Add support for Tesseract version 3.05.00

This is a bit more involved, because Tesseract 3.05.00 comes not only
with improvements but also with a few quirks we need to deal with.

The first quirk is that the order arguments of the `tesseract' command
now matters and the list of configurations has to be at the end of the
command line. So we add a new attribute tesseract_flags to the
BaseBuilder class that contains a list of all the flags to pass to
`tesseract', the tesseract_configs attribute however remains pretty much
the same but now only really contains a list of configs instead of being
mixed with flag arguments.

Another quirk has to do with Leptonica >= 1.74 which Tesseract 3.05.00
now requires. Leptonica has special handling of files that reside in
/tmp and assumes that it's an internal temporary file of Leptonica. In
order to deal with it, we now run Tesseract in a temporary directory,
which contains the input/output files and use the relative name of these
files because Leptonica only searches for path names beginning with
/tmp.

Fortunately the last item we need to address is not really a quirk, but
an API change. In Tesseract 3.05.00 there is now a new function called
TessBaseAPIDetectOrientationScript(), which doesn't fill the OSResults
object anymore but now allows to pass the values we're interested in
directly by reference. We need to use this new function because the old
function TessBaseAPIDetectOS() now *always* returns false.

Ran the test suite successfully with Python 3.5 and both Tesseract
3.04.01 and 3.05.00 except the following tests, which also didn't
succeed prior to this commit:

 * cuneiform:TestTxt.test_basic
 * cuneiform:TestTxt.test_european
 * cuneiform:TestTxt.test_french
 * cuneiform:TestWordBox.test_basic
 * cuneiform:TestWordBox.test_european
 * cuneiform:TestWordBox.test_french
 * libtesseract:TestBasicDoc.test_basic
 * libtesseract:TestDigitLineBox.test_digits
 * libtesseract:TestLineBox.test_japanese
 * libtesseract:TestTxt.test_japanese
 * libtesseract:TestWordBox.test_japanese
 * tesseract:TestDigitLineBox.test_digits
 * tesseract:TestTxt.test_japanese

The failure of these test cases is probably related to issue openpaperwork#52, but
from looking at the failures it doesn't seem to be related to this
change anyway.

Signed-off-by: aszlig <[email protected]>

Author: aszlig

src/pyocr/builders.py

@@ -240,8 +240,10 @@ class BaseBuilder(object):
         cuneiform_args : Arguments passed to the Cuneiform command line.
     """
 
-    def __init__(self, file_extensions, tesseract_configs, cuneiform_args):
+    def __init__(self, file_extensions, tesseract_flags, tesseract_configs,
+                 cuneiform_args):
         self.file_extensions = file_extensions
+        self.tesseract_flags = tesseract_flags
         self.tesseract_configs = tesseract_configs
         self.cuneiform_args = cuneiform_args
 
@@ -298,15 +300,15 @@ class TextBuilder(BaseBuilder):
     def __init__(self, tesseract_layout=3, cuneiform_dotmatrix=False,
                  cuneiform_fax=False, cuneiform_singlecolumn=False):
         file_ext = ["txt"]
-        tess_conf = ["-psm", str(tesseract_layout)]
+        tess_flags = ["-psm", str(tesseract_layout)]
         cun_args = ["-f", "text"]
         # Add custom cuneiform parameters if needed
         for par, arg in [(cuneiform_dotmatrix, "--dotmatrix"),
                          (cuneiform_fax, "--fax"),
                          (cuneiform_singlecolumn, "--singlecolumn")]:
             if par:
                 cun_args.append(arg)
-        super(TextBuilder, self).__init__(file_ext, tess_conf, cun_args)
+        super(TextBuilder, self).__init__(file_ext, tess_flags, [], cun_args)
         self.tesseract_layout = tesseract_layout
         self.built_text = []
 
@@ -540,9 +542,11 @@ class WordBoxBuilder(BaseBuilder):
 
     def __init__(self, tesseract_layout=1):
         file_ext = ["html", "hocr"]
-        tess_conf = ["hocr", "-psm", str(tesseract_layout)]
+        tess_flags = ["-psm", str(tesseract_layout)]
+        tess_conf = ["hocr"]
         cun_args = ["-f", "hocr"]
-        super(WordBoxBuilder, self).__init__(file_ext, tess_conf, cun_args)
+        super(WordBoxBuilder, self).__init__(file_ext, tess_flags, tess_conf,
+                                             cun_args)
         self.word_boxes = []
         self.tesseract_layout = tesseract_layout
 
@@ -614,9 +618,11 @@ class LineBoxBuilder(BaseBuilder):
 
     def __init__(self, tesseract_layout=1):
         file_ext = ["html", "hocr"]
-        tess_conf = ["hocr", "-psm", str(tesseract_layout)]
+        tess_flags = ["-psm", str(tesseract_layout)]
+        tess_conf = ["hocr"]
         cun_args = ["-f", "hocr"]
-        super(LineBoxBuilder, self).__init__(file_ext, tess_conf, cun_args)
+        super(LineBoxBuilder, self).__init__(file_ext, tess_flags, tess_conf,
+                                             cun_args)
         self.lines = []
         self.tesseract_layout = tesseract_layout
 

src/pyocr/libtesseract/tesseract_raw.py

@@ -263,11 +263,22 @@ class OSResults(ctypes.Structure):
     ]
     g_libtesseract.TessDeleteText.restype = None
 
-    g_libtesseract.TessBaseAPIDetectOS.argtypes = [
-        ctypes.c_void_p,  # TessBaseAPI*
-        ctypes.POINTER(OSResults),
-    ]
-    g_libtesseract.TessBaseAPIDetectOS.restype = ctypes.c_bool
+    if hasattr(g_libtesseract, 'TessBaseAPIDetectOrientationScript'):
+        g_libtesseract.TessBaseAPIDetectOrientationScript.argtypes = [
+            ctypes.c_void_p,  # TessBaseAPI*
+            ctypes.POINTER(ctypes.c_int),  # orient_deg
+            ctypes.POINTER(ctypes.c_float),  # orient_conf
+            ctypes.POINTER(ctypes.c_char_p),  # script_name
+            ctypes.POINTER(ctypes.c_float),  # script_conf
+        ]
+        g_libtesseract.TessBaseAPIDetectOrientationScript.restype = \
+            ctypes.c_bool
+    else:
+        g_libtesseract.TessBaseAPIDetectOS.argtypes = [
+            ctypes.c_void_p,  # TessBaseAPI*
+            ctypes.POINTER(OSResults),
+        ]
+        g_libtesseract.TessBaseAPIDetectOS.restype = ctypes.c_bool
 
 
 def init(lang=None):
@@ -526,15 +537,37 @@ def detect_os(handle):
     global g_libtesseract
     assert(g_libtesseract)
 
-    results = OSResults()
-    r = g_libtesseract.TessBaseAPIDetectOS(
-        ctypes.c_void_p(handle),
-        ctypes.pointer(results)
-    )
-    if not r:
-        raise TesseractError("detect_orientation failed",
-                             "TessBaseAPIDetectOS() failed")
-    return {
-        "orientation": results.best_orientation_id,
-        "confidence": results.best_oconfidence,
-    }
+    # Use the new API function if it is available, because since Tesseract
+    # 3.05.00 the old API function _always_ returns False.
+    if hasattr(g_libtesseract, 'TessBaseAPIDetectOrientationScript'):
+        orientation_deg = ctypes.c_int(0)
+        orientation_confidence = ctypes.c_float(0.0)
+
+        r = g_libtesseract.TessBaseAPIDetectOrientationScript(
+            ctypes.c_void_p(handle),
+            ctypes.byref(orientation_deg),
+            ctypes.byref(orientation_confidence),
+            None,  # script_name
+            None  # script_confidence
+        )
+
+        if not r:
+            raise TesseractError("detect_orientation failed",
+                                 "TessBaseAPIDetectOrientationScript() failed")
+        return {
+            "orientation": round(orientation_deg.value / 90),
+            "confidence": orientation_confidence.value,
+        }
+    else:  # old API (before Tesseract 3.05.00)
+        results = OSResults()
+        r = g_libtesseract.TessBaseAPIDetectOS(
+            ctypes.c_void_p(handle),
+            ctypes.pointer(results)
+        )
+        if not r:
+            raise TesseractError("detect_orientation failed",
+                                 "TessBaseAPIDetectOS() failed")
+        return {
+            "orientation": results.best_orientation_id,
+            "confidence": results.best_oconfidence,
+        }

src/pyocr/tesseract.py

@@ -22,6 +22,8 @@
 import subprocess
 import sys
 import tempfile
+import contextlib
+import shutil
 
 from . import builders
 from . import error
@@ -62,9 +64,11 @@ class CharBoxBuilder(builders.BaseBuilder):
 
     def __init__(self):
         file_ext = ["box"]
+        tess_flags = []
         tess_conf = ["batch.nochop", "makebox"]
         cun_args = []
-        super(CharBoxBuilder, self).__init__(file_ext, tess_conf, cun_args)
+        super(CharBoxBuilder, self).__init__(file_ext, tess_flags, tess_conf,
+                                             cun_args)
         self.tesseract_layout = 1
 
     @staticmethod
@@ -173,18 +177,19 @@ def detect_orientation(image, lang=None):
         TesseractError --- if no script detected on the image
     """
     _set_environment()
-    with temp_file(".bmp") as input_file:
-        command = [TESSERACT_CMD, input_file.name, 'stdout', "-psm", "0"]
+    with temp_dir() as tmpdir:
+        command = [TESSERACT_CMD, "input.bmp", 'stdout', "-psm", "0"]
         if lang is not None:
             command += ['-l', lang]
 
         if image.mode != "RGB":
             image = image.convert("RGB")
-        image.save(input_file.name)
+        image.save(os.path.join(tmpdir, "input.bmp"))
 
         proc = subprocess.Popen(command, stdin=subprocess.PIPE, shell=False,
                                 startupinfo=g_subprocess_startup_info,
                                 creationflags=g_creation_flags,
+                                cwd=tmpdir,
                                 stdout=subprocess.PIPE,
                                 stderr=subprocess.STDOUT)
         proc.stdin.close()
@@ -224,8 +229,8 @@ def get_available_builders():
     ]
 
 
-def run_tesseract(input_filename, output_filename_base, lang=None,
-                  configs=None):
+def run_tesseract(input_filename, output_filename_base, cwd=None, lang=None,
+                  flags=None, configs=None):
     '''
     Runs Tesseract:
         `TESSERACT_CMD` \
@@ -238,6 +243,8 @@ def run_tesseract(input_filename, output_filename_base, lang=None,
         input_filename --- image to read
         output_filename_base --- file name in which must be stored the result
             (without the extension)
+        cwd --- Run Tesseract in the specified working directory or use current
+            one if None
         lang --- Tesseract language to use (if None, none will be specified)
         config --- List of Tesseract configs to use (if None, none will be
             specified)
@@ -252,10 +259,13 @@ def run_tesseract(input_filename, output_filename_base, lang=None,
     if lang is not None:
         command += ['-l', lang]
 
+    if flags is not None:
+        command += flags
+
     if configs is not None:
         command += configs
 
-    proc = subprocess.Popen(command,
+    proc = subprocess.Popen(command, cwd=cwd,
                             startupinfo=g_subprocess_startup_info,
                             creationflags=g_creation_flags,
                             stdout=subprocess.PIPE,
@@ -301,11 +311,18 @@ def close(self):
             self.name = None
 
 
-def temp_file(suffix):
-    ''' Returns a temporary file '''
-    if os.name == 'nt':  # Windows
-        return ReOpenableTempfile(suffix)
-    return tempfile.NamedTemporaryFile(prefix='tess_', suffix=suffix)
+@contextlib.contextmanager
+def temp_dir():
+    """
+    A context manager for maintaining a temporary directory
+    """
+    # NOTE: Drop this as soon as we don't support Python 2.7 anymore, because
+    # since Python 3.2 there is a context manager called TemporaryDirectory().
+    path = tempfile.mkdtemp(prefix='tess_')
+    try:
+        yield path
+    finally:
+        shutil.rmtree(path)
 
 
 def image_to_string(image, lang=None, builder=None):
@@ -329,23 +346,20 @@ def image_to_string(image, lang=None, builder=None):
 
     if builder is None:
         builder = builders.TextBuilder()
-    with temp_file(".bmp") as input_file:
-        with temp_file('') as output_file:
-            output_file_name_base = output_file.name
-
+    with temp_dir() as tmpdir:
         if image.mode != "RGB":
             image = image.convert("RGB")
-        image.save(input_file.name)
-        (status, errors) = run_tesseract(input_file.name,
-                                         output_file_name_base,
+        image.save(os.path.join(tmpdir, "input.bmp"))
+        (status, errors) = run_tesseract("input.bmp", "output", cwd=tmpdir,
                                          lang=lang,
+                                         flags=builder.tesseract_flags,
                                          configs=builder.tesseract_configs)
         if status:
             raise TesseractError(status, errors)
 
         output_file_name = "ERROR"
         for file_extension in builder.file_extensions:
-            output_file_name = ('%s.%s' % (output_file_name_base,
+            output_file_name = ('%s.%s' % (os.path.join(tmpdir, "output"),
                                            file_extension))
             if not os.access(output_file_name, os.F_OK):
                 continue

tests/tests_libtesseract.py

@@ -33,8 +33,9 @@ def test_version(self):
             (3, 3, 0),
             (3, 4, 0),
             (3, 4, 1),
+            (3, 5, 0),
         ), ("Tesseract does not have the expected version"
-            " (3.4.0) ! Some tests will be skipped !"))
+            " (3.5.0) ! Some tests will be skipped !"))
 
     def test_langs(self):
         langs = libtesseract.get_available_languages()

tests/tests_tesseract.py

@@ -27,8 +27,9 @@ def test_version(self):
             (3, 3, 0),
             (3, 4, 0),
             (3, 4, 1),
+            (3, 5, 0),
         ), ("Tesseract does not have the expected version"
-            " (3.4.0) ! Some tests will be skipped !"))
+            " (3.5.0) ! Some tests will be skipped !"))
 
     def test_langs(self):
         langs = tesseract.get_available_languages()