def nom_and_gen_xml(content0, content1, content2, content3, content4):
result = []
tag_span = 0
sense_numbers_in_end = []
def remove_sense_numbers(text):
sense_numbers_in_end = []
if text.strip().endswith('1.'):
sense_numbers_in_end = create_nodes_for_sense_numbers('1.')
text = text[:text.index('1.')]
elif text.strip().endswith('1. a)'):
sense_numbers_in_end = create_nodes_for_sense_numbers('1. a)')
text = text[:text.index('1.')]
return text, sense_numbers_in_end
def create_nodes_for_sense_numbers(sense_number):
res = []
if sense_number == '1.':
res.append(nf.create_bold_hi_node('1.'))
elif sense_number == '1. a)':
res.append(nf.create_bold_hi_node('1.'))
res.append(nf.create_bold_hi_node(' a)'))
return res
if content0.rstrip()[-1] == ',':
form_lemma = nf.create_form_lemma_node(content0.rstrip()[:-1])
usg_gen = nf.create_usg_node('gen.')
if content2.strip() == '-':
content3, sense_numbers_in_end = remove_sense_numbers(content3)
form_inflected = nf.create_form_inflected_node(' -' + content3)
tag_span = 4
else:
content2, sense_numbers_in_end = remove_sense_numbers(content2)
form_inflected = nf.create_form_inflected_node(' ' + content2.lstrip())
tag_span = 3
else:
form_lemma = nf.create_form_lemma_node(content0)
usg_gen = nf.create_usg_node('gen.')
if content3.strip() == '-':
content4, sense_numbers_in_end = remove_sense_numbers(content4)
form_inflected = nf.create_form_inflected_node(' -' + content4)
tag_span = 5
else:
content3, sense_numbers_in_end = remove_sense_numbers(content3)
form_inflected = nf.create_form_inflected_node(' ' + content3.lstrip())
tag_span = 4
pc = nf.create_pc_node(', ')
result.extend([form_lemma, pc, *usg_gen, form_inflected])
return result, tag_span, sense_numbers_in_end
def adj_multiple_forms_xml(content0):
content0_split = content0.split(', ')
result = []
form_lemma = nf.create_form_lemma_node(content0_split[0])
result.append(form_lemma)
for word in content0_split[1:]:
form_inflected = nf.create_form_inflected_node(word)
pc = nf.create_pc_node(', ')
result.append(pc)
result.append(form_inflected)
return result
def unknown_initial_xml(content0):
res = []
content0_split = [
x for x in content0.split(', ') if not is_empty_string(x)
]
for i in range(len(content0_split)):
if i == 0:
form_lemma = nf.create_form_lemma_node(content0_split[i])
res.append(form_lemma)
if len(content0_split) > 1:
pc = nf.create_pc_node(', ')
res.append(pc)
# TODO: Extra comma / missing comma
elif i == len(content0_split) - 1:
form_inflected = nf.create_form_inflected_node(content0_split[i])
res.append(form_inflected)
else:
form_inflected = nf.create_form_inflected_node(content0_split[i])
pc = nf.create_pc_node(', ')
res.append(form_inflected)
res.append(pc)
return res
def noun_xml(content0, content1):
result = []
content0_split = content0.rsplit(', ', 1)
form_lemma = nf.create_form_lemma_node(content0_split[0])
result.append(form_lemma)
if content0_split[1]:
pc = nf.create_pc_node(', ')
form_inflected = nf.create_form_inflected_node(content0_split[1])
result.append(pc)
result.append(form_inflected)
gram_grp = nf.create_gram_grp(content1)
result.append(gram_grp)
return result
def verb_xml(entry_type, content0, content1):
result = []
content0_split = content0.split(', ')
form_lemma = nf.create_form_lemma_node(content0_split[0])
result.append(form_lemma)
if len(content0_split) > 1:
for word in content0_split[1:]:
pc = nf.create_pc_node(', ')
form_inflected = nf.create_form_inflected_node(word)
result.append(pc)
result.append(form_inflected)
if entry_type != 'special_verb':
gram_grp = nf.create_gram_grp(content1, "iType")
result.append(gram_grp)
return result
def unknown_entry_partially_encode(entry):
old_morph_part = copy(entry.encoded_parts['morph_part'])
entry.encoded_parts['morph_part'] = []
counter = 0
while old_morph_part:
content_node = [old_morph_part[0]]
node_content = SafeString(old_morph_part[0].text)
if counter == 0:
content_node = unknown_initial_xml(content_node[0].text)
elif node_content.strip().startswith(
'(') and node_content.strip().endswith(')'):
content_node = [nf.create_extra_morph(node_content)]
elif node_content.strip() in (punctuation + '–'):
content_node = [nf.create_pc_node(node_content)]
elif old_morph_part[0].get('rend') == "italic":
if node_content.strip() in ('m', 'f', 'n'):
content_node = [nf.create_gram_grp(node_content)]
elif len(entry.encoded_parts['morph_part']) == 1 and len(old_morph_part) >= 2 and old_morph_part[1].get('rend' ) == 'bold' and \
node_content.strip() == 'и' and entry.encoded_parts['morph_part'][0].tag == nf.get_ns('form'):
content_node = []
entry.encoded_parts['morph_part'][0].append(old_morph_part[0])
entry.encoded_parts['morph_part'][0].append(
nf.create_orth_node(SafeString(old_morph_part[1].text)))
old_morph_part.pop(0)
else:
content_node = nf.create_usg_node(node_content)
elif node_content.strip() in ('1', '2', '3', '4') and (
len(old_morph_part) == 1
or old_morph_part[1].text.strip() != '.'):
content_node = [nf.create_gram_grp(node_content, 'iType')]
[entry.encoded_parts['morph_part'].append(x) for x in content_node]
old_morph_part.pop(0)
counter += 1
def create_cit_nodes(node_content):
result = []
node_content, *separated_end_punctuation = separate_dash_dot_semi_colon_in_end_of_node_content(SafeString(node_content))
split_contents = node_content.split('; ')
split_contents = [x for x in split_contents if not is_empty_string(x)]
for y in range(len(split_contents)):
x = split_contents[y]
for i in range(len(x.split(' '))):
word = x.split(' ')[i]
if is_empty_string(word) and i != len(x.split(' '))-1:
continue
if has_more_cyrillic_than_latin(word) and not (len(word) == 1 and not has_more_cyrillic_than_latin(SafeString(x).split(' ')[i+1])):
if i > 0:
if not (x.split(' ')[0] == '' and i == 1):
cit_node = nf.assemble_cit_nodes('example', ' '.join((x.split(' ')[:i])) + ' ')
result.append(cit_node)
if x.split(' ')[0] == '' and i == 1:
cit_node = nf.assemble_cit_nodes('translation', ' '.join(x.split(' ')))
else:
cit_node = nf.assemble_cit_nodes('translation', ' '.join(x.split(' ')[i:]))
result.append(cit_node)
break
elif i == len(x.split(' ')) - 1:
cit_node = nf.assemble_cit_nodes('example', ' '.join((x.split(' '))))
result.append(cit_node)
break
if y < (len(split_contents)-1):
result.append(nf.create_pc_node('; '))
for punct_node in separated_end_punctuation:
result.append(punct_node)
return result
def separate_dash_dot_semi_colon_in_end_of_node_content(node_content):
dash_node = None
dot_node = None
s_colon_node = None
result = deque()
if node_content.strip().endswith('–'):
if node_content.endswith(' '):
node_content = node_content.rstrip()[:-1]
dash_node = nf.create_pc_node('— ')
else:
dash_node = nf.create_pc_node('—')
node_content = node_content[:-1]
result.appendleft(dash_node)
if node_content.strip().endswith('.'):
if node_content.endswith(' '):
node_content = node_content.rstrip()[:-1]
dot_node = nf.create_pc_node('. ')
else:
dot_node = nf.create_pc_node('.')
node_content = node_content[:-1]
result.appendleft(dot_node)
if node_content.strip().endswith(';'):
if node_content.endswith(' '):
node_content = node_content.rstrip()[:-1]
s_colon_node = nf.create_pc_node('; ')
else:
s_colon_node = nf.create_pc_node(';')
node_content = node_content[:-1]
result.appendleft(s_colon_node)
result.appendleft(node_content)
return result
def encode_senses(entry):
raw_senses = entry.raw_senses
title_lemma = entry.title_lemma
last_sense_container = None
numbers = []
if raw_senses:
fix_dot_in_next_node(raw_senses)
if one_is_missing(raw_senses):
add_missing_one(entry, title_lemma)
numbers.append('1')
if not is_numbered_entry(raw_senses):
entry.encoded_parts['senses'].append(nf.create_sense_container_non_numbered(title_lemma))
last_sense_container = entry.encoded_parts['senses'][0]
numbers.append('1')
else:
deal_with_completely_unknown_entry(entry)
while raw_senses:
initial = SafeString(raw_senses[0].text).strip()
initial = fix_cyrillic_letter(initial)
if is_subsense_number(initial):
if fix_mixed_numbers(entry, initial):
continue
sense_number = initial[0]
if sense_number == 'I':
sense_number = initial[:-1]
numbers.append(sense_number)
last_sense_container = create_subsense_number_node(title_lemma, numbers, SafeString(raw_senses[0].text))
append_sense_container_and_label(entry, last_sense_container)
else:
content_node = [raw_senses[0]]
current_text = SafeString(raw_senses[0].text)
if current_text.strip() in (punctuation + '–'):
content_node = [nf.create_pc_node(current_text)]
elif raw_senses[0].get('rend') == "italic":
content_node = nf.create_usg_node(current_text)
elif raw_senses[0].get('rend') == "bold" and has_more_cyrillic_than_latin(current_text):
content_node = nf.create_def_node(current_text)
elif (not last_sense_container or len([x for x in last_sense_container.getchildren() if x.tag in (nf.get_ns('cit'), nf.get_ns('quote'))]) == 0) and \
has_more_cyrillic_than_latin(current_text.strip().split(' ')[0]):
node_content, *separated_end_punctuation = separate_dash_dot_semi_colon_in_end_of_node_content(current_text)
content_node = []
found_latin = False
i = 0
for i in range(len(node_content.split(' '))):
word = node_content.split(' ')[i]
if is_empty_string(word):
continue
if has_more_cyrillic_than_latin(word) and not (len(word) == 1 and not has_more_cyrillic_than_latin(SafeString(node_content).split(' ')[i+1])):
pass
else:
found_latin = True
break
if not found_latin:
def_node = nf.create_def_node(node_content)
content_node.extend(def_node)
else:
def_node = nf.create_def_node(' '.join(node_content.split(' ')[:i]) + ' ')
content_node.extend(def_node)
cit_node = create_cit_nodes(' '.join(node_content.split(' ')[i:]))
content_node.extend(cit_node)
for punct_node in separated_end_punctuation:
content_node.append(punct_node)
else:
content_node = create_cit_nodes(current_text)
if last_sense_container is not None:
[last_sense_container.append(x) for x in content_node]
else:
[entry.encoded_parts['senses'].append(x) for x in content_node]
raw_senses.pop(0)