deepdoctection 0.30__py3-none-any.whl → 0.31__py3-none-any.whl

deepdoctection/utils/env_info.py

@@ -53,7 +53,7 @@ import re
 import subprocess
 import sys
 from collections import defaultdict
-from typing import List, Optional, Tuple
+from typing import List, Literal, Optional, Tuple
 
 import numpy as np
 from tabulate import tabulate
@@ -420,7 +420,7 @@ def collect_env_info() -> str:
     try:
         import prctl  # type: ignore
 
-        _ = prctl.set_pdeathsig  # noqa
+        _ = prctl.set_pdeathsig  # pylint: disable=E1101
     except ModuleNotFoundError:
         has_prctl = False
     data.append(("python-prctl", str(has_prctl)))
@@ -452,6 +452,20 @@ def collect_env_info() -> str:
     return env_str
 
 
+def set_env(name: str, value: str) -> None:
+    """
+    Set an environment variable if it is not already set.
+
+    :param name: The name of the environment variable
+    :param value: The value of the environment variable
+    """
+
+    if os.environ.get(name):
+        return
+    os.environ[name] = value
+    return
+
+
 def auto_select_lib_and_device() -> None:
     """
     Select the DL library and subsequently the device.
@@ -461,41 +475,60 @@ def auto_select_lib_and_device() -> None:
     is not installed raise ImportError.
     """
 
+    # USE_TF and USE_TORCH are env variables that steer DL library selection for Doctr.
     if tf_available() and tensorpack_available():
         from tensorpack.utils.gpu import get_num_gpu  # pylint: disable=E0401
 
         if get_num_gpu() >= 1:
-            os.environ["USE_TENSORFLOW"] = "True"
-            os.environ["USE_PYTORCH"] = "False"
-            os.environ["USE_CUDA"] = "True"
-            os.environ["USE_MPS"] = "False"
+            set_env("USE_TENSORFLOW", "True")
+            set_env("USE_PYTORCH", "False")
+            set_env("USE_CUDA", "True")
+            set_env("USE_MPS", "False")
+            set_env("USE_TF", "TRUE")
+            set_env("USE_TORCH", "False")
             return
         if pytorch_available():
-            os.environ["USE_TENSORFLOW"] = "False"
-            os.environ["USE_PYTORCH"] = "True"
-            os.environ["USE_CUDA"] = "False"
+            set_env("USE_TENSORFLOW", "False")
+            set_env("USE_PYTORCH", "True")
+            set_env("USE_CUDA", "False")
+            set_env("USE_TF", "False")
+            set_env("USE_TORCH", "TRUE")
             return
         logger.warning(
             LoggingRecord("You have Tensorflow installed but no GPU is available. All Tensorflow models require a GPU.")
         )
+    if tf_available():
+        set_env("USE_TENSORFLOW", "False")
+        set_env("USE_PYTORCH", "False")
+        set_env("USE_CUDA", "False")
+        set_env("USE_TF", "AUTO")
+        set_env("USE_TORCH", "AUTO")
+        return
+
     if pytorch_available():
         import torch
 
         if torch.cuda.is_available():
-            os.environ["USE_TENSORFLOW"] = "False"
-            os.environ["USE_PYTORCH"] = "True"
-            os.environ["USE_CUDA"] = "True"
+            set_env("USE_TENSORFLOW", "False")
+            set_env("USE_PYTORCH", "True")
+            set_env("USE_CUDA", "True")
+            set_env("USE_TF", "False")
+            set_env("USE_TORCH", "TRUE")
             return
         if torch.backends.mps.is_available():
-            os.environ["USE_TENSORFLOW"] = "False"
-            os.environ["USE_PYTORCH"] = "True"
-            os.environ["USE_CUDA"] = "False"
-            os.environ["USE_MPS"] = "True"
+            set_env("USE_TENSORFLOW", "False")
+            set_env("USE_PYTORCH", "True")
+            set_env("USE_CUDA", "False")
+            set_env("USE_MPS", "True")
+            set_env("USE_TF", "False")
+            set_env("USE_TORCH", "TRUE")
             return
-        os.environ["USE_TENSORFLOW"] = "False"
-        os.environ["USE_PYTORCH"] = "True"
-        os.environ["USE_CUDA"] = "False"
-        os.environ["USE_MPS"] = "False"
+        set_env("USE_TENSORFLOW", "False")
+        set_env("USE_PYTORCH", "True")
+        set_env("USE_CUDA", "False")
+        set_env("USE_MPS", "False")
+        set_env("USE_TF", "AUTO")
+        set_env("USE_TORCH", "AUTO")
         return
     logger.warning(
         LoggingRecord(
@@ -505,7 +538,7 @@ def auto_select_lib_and_device() -> None:
     )
 
 
-def get_device(ignore_cpu: bool = True) -> str:
+def get_device(ignore_cpu: bool = True) -> Literal["cuda", "mps", "cpu"]:
     """
     Device checks for running PyTorch with CUDA, MPS or optionall CPU.
     If nothing can be found and if `disable_cpu` is deactivated it will raise a `ValueError`
@@ -520,7 +553,7 @@ def get_device(ignore_cpu: bool = True) -> str:
         return "mps"
     if not ignore_cpu:
         return "cpu"
-    raise ValueError("Could not find either GPU nor MPS")
+    raise RuntimeWarning("Could not find either GPU nor MPS")
 
 
 def auto_select_viz_library() -> None:

deepdoctection/utils/error.py

@@ -0,0 +1,84 @@
+# -*- coding: utf-8 -*-
+# File: error.py
+
+# Copyright 2024 Dr. Janis Meyer. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""
+Module for custom exceptions
+"""
+
+
+class BoundingBoxError(BaseException):
+    """Special exception only for `datapoint.box.BoundingBox`"""
+
+
+class AnnotationError(BaseException):
+    """Special exception only for `datapoint.annotation.Annotation`"""
+
+
+class ImageError(BaseException):
+    """Special exception only for `datapoint.image.Image`"""
+
+
+class UUIDError(BaseException):
+    """Special exception only for `utils.identifier`"""
+
+
+class DependencyError(BaseException):
+    """Special exception only for missing dependencies. We do not use the internals ImportError or
+    ModuleNotFoundError."""
+
+
+class DataFlowTerminatedError(BaseException):
+    """
+    An exception indicating that the DataFlow is unable to produce any more
+    data, i.e. something wrong happened so that calling `__iter__`
+    cannot give a valid iterator anymore.
+    In most DataFlow this will never be raised.
+    """
+
+
+class DataFlowResetStateNotCalledError(BaseException):
+    """
+    An exception indicating that `reset_state()` has not been called before starting
+    iteration.
+    """
+
+    def __init__(self) -> None:
+        super().__init__("Iterating a dataflow requires .reset_state() to be called first")
+
+
+class MalformedData(BaseException):
+    """
+    Exception class for malformed data. Use this class if something does not look right with the data
+    """
+
+
+class FileExtensionError(BaseException):
+    """
+    Exception class for wrong file extensions.
+    """
+
+
+class TesseractError(RuntimeError):
+    """
+    Tesseract Error
+    """
+
+    def __init__(self, status: int, message: str) -> None:
+        super().__init__()
+        self.status = status
+        self.message = message
+        self.args = (status, message)

deepdoctection/utils/file_utils.py

@@ -22,6 +22,7 @@ import importlib_metadata
 from packaging import version
 
 from .detection_types import Requirement
+from .error import DependencyError
 from .logger import LoggingRecord, logger
 from .metacfg import AttrDict
 
@@ -263,7 +264,7 @@ def set_tesseract_path(tesseract_path: str) -> None:
     :param tesseract_path: Tesseract installation path.
     """
     if tesseract_path is None:
-        raise ValueError("tesseract_path is empty.")
+        raise TypeError("tesseract_path cannot be None")
 
     global _TESS_AVAILABLE  # pylint: disable=W0603
     global _TESS_PATH  # pylint: disable=W0603
@@ -288,12 +289,6 @@ def tesseract_available() -> bool:
 # copy paste from https://github.com/madmaze/pytesseract/blob/master/pytesseract/pytesseract.py
 
 
-class TesseractNotFound(BaseException):
-    """
-    Exception class for Tesseract being not found
-    """
-
-
 def get_tesseract_version() -> Union[int, version.Version]:
     """
     Returns Version object of the Tesseract version. We need at least Tesseract 3.05
@@ -306,7 +301,7 @@ def get_tesseract_version() -> Union[int, version.Version]:
             stdin=subprocess.DEVNULL,
         )
     except OSError:
-        raise TesseractNotFound(_TESS_ERR_MSG) from OSError
+        raise DependencyError(_TESS_ERR_MSG) from OSError
 
     raw_version = output.decode("utf-8")
     str_version, *_ = raw_version.lstrip(string.printable[10:]).partition(" ")
@@ -348,12 +343,6 @@ def pdf_to_cairo_available() -> bool:
     return bool(_PDF_TO_CAIRO_AVAILABLE)
 
 
-class PopplerNotFound(BaseException):
-    """
-    Exception class for Poppler being not found
-    """
-
-
 def get_poppler_version() -> Union[int, version.Version]:
     """
     Returns Version object of the Poppler version. We need at least Tesseract 3.05
@@ -371,7 +360,7 @@ def get_poppler_version() -> Union[int, version.Version]:
             [command, "-v"], stderr=subprocess.STDOUT, env=environ, stdin=subprocess.DEVNULL
         )
     except OSError:
-        raise PopplerNotFound() from OSError
+        raise DependencyError(_POPPLER_ERR_MSG) from OSError
 
     raw_version = output.decode("utf-8")
     list_version = raw_version.split("\n", maxsplit=1)[0].split(" ")[-1].split(".")

deepdoctection/utils/fs.py

@@ -34,7 +34,7 @@ from .logger import LoggingRecord, logger
 from .pdf_utils import get_pdf_file_reader, get_pdf_file_writer
 from .settings import CONFIGS, DATASET_DIR, MODEL_DIR, PATH
 from .tqdm import get_tqdm
-from .utils import FileExtensionError, is_file_extension
+from .utils import is_file_extension
 from .viz import viz_handler
 
 __all__ = [
@@ -44,9 +44,7 @@ __all__ = [
     "maybe_path_or_pdf",
     "download",
     "mkdir_p",
-    "is_file_extension",
     "load_json",
-    "FileExtensionError",
     "sub_path",
     "get_package_path",
     "get_configs_dir_path",
@@ -125,8 +123,8 @@ def download(url: str, directory: Pathlike, file_name: Optional[str] = None, exp
     assert size > 0, f"Downloaded an empty file from {url}!"
 
     if expect_size is not None and size != expect_size:
-        logger.error(LoggingRecord(f"File downloaded from {url} does not match the expected size!"))
-        logger.error(
+        logger.warning(LoggingRecord(f"File downloaded from {url} does not match the expected size!"))
+        logger.warning(
             LoggingRecord("You may have downloaded a broken file, or the upstream may have modified the file.")
         )
 
@@ -210,13 +208,15 @@ def get_load_image_func(
     :return: The function loading the file (and converting to its desired format)
     """
 
-    assert is_file_extension(path, [".png", ".jpeg", ".jpg", ".pdf", ".tif"]), f"image type not allowed: {path}"
+    assert is_file_extension(path, [".png", ".jpeg", ".jpg", ".pdf", ".tif"]), f"image type not allowed: " f"{path}"
 
     if is_file_extension(path, [".png", ".jpeg", ".jpg", ".tif"]):
         return load_image_from_file
     if is_file_extension(path, [".pdf"]):
         return load_bytes_from_pdf_file
-    return NotImplemented
+    raise NotImplementedError(
+        "File extension not supported by any loader. Please specify a file type and raise an issue"
+    )
 
 
 def maybe_path_or_pdf(path: Pathlike) -> int:

deepdoctection/utils/pdf_utils.py

@@ -32,9 +32,10 @@ from pypdf import PdfReader, PdfWriter, errors
 
 from .context import save_tmp_file, timeout_manager
 from .detection_types import ImageType, Pathlike
-from .file_utils import PopplerNotFound, pdf_to_cairo_available, pdf_to_ppm_available, qpdf_available
+from .error import DependencyError, FileExtensionError
+from .file_utils import pdf_to_cairo_available, pdf_to_ppm_available, qpdf_available
 from .logger import LoggingRecord, logger
-from .utils import FileExtensionError, is_file_extension
+from .utils import is_file_extension
 from .viz import viz_handler
 
 __all__ = ["decrypt_pdf_document", "get_pdf_file_reader", "get_pdf_file_writer", "PDFStreamer", "pdf_to_np_array"]
@@ -165,7 +166,7 @@ def _input_to_cli_str(
     elif pdf_to_cairo_available():
         command = "pdftocairo"
     else:
-        raise PopplerNotFound("Poppler not found. Please install or add to your PATH.")
+        raise DependencyError("Poppler not found. Please install or add to your PATH.")
 
     if platform.system() == "Windows":
         command = command + ".exe"
@@ -201,7 +202,7 @@ def _run_poppler(poppler_args: List[str]) -> None:
     except OSError as error:
         if error.errno != ENOENT:
             raise error from error
-        raise PopplerNotFound("Poppler not found. Please install or add to your PATH.") from error
+        raise DependencyError("Poppler not found. Please install or add to your PATH.") from error
 
     with timeout_manager(proc, 0):
         if proc.returncode:

deepdoctection/utils/settings.py

@@ -65,6 +65,7 @@ class PageType(ObjectTypes):
 
     document_type = "document_type"
     language = "language"
+    angle = "angle"
 
 
 @object_types_registry.register("SummaryType")
@@ -125,6 +126,7 @@ class LayoutType(ObjectTypes):
     column = "column"
     word = "word"
     line = "line"
+    background = "background"
 
 
 @object_types_registry.register("TableType")
@@ -324,7 +326,9 @@ def token_class_tag_to_token_class_with_tag(token: ObjectTypes, tag: ObjectTypes
     """
     if isinstance(token, TokenClasses) and isinstance(tag, BioTag):
         return _TOKEN_AND_TAG_TO_TOKEN_CLASS_WITH_TAG[(token, tag)]
-    raise TypeError("Token must be of type TokenClasses and tag must be of type BioTag")
+    raise TypeError(
+        f"Token must be of type TokenClasses, is of {type(token)} and tag " f"{type(tag)} must be of type BioTag"
+    )
 
 
 def token_class_with_tag_to_token_class_and_tag(

deepdoctection/utils/transform.py

@@ -47,7 +47,7 @@ class BaseTransform(ABC):
     @abstractmethod
     def apply_image(self, img: ImageType) -> ImageType:
         """The transformation that should be applied to the image"""
-        raise NotImplementedError
+        raise NotImplementedError()
 
 
 class ResizeTransform(BaseTransform):

deepdoctection/utils/utils.py

@@ -144,12 +144,6 @@ def get_rng(obj: Any = None) -> np.random.RandomState:
     return np.random.RandomState(seed)
 
 
-class FileExtensionError(BaseException):
-    """
-    An exception indicating that a file does not seem to have an expected type
-    """
-
-
 def is_file_extension(file_name: Pathlike, extension: Union[str, Sequence[str]]) -> bool:
     """
     Check if a given file name has a given extension

deepdoctection/utils/viz.py

@@ -38,6 +38,7 @@ from numpy import float32, uint8
 
 from .detection_types import ImageType
 from .env_info import auto_select_viz_library
+from .error import DependencyError
 from .file_utils import get_opencv_requirement, get_pillow_requirement, opencv_available, pillow_available
 
 if opencv_available():
@@ -307,6 +308,7 @@ class VizPackageHandler:
             "draw_text": "_cv2_draw_text",
             "interactive_imshow": "_cv2_interactive_imshow",
             "encode": "_cv2_encode",
+            "rotate_image": "_cv2_rotate_image",
         },
         "pillow": {
             "read_image": "_pillow_read_image",
@@ -319,6 +321,7 @@ class VizPackageHandler:
             "draw_text": "_pillow_draw_text",
             "interactive_imshow": "_pillow_interactive_imshow",
             "encode": "_pillow_encode",
+            "rotate_image": "_pillow_rotate_image",
         },
     }
 
@@ -352,12 +355,12 @@ class VizPackageHandler:
         if maybe_cv2:
             requirements = get_opencv_requirement()
             if not requirements[1]:
-                raise ImportError(requirements[2])
+                raise DependencyError(requirements[2])
             return maybe_cv2
 
         requirements = get_pillow_requirement()
         if not requirements[1]:
-            raise ImportError(requirements[2])
+            raise DependencyError(requirements[2])
         return "pillow"
 
     def _set_vars(self, package: str) -> None:
@@ -690,6 +693,45 @@ class VizPackageHandler:
         pil_image = Image.fromarray(np.uint8(np_image[:, :, ::-1]))
         pil_image.show(name)
 
+    def rotate_image(self, np_image: ImageType, angle: int) -> ImageType:
+        """Rotating an image by some angle"""
+        return getattr(self, self.pkg_func_dict["rotate_image"])(np_image, angle)
+
+    @staticmethod
+    def _cv2_rotate_image(np_image: ImageType, angle: float) -> ImageType:
+        # copy & paste from https://stackoverflow.com/questions/43892506
+        # /opencv-python-rotate-image-without-cropping-sides
+
+        height, width = np_image.shape[:2]
+        image_center = (width / 2, height / 2)
+        rotation_mat = cv2.getRotationMatrix2D(center=image_center, angle=angle, scale=1.0)
+
+        # rotation calculates the cos and sin, taking absolutes of those.
+        abs_cos = abs(rotation_mat[0, 0])
+        abs_sin = abs(rotation_mat[0, 1])
+
+        # find the new width and height bounds
+        bound_w = int(height * abs_sin + width * abs_cos)
+        bound_h = int(height * abs_cos + width * abs_sin)
+
+        # subtract old image center (bringing image back to origo) and adding the new image center coordinates
+        rotation_mat[0, 2] += bound_w / 2 - image_center[0]
+        rotation_mat[1, 2] += bound_h / 2 - image_center[1]
+
+        np_image = cv2.warpAffine(  # type: ignore
+            src=np_image,
+            M=rotation_mat,
+            dsize=(bound_w, bound_h),
+        )
+
+        return np_image
+
+    @staticmethod
+    def _pillow_rotate_image(np_image: ImageType, angle: int) -> ImageType:
+        pil_image = Image.fromarray(np.uint8(np_image[:, :, ::-1]))
+        pil_image_rotated = pil_image.rotate(angle, expand=True)
+        return np.array(pil_image_rotated)[:, :, ::-1]
+
 
 auto_select_viz_library()
 viz_handler = VizPackageHandler()