KevinHuSh
commited on
Commit
·
738c322
1
Parent(s):
f4456af
add docker compose (#8)
Browse files* add docker compose
* add docker compose
- docker/docker-compose.yml +68 -0
- python/README.md +22 -0
- python/conf/sys.cnf +4 -0
- python/nlp/huchunk.py +14 -2
- python/parser/pdf_parser.py +1 -1
- python/util/db_conn.py +44 -0
docker/docker-compose.yml
ADDED
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
version: '2.2'
|
2 |
+
services:
|
3 |
+
es01:
|
4 |
+
container_name: docass-es-01
|
5 |
+
image: docker.elastic.co/elasticsearch/elasticsearch:${STACK_VERSION}
|
6 |
+
volumes:
|
7 |
+
- esdata01:/usr/share/elasticsearch/data
|
8 |
+
ports:
|
9 |
+
- ${ES_PORT}:9200
|
10 |
+
environment:
|
11 |
+
- node.name=es01
|
12 |
+
- cluster.name=${CLUSTER_NAME}
|
13 |
+
- cluster.initial_master_nodes=es01
|
14 |
+
- ELASTIC_PASSWORD=${ELASTIC_PASSWORD}
|
15 |
+
- bootstrap.memory_lock=false
|
16 |
+
- xpack.security.enabled=false
|
17 |
+
mem_limit: ${MEM_LIMIT}
|
18 |
+
ulimits:
|
19 |
+
memlock:
|
20 |
+
soft: -1
|
21 |
+
hard: -1
|
22 |
+
networks:
|
23 |
+
- docass
|
24 |
+
restart: always
|
25 |
+
|
26 |
+
kibana:
|
27 |
+
depends_on:
|
28 |
+
- es01
|
29 |
+
image: docker.elastic.co/kibana/kibana:${STACK_VERSION}
|
30 |
+
container_name: docass-kibana
|
31 |
+
volumes:
|
32 |
+
- kibanadata:/usr/share/kibana/data
|
33 |
+
ports:
|
34 |
+
- ${KIBANA_PORT}:5601
|
35 |
+
environment:
|
36 |
+
- SERVERNAME=kibana
|
37 |
+
- ELASTICSEARCH_HOSTS=http://es01:9200
|
38 |
+
mem_limit: ${MEM_LIMIT}
|
39 |
+
networks:
|
40 |
+
- docass
|
41 |
+
|
42 |
+
postgres:
|
43 |
+
image: postgres
|
44 |
+
container_name: docass-postgres
|
45 |
+
environment:
|
46 |
+
- POSTGRES_USER=${POSTGRES_USER}
|
47 |
+
- POSTGRES_PASSWORD=${POSTGRES_PASSWORD}
|
48 |
+
- POSTGRES_DB=${POSTGRES_DB}
|
49 |
+
ports:
|
50 |
+
- 5455:5455
|
51 |
+
volumes:
|
52 |
+
- pg_data:/usr/share/elasticsearch/data
|
53 |
+
networks:
|
54 |
+
- docass
|
55 |
+
restart: always
|
56 |
+
|
57 |
+
|
58 |
+
volumes:
|
59 |
+
esdata01:
|
60 |
+
driver: local
|
61 |
+
kibanadata:
|
62 |
+
driver: local
|
63 |
+
pg_data:
|
64 |
+
driver: local
|
65 |
+
|
66 |
+
networks:
|
67 |
+
docass:
|
68 |
+
driver: bridge
|
python/README.md
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
```shell
|
3 |
+
|
4 |
+
docker pull postgres
|
5 |
+
|
6 |
+
LOCAL_POSTGRES_DATA=./postgres-data
|
7 |
+
|
8 |
+
docker run
|
9 |
+
--name docass-postgres
|
10 |
+
-p 5455:5432
|
11 |
+
-v $LOCAL_POSTGRES_DATA:/var/lib/postgresql/data
|
12 |
+
-e POSTGRES_USER=root
|
13 |
+
-e POSTGRES_PASSWORD=infiniflow_docass
|
14 |
+
-e POSTGRES_DB=docass
|
15 |
+
-d
|
16 |
+
postgres
|
17 |
+
|
18 |
+
docker network create elastic
|
19 |
+
docker pull elasticsearch:8.11.3;
|
20 |
+
docker pull docker.elastic.co/kibana/kibana:8.11.3
|
21 |
+
|
22 |
+
```
|
python/conf/sys.cnf
CHANGED
@@ -1,4 +1,8 @@
|
|
1 |
[online]
|
2 |
es=127.0.0.1:9200
|
3 |
idx_nm=toxic
|
|
|
|
|
|
|
|
|
4 |
|
|
|
1 |
[online]
|
2 |
es=127.0.0.1:9200
|
3 |
idx_nm=toxic
|
4 |
+
pgdb_usr=root
|
5 |
+
pgdb_pwd=infiniflow_docass
|
6 |
+
pgdb_host=127.0.0.1
|
7 |
+
pgdb_port=5432
|
8 |
|
python/nlp/huchunk.py
CHANGED
@@ -291,6 +291,12 @@ class PdfChunker(HuChunker):
|
|
291 |
|
292 |
|
293 |
class DocxChunker(HuChunker):
|
|
|
|
|
|
|
|
|
|
|
|
|
294 |
def __init__(self, doc_parser):
|
295 |
self.doc = doc_parser
|
296 |
super().__init__()
|
@@ -336,6 +342,12 @@ class DocxChunker(HuChunker):
|
|
336 |
|
337 |
|
338 |
class ExcelChunker(HuChunker):
|
|
|
|
|
|
|
|
|
|
|
|
|
339 |
def __init__(self, excel_parser):
|
340 |
self.excel = excel_parser
|
341 |
super().__init__()
|
@@ -354,10 +366,10 @@ if __name__ == "__main__":
|
|
354 |
from parser import PdfParser
|
355 |
ckr = PdfChunker(PdfParser())
|
356 |
if sys.argv[1].split(".")[-1].lower().find("doc") >= 0:
|
357 |
-
from
|
358 |
ckr = DocxChunker(DocxParser())
|
359 |
if sys.argv[1].split(".")[-1].lower().find("xlsx") >= 0:
|
360 |
-
from
|
361 |
ckr = ExcelChunker(ExcelParser())
|
362 |
|
363 |
# ckr.html(sys.argv[1])
|
|
|
291 |
|
292 |
|
293 |
class DocxChunker(HuChunker):
|
294 |
+
|
295 |
+
@dataclass
|
296 |
+
class Fields:
|
297 |
+
text_chunks: List = None
|
298 |
+
table_chunks: List = None
|
299 |
+
|
300 |
def __init__(self, doc_parser):
|
301 |
self.doc = doc_parser
|
302 |
super().__init__()
|
|
|
342 |
|
343 |
|
344 |
class ExcelChunker(HuChunker):
|
345 |
+
|
346 |
+
@dataclass
|
347 |
+
class Fields:
|
348 |
+
text_chunks: List = None
|
349 |
+
table_chunks: List = None
|
350 |
+
|
351 |
def __init__(self, excel_parser):
|
352 |
self.excel = excel_parser
|
353 |
super().__init__()
|
|
|
366 |
from parser import PdfParser
|
367 |
ckr = PdfChunker(PdfParser())
|
368 |
if sys.argv[1].split(".")[-1].lower().find("doc") >= 0:
|
369 |
+
from parser import DocxParser
|
370 |
ckr = DocxChunker(DocxParser())
|
371 |
if sys.argv[1].split(".")[-1].lower().find("xlsx") >= 0:
|
372 |
+
from parser import ExcelParser
|
373 |
ckr = ExcelChunker(ExcelParser())
|
374 |
|
375 |
# ckr.html(sys.argv[1])
|
python/parser/pdf_parser.py
CHANGED
@@ -323,7 +323,7 @@ class HuParser:
|
|
323 |
return layouts
|
324 |
|
325 |
def __table_paddle(self, images):
|
326 |
-
tbls = self.tbl_det([
|
327 |
res = []
|
328 |
# align left&right for rows, align top&bottom for columns
|
329 |
for tbl in tbls:
|
|
|
323 |
return layouts
|
324 |
|
325 |
def __table_paddle(self, images):
|
326 |
+
tbls = self.tbl_det([img for img in images], threshold=0.5)
|
327 |
res = []
|
328 |
# align left&right for rows, align top&bottom for columns
|
329 |
for tbl in tbls:
|
python/util/db_conn.py
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import logging
|
2 |
+
import time
|
3 |
+
from util import config
|
4 |
+
import pandas as pd
|
5 |
+
|
6 |
+
class Postgre(object):
|
7 |
+
def __init__(self, env, dbnm):
|
8 |
+
self.config = config.init(env)
|
9 |
+
self.conn = None
|
10 |
+
self.dbnm = dbnm
|
11 |
+
self.__open__()
|
12 |
+
|
13 |
+
def __open__(self):
|
14 |
+
import psycopg2
|
15 |
+
try:
|
16 |
+
if self.conn:self.__close__()
|
17 |
+
del self.conn
|
18 |
+
except Exception as e:
|
19 |
+
pass
|
20 |
+
|
21 |
+
try:
|
22 |
+
self.conn = psycopg2.connect(f"dbname={self.dbnm} user={self.config.get('pgdb_usr')} password={self.config.get('pgdb_pwd')} host={self.config.get('pgdb_host')} port={self.config.get('pgdb_port')}")
|
23 |
+
except Exception as e:
|
24 |
+
logging.error("Fail to connect %s "%self.config.get("pgdb_host") + str(e))
|
25 |
+
|
26 |
+
|
27 |
+
def __close__(self):
|
28 |
+
try:
|
29 |
+
self.conn.close()
|
30 |
+
except Exception as e:
|
31 |
+
logging.error("Fail to close %s "%self.config.get("pgdb_host") + str(e))
|
32 |
+
|
33 |
+
|
34 |
+
def select(self, sql):
|
35 |
+
for _ in range(10):
|
36 |
+
try:
|
37 |
+
return pd.read_sql(sql, self.conn)
|
38 |
+
except Exception as e:
|
39 |
+
logging.error(f"Fail to exec {sql}l "+str(e))
|
40 |
+
self.__open__()
|
41 |
+
time.sleep(1)
|
42 |
+
|
43 |
+
return pd.DataFrame()
|
44 |
+
|