mirror of
https://github.com/opendatalab/MinerU.git
synced 2026-03-27 11:08:32 +07:00
demo重构
This commit is contained in:
@@ -28,7 +28,9 @@ pip install -r requirements.txt
|
||||
3.Run the main script
|
||||
|
||||
```sh
|
||||
use demo/demo_test.py
|
||||
use demo/text_demo.py
|
||||
or
|
||||
use demo/ocr_demo.py
|
||||
```
|
||||
|
||||
### 版权说明
|
||||
|
||||
32
demo/demo_commons.py
Normal file
32
demo/demo_commons.py
Normal file
@@ -0,0 +1,32 @@
|
||||
import json
|
||||
|
||||
from app.common.s3 import get_s3_config
|
||||
from magic_pdf.libs.commons import join_path, read_file, json_dump_path
|
||||
|
||||
|
||||
local_json_path = "Z:/format.json"
|
||||
local_jsonl_path = "Z:/format.jsonl"
|
||||
|
||||
def get_json_from_local_or_s3(book_name=None):
|
||||
if book_name is None:
|
||||
with open(local_json_path, "r", encoding="utf-8") as json_file:
|
||||
json_line = json_file.read()
|
||||
json_object = json.loads(json_line)
|
||||
else:
|
||||
# error_log_path & json_dump_path
|
||||
# 可配置从上述两个地址获取源json
|
||||
json_path = join_path(json_dump_path, book_name + ".json")
|
||||
s3_config = get_s3_config(json_path)
|
||||
file_content = read_file(json_path, s3_config)
|
||||
json_str = file_content.decode("utf-8")
|
||||
# logger.info(json_str)
|
||||
json_object = json.loads(json_str)
|
||||
return json_object
|
||||
|
||||
|
||||
def write_json_to_local(jso, book_name=None):
|
||||
if book_name is None:
|
||||
with open(local_json_path, "w", encoding="utf-8") as file:
|
||||
file.write(json.dumps(jso, ensure_ascii=False))
|
||||
else:
|
||||
pass
|
||||
@@ -5,7 +5,7 @@ from loguru import logger
|
||||
from pathlib import Path
|
||||
|
||||
from app.common.s3 import get_s3_config
|
||||
from demo.demo_test import get_json_from_local_or_s3
|
||||
from demo.demo_commons import get_json_from_local_or_s3
|
||||
from magic_pdf.dict2md.ocr_mkcontent import (
|
||||
ocr_mk_mm_markdown_with_para,
|
||||
ocr_mk_nlp_markdown,
|
||||
@@ -14,7 +14,7 @@ from magic_pdf.dict2md.ocr_mkcontent import (
|
||||
ocr_mk_mm_markdown_with_para_and_pagination,
|
||||
make_standard_format_with_para
|
||||
)
|
||||
from magic_pdf.libs.commons import join_path
|
||||
from magic_pdf.libs.commons import join_path, read_file
|
||||
from magic_pdf.pdf_parse_by_ocr import parse_pdf_by_ocr
|
||||
|
||||
|
||||
@@ -43,7 +43,8 @@ def ocr_local_parse(ocr_pdf_path, ocr_json_file_path):
|
||||
ocr_pdf_model_info = read_json_file(ocr_json_file_path)
|
||||
pth = Path(ocr_json_file_path)
|
||||
book_name = pth.name
|
||||
ocr_parse_core(book_name, ocr_pdf_path, ocr_pdf_model_info)
|
||||
pdf_bytes = read_file(ocr_pdf_path, None)
|
||||
ocr_parse_core(book_name, pdf_bytes, ocr_pdf_model_info)
|
||||
except Exception as e:
|
||||
logger.exception(e)
|
||||
|
||||
@@ -54,20 +55,20 @@ def ocr_online_parse(book_name, start_page_id=0, debug_mode=True):
|
||||
# logger.info(json_object)
|
||||
s3_pdf_path = json_object["file_location"]
|
||||
s3_config = get_s3_config(s3_pdf_path)
|
||||
pdf_bytes = read_file(s3_pdf_path, s3_config)
|
||||
ocr_pdf_model_info = json_object.get("doc_layout_result")
|
||||
ocr_parse_core(book_name, s3_pdf_path, ocr_pdf_model_info, s3_config=s3_config)
|
||||
ocr_parse_core(book_name, pdf_bytes, ocr_pdf_model_info)
|
||||
except Exception as e:
|
||||
logger.exception(e)
|
||||
|
||||
|
||||
def ocr_parse_core(book_name, ocr_pdf_path, ocr_pdf_model_info, start_page_id=0, s3_config=None):
|
||||
def ocr_parse_core(book_name, pdf_bytes, ocr_pdf_model_info, start_page_id=0):
|
||||
save_tmp_path = os.path.join(os.path.dirname(__file__), "../..", "tmp", "unittest")
|
||||
save_path = join_path(save_tmp_path, "md")
|
||||
save_path_with_bookname = os.path.join(save_path, book_name)
|
||||
text_content_save_path = f"{save_path_with_bookname}/book.md"
|
||||
pdf_info_dict = parse_pdf_by_ocr(
|
||||
ocr_pdf_path,
|
||||
s3_config,
|
||||
pdf_bytes,
|
||||
ocr_pdf_model_info,
|
||||
save_path,
|
||||
book_name,
|
||||
|
||||
@@ -5,6 +5,7 @@ from pathlib import Path
|
||||
|
||||
import click
|
||||
|
||||
from demo.demo_commons import get_json_from_local_or_s3, write_json_to_local, local_jsonl_path, local_json_path
|
||||
from magic_pdf.dict2md.mkcontent import mk_mm_markdown
|
||||
from magic_pdf.pipeline import (
|
||||
meta_scan,
|
||||
@@ -13,38 +14,10 @@ from magic_pdf.pipeline import (
|
||||
pdf_intermediate_dict_to_markdown,
|
||||
save_tables_to_s3,
|
||||
)
|
||||
from magic_pdf.libs.commons import join_path, read_file, json_dump_path
|
||||
from app.common.s3 import get_s3_config
|
||||
from magic_pdf.libs.commons import join_path
|
||||
from loguru import logger
|
||||
|
||||
|
||||
local_json_path = "Z:/format.json"
|
||||
local_jsonl_path = "Z:/format.jsonl"
|
||||
|
||||
|
||||
def get_json_from_local_or_s3(book_name=None):
|
||||
if book_name is None:
|
||||
with open(local_json_path, "r", encoding="utf-8") as json_file:
|
||||
json_line = json_file.read()
|
||||
json_object = json.loads(json_line)
|
||||
else:
|
||||
# error_log_path & json_dump_path
|
||||
# 可配置从上述两个地址获取源json
|
||||
json_path = join_path(json_dump_path, book_name + ".json")
|
||||
s3_config = get_s3_config(json_path)
|
||||
file_content = read_file(json_path, s3_config)
|
||||
json_str = file_content.decode("utf-8")
|
||||
# logger.info(json_str)
|
||||
json_object = json.loads(json_str)
|
||||
return json_object
|
||||
|
||||
|
||||
def write_json_to_local(jso, book_name=None):
|
||||
if book_name is None:
|
||||
with open(local_json_path, "w", encoding="utf-8") as file:
|
||||
file.write(json.dumps(jso, ensure_ascii=False))
|
||||
else:
|
||||
pass
|
||||
|
||||
|
||||
def demo_parse_pdf(book_name=None, start_page_id=0, debug_mode=True):
|
||||
Reference in New Issue
Block a user