Hello community,
here is the log from the commit of package python-tesserocr for
openSUSE:Factory checked in at 2019-12-03 15:21:47
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-tesserocr (Old)
and /work/SRC/openSUSE:Factory/.python-tesserocr.new.4691 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Package is "python-tesserocr"
Tue Dec 3 15:21:47 2019 rev:8 rq:753070 version:2.5.0
Changes:
--------
--- /work/SRC/openSUSE:Factory/python-tesserocr/python-tesserocr.changes
2019-08-24 18:47:34.657750513 +0200
+++
/work/SRC/openSUSE:Factory/.python-tesserocr.new.4691/python-tesserocr.changes
2019-12-03 15:21:50.130521845 +0100
@@ -1,0 +2,9 @@
+Tue Nov 26 00:49:44 UTC 2019 - Martin Herkt <9+suse@cirno.systems>
+
+- Update to version 2.5.0
+ * Support for RowAttributes method in LTRResultIterator
+ * SetImage: use PNG instead of JPEG fallback
+ * Replace STRING::string() by c_str()
+ * Don't use assignment operator for TessBaseAPI
+
+-------------------------------------------------------------------
Old:
----
tesserocr-2.4.1.tar.gz
New:
----
tesserocr-2.5.0.tar.gz
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Other differences:
------------------
++++++ python-tesserocr.spec ++++++
--- /var/tmp/diff_new_pack.VK66EF/_old 2019-12-03 15:21:50.822521528 +0100
+++ /var/tmp/diff_new_pack.VK66EF/_new 2019-12-03 15:21:50.826521526 +0100
@@ -1,7 +1,7 @@
#
# spec file for package python-tesserocr
#
-# Copyright (c) 2019 SUSE LINUX GmbH, Nuernberg, Germany.
+# Copyright (c) 2019 SUSE LLC
#
# All modifications and additions to the file contributed by third parties
# remain the property of their copyright owners, unless otherwise agreed
@@ -18,7 +18,7 @@
%{?!python_module:%define python_module() python-%{**} python3-%{**}}
Name: python-tesserocr
-Version: 2.4.1
+Version: 2.5.0
Release: 0
Summary: A Python wrapper around tesseract-ocr
License: MIT
++++++ tesserocr-2.4.1.tar.gz -> tesserocr-2.5.0.tar.gz ++++++
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/PKG-INFO new/tesserocr-2.5.0/PKG-INFO
--- old/tesserocr-2.4.1/PKG-INFO 2019-08-23 18:03:12.000000000 +0200
+++ new/tesserocr-2.5.0/PKG-INFO 2019-11-09 00:11:54.000000000 +0100
@@ -1,6 +1,6 @@
Metadata-Version: 2.1
Name: tesserocr
-Version: 2.4.1
+Version: 2.5.0
Summary: A simple, Pillow-friendly, Python wrapper around tesseract-ocr API
using Cython
Home-page: https://github.com/sirfz/tesserocr
Author: Fayez Zouheiry
@@ -88,20 +88,16 @@
The proposed downloads consist of stand-alone packages containing all
the Windows libraries needed for execution. This means that no additional
installation of tesseract is required on your system.
+ The recommended method of installation is via Conda as described below.
+
Conda
`````
- You can use the channel `simonflueckiger
<https://anaconda.org/simonflueckiger/tesserocr>`_ to install from Conda:
-
- ::
-
- > conda install -c simonflueckiger tesserocr
-
- or to get **tesserocr** compiled with **tesseract 4.0.0**:
+ You can use the `conda-forge
<https://anaconda.org/conda-forge/tesserocr>`_ channel to install from Conda:
::
- > conda install -c simonflueckiger/label/tesseract-4.0.0-master
tesserocr
+ > conda install -c conda-forge tesserocr
pip
```
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/README.rst
new/tesserocr-2.5.0/README.rst
--- old/tesserocr-2.4.1/README.rst 2019-08-23 18:01:21.000000000 +0200
+++ new/tesserocr-2.5.0/README.rst 2019-11-09 00:11:39.000000000 +0100
@@ -80,20 +80,16 @@
The proposed downloads consist of stand-alone packages containing all the
Windows libraries needed for execution. This means that no additional
installation of tesseract is required on your system.
+The recommended method of installation is via Conda as described below.
+
Conda
`````
-You can use the channel `simonflueckiger
<https://anaconda.org/simonflueckiger/tesserocr>`_ to install from Conda:
-
-::
-
- > conda install -c simonflueckiger tesserocr
-
-or to get **tesserocr** compiled with **tesseract 4.0.0**:
+You can use the `conda-forge <https://anaconda.org/conda-forge/tesserocr>`_
channel to install from Conda:
::
- > conda install -c simonflueckiger/label/tesseract-4.0.0-master tesserocr
+ > conda install -c conda-forge tesserocr
pip
```
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/tesseract.pxd
new/tesserocr-2.5.0/tesseract.pxd
--- old/tesserocr-2.4.1/tesseract.pxd 2019-08-23 18:01:21.000000000 +0200
+++ new/tesserocr-2.5.0/tesseract.pxd 2019-11-08 23:49:38.000000000 +0100
@@ -110,7 +110,7 @@
cdef extern from "tesseract/strngs.h" nogil:
cdef cppclass STRING:
- cchar_t *string() const
+ cchar_t *c_str() const
STRING &operator=(cchar_t *)
cdef extern from "tesseract/ocrclass.h" nogil:
@@ -149,6 +149,7 @@
void SetLineSeparator(cchar_t *)
void SetParagraphSeparator(cchar_t *)
float Confidence(PageIteratorLevel) const
+ void RowAttributes(float *, float *, float *) const
cchar_t *WordFontAttributes(bool *, bool *, bool *, bool *, bool
*, bool *, int *, int *) const
cchar_t *WordRecognitionLanguage() const
StrongScriptDirection WordDirection() const
@@ -157,6 +158,29 @@
bool WordIsNumeric() const
bool HasBlamerInfo() const
cchar_t *GetBlamerDebug() const
+ cchar_t *GetBlamerMisadaptionDebug() const
+ bool HasTruthString() const
+ bool EquivalentToTruth(cchar_t *) const
+ char *WordTruthUTF8Text() const
+ char *WordNormedUTF8Text() const
+ cchar_t *WordLattice(int *) const
+ bool SymbolIsSuperscript() const
+ bool SymbolIsSubscript() const
+ bool SymbolIsDropcap() const
+ ELIF TESSERACT_VERSION >= 0x3040100:
+ cdef cppclass LTRResultIterator(PageIterator):
+ char *GetUTF8Text(PageIteratorLevel) const
+ void SetLineSeparator(cchar_t *)
+ void SetParagraphSeparator(cchar_t *)
+ float Confidence(PageIteratorLevel) const
+ void RowAttributes(float *, float *, float *) const
+ cchar_t *WordFontAttributes(bool *, bool *, bool *, bool *, bool
*, bool *, int *, int *) const
+ cchar_t *WordRecognitionLanguage() const
+ StrongScriptDirection WordDirection() const
+ bool WordIsFromDictionary() const
+ bool WordIsNumeric() const
+ bool HasBlamerInfo() const
+ cchar_t *GetBlamerDebug() const
cchar_t *GetBlamerMisadaptionDebug() const
bool HasTruthString() const
bool EquivalentToTruth(cchar_t *) const
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/tesserocr.egg-info/PKG-INFO
new/tesserocr-2.5.0/tesserocr.egg-info/PKG-INFO
--- old/tesserocr-2.4.1/tesserocr.egg-info/PKG-INFO 2019-08-23
18:03:11.000000000 +0200
+++ new/tesserocr-2.5.0/tesserocr.egg-info/PKG-INFO 2019-11-09
00:11:54.000000000 +0100
@@ -1,6 +1,6 @@
Metadata-Version: 2.1
Name: tesserocr
-Version: 2.4.1
+Version: 2.5.0
Summary: A simple, Pillow-friendly, Python wrapper around tesseract-ocr API
using Cython
Home-page: https://github.com/sirfz/tesserocr
Author: Fayez Zouheiry
@@ -88,20 +88,16 @@
The proposed downloads consist of stand-alone packages containing all
the Windows libraries needed for execution. This means that no additional
installation of tesseract is required on your system.
+ The recommended method of installation is via Conda as described below.
+
Conda
`````
- You can use the channel `simonflueckiger
<https://anaconda.org/simonflueckiger/tesserocr>`_ to install from Conda:
-
- ::
-
- > conda install -c simonflueckiger tesserocr
-
- or to get **tesserocr** compiled with **tesseract 4.0.0**:
+ You can use the `conda-forge
<https://anaconda.org/conda-forge/tesserocr>`_ channel to install from Conda:
::
- > conda install -c simonflueckiger/label/tesseract-4.0.0-master
tesserocr
+ > conda install -c conda-forge tesserocr
pip
```
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/tesserocr.pyx
new/tesserocr-2.5.0/tesserocr.pyx
--- old/tesserocr-2.4.1/tesserocr.pyx 2019-08-23 18:01:38.000000000 +0200
+++ new/tesserocr-2.5.0/tesserocr.pyx 2019-11-08 23:49:38.000000000 +0100
@@ -18,7 +18,7 @@
['eng', 'osd', 'equ'])
"""
-__version__ = '2.4.1'
+__version__ = '2.5.0'
import os
from io import BytesIO
@@ -45,7 +45,7 @@
# default parameters
setMsgSeverity(L_SEVERITY_NONE) # suppress leptonica error messages
-cdef TessBaseAPI _api = TessBaseAPI()
+cdef TessBaseAPI _api
_api.SetVariable('debug_file', '/dev/null') # suppress tesseract debug
messages
_api.Init(NULL, NULL)
IF TESSERACT_VERSION >= 0x3999800:
@@ -67,7 +67,7 @@
cdef class OEM(_Enum):
- """An enum that defines avaialble OCR engine modes.
+ """An enum that defines available OCR engine modes.
Attributes:
TESSERACT_ONLY: Run Tesseract only - fastest
@@ -323,7 +323,7 @@
cdef bytes _image_buffer(image):
"""Return raw bytes of a PIL Image"""
with BytesIO() as f:
- image.save(f, image.format or 'JPEG')
+ image.save(f, image.format or 'PNG')
return f.getvalue()
@@ -831,6 +831,21 @@
"""
return self._ltrriter.Confidence(level)
+ IF TESSERACT_VERSION >= 0x3040100:
+ def RowAttributes(self):
+ """Return row_height, descenders and ascenders in a dict"""
+ cdef:
+ float row_height
+ float descenders
+ float ascenders
+
+ self._ltrriter.RowAttributes(&row_height, &descenders, &ascenders)
+ return {
+ 'row_height': row_height,
+ 'descenders': descenders,
+ 'ascenders': ascenders
+ }
+
def WordFontAttributes(self):
"""Return the font attributes of the current word.
@@ -1138,7 +1153,7 @@
applicable language, and there is more chance of hallucinating
incorrect
words.
psm (int): Page segmentation mode. Defaults to :attr:`PSM.AUTO`.
- See :class:`PSM` for avaialble psm values.
+ See :class:`PSM` for available psm values.
init (bool): If ``False``, :meth:`Init` will not be called and has to
be called
after initialization.
oem (int): OCR engine mode. Defaults to :attr:`OEM.DEFAULT`.
@@ -1257,7 +1272,7 @@
def GetIntVariable(self, name):
"""Return the value of the given int parameter if it exists among
Tesseract parameters.
- Returns ``None`` if the paramter was not found.
+ Returns ``None`` if the parameter was not found.
"""
cdef:
bytes py_name = _b(name)
@@ -1269,7 +1284,7 @@
def GetBoolVariable(self, name):
"""Return the value of the given bool parameter if it exists among
Tesseract parameters.
- Returns ``None`` if the paramter was not found.
+ Returns ``None`` if the parameter was not found.
"""
cdef:
bytes py_name = _b(name)
@@ -1281,7 +1296,7 @@
def GetDoubleVariable(self, name):
"""Return the value of the given double parameter if it exists among
Tesseract parameters.
- Returns ``None`` if the paramter was not found.
+ Returns ``None`` if the parameter was not found.
"""
cdef:
bytes py_name = _b(name)
@@ -1293,7 +1308,7 @@
def GetStringVariable(self, name):
"""Return the value of the given string parameter if it exists among
Tesseract parameters.
- Returns ``None`` if the paramter was not found.
+ Returns ``None`` if the parameter was not found.
"""
cdef:
bytes py_name = _b(name)
@@ -1306,13 +1321,13 @@
"""Return the value of named variable as a string (regardless of type),
if it exists.
- Returns ``None`` if paramter was not found.
+ Returns ``None`` if parameter was not found.
"""
cdef:
bytes py_name = _b(name)
STRING val
if self._baseapi.GetVariableAsString(py_name, &val):
- return val.string()
+ return val.c_str()
return None
def InitFull(self, path=_DEFAULT_PATH, lang=_DEFAULT_LANG,
@@ -1387,7 +1402,7 @@
OcrEngineMode oem=OEM_DEFAULT):
"""Initialize the API with the given data path, language and OCR
engine mode.
- See :meth:`InitFull` for more intialization info and options.
+ See :meth:`InitFull` for more initialization info and options.
Args:
path (str): The name of the parent directory of tessdata.
@@ -1426,7 +1441,7 @@
"""
cdef GenericVector[STRING] langs
self._baseapi.GetLoadedLanguagesAsVector(&langs)
- return [langs[i].string() for i in xrange(langs.size())]
+ return [langs[i].c_str() for i in xrange(langs.size())]
def GetAvailableLanguages(self):
"""Return list of available languages in the init data path"""
@@ -1435,7 +1450,7 @@
int i
langs = []
self._baseapi.GetAvailableLanguagesAsVector(&v)
- langs = [v[i].string() for i in xrange(v.size())]
+ langs = [v[i].c_str() for i in xrange(v.size())]
return langs
def InitForAnalysePage(self):
@@ -2416,7 +2431,7 @@
path (str): The name of the parent directory of tessdata.
Must end in /.
oem (int): OCR engine mode. Defaults to :attr:`OEM.DEFAULT`.
- see :class:`OEM` for all avaialble oem options.
+ see :class:`OEM` for all available oem options.
Returns:
unicode: The text extracted from the image.
@@ -2466,7 +2481,7 @@
path (str): The name of the parent directory of tessdata.
Must end in /.
oem (int): OCR engine mode. Defaults to :attr:`OEM.DEFAULT`.
- see :class:`OEM` for all avaialble oem options.
+ see :class:`OEM` for all available oem options.
Returns:
unicode: The text extracted from the image.
@@ -2514,7 +2529,7 @@
Must end in /. Default tesseract-ocr datapath is used
if no path is provided.
- Retruns
+ Returns
tuple: Tuple with two elements:
- path (str): tessdata parent directory path
- languages (list): list of available languages as ISO 639-3
strings.
@@ -2527,6 +2542,6 @@
baseapi.Init(py_path, NULL)
path = baseapi.GetDatapath()
baseapi.GetAvailableLanguagesAsVector(&v)
- langs = [v[i].string() for i in xrange(v.size())]
+ langs = [v[i].c_str() for i in xrange(v.size())]
baseapi.End()
return path, langs
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/tesserocr_experiment.pyx
new/tesserocr-2.5.0/tesserocr_experiment.pyx
--- old/tesserocr-2.4.1/tesserocr_experiment.pyx 2016-08-22
15:27:56.000000000 +0200
+++ new/tesserocr-2.5.0/tesserocr_experiment.pyx 2019-11-08
23:49:38.000000000 +0100
@@ -1,4 +1,4 @@
-# An attemp to address the PIL.Image buffer directly without copying it.
+# An attempt to address the PIL.Image buffer directly without copying it.
#
# This is achieved by extracting the buffer ptr from Image.im.unsafe_ptrs
# the xsize, ysize, pixelsize and linesize are extracted as well to be used
diff -urN '--exclude=CVS' '--exclude=.cvsignore' '--exclude=.svn'
'--exclude=.svnignore' old/tesserocr-2.4.1/tests/test_api.py
new/tesserocr-2.5.0/tests/test_api.py
--- old/tesserocr-2.4.1/tests/test_api.py 2019-08-23 18:01:21.000000000
+0200
+++ new/tesserocr-2.5.0/tests/test_api.py 2019-11-08 23:49:38.000000000
+0100
@@ -326,6 +326,16 @@
res = self._api.Recognize()
self.assertTrue(res)
+ @unittest.skipIf(_TESSERACT_VERSION < 0x3040100, "tesseract < 4")
+ def test_row_attributes(self):
+ self._api.SetImageFile(self._image_file)
+ self._api.Recognize()
+ it = self._api.GetIterator()
+ attrs = it.RowAttributes()
+ self.assertIsInstance(attrs['row_height'], float)
+ self.assertIsInstance(attrs['ascenders'], float)
+ self.assertIsInstance(attrs['descenders'], float)
+
if __name__ == '__main__':
unittest.main()
