ocr-stream 2.2__tar.gz → 2.2.7__tar.gz

{ocr_stream-2.2 → ocr_stream-2.2.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 2.2
+Version: 2.2.7
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires-Python: >=3.11
@@ -11,7 +11,7 @@ Requires-Dist: openpyxl
 Requires-Dist: pandas
 Requires-Dist: requests>=2.32.4
 Requires-Dist: opencv-python
-Requires-Dist: convert_stream>=2.2.4
+Requires-Dist: convert_stream>=2.2.7
 Requires-Dist: soup_files
 
 # ocrlib

{ocr_stream-2.2 → ocr_stream-2.2.7}/ocr_stream/ocr/__init__.py

@@ -150,7 +150,6 @@ class IPytesseract(ABCOcrTesseract):
 # Modulo OCR pyocr
 # ======================================================================#
 
-
 class IPyOcr(ABCOcrTesseract):
 
     def __init__(self, bin_tess: BinTesseract, *, lang: str = None, tess_data_dir: str = None):

{ocr_stream-2.2 → ocr_stream-2.2.7}/ocr_stream/recognize.py

@@ -13,7 +13,11 @@ class RecognizeImage(object):
         Reconhecer textos em imagens com pytesseract|pyocr usando o tesseract
     """
 
-    def __init__(self, bin_tess: BinTesseract, *, lib_ocr: LibOcr = DEFAULT_LIB_OCR):
+    def __init__(
+                self,
+                bin_tess: BinTesseract = BinTesseract(), *,
+                lib_ocr: LibOcr = DEFAULT_LIB_OCR
+            ):
         super().__init__()
         self.module_ocr: TesseractOcr = TesseractOcr.create(
             bin_tess,
@@ -65,28 +69,36 @@ class RecognizeImage(object):
 
 class RecognizePdf(object):
 
-    def __init__(self, bin_tess: BinTesseract, *, lib_ocr: LibOcr = DEFAULT_LIB_OCR):
+    def __init__(
+                self,
+                bin_tess: BinTesseract = BinTesseract(), *,
+                lib_ocr: LibOcr = DEFAULT_LIB_OCR,
+            ):
         self.recognize_image: RecognizeImage = RecognizeImage(bin_tess, lib_ocr=lib_ocr)
         self.pbar: ProgressBarAdapter = CreatePbar().get()
 
     def set_pbar(self, p: ProgressBarAdapter):
         self.pbar = p
 
-    def recognize_page_pdf(self, page: PageDocumentPdf) -> DocumentPdf:
+    def recognize_page_pdf(self, page: PageDocumentPdf, *, dpi: int = 350) -> DocumentPdf:
         """
             Converte a página em Imagem, reconhece o texto com OCR e
         retorna uma nova página com o texto reconhecido.
         """
         _doc = DocumentPdf.create_from_pages([page])
         _conv = ConvertPdfToImages.create(_doc)
-        img: ImageObject = _conv.to_images()[0]
+        img: ImageObject = _conv.to_images(dpi=dpi)[0]
         text_recognized: TextRecognized = self.recognize_image.image_recognize(img)
+        del _doc
+        del _conv
+        del img
         return text_recognized.to_document()
 
-    def recognize_document(self, doc: DocumentPdf) -> DocumentPdf:
-        _stream = PdfStream()
+    def recognize_document(self, doc: DocumentPdf, *, dpi: int = 350) -> DocumentPdf:
+        _pages: list[PageDocumentPdf] = []
+        _pages.clear()
         _conv = ConvertPdfToImages.create(doc)
-        images: list[ImageObject] = _conv.to_images()
+        images: list[ImageObject] = _conv.to_images(dpi=dpi)
         max_num: int = len(images)
         for num, img in enumerate(images):
             self.pbar.update(
@@ -94,12 +106,15 @@ class RecognizePdf(object):
                 f'Aplicando OCR [{num+1} de {max_num}]'
             )
             text_recognized = self.recognize_image.image_recognize(img)
-            _stream.add_page(text_recognized.to_page_pdf())
+            _pages.append(text_recognized.to_page_pdf())
         print()
         self.pbar.stop()
-        return _stream.to_document()
-
-    def recognize_file_pdf(self, file: File) -> DocumentPdf:
-        _doc = DocumentPdf.create_from_file(file)
-        return self.recognize_document(_doc)
+        new_doc = DocumentPdf.create_from_pages(_pages)
+        return new_doc
+
+    def recognize_file_pdf(self, file: File, *, dpi: int = 300) -> DocumentPdf:
+        file_doc = DocumentPdf.create_from_file(file)
+        recognized_doc = self.recognize_document(file_doc, dpi=dpi)
+        del file_doc
+        return recognized_doc
 

{ocr_stream-2.2 → ocr_stream-2.2.7}/ocr_stream.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 2.2
+Version: 2.2.7
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires-Python: >=3.11
@@ -11,7 +11,7 @@ Requires-Dist: openpyxl
 Requires-Dist: pandas
 Requires-Dist: requests>=2.32.4
 Requires-Dist: opencv-python
-Requires-Dist: convert_stream>=2.2.4
+Requires-Dist: convert_stream>=2.2.7
 Requires-Dist: soup_files
 
 # ocrlib

{ocr_stream-2.2 → ocr_stream-2.2.7}/ocr_stream.egg-info/requires.txt

@@ -4,5 +4,5 @@ openpyxl
 pandas
 requests>=2.32.4
 opencv-python
-convert_stream>=2.2.4
+convert_stream>=2.2.7
 soup_files

{ocr_stream-2.2 → ocr_stream-2.2.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ocr_stream"
-version = "2.2"
+version = "2.2.7"
 authors = [
   { name="Bruno Chaves"},
 ]
@@ -20,7 +20,7 @@ dependencies = [
     "pandas",
     "requests>=2.32.4",
     "opencv-python",
-    "convert_stream>=2.2.4",
+    "convert_stream>=2.2.7",
     "soup_files",
 ]