ocr-stream 2.5__tar.gz → 2.5.2__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: ocr_stream
3
- Version: 2.5
3
+ Version: 2.5.2
4
4
  Summary: Extrair textos de documentos digitalizados e imagens.
5
5
  Author: Bruno Chaves
6
6
  Requires-Python: >=3.11
@@ -12,7 +12,7 @@ Requires-Dist: pandas
12
12
  Requires-Dist: requests>=2.32.4
13
13
  Requires-Dist: opencv-python
14
14
  Requires-Dist: pytesseract
15
- Requires-Dist: convert_stream>=2.5
15
+ Requires-Dist: convert_stream>=2.5.1
16
16
  Requires-Dist: soup_files>=1.4.2
17
17
 
18
18
  # ocrlib
@@ -6,10 +6,10 @@ from abc import abstractmethod, ABC
6
6
  from typing import Union
7
7
  from pandas import DataFrame
8
8
  from soup_files import File
9
- from convert_stream import DocumentPdf, PageDocumentPdf, DictTextTable
9
+ from convert_stream import DocumentPdf, PageDocumentPdf
10
10
  from convert_stream.mod_types.modules import DEFAULT_LIB_IMAGE, LibImage
11
11
  from convert_stream.image import ImageObject
12
- from sheet_stream.type_utils import MetaDataFile
12
+ from sheet_stream import MetaDataFile, TableDocuments
13
13
  from ocr_stream.modules import (
14
14
  LibOcr, DEFAULT_LIB_OCR, pytesseract, pyocr, easyocr
15
15
  )
@@ -77,7 +77,7 @@ class TextRecognized(object):
77
77
  tmp_doc.metadata.dir_path = self.metadata.dir_path
78
78
  return tmp_doc
79
79
 
80
- def to_dict(self, separator: str = '\n') -> DictTextTable:
80
+ def to_dict(self, separator: str = '\n') -> TableDocuments:
81
81
  return self.to_page_pdf().to_dict(separator=separator)
82
82
 
83
83
  def to_dataframe(self, separator='\n') -> DataFrame:
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: ocr_stream
3
- Version: 2.5
3
+ Version: 2.5.2
4
4
  Summary: Extrair textos de documentos digitalizados e imagens.
5
5
  Author: Bruno Chaves
6
6
  Requires-Python: >=3.11
@@ -12,7 +12,7 @@ Requires-Dist: pandas
12
12
  Requires-Dist: requests>=2.32.4
13
13
  Requires-Dist: opencv-python
14
14
  Requires-Dist: pytesseract
15
- Requires-Dist: convert_stream>=2.5
15
+ Requires-Dist: convert_stream>=2.5.1
16
16
  Requires-Dist: soup_files>=1.4.2
17
17
 
18
18
  # ocrlib
@@ -5,5 +5,5 @@ pandas
5
5
  requests>=2.32.4
6
6
  opencv-python
7
7
  pytesseract
8
- convert_stream>=2.5
8
+ convert_stream>=2.5.1
9
9
  soup_files>=1.4.2
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "ocr_stream"
7
- version = "2.5"
7
+ version = "2.5.2"
8
8
  authors = [
9
9
  { name="Bruno Chaves"},
10
10
  ]
@@ -21,7 +21,7 @@ dependencies = [
21
21
  "requests>=2.32.4",
22
22
  "opencv-python",
23
23
  "pytesseract",
24
- "convert_stream>=2.5",
24
+ "convert_stream>=2.5.1",
25
25
  "soup_files>=1.4.2",
26
26
  ]
27
27
 
File without changes
File without changes