Merge branch 'master' into pr

ocrolib/common.py

@@ -11,100 +11,33 @@
 import sys
 import sysconfig
 import unicodedata
-import warnings
 import inspect
 import glob
 import cPickle
+from ocrolib.exceptions import (BadClassLabel, BadInput, FileNotFound,
+                                OcropusException)
 
 import numpy
-from numpy import *
+from numpy import (amax, amin, array, bitwise_and, clip, dtype, mean, minimum,
+                   nan, sin, sqrt, zeros)
 import pylab
-from pylab import imshow
-from scipy.ndimage import morphology,measurements
+from pylab import (clf, cm, ginput, gray, imshow, ion, subplot, where)
+from scipy.ndimage import morphology, measurements
 import PIL
 
 from default import getlocal
-from toplevel import *
+from toplevel import (checks, ABINARY2, AINT2, AINT3, BOOL, DARKSEG, GRAYSCALE,
+                      LIGHTSEG, LINESEG, PAGESEG)
 import chars
 import codecs
 import ligatures
 import lstm
 import morph
 import multiprocessing
-
-################################################################
-### exceptions
-################################################################
-
-class OcropusException(Exception):
-    trace = 1
-    def __init__(self,*args,**kw):
-        Exception.__init__(self,*args,**kw)
-
-class Unimplemented(OcropusException):
-    trace = 1
-    "Exception raised when a feature is unimplemented."
-    def __init__(self,s):
-        Exception.__init__(self,inspect.stack()[1][3])
-
-class Internal(OcropusException):
-    trace = 1
-    "Exception raised when a feature is unimplemented."
-    def __init__(self,s):
-        Exception.__init__(self,inspect.stack()[1][3])
-
-class RecognitionError(OcropusException):
-    trace = 1
-    "Some kind of error during recognition."
-    def __init__(self,explanation,**kw):
-        self.context = kw
-        s = [explanation]
-        s += ["%s=%s"%(k,summary(kw[k])) for k in kw]
-        message = " ".join(s)
-        Exception.__init__(self,message)
-
-class Warning(OcropusException):
-    trace = 0
-    def __init__(self,*args,**kw):
-        OcropusException.__init__(self,*args,**kw)
-
-class BadClassLabel(OcropusException):
-    trace = 0
-    "Exception for bad class labels in a dataset or input."
-    def __init__(self,s):
-        Exception.__init__(self,s)
-
-class BadImage(OcropusException):
-    trace = 0
-    def __init__(self,*args,**kw):
-        OcropusException.__init__(self,*args)
-
-class BadInput(OcropusException):
-    trace = 0
-    def __init__(self,*args,**kw):
-        OcropusException.__init__(self,*args,**kw)
-
-class FileNotFound(OcropusException):
-    trace = 0
-    """Some file-not-found error during OCRopus processing."""
-    def __init__(self,fname):
-        self.fname = fname
-    def __str__(self):
-        return "file not found %s"%(self.fname,)
+import sl
 
 pickle_mode = 2
 
-def deprecated(f):
-    def _wrap(f):
-        warned = 0
-        def _wrapper(*args,**kw):
-            if not warned:
-                print(f, "has been DEPRECATED")
-                warned = 1
-            return f(*args,**kw)
-    return _wrap
-
-
 
 ################################################################
 # text normalization
@@ -466,10 +399,10 @@ def extractMasked(self,image,index,grow=0,bg=None,margin=0,dtype=None):
         mh,mw = mask.shape
         box = self.bbox(index)
         r0,c0,r1,c1 = box
-        subimage = improc.cut(image,(r0,c0,r0+mh-2*margin,c0+mw-2*margin),margin,bg=bg)
+        subimage = sl.cut(image,(r0,c0,r0+mh-2*margin,c0+mw-2*margin),margin,bg=bg)
         return where(mask,subimage,bg)
 
-
+
 
 ################################################################
 ### Object reading and writing
@@ -517,7 +450,7 @@ class names that have changed."""
             unpickler.find_global = unpickle_find_global
             return unpickler.load()
 
-
+
 
 ################################################################
 ### Simple record object.
@@ -581,16 +514,6 @@ def check_valid_class_label(s):
     else:
         raise BadClassLabel(s)
 
-def summary(x):
-    """Summarize a datatype as a string (for display and debugging)."""
-    if type(x)==numpy.ndarray:
-        return "<ndarray %s %s>"%(x.shape,x.dtype)
-    if type(x)==str and len(x)>10:
-        return '"%s..."'%x
-    if type(x)==list and len(x)>10:
-        return '%s...'%x
-    return str(x)
-
 ################################################################
 ### file name manipulation
 ################################################################
@@ -824,19 +747,6 @@ def quick_check_line_components(line_bin,dpi):
     there is probably something wrong."""
     return 1.0
 
-def deprecated(func):
-    """This is a decorator which can be used to mark functions
-    as deprecated. It will result in a warning being emitted
-    when the function is used."""
-    def newFunc(*args, **kwargs):
-        warnings.warn("Call to deprecated function %s." % func.__name__,
-                      category=DeprecationWarning,stacklevel=2)
-        return func(*args, **kwargs)
-    newFunc.__name__ = func.__name__
-    newFunc.__doc__ = func.__doc__
-    newFunc.__dict__.update(func.__dict__)
-    return newFunc
-
 ################################################################
 ### conversion functions
 ################################################################
@@ -904,67 +814,6 @@ def obinfo(ob):
         result += str(ob.shape)
     return result
 
-def save_component(file,object,verbose=0,verify=0):
-    """Save an object to disk in an appropriate format.  If the object
-    is a wrapper for a native component (=inherits from
-    CommonComponent and has a comp attribute, or is in package
-    ocropus), write it using ocropus.save_component in native format.
-    Otherwise, write it using Python's pickle.  We could use pickle
-    for everything (since the native components pickle), but that
-    would be slower and more confusing."""
-    if hasattr(object,"save_component"):
-        object.save_component(file)
-        return
-    if object.__class__.__name__=="CommonComponent" and hasattr(object,"comp"):
-        # FIXME -- get rid of this eventually
-        import ocropus
-        ocropus.save_component(file,object.comp)
-        return
-    if type(object).__module__=="ocropus":
-        import ocropus
-        ocropus.save_component(file,object)
-        return
-    if verbose:
-        print("[save_component]")
-    if verbose:
-        for k,v in object.__dict__.items():
-            print(":", k, obinfo(v))
-    with open(file,"wb") as stream:
-        pickle.dump(object,stream,pickle_mode)
-    if verify:
-        if verbose:
-            print("[trying to read it again]")
-        with open(file,"rb") as stream:
-            pickle.load(stream)
-
-def load_component(file):
-    """Load a component. This handles various special cases,
-    including old-style C++ recognizers (soon to be gotten rid of),
-    python expressions ("=package.ObjectName(arg1,arg2)"),
-    and simple pickled Python objects (default)."""
-    if file[0]=="=":
-        return pyconstruct(file[1:])
-    elif file[0]=="@":
-        file = file[1:]
-    with open(file,"r") as stream:
-        # FIXME -- get rid of this eventually
-        start = stream.read(128)
-    if start.startswith("<object>\nlinerec\n"):
-        # FIXME -- get rid of this eventually
-        warnings.warn("loading old-style linerec: %s"%file)
-        result = RecognizeLine()
-        import ocropus
-        result.comp = ocropus.load_IRecognizeLine(file)
-        return result
-    if start.startswith("<object>"):
-        # FIXME -- get rid of this eventually
-        warnings.warn("loading old-style cmodel: %s"%file)
-        import ocroold
-        result = ocroold.Model()
-        import ocropus
-        result.comp = ocropus.load_IModel(file)
-        return result
-    return load_object(file)
 
 def binarize_range(image,dtype='B',threshold=0.5):
     """Binarize an image by its range."""
@@ -973,39 +822,6 @@ def binarize_range(image,dtype='B',threshold=0.5):
     if dtype=='B': scale = 255
     return array(scale*(image>threshold),dtype=dtype)
 
-def draw_pseg(pseg,axis=None):
-    if axis is None:
-        axis = subplot(111)
-    h = pseg.dim(1)
-    regions = ocropy.RegionExtractor()
-    regions.setPageLines(pseg)
-    for i in range(1,regions.length()):
-        x0,y0,x1,y1 = (regions.x0(i),regions.y0(i),regions.x1(i),regions.y1(i))
-        p = patches.Rectangle((x0,h-y1-1),x1-x0,y1-y0,edgecolor="red",fill=0)
-        axis.add_patch(p)
-
-def draw_aligned(result,axis=None):
-    raise Unimplemented("FIXME draw_aligned")
-    if axis is None:
-        axis = subplot(111)
-    axis.imshow(NI(result.image),cmap=cm.gray)
-    cseg = result.cseg
-    if type(cseg)==numpy.ndarray: cseg = common.lseg2narray(cseg)
-    ocropy.make_line_segmentation_black(cseg)
-    ocropy.renumber_labels(cseg,1)
-    bboxes = ocropy.rectarray()
-    ocropy.bounding_boxes(bboxes,cseg)
-    s = re.sub(r'\s+','',result.output)
-    h = cseg.dim(1)
-    for i in range(1,bboxes.length()):
-        r = bboxes.at(i)
-        x0,y0,x1,y1 = (r.x0,r.y0,r.x1,r.y1)
-        p = patches.Rectangle((x0,h-y1-1),x1-x0,y1-y0,edgecolor=(0.0,0.0,1.0,0.5),fill=0)
-        axis.add_patch(p)
-        if i>0 and i-1<len(s):
-            axis.text(x0,h-y0-1,s[i-1],color="red",weight="bold",fontsize=14)
-    draw()
-
 def plotgrid(data,d=10,shape=(30,30)):
     """Plot a list of images on a grid."""
     ion()
@@ -1024,13 +840,13 @@ def showrgb(r,g=None,b=None):
     imshow(array([r,g,b]).transpose([1,2,0]))
 
 def showgrid(l,cols=None,n=400,titles=None,xlabels=None,ylabels=None,**kw):
-    if "cmap" not in kw: kw["cmap"] = pylab.cm.gray
+    if "cmap" not in kw: kw["cmap"] = cm.gray
     if "interpolation" not in kw: kw["interpolation"] = "nearest"
     n = minimum(n,len(l))
     if cols is None: cols = int(sqrt(n))
     rows = (n+cols-1)//cols
     for i in range(n):
-        pylab.xticks([]); pylab.yticks([])
+        pylab.xticks([]) ;pylab.yticks([])
         pylab.subplot(rows,cols,i+1)
         pylab.imshow(l[i],**kw)
         if titles is not None: pylab.title(str(titles[i]))

ocrolib/edist.py

@@ -21,6 +21,8 @@ def xlevenshtein(a,b,context=1):
     """Calculates the Levensthein distance between a and b
     and generates a list of differences by context."""
     n, m = len(a), len(b)
+    assert m>0 # xlevenshtein should only be called with non-empty b string (ground truth)
+    if a == b: return 0,[] # speed up for the easy case
     sources = empty((m+1,n+1),object)
     sources[:,:] = None
     dists = 99999*ones((m+1,n+1))

ocrolib/exceptions.py

@@ -0,0 +1,73 @@
+import inspect
+import numpy
+
+def summary(x):
+    """Summarize a datatype as a string (for display and debugging)."""
+    if type(x)==numpy.ndarray:
+        return "<ndarray %s %s>"%(x.shape,x.dtype)
+    if type(x)==str and len(x)>10:
+        return '"%s..."'%x
+    if type(x)==list and len(x)>10:
+        return '%s...'%x
+    return str(x)
+
+
+################################################################
+### Ocropy exceptions
+################################################################
+
+class OcropusException(Exception):
+    trace = 1
+    def __init__(self,*args,**kw):
+        Exception.__init__(self,*args,**kw)
+
+class Unimplemented(OcropusException):
+    trace = 1
+    "Exception raised when a feature is unimplemented."
+    def __init__(self,s):
+        Exception.__init__(self,inspect.stack()[1][3])
+
+class Internal(OcropusException):
+    trace = 1
+    "Exception raised when a feature is unimplemented."
+    def __init__(self,s):
+        Exception.__init__(self,inspect.stack()[1][3])
+
+class RecognitionError(OcropusException):
+    trace = 1
+    "Some kind of error during recognition."
+    def __init__(self,explanation,**kw):
+        self.context = kw
+        s = [explanation]
+        s += ["%s=%s"%(k,summary(kw[k])) for k in kw]
+        message = " ".join(s)
+        Exception.__init__(self,message)
+
+class Warning(OcropusException):
+    trace = 0
+    def __init__(self,*args,**kw):
+        OcropusException.__init__(self,*args,**kw)
+
+class BadClassLabel(OcropusException):
+    trace = 0
+    "Exception for bad class labels in a dataset or input."
+    def __init__(self,s):
+        Exception.__init__(self,s)
+
+class BadImage(OcropusException):
+    trace = 0
+    def __init__(self,*args,**kw):
+        OcropusException.__init__(self,*args)
+
+class BadInput(OcropusException):
+    trace = 0
+    def __init__(self,*args,**kw):
+        OcropusException.__init__(self,*args,**kw)
+
+class FileNotFound(OcropusException):
+    trace = 0
+    """Some file-not-found error during OCRopus processing."""
+    def __init__(self,fname):
+        self.fname = fname
+    def __str__(self):
+        return "file not found %s"%(self.fname,)