ocr-stream 2.2.8__tar.gz → 2.3__tar.gz

{ocr_stream-2.2.8 → ocr_stream-2.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 2.2.8
+Version: 2.3
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires-Python: >=3.11
@@ -11,8 +11,9 @@ Requires-Dist: openpyxl
 Requires-Dist: pandas
 Requires-Dist: requests>=2.32.4
 Requires-Dist: opencv-python
-Requires-Dist: convert_stream>=2.2.7
-Requires-Dist: soup_files
+Requires-Dist: pytesseract
+Requires-Dist: convert_stream>=2.3
+Requires-Dist: soup_files>=1.4
 
 # ocrlib
 

ocr_stream-2.3/ocr_stream/__init__.py

@@ -0,0 +1,51 @@
+#!/usr/bin/env python3
+
+from ocr_stream.modules import *
+from ocr_stream.bin_tess import BinTesseract, get_path_tesseract_sys
+from ocr_stream.ocr import TesseractOcr, TextRecognized, DEFAULT_LIB_OCR
+from ocr_stream.recognize import RecognizeImage, RecognizePdf
+from soup_files import File, Directory, InputFiles, LibraryDocs, CreatePbar, ProgressBarAdapter
+from convert_stream import ImageObject, PdfStream, DocumentPdf, ConvertPdfToImages
+
+
+def read_images(
+            images: list[ImageObject], *,
+            pbar: ProgressBarAdapter = CreatePbar().get()
+        ) -> DocumentPdf:
+    pbar.start()
+    print()
+    max_num = len(images)
+    pages = []
+    rec = RecognizeImage()
+    for idx, image in enumerate(images):
+        pbar.update(
+            ((idx + 1) / max_num) * 100,
+            f'[OCR] {idx + 1} / {max_num}',
+        )
+        out = rec.image_recognize(image).to_page_pdf()
+        pages.append(out)
+    tmp_doc = DocumentPdf.create_from_pages(pages)
+    print()
+    pbar.stop()
+    return tmp_doc
+
+
+def read_document(
+            document: DocumentPdf, *,
+            dpi: int = 300,
+            pbar: ProgressBarAdapter = CreatePbar().get()
+        ) -> DocumentPdf:
+    pbar.start()
+    print()
+    convert = ConvertPdfToImages.create(document)
+    convert.set_pbar(pbar)
+    images = convert.to_images(dpi=dpi)
+    print()
+    return read_images(images, pbar=pbar)
+
+
+
+
+
+
+

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream/bin_tess/__init__.py

@@ -15,6 +15,8 @@ def __get_path_tesseract_unix() -> File | None:
 def __get_path_tesseract_windows() -> File | None:
     out = which('tesseract.exe')
     if out is None:
+        if os.path.isfile("C:\\Program Files (x86)\\Tesseract-OCR\\tesseract.exe"):
+            return File("C:\\Program Files (x86)\\Tesseract-OCR\\tesseract.exe")
         return None
     return File(out)
 

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream/modules/__init__.py

@@ -26,8 +26,8 @@ try:
     import pyocr.tesseract
     MOD_PYOCR = True
 except:
-    pyocr = None
-    pyocr.tesseract = None
+    pyocr = object
+    #pyocr.tesseract = object
 
 if MOD_PYTESSERACT and MOD_PYOCR:
     ModuleOcr: TypeAlias = Union[pyocr, pytesseract]

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream/ocr/__init__.py

@@ -49,7 +49,7 @@ class TextRecognized(object):
         return doc.to_pages()[0]
 
     def to_document(self) -> DocumentPdf:
-        return DocumentPdf.create_from_bytes(BytesIO(self.bytes_recognized))
+        return DocumentPdf(BytesIO(self.bytes_recognized))
 
     def to_map(self, separator: str = '\n') -> MapText:
         text = self.to_string()

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream/recognize.py

@@ -59,7 +59,11 @@ class RecognizeImage(object):
         """
         return self.image_recognize(img).to_dataframe()
 
-    def image_recognize(self, img: ImageObject | File) -> TextRecognized:
+    def image_recognize(self, img: ImageObject | File, *, land_scape: bool = False) -> TextRecognized:
+        if isinstance(img, File):
+            img = ImageObject(img)
+        if land_scape:
+            img.set_landscape()
         rec: TextRecognized = self.module_ocr.to_reconize(img)
         return rec
 
@@ -80,7 +84,7 @@ class RecognizePdf(object):
     def set_pbar(self, p: ProgressBarAdapter):
         self.pbar = p
 
-    def recognize_page_pdf(self, page: PageDocumentPdf, *, dpi: int = 350) -> DocumentPdf:
+    def recognize_page_pdf(self, page: PageDocumentPdf, *, dpi: int = 300) -> DocumentPdf:
         """
             Converte a página em Imagem, reconhece o texto com OCR e
         retorna uma nova página com o texto reconhecido.
@@ -94,7 +98,7 @@ class RecognizePdf(object):
         del img
         return text_recognized.to_document()
 
-    def recognize_document(self, doc: DocumentPdf, *, dpi: int = 350) -> DocumentPdf:
+    def recognize_document(self, doc: DocumentPdf, *, dpi: int = 300) -> DocumentPdf:
         _pages: list[PageDocumentPdf] = []
         _pages.clear()
         _conv = ConvertPdfToImages.create(doc)

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ocr_stream
-Version: 2.2.8
+Version: 2.3
 Summary: Extrair textos de documentos digitalizados e imagens.
 Author: Bruno Chaves
 Requires-Python: >=3.11
@@ -11,8 +11,9 @@ Requires-Dist: openpyxl
 Requires-Dist: pandas
 Requires-Dist: requests>=2.32.4
 Requires-Dist: opencv-python
-Requires-Dist: convert_stream>=2.2.7
-Requires-Dist: soup_files
+Requires-Dist: pytesseract
+Requires-Dist: convert_stream>=2.3
+Requires-Dist: soup_files>=1.4
 
 # ocrlib
 

{ocr_stream-2.2.8 → ocr_stream-2.3}/ocr_stream.egg-info/requires.txt

@@ -4,5 +4,6 @@ openpyxl
 pandas
 requests>=2.32.4
 opencv-python
-convert_stream>=2.2.7
-soup_files
+pytesseract
+convert_stream>=2.3
+soup_files>=1.4

{ocr_stream-2.2.8 → ocr_stream-2.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ocr_stream"
-version = "2.2.8"
+version = "2.3"
 authors = [
   { name="Bruno Chaves"},
 ]
@@ -20,8 +20,9 @@ dependencies = [
     "pandas",
     "requests>=2.32.4",
     "opencv-python",
-    "convert_stream>=2.2.7",
-    "soup_files",
+    "pytesseract",
+    "convert_stream>=2.3",
+    "soup_files>=1.4",
 ]
 
 [tool.setuptools]

ocr_stream-2.2.8/ocr_stream/__init__.py

@@ -1,5 +0,0 @@
-#!/usr/bin/env python3
-from ocr_stream.modules import *
-from ocr_stream.bin_tess import BinTesseract, get_path_tesseract_sys
-from ocr_stream.ocr import TesseractOcr, TextRecognized, DEFAULT_LIB_OCR
-from ocr_stream.recognize import RecognizeImage, RecognizePdf