refactor code (#583)

### What problem does this PR solve?

### Type of change

- [x] Refactoring

deepdoc/parser/__init__.py

@@ -1,6 +1,6 @@
 
 
-from .pdf_parser import HuParser as PdfParser, PlainParser
-from .docx_parser import HuDocxParser as DocxParser
-from .excel_parser import HuExcelParser as ExcelParser
-from .ppt_parser import HuPptParser as PptParser
+from .pdf_parser import RAGFlowPdfParser as PdfParser, PlainParser
+from .docx_parser import RAGFlowDocxParser as DocxParser
+from .excel_parser import RAGFlowExcelParser as ExcelParser
+from .ppt_parser import RAGFlowPptParser as PptParser

deepdoc/parser/docx_parser.py

@@ -7,7 +7,7 @@ from rag.nlp import huqie
 from io import BytesIO
 
 
-class HuDocxParser:
+class RAGFlowDocxParser:
 
     def __extract_table_content(self, tb):
         df = []

deepdoc/parser/excel_parser.py

@@ -6,7 +6,7 @@ from io import BytesIO
 from rag.nlp import find_codec
 
 
-class HuExcelParser:
+class RAGFlowExcelParser:
     def html(self, fnm):
         if isinstance(fnm, str):
             wb = load_workbook(fnm)
@@ -74,5 +74,5 @@ class HuExcelParser:
 
 
 if __name__ == "__main__":
-    psr = HuExcelParser()
+    psr = RAGFlowExcelParser()
     psr(sys.argv[1])

deepdoc/parser/pdf_parser.py

@@ -23,7 +23,7 @@ from huggingface_hub import snapshot_download
 logging.getLogger("pdfminer").setLevel(logging.WARNING)
 
 
-class HuParser:
+class RAGFlowPdfParser:
     def __init__(self):
         self.ocr = OCR()
         if hasattr(self, "model_speciess"):

deepdoc/parser/ppt_parser.py

@@ -14,7 +14,7 @@ from io import BytesIO
 from pptx import Presentation
 
 
-class HuPptParser(object):
+class RAGFlowPptParser(object):
     def __init__(self):
         super().__init__()
 

deepdoc/vision/t_ocr.py

@@ -11,10 +11,6 @@
 #  limitations under the License.
 #
 
-from deepdoc.vision.seeit import draw_box
-from deepdoc.vision import OCR, init_in_out
-import argparse
-import numpy as np
 import os
 import sys
 sys.path.insert(
@@ -25,6 +21,11 @@ sys.path.insert(
                 os.path.abspath(__file__)),
             '../../')))
 
+from deepdoc.vision.seeit import draw_box
+from deepdoc.vision import OCR, init_in_out
+import argparse
+import numpy as np
+
 
 def main(args):
     ocr = OCR()

deepdoc/vision/t_recognizer.py

@@ -10,17 +10,7 @@
 #  See the License for the specific language governing permissions and
 #  limitations under the License.
 #
-
-from deepdoc.vision.seeit import draw_box
-from deepdoc.vision import Recognizer, LayoutRecognizer, TableStructureRecognizer, OCR, init_in_out
-from api.utils.file_utils import get_project_base_directory
-import argparse
-import os
-import sys
-import re
-
-import numpy as np
-
+import os, sys
 sys.path.insert(
     0,
     os.path.abspath(
@@ -29,6 +19,13 @@ sys.path.insert(
                 os.path.abspath(__file__)),
             '../../')))
 
+from deepdoc.vision.seeit import draw_box
+from deepdoc.vision import Recognizer, LayoutRecognizer, TableStructureRecognizer, OCR, init_in_out
+from api.utils.file_utils import get_project_base_directory
+import argparse
+import re
+import numpy as np
+
 
 def main(args):
     images, outputs = init_in_out(args)