text consistency errors are too hard to debug currently, added more f…

…eedback in error/warning

pynlpl/__init__.py

@@ -2,4 +2,4 @@
 
 The library is divided into several packages and modules. It is designed for Python 2.6 and upwards. Including Python 3."""
 
-VERSION = "1.2.6"
+VERSION = "1.2.7"

pynlpl/formats/folia.py

@@ -851,7 +851,12 @@ def textvalidation(self, warnonly=None):
  deepnormtext = self.text(cls,retaintokenisation=False,strict=False, normalize_spaces=True)
  if strictnormtext != deepnormtext:
  valid = False
- msg = "Text for " + self.__class__.__name__ + ", ID " + str(self.id) + ", class " + cls + ", is inconsistent: expected (after normalization): '" + deepnormtext + "', got (after normalization): '" + strictnormtext + "'"
+ deviation = 0
+ for i, (c1,c2) in enumerate(zip(strictnormtext,deepnormtext)):
+ if c1 != c2:
+ deviation = i
+ break
+ msg = "Text for " + self.__class__.__name__ + ", ID " + str(self.id) + ", class " + cls + ", is inconsistent: EXPECTED (after normalization) *****>\n" + deepnormtext + "\n****> BUT FOUND (after normalization) ****>\n" + strictnormtext + "\n******* DEVIATION POINT: " + strictnormtext[min(0,deviation-10):deviation] + "<*HERE*>" + strictnormtext[deviation:deviation+10]
  if warnonly:
  print("TEXT VALIDATION ERROR: " + msg,file=sys.stderr)
  else:

setup.py

@@ -23,7 +23,7 @@ def read(fname):
 
 setup(
  name = "PyNLPl",
- version = "1.2.6", #edit version in __init__.py as well and ensure tests/folia.py FOLIARELEASE points to the right version and is not set to None!
+ version = "1.2.7", #edit version in __init__.py as well and ensure tests/folia.py FOLIARELEASE points to the right version and is not set to None!
  author = "Maarten van Gompel",
  author_email = "proycon@anaproy.nl",
  description = ("PyNLPl, pronounced as 'pineapple', is a Python library for Natural Language Processing. It contains various modules useful for common, and less common, NLP tasks. PyNLPl contains modules for basic tasks, clients for interfacting with server, and modules for parsing several file formats common in NLP, most notably FoLiA."),