapping.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"settings": {
|
3 |
"index": {
|
4 |
-
"number_of_shards":
|
5 |
"number_of_replicas": 0,
|
6 |
"refresh_interval" : "1000ms"
|
7 |
},
|
|
|
1 |
{
|
2 |
"settings": {
|
3 |
"index": {
|
4 |
+
"number_of_shards": 2,
|
5 |
"number_of_replicas": 0,
|
6 |
"refresh_interval" : "1000ms"
|
7 |
},
|
conf/service_conf.yaml
CHANGED
@@ -13,11 +13,16 @@ minio:
|
|
13 |
user: 'rag_flow'
|
14 |
password: 'infini_rag_flow'
|
15 |
host: 'minio:9000'
|
|
|
|
|
|
|
|
|
16 |
es:
|
17 |
hosts: 'http://es01:9200'
|
18 |
user_default_llm:
|
19 |
factory: 'Tongyi-Qianwen'
|
20 |
api_key: 'sk-xxxxxxxxxxxxx'
|
|
|
21 |
oauth:
|
22 |
github:
|
23 |
client_id: xxxxxxxxxxxxxxxxxxxxxxxxx
|
|
|
13 |
user: 'rag_flow'
|
14 |
password: 'infini_rag_flow'
|
15 |
host: 'minio:9000'
|
16 |
+
redis:
|
17 |
+
db: 1
|
18 |
+
password: 'infini_rag_flow'
|
19 |
+
host: 'redis:6379'
|
20 |
es:
|
21 |
hosts: 'http://es01:9200'
|
22 |
user_default_llm:
|
23 |
factory: 'Tongyi-Qianwen'
|
24 |
api_key: 'sk-xxxxxxxxxxxxx'
|
25 |
+
base_url: ''
|
26 |
oauth:
|
27 |
github:
|
28 |
client_id: xxxxxxxxxxxxxxxxxxxxxxxxx
|
docker/service_conf.yaml
CHANGED
@@ -13,6 +13,10 @@ minio:
|
|
13 |
user: 'rag_flow'
|
14 |
password: 'infini_rag_flow'
|
15 |
host: 'minio:9000'
|
|
|
|
|
|
|
|
|
16 |
es:
|
17 |
hosts: 'http://es01:9200'
|
18 |
user_default_llm:
|
|
|
13 |
user: 'rag_flow'
|
14 |
password: 'infini_rag_flow'
|
15 |
host: 'minio:9000'
|
16 |
+
redis:
|
17 |
+
db: 1
|
18 |
+
password: 'infini_rag_flow'
|
19 |
+
host: 'redis:6379'
|
20 |
es:
|
21 |
hosts: 'http://es01:9200'
|
22 |
user_default_llm:
|
rag/app/naive.py
CHANGED
@@ -66,6 +66,8 @@ class Docx(DocxParser):
|
|
66 |
class Pdf(PdfParser):
|
67 |
def __call__(self, filename, binary=None, from_page=0,
|
68 |
to_page=100000, zoomin=3, callback=None):
|
|
|
|
|
69 |
callback(msg="OCR is running...")
|
70 |
self.__images__(
|
71 |
filename if not binary else binary,
|
@@ -75,8 +77,8 @@ class Pdf(PdfParser):
|
|
75 |
callback
|
76 |
)
|
77 |
callback(msg="OCR finished")
|
|
|
78 |
|
79 |
-
from timeit import default_timer as timer
|
80 |
start = timer()
|
81 |
self._layouts_rec(zoomin)
|
82 |
callback(0.63, "Layout analysis finished.")
|
@@ -90,7 +92,7 @@ class Pdf(PdfParser):
|
|
90 |
self._concat_downward()
|
91 |
#self._filter_forpages()
|
92 |
|
93 |
-
cron_logger.info("paddle layouts:".format(
|
94 |
(timer() - start) / (self.total_page + 0.1)))
|
95 |
return [(b["text"], self._line_tag(b, zoomin))
|
96 |
for b in self.boxes], tbls
|
|
|
66 |
class Pdf(PdfParser):
|
67 |
def __call__(self, filename, binary=None, from_page=0,
|
68 |
to_page=100000, zoomin=3, callback=None):
|
69 |
+
from timeit import default_timer as timer
|
70 |
+
start = timer()
|
71 |
callback(msg="OCR is running...")
|
72 |
self.__images__(
|
73 |
filename if not binary else binary,
|
|
|
77 |
callback
|
78 |
)
|
79 |
callback(msg="OCR finished")
|
80 |
+
cron_logger.info("OCR: {}".format(timer() - start))
|
81 |
|
|
|
82 |
start = timer()
|
83 |
self._layouts_rec(zoomin)
|
84 |
callback(0.63, "Layout analysis finished.")
|
|
|
92 |
self._concat_downward()
|
93 |
#self._filter_forpages()
|
94 |
|
95 |
+
cron_logger.info("paddle layouts: {}".format(
|
96 |
(timer() - start) / (self.total_page + 0.1)))
|
97 |
return [(b["text"], self._line_tag(b, zoomin))
|
98 |
for b in self.boxes], tbls
|
rag/settings.py
CHANGED
@@ -25,6 +25,11 @@ SUBPROCESS_STD_LOG_NAME = "std.log"
|
|
25 |
|
26 |
ES = get_base_config("es", {})
|
27 |
MINIO = decrypt_database_config(name="minio")
|
|
|
|
|
|
|
|
|
|
|
28 |
DOC_MAXIMUM_SIZE = 128 * 1024 * 1024
|
29 |
|
30 |
# Logger
|
@@ -39,5 +44,6 @@ LoggerFactory.LEVEL = 30
|
|
39 |
es_logger = getLogger("es")
|
40 |
minio_logger = getLogger("minio")
|
41 |
cron_logger = getLogger("cron_logger")
|
|
|
42 |
chunk_logger = getLogger("chunk_logger")
|
43 |
database_logger = getLogger("database")
|
|
|
25 |
|
26 |
ES = get_base_config("es", {})
|
27 |
MINIO = decrypt_database_config(name="minio")
|
28 |
+
try:
|
29 |
+
REDIS = decrypt_database_config(name="redis")
|
30 |
+
except Exception as e:
|
31 |
+
REDIS = {}
|
32 |
+
pass
|
33 |
DOC_MAXIMUM_SIZE = 128 * 1024 * 1024
|
34 |
|
35 |
# Logger
|
|
|
44 |
es_logger = getLogger("es")
|
45 |
minio_logger = getLogger("minio")
|
46 |
cron_logger = getLogger("cron_logger")
|
47 |
+
cron_logger.setLevel(20)
|
48 |
chunk_logger = getLogger("chunk_logger")
|
49 |
database_logger = getLogger("database")
|
rag/svr/task_broker.py
CHANGED
@@ -32,6 +32,7 @@ from api.db.services.document_service import DocumentService
|
|
32 |
from api.settings import database_logger
|
33 |
from api.utils import get_format_time, get_uuid
|
34 |
from api.utils.file_utils import get_project_base_directory
|
|
|
35 |
|
36 |
|
37 |
def collect(tm):
|
@@ -84,10 +85,16 @@ def dispatch():
|
|
84 |
|
85 |
tsks = []
|
86 |
try:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
87 |
if r["type"] == FileType.PDF.value:
|
88 |
do_layout = r["parser_config"].get("layout_recognize", True)
|
89 |
-
pages = PdfParser.total_page_number(
|
90 |
-
r["name"], MINIO.get(r["kb_id"], r["location"]))
|
91 |
page_size = r["parser_config"].get("task_page_size", 12)
|
92 |
if r["parser_id"] == "paper":
|
93 |
page_size = r["parser_config"].get("task_page_size", 22)
|
@@ -110,8 +117,7 @@ def dispatch():
|
|
110 |
|
111 |
elif r["parser_id"] == "table":
|
112 |
rn = HuExcelParser.row_number(
|
113 |
-
r["name"],
|
114 |
-
r["kb_id"], r["location"]))
|
115 |
for i in range(0, rn, 3000):
|
116 |
task = new_task()
|
117 |
task["from_page"] = i
|
|
|
32 |
from api.settings import database_logger
|
33 |
from api.utils import get_format_time, get_uuid
|
34 |
from api.utils.file_utils import get_project_base_directory
|
35 |
+
from rag.utils.redis_conn import REDIS_CONN
|
36 |
|
37 |
|
38 |
def collect(tm):
|
|
|
85 |
|
86 |
tsks = []
|
87 |
try:
|
88 |
+
file_bin = MINIO.get(r["kb_id"], r["location"])
|
89 |
+
if REDIS_CONN.is_alive():
|
90 |
+
try:
|
91 |
+
REDIS_CONN.set("{}/{}".format(r["kb_id"], r["location"]), file_bin, 12*60)
|
92 |
+
except Exception as e:
|
93 |
+
cron_logger.warning("Put into redis[EXCEPTION]:" + str(e))
|
94 |
+
|
95 |
if r["type"] == FileType.PDF.value:
|
96 |
do_layout = r["parser_config"].get("layout_recognize", True)
|
97 |
+
pages = PdfParser.total_page_number(r["name"], file_bin)
|
|
|
98 |
page_size = r["parser_config"].get("task_page_size", 12)
|
99 |
if r["parser_id"] == "paper":
|
100 |
page_size = r["parser_config"].get("task_page_size", 22)
|
|
|
117 |
|
118 |
elif r["parser_id"] == "table":
|
119 |
rn = HuExcelParser.row_number(
|
120 |
+
r["name"], file_bin)
|
|
|
121 |
for i in range(0, rn, 3000):
|
122 |
task = new_task()
|
123 |
task["from_page"] = i
|
rag/svr/task_executor.py
CHANGED
@@ -19,13 +19,12 @@ import logging
|
|
19 |
import os
|
20 |
import hashlib
|
21 |
import copy
|
22 |
-
import random
|
23 |
import re
|
24 |
import sys
|
25 |
import time
|
26 |
import traceback
|
27 |
from functools import partial
|
28 |
-
|
29 |
from api.db.db_models import close_connection
|
30 |
from rag.settings import database_logger
|
31 |
from rag.settings import cron_logger, DOC_MAXIMUM_SIZE
|
@@ -35,7 +34,7 @@ from elasticsearch_dsl import Q
|
|
35 |
from multiprocessing.context import TimeoutError
|
36 |
from api.db.services.task_service import TaskService
|
37 |
from rag.utils import ELASTICSEARCH
|
38 |
-
from
|
39 |
from rag.utils import rmSpace, findMaxTm
|
40 |
|
41 |
from rag.nlp import search
|
@@ -48,6 +47,7 @@ from api.db import LLMType, ParserType
|
|
48 |
from api.db.services.document_service import DocumentService
|
49 |
from api.db.services.llm_service import LLMBundle
|
50 |
from api.utils.file_utils import get_project_base_directory
|
|
|
51 |
|
52 |
BATCH_SIZE = 64
|
53 |
|
@@ -105,11 +105,16 @@ def collect(comm, mod, tm):
|
|
105 |
|
106 |
def get_minio_binary(bucket, name):
|
107 |
global MINIO
|
|
|
|
|
|
|
|
|
|
|
|
|
108 |
return MINIO.get(bucket, name)
|
109 |
|
110 |
|
111 |
def build(row):
|
112 |
-
from timeit import default_timer as timer
|
113 |
if row["size"] > DOC_MAXIMUM_SIZE:
|
114 |
set_progress(row["id"], prog=-1, msg="File size exceeds( <= %dMb )" %
|
115 |
(int(DOC_MAXIMUM_SIZE / 1024 / 1024)))
|
@@ -265,6 +270,7 @@ def main(comm, mod):
|
|
265 |
callback(
|
266 |
msg="Finished slicing files(%d). Start to embedding the content." %
|
267 |
len(cks))
|
|
|
268 |
try:
|
269 |
tk_count = embedding(cks, embd_mdl, r["parser_config"], callback)
|
270 |
except Exception as e:
|
@@ -272,9 +278,10 @@ def main(comm, mod):
|
|
272 |
cron_logger.error(str(e))
|
273 |
tk_count = 0
|
274 |
|
275 |
-
callback(msg="Finished embedding! Start to build index!")
|
276 |
init_kb(r)
|
277 |
chunk_count = len(set([c["_id"] for c in cks]))
|
|
|
278 |
es_r = ELASTICSEARCH.bulk(cks, search.index_name(r["tenant_id"]))
|
279 |
if es_r:
|
280 |
callback(-1, "Index failure!")
|
@@ -290,8 +297,8 @@ def main(comm, mod):
|
|
290 |
DocumentService.increment_chunk_num(
|
291 |
r["doc_id"], r["kb_id"], tk_count, chunk_count, 0)
|
292 |
cron_logger.info(
|
293 |
-
"Chunk doc({}), token({}), chunks({})".format(
|
294 |
-
r["id"], tk_count, len(cks)))
|
295 |
|
296 |
tmf.write(str(r["update_time"]) + "\n")
|
297 |
tmf.close()
|
|
|
19 |
import os
|
20 |
import hashlib
|
21 |
import copy
|
|
|
22 |
import re
|
23 |
import sys
|
24 |
import time
|
25 |
import traceback
|
26 |
from functools import partial
|
27 |
+
from rag.utils import MINIO
|
28 |
from api.db.db_models import close_connection
|
29 |
from rag.settings import database_logger
|
30 |
from rag.settings import cron_logger, DOC_MAXIMUM_SIZE
|
|
|
34 |
from multiprocessing.context import TimeoutError
|
35 |
from api.db.services.task_service import TaskService
|
36 |
from rag.utils import ELASTICSEARCH
|
37 |
+
from timeit import default_timer as timer
|
38 |
from rag.utils import rmSpace, findMaxTm
|
39 |
|
40 |
from rag.nlp import search
|
|
|
47 |
from api.db.services.document_service import DocumentService
|
48 |
from api.db.services.llm_service import LLMBundle
|
49 |
from api.utils.file_utils import get_project_base_directory
|
50 |
+
from rag.utils.redis_conn import REDIS_CONN
|
51 |
|
52 |
BATCH_SIZE = 64
|
53 |
|
|
|
105 |
|
106 |
def get_minio_binary(bucket, name):
|
107 |
global MINIO
|
108 |
+
if REDIS_CONN.is_alive():
|
109 |
+
try:
|
110 |
+
r = REDIS_CONN.get("{}/{}".format(bucket, name))
|
111 |
+
if r: return r
|
112 |
+
except Exception as e:
|
113 |
+
cron_logger.warning("Get redis[EXCEPTION]:" + str(e))
|
114 |
return MINIO.get(bucket, name)
|
115 |
|
116 |
|
117 |
def build(row):
|
|
|
118 |
if row["size"] > DOC_MAXIMUM_SIZE:
|
119 |
set_progress(row["id"], prog=-1, msg="File size exceeds( <= %dMb )" %
|
120 |
(int(DOC_MAXIMUM_SIZE / 1024 / 1024)))
|
|
|
270 |
callback(
|
271 |
msg="Finished slicing files(%d). Start to embedding the content." %
|
272 |
len(cks))
|
273 |
+
st = timer()
|
274 |
try:
|
275 |
tk_count = embedding(cks, embd_mdl, r["parser_config"], callback)
|
276 |
except Exception as e:
|
|
|
278 |
cron_logger.error(str(e))
|
279 |
tk_count = 0
|
280 |
|
281 |
+
callback(msg="Finished embedding({})! Start to build index!".format(timer()-st))
|
282 |
init_kb(r)
|
283 |
chunk_count = len(set([c["_id"] for c in cks]))
|
284 |
+
st = timer()
|
285 |
es_r = ELASTICSEARCH.bulk(cks, search.index_name(r["tenant_id"]))
|
286 |
if es_r:
|
287 |
callback(-1, "Index failure!")
|
|
|
297 |
DocumentService.increment_chunk_num(
|
298 |
r["doc_id"], r["kb_id"], tk_count, chunk_count, 0)
|
299 |
cron_logger.info(
|
300 |
+
"Chunk doc({}), token({}), chunks({}), elapsed:{}".format(
|
301 |
+
r["id"], tk_count, len(cks), timer()-st))
|
302 |
|
303 |
tmf.write(str(r["update_time"]) + "\n")
|
304 |
tmf.close()
|
rag/utils/redis_conn.py
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
|
3 |
+
import redis
|
4 |
+
import logging
|
5 |
+
from rag import settings
|
6 |
+
from rag.utils import singleton
|
7 |
+
|
8 |
+
@singleton
|
9 |
+
class RedisDB:
|
10 |
+
def __init__(self):
|
11 |
+
self.REDIS = None
|
12 |
+
self.config = settings.REDIS
|
13 |
+
self.__open__()
|
14 |
+
|
15 |
+
def __open__(self):
|
16 |
+
try:
|
17 |
+
self.REDIS = redis.Redis(host=self.config.get("host", "redis").split(":")[0],
|
18 |
+
port=int(self.config.get("host", ":6379").split(":")[1]),
|
19 |
+
db=int(self.config.get("db", 1)),
|
20 |
+
password=self.config.get("password"))
|
21 |
+
except Exception as e:
|
22 |
+
logging.warning("Redis can't be connected.")
|
23 |
+
return self.REDIS
|
24 |
+
|
25 |
+
def is_alive(self):
|
26 |
+
return self.REDIS is not None
|
27 |
+
|
28 |
+
def get(self, k):
|
29 |
+
if not self.REDIS: return
|
30 |
+
try:
|
31 |
+
return self.REDIS.get(k)
|
32 |
+
except Exception as e:
|
33 |
+
logging.warning("[EXCEPTION]get" + str(k) + "||" + str(e))
|
34 |
+
self.__open__()
|
35 |
+
|
36 |
+
def set_obj(self, k, obj, exp=3600):
|
37 |
+
try:
|
38 |
+
self.REDIS.set(k, json.dumps(obj, ensure_ascii=False), exp)
|
39 |
+
return True
|
40 |
+
except Exception as e:
|
41 |
+
logging.warning("[EXCEPTION]set_obj" + str(k) + "||" + str(e))
|
42 |
+
self.__open__()
|
43 |
+
return False
|
44 |
+
|
45 |
+
def set(self, k, v, exp=3600):
|
46 |
+
try:
|
47 |
+
self.REDIS.set(k, v, exp)
|
48 |
+
return True
|
49 |
+
except Exception as e:
|
50 |
+
logging.warning("[EXCEPTION]set" + str(k) + "||" + str(e))
|
51 |
+
self.__open__()
|
52 |
+
return False
|
53 |
+
|
54 |
+
|
55 |
+
REDIS_CONN = RedisDB()
|