ocr-stream 1.2__tar.gz → 1.4__tar.gz

{ocr_stream-1.2 → ocr_stream-1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 1.2
+Version: 1.4
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires: PyPDF2
@@ -17,14 +17,14 @@ Requires: soup_files
 Requires: convert_stream
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown
-Requires-Dist: PyPDF2==3.0.1
-Requires-Dist: PyMuPDF==1.25.5
-Requires-Dist: reportlab==4.3.1
-Requires-Dist: openpyxl==3.1.5
-Requires-Dist: pandas==2.2.3
-Requires-Dist: Pillow==11.2.1
-Requires-Dist: opencv-python==4.11.0.86
-Requires-Dist: convert_stream
+Requires-Dist: PyMuPDF
+Requires-Dist: reportlab
+Requires-Dist: openpyxl
+Requires-Dist: pandas
+Requires-Dist: Pillow>=11.3.0
+Requires-Dist: requests>=2.32.4
+Requires-Dist: opencv-python
+Requires-Dist: convert_stream>=1.4
 Requires-Dist: soup_files
 Dynamic: requires
 

{ocr_stream-1.2 → ocr_stream-1.4}/ocr_stream/__init__.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python3
 
 from .models.m_modules_ocr import LibraryOCR
-from .modules_ocr import TextRecognized
+from .modules_ocr import TextRecognized, BinaryTesseract
 from .utils import File, Directory, InputFiles, UserFileSystem
 from .extractors import RecognizeImage, RecognizePdf
 from ._version import __version__, __modify_data__

ocr_stream-1.4/ocr_stream/_version.py

@@ -0,0 +1,5 @@
+#!/usr/bin/env python3
+#
+
+__version__ = '1.4'
+__modify_data__ = '2025-07-11'

ocr_stream-1.4/ocr_stream/extractors.py

@@ -0,0 +1,126 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+from typing import List
+from ocr_stream.utils import (
+    File, ImageObject,
+    DataFrame, PageDocumentPdf, DocumentPdf,
+)
+from ocr_stream.models import LibraryOCR
+from ocr_stream.modules_ocr import ModuleOcr, TextRecognized, BinaryTesseract
+from convert_stream import ConvertPdfToImage, CreatePbar
+from soup_files import ProgressBarAdapter
+
+
+class RecognizeImage(object):
+    """
+        Reconhecer textos em imagens com pytesseract|pyocr usando o tesseract
+    """
+
+    def __init__(self, module_ocr):
+        super().__init__()
+        self.module_ocr: ModuleOcr = module_ocr
+
+    def image_content_data(self, img: ImageObject | File) -> DataFrame:
+        """
+        Retorna uma tabela estruturada com os dados reconhecidos da imagem.
+
+        Este método extrai as informações contidas em uma imagem e as organiza em
+        um DataFrame com colunas como o texto detectado, coordenadas da caixa delimitadora,
+        nível do OCR e confiabilidade, entre outras.
+
+        Parâmetros:
+        img (ImageObject | File): A imagem a ser processada, podendo ser um objeto
+        `ImageObject` ou um arquivo `File`.
+
+        Retorna:
+        DataFrame: Uma tabela contendo os dados reconhecidos na imagem.
+
+        Lança:
+        ValueError: Se o tipo do parâmetro `img` não for `ImageObject` ou `File`.
+
+        @param img: Imagem a ser processada, podendo ser uma instância de `ImageObject`
+            ou um arquivo `File`.
+        @type img: ImageObject | File
+
+        @rtype: pandas.DataFrame
+        @return: Um DataFrame com os dados reconhecidos na imagem.
+
+        @raise ValueError: Caso o tipo de `img` não seja `ImageObject` nem `File`.
+        """
+        if isinstance(img, File):
+            return self.module_ocr.image_content_data(ImageObject.create_from_file(img))
+        elif isinstance(img, ImageObject):
+            return self.module_ocr.image_content_data(img)
+        else:
+            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
+
+    def image_recognize(self, img: ImageObject | File) -> TextRecognized:
+        if isinstance(img, File):
+            return self.module_ocr.image_recognize(ImageObject.create_from_file(img))
+        elif isinstance(img, ImageObject):
+            return self.module_ocr.image_recognize(img)
+        else:
+            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
+
+    def image_to_string(self, img: ImageObject | File) -> str:
+        if isinstance(img, File):
+            return self.module_ocr.imag_to_string(ImageObject.create_from_file(img))
+        elif isinstance(img, ImageObject):
+            return self.module_ocr.imag_to_string(img)
+        else:
+            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
+
+    @classmethod
+    def create(
+            cls,
+            library_ocr: LibraryOCR = LibraryOCR.PYTESSERACT,
+            *,
+            binary_tess: BinaryTesseract = BinaryTesseract(),
+    ) -> RecognizeImage:
+        #
+        module_ocr = ModuleOcr.create(library_ocr, binary_tess=binary_tess)
+        return cls(module_ocr)
+
+
+class RecognizePdf(object):
+    def __init__(
+                self,
+                recognize_image: RecognizeImage, *,
+                pbar: ProgressBarAdapter = CreatePbar().get()
+            ):
+        self.recognize_image: RecognizeImage = recognize_image
+        self.pdf_to_image: ConvertPdfToImage = ConvertPdfToImage.create()
+        self.pbar: ProgressBarAdapter = pbar
+
+    def recognize_page_pdf(self, page: PageDocumentPdf) -> PageDocumentPdf:
+        """
+            Converte a página em Imagem, reconhece o texto com OCR e
+        retorna uma nova página com o texto reconhecido.
+        """
+        img: ImageObject = self.pdf_to_image.from_page_pdf(page)
+        text_recognized: TextRecognized = self.recognize_image.image_recognize(img)
+        return text_recognized.to_page_pdf()
+
+    def recognize_document(self, doc: DocumentPdf) -> DocumentPdf:
+        new_pages: List[PageDocumentPdf] = []
+
+        for n, page in enumerate(doc.get_pages()):
+            self.pbar.update(
+                ((n+1)/doc.get_num_pages()) * 100,
+                f'Reconhecendo página PDF [{n+1} de {doc.get_num_pages()}]',
+            )
+            recognized_page: PageDocumentPdf = self.recognize_page_pdf(page)
+            new_pages.append(recognized_page)
+        return DocumentPdf(pages=new_pages)
+
+    @classmethod
+    def create(
+            cls,
+            library_ocr=LibraryOCR.PYTESSERACT,
+            *,
+            binary_tess: BinaryTesseract = BinaryTesseract(),
+            pbar: ProgressBarAdapter = CreatePbar().get(),
+    ) -> RecognizePdf:
+        #
+        recognize_image = RecognizeImage.create(library_ocr, binary_tess=binary_tess)
+        return cls(recognize_image, pbar=pbar)

{ocr_stream-1.2 → ocr_stream-1.4}/ocr_stream/models/__init__.py

@@ -1,3 +1,2 @@
 #!/usr/bin/env python3
 from .m_modules_ocr import LibraryOCR, ABCTextRecognized, ABCModuleOcr
-from .m_extractors import ABCRecognizeImage

ocr_stream-1.4/ocr_stream/models/m_extractors.py

@@ -0,0 +1,2 @@
+#!/usr/bin/env python3
+

{ocr_stream-1.2 → ocr_stream-1.4}/ocr_stream/models/m_modules_ocr.py

@@ -8,72 +8,61 @@ from pandas import DataFrame
 from soup_files import File, Directory
 from convert_stream import ImageObject, DocumentPdf, PageDocumentPdf, LibraryImage
 
+
 class LibraryOCR(Enum):
     PYTESSERACT = 'pytesseract'
     PYOCR = 'pyocr'
-    
-    
+
+
 class ABCTextRecognized(ABC):
-    def __init__(self, bytes_recognized:bytes):
+    def __init__(self, bytes_recognized: bytes):
         super().__init__()
-        self.bytes_recognized:bytes = bytes_recognized
-        
+        self.bytes_recognized: bytes = bytes_recognized
+
     @abstractmethod
     def to_string(self) -> str:
         pass
-    
+
     @abstractmethod
     def to_page_pdf(self) -> PageDocumentPdf:
         pass
-    
+
     @abstractmethod
     def to_document(self) -> DocumentPdf:
         pass
-    
+
     @abstractmethod
-    def to_dataframe(self, separator:str='\n', column_text: str = 'TEXT') -> DataFrame:
+    def to_dataframe(self, separator: str = '\n', column_text: str = 'TEXT') -> DataFrame:
         pass
-    
+
     @abstractmethod
-    def to_dict(self, separator:str='\n', column_text: str = 'TEXT') -> Dict[str, List[str]]:
+    def to_dict(self, separator: str = '\n', column_text: str = 'TEXT') -> Dict[str, List[str]]:
         pass
-  
-  
+
+
 #======================================================================#
 # Abstração para o módulo que extrai o texto das imagens (OCR)
 # pode ser implementado com pytesseract ou outros.
 #======================================================================#
+
+
 class ABCModuleOcr(ABC):
-    def __init__(self, *, cmd_executable:File, lang:str=None, tess_data_dir:Directory=None):
-        self.cmd_executable:File = cmd_executable
-        self.lang:str = lang
-        self.tess_data_dir:Directory = tess_data_dir
-        self.current_library_ocr:LibraryOCR = None
-    
+
+    def __init__(self, *, cmd_executable: File, lang: str = None, tess_data_dir: Directory = None):
+        self.cmd_executable: File = cmd_executable
+        self.lang: str = lang
+        self.tess_data_dir: Directory = tess_data_dir
+        self.current_library_ocr: LibraryOCR = None
+
     @abstractmethod
-    def image_recognize(self, img:ImageObject) -> ABCTextRecognized:
+    def image_recognize(self, img: ImageObject) -> ABCTextRecognized:
         pass
-    
+
     @abstractmethod
     def image_content_data(self, img: ImageObject) -> DataFrame:
         pass
-    
+
     @abstractmethod
-    def imag_to_string(self, img:ImageObject) -> str:
-        pass
-    
-    @classmethod
-    def create(
-                cls, 
-                library_ocr:LibraryOCR=LibraryOCR.PYTESSERACT,
-                *,
-                path_tesseract:File,
-                lang:str=None,
-                tess_data_dir:Directory=None,
-            ) -> ABCModuleOcr:
+    def imag_to_string(self, img: ImageObject) -> str:
         pass
-    
-    
-  
 
-    

{ocr_stream-1.2 → ocr_stream-1.4}/ocr_stream/modules_ocr.py

@@ -2,99 +2,93 @@
 from __future__ import annotations
 from types import ModuleType
 from typing import List, Dict
-from pathlib import Path
+from shutil import which
 import re
 import os
 from pandas import DataFrame
 from pytesseract import pytesseract, Output
-from ocr_stream.models import ( 
+
+from ocr_stream.models import (
     ABCModuleOcr, ABCTextRecognized, LibraryOCR,
 )
+
 from ocr_stream.utils import (
     PageDocumentPdf, DocumentPdf, LibraryImage, LibraryPDF,
-    ImageObject, ArrayString, DataString, File, Directory
+    ImageObject, DataString, File, Directory,
+    ColumnTable, KERNEL_TYPE, ArrayString,
 )
 
+try:
+    import pyocr
+except Exception as e:
+    print(e)
 
-
-def get_temp_dir(prefix=Path().home()) -> str:
-    prefix = Path().home()
-    sys_temp_dir = os.environ.get('TMP', os.path.join(prefix, 'var', 'temp'))
-    if not sys_temp_dir:
-        sys_temp_dir = os.path.join(prefix, 'var', 'temp')
-    if not os.path.exists(sys_temp_dir):
-        try:
-            os.makedirs(sys_temp_dir)
-        except Exception as e:
-            print(e)
-    os.environ["TEMP"] = sys_temp_dir
-    print('-' * 30)
-    print(f'Diretório temporário: {sys_temp_dir}')
-    print('-' * 30)
-    return sys_temp_dir
-
-  
 class TextRecognizedToi(object):
     """
         Extração/Filtro de texto em documentos do tipo carta TOI.
     """
-    def __init__(self, value:str=None, *, separator:str='\n', column_text='TEXT', file:File):
-        # Todas as linhas que foram reconhecidas em uma imagem/página PDF.
+
+    def __init__(self, value: str = None, *, separator: str = '\n', column_text='TEXT', file: File):
+        # Todas as linhas reconhecidas em imagem/página PDF.
         if not isinstance(value, str):
             raise ValueError(f'{__class__.__name__}Use: str, não {type(value)}')
-        self.text_document:str = DataString(value).to_upper().value if value is not None else None
-        self.separator:str = separator
-        self.file:File = file
-        self.column_text:str = column_text
-        
+        self.text_document: str = DataString(value).to_upper().value if value is not None else None
+        self.separator: str = separator
+        self.file: File = file
+        self.column_text: str = column_text
+        self.listText = []
         self.pattern_line_uc_toi = r'UC:\s*\d+\s*TOI:\s*\d+'
         self.pattern_uc = r'\b(?:UC:?|UC):?\s*(\d+)'
         self.pattern_toi = r'\b(?:TOI?|TO):?\s*(\d+)'
         self.pattern_localidade = r'LOCALIDADE:\s*(.*)'
-        
+
         self.words_toi = (
             'TO]!', 'TO]', 'TOl', 'TOI', 'TOI:'
         )
-        
+
         self.words_uc = (
             'UE', 'UE:' 'UC', 'UC:', 'JC;'
         )
-        
+
     def __replace_chars(self, text, char, new_char) -> str:
         return re.sub(r'{}'.format(char), new_char, text)
-    
-    def __get_match_groups(self, pattern, text:str) -> object | None:
+
+    def __get_match_groups(self, pattern, text: str) -> object | None:
         i = re.search(pattern, text)
         if i is None:
             return None
         return i.groups()
-        
-    def __get_match_group(self, pattern, text:str) -> object | None:
+
+    def __get_match_group(self, pattern, text: str) -> object | None:
         i = re.search(pattern, text)
         if i is None:
             return None
         return i.group()
-    
+
     def is_null(self) -> bool:
         if self.text_document is None:
             return True
         if self.text_document == '':
             return True
         return False
-        
+
     def to_string(self) -> str | None:
         return DataString(self.text_document).to_upper().value if not self.is_null() else None
-    
+
     def to_list(self) -> List[str]:
         return self.text_document.split(self.separator)
-    
+
     def to_array_string(self) -> ArrayString:
-        values:List[str] = self.to_list()
-    
+        try:
+            return ArrayString(self.to_list())
+        except Exception as e:
+            print(e)
+            return ArrayString([])
+
     def to_upper(self) -> TextRecognizedToi:
         self.text_document = self.text_document.upper()
         return self
-    
+
     def to_dataframe(self) -> DataFrame:
         uc = self.uc()
         toi = self.toi()
@@ -107,8 +101,9 @@ class TextRecognizedToi(object):
             toi = 'nan'
         else:
             toi = DataString(toi).replace_bad_chars()
-            toi = toi.replace_all('- ', '-').replace_all('_', ' ').replace_all('-', ' ').replace_all('TOI', '').value.strip()
-        
+            toi = toi.replace_all('- ', '-').replace_all('_', ' ').replace_all('-', ' ').replace_all('TOI',
+                                                                                                     '').value.strip()
+
         df = DataFrame()
         df[self.column_text] = self.to_list()
         num = len(df)
@@ -118,7 +113,7 @@ class TextRecognizedToi(object):
         df['ARQUIVO'] = [self.file.absolute()] * num
         df['MD5'] = [self.file.md5()] * num
         return df
-    
+
     def remove_bad_chars(self) -> TextRecognizedToi:
         values = []
         for item in self.to_list():
@@ -127,21 +122,22 @@ class TextRecognizedToi(object):
             )
         self.listText = values
         return self
-    
-    def uc(self) -> str:
+
+    def uc(self) -> str | None:
         """
             Extrair o número da UC em forma de string, se a operação falhar, será retornado (nan)
         """
         line_uc_toi = self.line_uc_and_toi()
-        uc_value:str = None
+
+        uc_value: str = None
         if line_uc_toi is not None:
-            uc_value = self.__get_match_group(self.pattern_uc, line_uc_toi)
+            uc_value: str = self.__get_match_group(self.pattern_uc, line_uc_toi.value)
         if uc_value is not None:
             return uc_value.replace('UC', '')
-        
+
         arr = ArrayString(self.text_document.split(' '))
         for w in self.words_uc:
-            out = arr.get_next(w).value
+            out: str = arr.get_next(w).value
             try:
                 int(out)
             except:
@@ -149,16 +145,19 @@ class TextRecognizedToi(object):
             else:
                 return out
         return None
-    
-    def toi(self) -> str:
+
+    def toi(self) -> str | None:
         line_uc_toi = self.line_uc_and_toi()
-        toi_value:str = None
+        if line_uc_toi is None:
+            return None
+
+        toi_value: str = None
         if line_uc_toi is not None:
-            toi_value = self.__get_match_group(self.pattern_toi, line_uc_toi)
+            toi_value = self.__get_match_group(self.pattern_toi, line_uc_toi.value)
         if toi_value is not None:
             return toi_value.replace('TOI', '')
         arr = ArrayString(self.text_document.split(' '))
-        
+
         for w in self.words_toi:
             out = arr.get_next(w).value
             try:
@@ -168,7 +167,7 @@ class TextRecognizedToi(object):
             else:
                 return out
         return None
-        
+
     def roteiro(self) -> str:
         self.remove_bad_chars()
         pass
@@ -184,10 +183,10 @@ class TextRecognizedToi(object):
         loc = DataString(loc).replace_all(' ').value
         # Remover "_" do final
         return re.sub(r"_$", "", loc)
-    
+
     def line_postagem(self) -> str:
         pass
-        
+
     def line_uc_and_toi(self) -> DataString | None:
         line_toi = None
         lines = self.to_list()
@@ -198,9 +197,9 @@ class TextRecognizedToi(object):
             line_toi = out
             break
         return line_toi
-    
+
     def line_roteiro(self) -> DataString:
-        return DataString(None) 
+        return DataString(None)
 
 
 #======================================================================#
@@ -216,66 +215,125 @@ class TextRecognized(ABCTextRecognized):
     -> str
     -> PageDocumentPdf()
     """
+
     def __init__(self, bytes_recognized: bytes):
         super().__init__(bytes_recognized)
-        
+
         self.list_bad_char: List[str] = [
-                            ':', ',', ';', '$', '=', 
-                            '!', '}', '{', '(', ')', 
-                            '|', '\\', '‘', '*'
-                            '¢', '“', '\'', '¢', '"', 
-                            '#', '.', '<', '?', '>', 
-                            '»', '@', '+', '[', ']',
-                            '%', '~', '¥', '♀',
+            ':', ',', ';', '$', '=',
+            '!', '}', '{', '(', ')',
+            '|', '\\', '‘', '*'
+                            '¢', '“', '\'', '¢', '"',
+            '#', '.', '<', '?', '>',
+            '»', '@', '+', '[', ']',
+            '%', '~', '¥', '♀',
         ]
-    
+
     def to_string(self) -> str | None:
         return self.to_page_pdf().to_string()
-    
+
     def to_page_pdf(self) -> PageDocumentPdf:
         _page = PageDocumentPdf.create_from_page_bytes(
-            self.bytes_recognized, 
-            library = LibraryPDF.FITZ
+            self.bytes_recognized,
+            library=LibraryPDF.FITZ
         )
         return _page
-    
+
     def to_document(self) -> DocumentPdf:
-        doc = DocumentPdf(pages=self.to_page_pdf())
-        return doc
-    
-    def to_dict(self, separator = '\n', column_text = 'TEXT') -> Dict[str, List[str]]:
-        image_dict:Dict[str, List[str]] = {
-            column_text: [],
-            'NUM_LINHA': [],
+        return DocumentPdf(pages=[self.to_page_pdf()])
+
+    def to_dict(
+                self,
+                separator='\n',
+                column_text: ColumnTable.TEXT = ColumnTable.TEXT
+            ) -> Dict[str, List[str]]:
+        #
+        image_dict: Dict[str, List[str]] = {
+            column_text.value: [],
+            ColumnTable.NUMLINHA.value: [],
         }
-        image_text:str = self.to_string()
+        image_text: str = self.to_string()
         if (image_text is None) or (image_text == ''):
             return {
-                    column_text: ['nan'],
-                    'NUM_LINHA': ['nan'],
-                }
-        
+                column_text.value: ['nan'],
+                ColumnTable.NUMLINHA.value: ['nan'],
+            }
+
         image_lines: List[str] = image_text.split(separator)
         for num, line in enumerate(image_lines):
-            image_dict[column_text].append(line)
-            image_dict['NUM_LINHA'].append(num+1)
-        return image_dict    
-    
-    def to_dataframe(self, separator = '\n', column_text = 'TEXT') -> DataFrame:
+            image_dict[column_text.value].append(line)
+            image_dict[ColumnTable.NUMLINHA.value].append(f'{num + 1}')
+        return image_dict
+
+    def to_dataframe(self, separator='\n', column_text: ColumnTable.TEXT = ColumnTable.TEXT) -> DataFrame:
         return DataFrame(self.to_dict(separator, column_text))
-    
-    
+
+
+class BinaryTesseract(object):
+    """
+        Fornece o caminho absoluto do tesseract instalado no sistema, se
+    disponível. Você pode usar um binário alternativo, basta informar
+    o caminho do binário desejado no construtor.
+    """
+    _instance = None  # Atributo de classe para armazenar a instância singleton
+
+    def __new__(cls, *args, **kwargs):
+        if cls._instance is None:
+            cls._instance = super(BinaryTesseract, cls).__new__(cls)
+        return cls._instance
+
+    def __init__(
+                    self,
+                    path: File = None, *,
+                    lang: str = None,
+                    tess_data_dir: Directory = None
+            ):
+        #
+        # Garante que __init__ não será executado mais de uma vez
+        if hasattr(self, '_initialized') and self._initialized:
+            return
+        self._initialized = True
+
+        self.tess_data_dir: Directory = tess_data_dir
+        self.lang: str = lang
+        self.path_tesseract: File = path
+        if self.path_tesseract is None:
+            if KERNEL_TYPE == 'Windows':
+                name = 'tesseract.exe'
+            else:
+                name = 'tesseract'
+            output = which(name)
+            if output is not None:
+                if os.path.isfile(output):
+                    print(f'[binary]: {output}')
+                    self.path_tesseract = File(output)
+
+    def exists(self) -> bool:
+        """Verifica se o binário tesseract existe"""
+        if self.path_tesseract is None:
+            return False
+        return self.path_tesseract.exists()
+
+
 #======================================================================#
 # Modulo OCR pytesseract
 #======================================================================#
 
-class IModulePytesseract(ABCModuleOcr):
-    def __init__(self, *, cmd_executable, lang = None, tess_data_dir = None):
+
+class IPytesseract(ABCModuleOcr):
+
+    def __init__(
+                    self, *,
+                    cmd_executable: File,
+                    lang: str = None,
+                    tess_data_dir: Directory = None,
+            ):
+        #
         super().__init__(cmd_executable=cmd_executable, lang=lang, tess_data_dir=tess_data_dir)
-        self.current_library:LibraryOCR = LibraryOCR.PYTESSERACT
-        self._pyTesseract:pytesseract = pytesseract
+        self.current_library: LibraryOCR = LibraryOCR.PYTESSERACT
+        self._pyTesseract: pytesseract = pytesseract
         self._pyTesseract.tesseract_cmd = self.cmd_executable.absolute()
-        
+
     def __get_tess_dir_config(self) -> str | None:
         """
         https://github.com/h/pytesseract
@@ -292,143 +350,142 @@ class IModulePytesseract(ABCModuleOcr):
         if not self.tess_data_dir.path.exists():
             return ''
         return r'--tessdata-dir "{}"'.format(self.tess_data_dir.absolute())
-        
+
     def imag_to_string(self, img) -> str:
         if img.current_library == LibraryImage.OPENCV:
-            text_img:str = self._pyTesseract.image_to_string(
+            text_img: str = self._pyTesseract.image_to_string(
                 img.to_image_opencv(),
                 lang=self.lang,
                 config=self.__get_tess_dir_config()
             )
         else:
-            text_img:str = self._pyTesseract.image_to_string(
-                    img.to_image_pil(),
-                    lang=self.lang,
-                    config=self.__get_tess_dir_config()
-                )
+            text_img: str = self._pyTesseract.image_to_string(
+                img.to_image_pil(),
+                lang=self.lang,
+                config=self.__get_tess_dir_config()
+            )
         return text_img
-        
+
     def image_recognize(self, img) -> TextRecognized:
+        """
+
+        @rtype: TextRecognized
+        """
         if img.current_library == LibraryImage.OPENCV:
-            pdf_bytes:bytes = self._pyTesseract.image_to_pdf_or_hocr(
+            pdf_bytes: bytes = self._pyTesseract.image_to_pdf_or_hocr(
                 img.to_image_opencv(),
                 lang=self.lang,
                 config=self.__get_tess_dir_config()
             )
         else:
-            pdf_bytes:bytes = self._pyTesseract.image_to_pdf_or_hocr(
-                    img.to_image_pil(),
-                    lang=self.lang,
-                    config=self.__get_tess_dir_config()
-                )
+            pdf_bytes: bytes = self._pyTesseract.image_to_pdf_or_hocr(
+                img.to_image_pil(),
+                lang=self.lang,
+                config=self.__get_tess_dir_config()
+            )
         return TextRecognized(pdf_bytes)
-        
-    def image_content_data(self, img) -> DataFrame:
+
+    def image_content_data(self, img: ImageObject) -> DataFrame:
+        """
+            Retorna um dataframe gerado por pytesseract
+
+        @rtype: DataFrame
+        """
         if img.current_library == LibraryImage.OPENCV:
-            text_data:DataFrame = self._pyTesseract.image_to_data(
+            text_data: DataFrame = self._pyTesseract.image_to_data(
                 img.to_image_opencv(),
                 lang=self.lang,
                 config=self.__get_tess_dir_config(),
                 output_type=Output.DATAFRAME,
             )
         else:
-            text_data:DataFrame = self._pyTesseract.image_to_data(
+            text_data: DataFrame = self._pyTesseract.image_to_data(
                 img.to_image_pil(),
                 lang=self.lang,
                 config=self.__get_tess_dir_config(),
                 output_type=Output.DATAFRAME,
             )
         return text_data
-    
-   
+
+
 #======================================================================#
 # Modulo OCR pyocr
 #======================================================================#
 
-class IModulePyOcr(ABCModuleOcr):
-    def __init__(self, *, cmd_executable, lang = None, tess_data_dir = None):
+
+class IPyOcr(ABCModuleOcr):
+    def __init__(self, *, cmd_executable, lang=None, tess_data_dir=None):
         super().__init__(cmd_executable=cmd_executable, lang=lang, tess_data_dir=tess_data_dir)
-        self.current_library:LibraryOCR = LibraryOCR.PYOCR
-        self.cmd_executable:File = cmd_executable
-        self.tess_data_dir:Directory = tess_data_dir
+        self.current_library: LibraryOCR = LibraryOCR.PYOCR
+        self.cmd_executable: File = cmd_executable
+        self.tess_data_dir: Directory = tess_data_dir
         try:
             import pyocr
             import pyocr.tesseract
         except Exception as e:
-            raise (e)
-        
-        pyocr_modules:List[ModuleType] = pyocr.get_available_tools()
+            raise e
+
+        pyocr_modules: List[ModuleType] = pyocr.get_available_tools()
         if len(pyocr_modules) == 0:
             raise ValueError(f"{__class__.__name__} No OCR tool found")
         # The tools are returned in the recommended order of usage
-        self._pyOcr:pyocr.tesseract = pyocr_modules[0]
-        langs:List[str] = self._pyOcr.get_available_languages()
+        self._pyOcr: pyocr.tesseract = pyocr_modules[0]
+        langs: List[str] = self._pyOcr.get_available_languages()
         if lang in lang:
             self.lang = lang
         else:
             self.lang = langs[0]
-        
+
         # Ex: Will use tool 'libtesseract'
         print(f"Will use tool {self._pyOcr.get_name()}")
-        
+
     def imag_to_string(self, img) -> str:
         return self._pyOcr.image_to_string(
             img.to_image_pil(),
             lang=self.lang,
             builder=pyocr.builders.TextBuilder()
         )
-        
+
     def image_recognize(self, img) -> TextRecognized:
         raise NotImplementedError(f'{__class__.__name__} método não implementado')
-        
+
     def image_content_data(self, img) -> DataFrame:
         raise NotImplementedError(f'{__class__.__name__} método não implementado')
-    
-
-class ModuleOcr(ABCModuleOcr):
-    def __init__(self, module_ocr: ABCModuleOcr):
-        super().__init__(
-                cmd_executable=module_ocr.cmd_executable, 
-                lang=module_ocr.lang, 
-                tess_data_dir=module_ocr.tess_data_dir
-            )
-        self.module_ocr:ABCModuleOcr = module_ocr
-        
+
+
+class ModuleOcr(object):
+    def __init__(self, module_ocr: IPytesseract | IPyOcr):
+        self.module_ocr: IPytesseract | IPyOcr = module_ocr
+
     def image_content_data(self, img: ImageObject) -> DataFrame:
         return self.module_ocr.image_content_data(img)
-    
-    def image_recognize(self, img:ImageObject) -> TextRecognized:
+
+    def image_recognize(self, img: ImageObject) -> TextRecognized:
         return self.module_ocr.image_recognize(img)
-    
+
     def imag_to_string(self, img) -> str:
         return self.module_ocr.imag_to_string(img)
-    
+
     @classmethod
     def create(
-                cls, 
-                library_ocr = LibraryOCR.PYTESSERACT, 
-                *, 
-                path_tesseract: File, 
-                lang = None, 
-                tess_data_dir = None
-            ) -> ModuleOcr:
+            cls,
+            library_ocr=LibraryOCR.PYTESSERACT,
+            *,
+            binary_tess: BinaryTesseract = BinaryTesseract(),
+    ) -> ModuleOcr:
         #
         if library_ocr == LibraryOCR.PYTESSERACT:
-            _module_ocr = IModulePytesseract(cmd_executable=path_tesseract, lang=lang, tess_data_dir=tess_data_dir)
+            _module_ocr = IPytesseract(
+                    cmd_executable=binary_tess.path_tesseract,
+                    lang=binary_tess.lang,
+                    tess_data_dir=binary_tess.tess_data_dir,
+            )
         elif library_ocr == LibraryOCR.PYOCR:
-            _module_ocr = IModulePyOcr(cmd_executable=path_tesseract, lang=lang, tess_data_dir=tess_data_dir)
+            _module_ocr = IPyOcr(
+                cmd_executable=binary_tess.path_tesseract,
+                lang=binary_tess.lang,
+                tess_data_dir=binary_tess.tess_data_dir,
+            )
         else:
             raise NotImplementedError(f'{__class__.__name__}\nLibraryOcr não implementada: {library_ocr}')
         return cls(_module_ocr)
-        
-        
-      
-   
-
-            
-        
-        
-        
-        
-        
-

ocr_stream-1.4/ocr_stream/utils/__init__.py

@@ -0,0 +1,16 @@
+#!/usr/bin/env python3
+#
+from pandas import DataFrame
+from soup_files import (
+    File, Directory, UserAppDir, UserFileSystem,
+    InputFiles, JsonConvert, JsonData, LibraryDocs,
+    KERNEL_TYPE,
+)
+
+from convert_stream import (
+    LibraryDates, LibraryImage, LibraryPDF, LibImageToPDF, LibConvertPDF,
+    ImageObject, PdfStream, ImageStream, PageDocumentPdf, DocumentPdf,
+    ArrayString, DataString, print_line, print_title, FileToTable,
+)
+from convert_stream.models import ConvertStringDate
+from convert_stream.table_files import ColumnTable

{ocr_stream-1.2 → ocr_stream-1.4}/ocr_stream.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 1.2
+Version: 1.4
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires: PyPDF2
@@ -17,14 +17,14 @@ Requires: soup_files
 Requires: convert_stream
 Requires-Python: >=3.11
 Description-Content-Type: text/markdown
-Requires-Dist: PyPDF2==3.0.1
-Requires-Dist: PyMuPDF==1.25.5
-Requires-Dist: reportlab==4.3.1
-Requires-Dist: openpyxl==3.1.5
-Requires-Dist: pandas==2.2.3
-Requires-Dist: Pillow==11.2.1
-Requires-Dist: opencv-python==4.11.0.86
-Requires-Dist: convert_stream
+Requires-Dist: PyMuPDF
+Requires-Dist: reportlab
+Requires-Dist: openpyxl
+Requires-Dist: pandas
+Requires-Dist: Pillow>=11.3.0
+Requires-Dist: requests>=2.32.4
+Requires-Dist: opencv-python
+Requires-Dist: convert_stream>=1.4
 Requires-Dist: soup_files
 Dynamic: requires
 

ocr_stream-1.4/ocr_stream.egg-info/requires.txt

@@ -0,0 +1,9 @@
+PyMuPDF
+reportlab
+openpyxl
+pandas
+Pillow>=11.3.0
+requests>=2.32.4
+opencv-python
+convert_stream>=1.4
+soup_files

{ocr_stream-1.2 → ocr_stream-1.4}/pyproject.toml

@@ -4,23 +4,24 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ocr_stream"
-version = "1.2" 
+version = "1.4"
 authors = [
   { name="Bruno Chaves"},
 ]
+
 description = "Extrair textos de documentos digitalizados e imagens."
 readme = "README.md"
 requires-python = ">=3.11"
 
 dependencies = [
-    "PyPDF2==3.0.1",
-    "PyMuPDF==1.25.5",
-    "reportlab==4.3.1",
-    "openpyxl==3.1.5",
-    "pandas==2.2.3",
-    "Pillow==11.2.1",
-    "opencv-python==4.11.0.86",
-    "convert_stream",
+    "PyMuPDF",
+    "reportlab",
+    "openpyxl",
+    "pandas",
+    "Pillow>=11.3.0",
+    "requests>=2.32.4",
+    "opencv-python",
+    "convert_stream>=1.4",
     "soup_files",
 ]
 

ocr_stream-1.2/ocr_stream/_version.py

@@ -1,5 +0,0 @@
-#!/usr/bin/env python3
-#
-
-__version__ = '1.2'
-__modify_data__ = '2025-06-22'

ocr_stream-1.2/ocr_stream/extractors.py

@@ -1,108 +0,0 @@
-#!/usr/bin/env python3
-from __future__ import annotations
-from ocr_stream.utils import (
-    File, Directory, ImageObject, 
-    DataFrame, PageDocumentPdf, 
-    DocumentPdf, PdfStream, LibraryPDF,    
-)
-from ocr_stream.models import ABCRecognizeImage, LibraryOCR
-from ocr_stream.modules_ocr import ModuleOcr, TextRecognized
-
-
-class RecognizeImage(ABCRecognizeImage):
-    def __init__(self, module_ocr):
-        super().__init__(module_ocr)
-
-    def image_content_data(self, img: ImageObject | File) -> DataFrame:
-        if isinstance(img, File):
-            return self.module_ocr.image_content_data(ImageObject.create_from_file(img))
-        elif isinstance(img, ImageObject):
-            return self.module_ocr.image_content_data(img)
-        else:
-            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
-    
-    def image_recognize(self, img: ImageObject | File) -> TextRecognized:
-        if isinstance(img, File):
-            return self.module_ocr.image_recognize(ImageObject.create_from_file(img))
-        elif isinstance(img, ImageObject):
-            return self.module_ocr.image_recognize(img)
-        else:
-            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
-    
-    def image_to_string(self, img: ImageObject | File) -> str:
-        if isinstance(img, File):
-            return self.module_ocr.imag_to_string(ImageObject.create_from_file(img))
-        elif isinstance(img, ImageObject):
-            return self.module_ocr.imag_to_string(img)
-        else:
-            raise ValueError(f'{__class__.__name__}\nUse: ImageObject() ou File(), não {type(img)}')
-    
-    @classmethod
-    def create(
-                cls, 
-                library_ocr = LibraryOCR.PYTESSERACT, 
-                *, 
-                path_tesseract, 
-                lang=None, 
-                tess_data_dir=None
-            ) -> RecognizeImage:
-        #
-        module_ocr: ModuleOcr = ModuleOcr.create(
-            library_ocr,
-            path_tesseract=path_tesseract,
-            lang=lang,
-            tess_data_dir=tess_data_dir
-        )
-        return cls(module_ocr)
-  
-  
-class RecognizePdf(object):
-    def __init__(self, recognize_image: RecognizeImage):
-        self.recognize_image: RecognizeImage = recognize_image
-
-    def recognize_page_pdf(self, page:PageDocumentPdf) -> PageDocumentPdf:
-        """
-            Converte a página em Imagem, reconhece o texto com OCR e
-        retorna uma nova página com o texto reconhecido.
-        """
-        pdf_stream = PdfStream(library_pdf=LibraryPDF.FITZ)
-        pdf_stream.add_page(page)
-        img: ImageObject = pdf_stream.to_images()[0]
-        text_recognized:TextRecognized = self.recognize_image.image_recognize(img)
-        return text_recognized.to_page_pdf()
-    
-    def recognize_document(self, doc:DocumentPdf) -> DocumentPdf:
-        new: DocumentPdf = DocumentPdf()
-        for page in doc.pages:
-            recognized_page: PageDocumentPdf = self.recognize_page_pdf(page)
-            new.add_page(recognized_page)
-        return new
-
-    @classmethod
-    def create(
-                cls, 
-                library_ocr = LibraryOCR.PYTESSERACT, 
-                *,
-                path_tesseract: File, 
-                lang:str=None,
-                tess_data_dir: Directory=None,
-            ) -> RecognizePdf:
-        #
-        #
-        recognize_image = RecognizeImage.create(
-                    library_ocr, 
-                    path_tesseract=path_tesseract, 
-                    lang=lang, 
-                    tess_data_dir=tess_data_dir
-                )
-        return cls(recognize_image)
-
-   
-
-            
-        
-        
-        
-        
-        
-

ocr_stream-1.2/ocr_stream/models/m_extractors.py

@@ -1,39 +0,0 @@
-#!/usr/bin/env python3
-#
-from __future__ import annotations
-from abc import ABC, abstractmethod
-
-from pandas import DataFrame
-from soup_files import File, Directory
-from convert_stream import ImageObject
-from ocr_stream.models.m_modules_ocr import ABCModuleOcr, ABCTextRecognized, LibraryOCR
-
-class ABCRecognizeImage(ABC):
-    def __init__(self, module_ocr:ABCModuleOcr):
-        super().__init__()
-        self.module_ocr:ABCModuleOcr = module_ocr
-        
-    @abstractmethod
-    def image_to_string(self, img:ImageObject | File) -> str:
-        pass
-    
-    @abstractmethod
-    def image_recognize(self, img:ImageObject | File) -> ABCTextRecognized:
-        pass
-        
-    @abstractmethod
-    def image_content_data(self, img:ImageObject | File) -> DataFrame:
-        pass
-        
-    @classmethod
-    def create(
-                cls, 
-                library_ocr:LibraryOCR = LibraryOCR.PYTESSERACT,
-                *,
-                path_tesseract:File, 
-                lang:str=None, 
-                tess_data_dir:Directory=None
-            ) -> ABCRecognizeImage:
-        pass
-
-    

ocr_stream-1.2/ocr_stream/utils/__init__.py

@@ -1,13 +0,0 @@
-#!/usr/bin/env python3
-#
-from pandas import DataFrame
-from soup_files import (
-    File, Directory, UserAppDir, UserFileSystem, InputFiles,
-    JsonConvert, JsonData
-)
-
-from convert_stream import (
-    LibraryDates, LibraryImage, LibraryPDF, ImageObject,
-    PdfStream, ImageStream,PageDocumentPdf, DocumentPdf, 
-    ArrayString, DataString, ConvertDate, print_line, print_title
-)

ocr_stream-1.2/ocr_stream.egg-info/requires.txt

@@ -1,9 +0,0 @@
-PyPDF2==3.0.1
-PyMuPDF==1.25.5
-reportlab==4.3.1
-openpyxl==3.1.5
-pandas==2.2.3
-Pillow==11.2.1
-opencv-python==4.11.0.86
-convert_stream
-soup_files