Cleanup the exception mess:

- All tools raise exceptions inheriting from pyocr.PyocrException
- There is now one and only one TesseractError

Signed-off-by: Jerome Flesch <jflesch@openpaper.work>

README.markdown

@@ -113,18 +113,28 @@ the tool used.
 Argument 'builder' is optional. Default value is
 builders.TextBuilder().
 
+If the OCR fails, an exception ```pyocr.PyocrException```
+will be raised.
+
+An exception MAY be raised if the input image contains no
+text at all (depends on the OCR tool behavior).
+
 
 ### Orientation detection
 
 Currently only available with Tesseract or Libtesseract.
 
 ```Python
 if tool.can_detect_orientation():
-    orientation = tool.detect_orientation(
-        Image.open('test.png'),
-        lang='fra'
-    )
-    pprint("Orientation: {}".format(orientation))
+    try:
+        orientation = tool.detect_orientation(
+            Image.open('test.png'),
+            lang='fra'
+        )
+    except pyocr.PyocrException as exc:
+        print("Orientation detection failed: {}".format(exc))
+        return
+    print("Orientation: {}".format(orientation))
 # Ex: Orientation: {
 #   'angle': 90,
 #   'confidence': 123.4,

src/pyocr/__init__.py

@@ -1 +1,9 @@
 from .pyocr import *
+from .error import PyocrException
+
+__all__ = [
+    'get_available_tools',
+    'PyocrException',
+    'TOOLS',
+    'VERSION',
+]

src/pyocr/cuneiform.py

@@ -23,6 +23,7 @@
 import tempfile
 
 from . import builders
+from . import error
 from . import util
 
 
@@ -65,9 +66,9 @@ def get_available_builders():
     ]
 
 
-class CuneiformError(Exception):
+class CuneiformError(error.PyocrException):
     def __init__(self, status, message):
-        Exception.__init__(self, message)
+        error.PyocrException.__init__(self, message)
         self.status = status
         self.message = message
         self.args = (status, message)

src/pyocr/error.py

@@ -0,0 +1,13 @@
+class PyocrException(Exception):
+    pass
+
+
+class TesseractError(PyocrException):
+    """
+    Obsolete. You should look for PyocrException
+    """
+    def __init__(self, status, message):
+        PyocrException.__init__(self, message)
+        self.status = status
+        self.message = message
+        self.args = (status, message)

src/pyocr/libtesseract/__init__.py

@@ -16,6 +16,8 @@
 '''
 from .. import builders
 
+from ..error import TesseractError
+
 from . import tesseract_raw
 
 
@@ -45,7 +47,7 @@ def detect_orientation(image, lang=None):
         tesseract_raw.set_image(handle, image)
         os = tesseract_raw.detect_os(handle)
         if os['confidence'] <= 0:
-            raise tesseract_raw.TesseractError(
+            raise TesseractError(
                 "no script", "no script detected"
             )
         orientation = {
@@ -171,7 +173,7 @@ def get_available_languages():
 def get_version():
     version = tesseract_raw.get_version()
     version = version.split(" ", 1)[0]
-    
+
     # cut off "dev" string if exists for proper int conversion
     index = version.find("dev")
     if index != -1:

src/pyocr/libtesseract/tesseract_raw.py

@@ -3,6 +3,8 @@
 import os
 import sys
 
+from ..error import TesseractError
+
 
 logger = logging.getLogger(__name__)
 
@@ -46,17 +48,6 @@
         pass
 
 
-class TesseractError(Exception):
-    """
-    Exception raised when Tesseract fails.
-    """
-    def __init__(self, status, message):
-        Exception.__init__(self, message)
-        self.status = status
-        self.message = message
-        self.args = (status, message)
-
-
 class PageSegMode(object):
     OSD_ONLY = 0
     AUTO_OSD = 1

src/pyocr/tesseract.py

@@ -24,8 +24,10 @@
 import tempfile
 
 from . import builders
+from . import error
 from . import util
 from .builders import DigitBuilder  # backward compatibility
+from .error import TesseractError  # backward compatibility
 
 # CHANGE THIS IF TESSERACT IS NOT IN YOUR PATH, OR IS NAMED DIFFERENTLY
 TESSERACT_CMD = 'tesseract.exe' if os.name == 'nt' else 'tesseract'
@@ -306,17 +308,6 @@ def temp_file(suffix):
     return tempfile.NamedTemporaryFile(prefix='tess_', suffix=suffix)
 
 
-class TesseractError(Exception):
-    """
-    Exception raised when Tesseract fails.
-    """
-    def __init__(self, status, message):
-        Exception.__init__(self, message)
-        self.status = status
-        self.message = message
-        self.args = (status, message)
-
-
 def image_to_string(image, lang=None, builder=None):
     '''
     Runs tesseract on the specified image. First, the image is written to disk,