Page MenuHomec4science
Files F62826909

word.py
No OneTemporary
Actions

File Metadata

Created
Wed, May 15, 22:34

word.py
View Options

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
""" This class can be used to represent a word.
"""
# Copyright (C) University of Basel 2019 {{{1
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/> 1}}}
__author__ = "Christian Steiner"
__maintainer__ = __author__
__copyright__ = 'University of Basel'
__email__ = "christian.steiner@unibas.ch"
__status__ = "Development"
__license__ = "GPL v3"
__version__ = "0.0.1"
from lxml import etree as ET
import warnings
from .class_spec import SemanticClass
from .matrix import Matrix
from .word_position import WordPosition
from .transkription_position import TranskriptionPosition
class Word(SemanticClass):
"""
This class represents a word.
"""
DATA = 'debug-data'
def __init__(self, id=0, text='', line_number=-1, transkription_positions=[], faksimile_positions=[], word_part_objs=[]):
self.id = id
self.text = text
self.line_number = line_number
self.transkription_positions = transkription_positions
self.faksimile_positions = faksimile_positions
self.word_part_objs = word_part_objs
self.is_head_of_inserted_words = False
self.is_tail_of_inserted_words = False
self.is_before_inserted_words = False
self.is_after_inserted_words = False
self.word_insertion_mark = None
self.debug_msg = None
@classmethod
def get_semantic_dictionary(cls):
""" Creates and returns a semantic dictionary as specified by SemanticClass.
"""
dictionary = {}
class_dict = cls.get_class_dictionary()
properties = {'id': (int, 1, 'word/@id'), 'text': (str, 1, 'word/@text'), 'line_number': (int, 1, 'word/@line-number'),\
'transkription_positions': (TranskriptionPosition, SemanticClass.LIST, 'word/@id'),\
'faksimile_positions': (WordPosition, SemanticClass.LIST, 'word/@id')}
dictionary.update({'class': class_dict})
dictionary.update({'properties': properties})
return dictionary
def set_word_insertion_mark(self, word_insertion_mark):
"""Sets word_insertion_mark
"""
self.word_insertion_mark = word_insertion_mark
def attach_word_to_tree(self, target_tree):
"""Attaches word to tree target_tree.
"""
word_node = target_tree.getroot().xpath('//word[@id="%s"]' % self.id)[0] \
if(len(target_tree.getroot().xpath('//word[@id="%s"]' % self.id)) > 0) \
else ET.SubElement(target_tree.getroot(), 'word', attrib={'id': str(self.id)})
word_node.set('text', self.text)
if self.line_number > -1:
word_node.set('line-number', str(self.line_number))
for transkription_position in self.transkription_positions:
transkription_position.attach_object_to_tree(word_node)
"""
data_node = word_node.find(self.DATA) if bool(word_node.find(self.DATA)) else ET.SubElement(word_node, self.DATA)
for part_index, word_part in enumerate(self.word_part_objs):
part_node = data_node.xpath('./part[@index="%s"]' % part_index)[0] \
if(len(data_node.xpath('./part[@index="%s"]' % part_index)) > 0) \
else ET.SubElement(data_node, 'part', attrib={'index': str(part_index)})
part_node.set('text', word_part['text'])
part_node.set('class', word_part['class'])
part_node.set('x', str(round(float(word_part['x']), 3)))
part_node.set('y', str(round(float(word_part['y']), 3)))
if self.debug_msg is not None:
ET.SubElement(data_node, 'end', attrib={'debug-msg': self.debug_msg})
"""
def split(self, page, split_string, start_id=0):
"""Splits the word and returns an 3-tuple of new words.
"""
previousString, currentString, nextString = self.text.partition(split_string)
currentWord = None
previousWord = None
nextWord = None
previousIndex = 0
current_id = start_id
all_positional_word_parts = []
for position in self.transkription_positions:
all_positional_word_parts += position.positional_word_parts
if len(all_positional_word_parts) == 0:
warnings.warn('ATTENTION: Word: {} {} with Strings "{}, {}, {}": there are no parts!'.format(self.id, self.text, previousString, currentString, nextString))
if len(previousString) > 0:
previous_pwps = []
while previousIndex < len(all_positional_word_parts) and previousString != ''.join([ pwp.text for pwp in previous_pwps ]):
previous_pwps.append(all_positional_word_parts[previousIndex])
previousIndex += 1
if previousString != ''.join([ pwp.text for pwp in previous_pwps ]):
warnings.warn('ATTENTION: "{}" does not match a word_part_obj!'.format(previousString))
else:
previous_transkription_positions = TranskriptionPosition.CREATE_TRANSKRIPTION_POSITION_LIST_FROM_PWPS(page, previous_pwps, debug_msg_string='word.split')
previous_text = ''.join([ pwp.text for pwp in previous_pwps ])
previousWord = Word(text=previous_text, id=current_id, line_number=self.line_number, transkription_positions=previous_transkription_positions)
current_id += 1
all_positional_word_parts = all_positional_word_parts[previousIndex:]
if len(nextString) > 0:
tmp_pwps = []
index = 0
while index < len(all_positional_word_parts) and currentString != ''.join([ pwp.text for pwp in tmp_pwps ]):
tmp_pwps.append(all_positional_word_parts[index])
index += 1
if currentString != ''.join([ pwp.text for pwp in tmp_pwps ]):
warnings.warn('ATTENTION: "{}" does not match a word_part_obj!'.format(currentString))
else:
next_pwps = all_positional_word_parts[index:]
next_transkription_positions = TranskriptionPosition.CREATE_TRANSKRIPTION_POSITION_LIST_FROM_PWPS(page, next_pwps, debug_msg_string='word.split')
next_text = ''.join([ pwp.text for pwp in next_pwps ])
nextWord = Word(text=next_text, id=current_id+1, line_number=self.line_number, transkription_positions=next_transkription_positions)
all_positional_word_parts = all_positional_word_parts[:index]
current_transkription_positions = TranskriptionPosition.CREATE_TRANSKRIPTION_POSITION_LIST_FROM_PWPS(page, all_positional_word_parts, debug_msg_string='word.split')
current_text = ''.join([ pwp.text for pwp in all_positional_word_parts ])
currentWord = Word(text=current_text, id=current_id, line_number=self.line_number, transkription_positions=current_transkription_positions)
return previousWord, currentWord, nextWord
def join(self, other_word, append_at_end_of_new_word=True):
"""Joins other_word with this word by changing the text of current word and adding other_word.transkription_positions.
"""
if append_at_end_of_new_word:
self.text = self.text + other_word.text
for position in other_word.transkription_positions:
position.id = str(len(self.transkription_positions))
self.transkription_positions.append(position)
else:
self.text = other_word.text + self.text
index = 0
for position in other_word.transkription_positions:
self.transkription_positions.insert(index, position)
index += 1
while index < len(self.transkription_positions):
self.transkription_positions[index].id = str(index)
index += 1
@staticmethod
def CREATE_WORD(word_node=None, page=None, word_part_objs=[], id=0, height=0, endX=0, endSign=None, matrix=None, line_number=-1, debug_msg=None):
"""Creates a word from a (lxml.Element) node or word_part_objs.
[:return:] Word
"""
if word_node is not None: # init word from xml node
id = int(word_node.get('id'))
line_number = int(word_node.get('line-number')) if bool(word_node.get('line-number')) else line_number
text = word_node.get('text')
transkription_positions = [ TranskriptionPosition(node=node) for node in word_node.findall('.//' + WordPosition.TRANSKRIPTION) ]
faksimile_positions = [ WordPosition(node=node) for node in word_node.findall('.//' + WordPosition.FAKSIMILE) ]
word_part_objs = [ item.attrib for item in word_node.findall('.//' + Word.DATA + '/part')]\
if len(word_node.findall('.//' + Word.DATA)) > 0\
else [ item.attrib for item in word_node.findall('.//part')]
return Word(id=id, text=text, line_number=line_number, transkription_positions=transkription_positions,\
faksimile_positions=faksimile_positions, word_part_objs=word_part_objs)
elif len(word_part_objs) > 0: # init word from word_part_obj that has been extracted from svg file
WIDTH = 5
TOPCORRECTION = 2.0
FONTWIDTHFACTOR = 0.7 # factor that multiplies lastCharFontSize
height = height
x = round(float(word_part_objs[0]['x']), 3)
if(page is not None and bool(page.style_dict)):
HEIGHT_FACTOR = 1.1 # factor that multiplies biggest_font_size -> height
style_set = set(' '.join(set( dict['class'] for dict in word_part_objs)).split(' '))
biggest_font_size = page.get_biggest_fontSize4styles(style_set=style_set)
height = round(biggest_font_size * HEIGHT_FACTOR + HEIGHT_FACTOR / biggest_font_size, 3)
TOPCORRECTION = 1 + HEIGHT_FACTOR / biggest_font_size
if endSign is not None and '%' in endSign:
lastCharFontSizeList = [ float(page.style_dict[key]['font-size'].replace('px',''))\
for key in word_part_objs[len(word_part_objs)-1]['class'].split(' ')\
if bool(page.style_dict[key].get('font-size'))]
lastCharFontSize = lastCharFontSizeList[0] if len(lastCharFontSizeList) > 0 else 1
endX = float(endX) + lastCharFontSize * FONTWIDTHFACTOR
elif endSign is not None and '%' in endSign:
endX = float(endX) + WIDTH
bottom = round(float(word_part_objs[0]['y']), 3)
y = round(bottom - height + TOPCORRECTION, 3)
width = round(float(endX) - x, 3)
transkription_positions = [ WordPosition(height=height, width=width, x=x, y=y, matrix=matrix, tag=WordPosition.TRANSKRIPTION) ]
text = ''.join([ dict['text'] for dict in word_part_objs])
line_number = page.get_line_number( (y + bottom)/2) if page is not None else line_number
word = Word(id=id, text=text, line_number=line_number, transkription_positions=transkription_positions, word_part_objs=word_part_objs)
word.debug_msg = debug_msg
return word
else:
error_msg = 'word_node has not been defined' if (word_node is None) else 'word_part_objs is empty'
raise Exception('Error: {}'.format(error_msg))

Event Timeline

c4science · Help