class KVGXmlDictionaryReader(_XmlBase):
def __init__(self):
self._charcol = CharacterCollection()
def get_character_collection(self):
return self._charcol
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "kanjis":
raise ValueError, "The very first tag should be <kanjis>"
if self._tag == "kanji":
self._writing = Writing()
self._utf8 = attrs["midashi"].encode("UTF-8")
if self._tag == "stroke":
self._stroke = Stroke()
if attrs.has_key("path"):
self._stroke_svg = attrs["path"].encode("UTF-8")
try:
svg_parser = SVG_Parser(self._stroke_svg)
svg_parser.parse()
self._stroke.append_points(svg_parser.get_points())
except:
sys.stderr.write("Something went wrong in this character: " + self._utf8 + "\n")
else:
print "Missing path in <stroke> element: " + self._utf8
def _end_element(self, name):
if name == "kanji":
char = Character()
char.set_utf8(self._utf8)
char.set_writing(self._writing)
self._charcol.add_set(self._utf8)
self._charcol.append_character(self._utf8, char)
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "stroke":
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
class KVGXmlDictionaryReader(_XmlBase):
def __init__(self):
self._charcol = CharacterCollection()
def get_character_collection(self):
return self._charcol
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "kanjivg":
raise ValueError, "The very first tag should be <kanjivg>"
if self._tag == "kanji":
self._writing = Writing()
self._utf8 = unichr(int(attrs["id"].split('_')[1],
16)).encode("UTF-8")
if self._tag == "path":
self._stroke = Stroke()
if attrs.has_key("d"):
self._stroke_svg = attrs["d"].encode("UTF-8")
svg_parser = SVG_Parser(self._stroke_svg)
svg_parser.parse()
self._stroke.append_points(svg_parser.get_points())
else:
sys.stderr.write("Missing data in <path> element: " +
self._utf8 + "\n")
def _end_element(self, name):
if name == "kanji":
char = Character()
char.set_utf8(self._utf8)
char.set_writing(self._writing)
self._charcol.add_set(self._utf8)
self._charcol.append_character(self._utf8, char)
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "path":
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
class KVGXmlDictionaryReader(_XmlBase):
def __init__(self):
self._charcol = CharacterCollection()
def get_character_collection(self):
return self._charcol
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "kanjivg":
raise ValueError, "The very first tag should be <kanjivg>"
if self._tag == "kanji":
self._writing = Writing()
self._utf8 = unichr(int(attrs["id"].split("_")[1], 16)).encode("UTF-8")
if self._tag == "path":
self._stroke = Stroke()
if attrs.has_key("d"):
self._stroke_svg = attrs["d"].encode("UTF-8")
svg_parser = SVG_Parser(self._stroke_svg)
svg_parser.parse()
self._stroke.append_points(svg_parser.get_points())
else:
sys.stderr.write("Missing data in <path> element: " + self._utf8 + "\n")
def _end_element(self, name):
if name == "kanji":
char = Character()
char.set_utf8(self._utf8)
char.set_writing(self._writing)
self._charcol.add_set(self._utf8)
self._charcol.append_character(self._utf8, char)
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "path":
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
def get_character_collection(self):
charcol = CharacterCollection()
# group characters with the same label into sets
sets = {}
for i in range(len(self._labels)):
# Create Character
writing = Writing()
if self.height and self.width:
writing.set_height(self.height)
writing.set_width(self.width)
for delin_range in self._delineations[i]:
if delin_range.start_comp == (delin_range.end_comp - 1):
stroke_points = self._strokes[delin_range.start_comp][delin_range.start_point:delin_range.end_point]
writing.append_stroke(Stroke.from_list(stroke_points))
else:
# add first stroke to writing
start_stroke_points = self._strokes[delin_range.start_comp][delin_range.start_point:-1]
if len(start_stroke_points) > 0:
writing.append_stroke(Stroke.from_list(start_stroke_points))
# add last stroke to writing
end_stroke_points = self._strokes[delin_range.end_comp - 1][0:delin_range.end_point]
if len(end_stroke_points) > 0:
writing.append_stroke(Stroke.from_list(end_stroke_points))
# add the remaining strokes to writing
for stroke in self._strokes[delin_range.start_comp + 1:delin_range.end_comp - 1]:
writing.append_stroke(stroke)
character = Character()
character.set_writing(writing)
utf8 = self._labels[i]
character.set_utf8(utf8)
sets[utf8] = sets.get(utf8, []) + [character]
charcol.add_sets(sets.keys())
for set_name, characters in sets.items():
charcol.append_characters(set_name, characters)
return charcol
class TomoeXmlDictionaryReader(_XmlBase):
def __init__(self):
self._charcol = CharacterCollection()
def get_character_collection(self):
return self._charcol
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "dictionary":
raise ValueError, "The very first tag should be <dictionary>"
if self._tag == "character":
self._writing = Writing()
if self._tag == "stroke":
self._stroke = Stroke()
elif self._tag == "point":
point = Point()
for key in ("x", "y", "pressure", "xtilt", "ytilt", "timestamp"):
if attrs.has_key(key):
value = attrs[key].encode("UTF-8")
if key in ("pressure", "xtilt", "ytilt"):
value = float(value)
else:
value = int(float(value))
else:
value = None
setattr(point, key, value)
self._stroke.append_point(point)
def _end_element(self, name):
if name == "character":
char = Character()
char.set_utf8(self._utf8)
char.set_writing(self._writing)
self._charcol.add_set(self._utf8)
self._charcol.append_character(self._utf8, char)
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "stroke":
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
class TomoeXmlDictionaryReader(_XmlBase):
def __init__(self):
self._charcol = CharacterCollection()
def get_character_collection(self):
return self._charcol
def _start_element(self, name, attrs):
self._tag = name
if self._first_tag:
self._first_tag = False
if self._tag != "dictionary":
raise ValueError, "The very first tag should be <dictionary>"
if self._tag == "character":
self._writing = Writing()
if self._tag == "stroke":
self._stroke = Stroke()
elif self._tag == "point":
point = Point()
for key in ("x", "y", "pressure", "xtilt", "ytilt", "timestamp"):
if attrs.has_key(key):
value = attrs[key].encode("UTF-8")
if key in ("pressure", "xtilt", "ytilt"):
value = float(value)
else:
value = int(float(value))
else:
value = None
setattr(point, key, value)
self._stroke.append_point(point)
def _end_element(self, name):
if name == "character":
char = Character()
char.set_utf8(self._utf8)
char.set_writing(self._writing)
self._charcol.add_set(self._utf8)
self._charcol.append_character(self._utf8, char)
for s in ["_tag", "_stroke"]:
if s in self.__dict__:
del self.__dict__[s]
if name == "stroke":
self._writing.append_stroke(self._stroke)
self._stroke = None
self._tag = None
def _char_data(self, data):
if self._tag == "utf8":
self._utf8 = data.encode("UTF-8")
elif self._tag == "width":
self._writing.set_width(int(data))
elif self._tag == "height":
self._writing.set_height(int(data))
