def _reply_buttons(self):
buttons = []
if len(self.lang_keys) > 1:
values = [
(f'{value}' if value != self.lang_key else f'• {value}',
f'{self.title}|{value}|0')
for value in self.lang_keys
]
for chunk in chunks(values, 4):
buttons.append([
telegram.InlineKeyboardButton(text, callback_data=data)
for text, data in chunk
])
if self.homonyms_count > 1:
values = [
(f'{i+1}' if i != self.homonym_index else f'• {i+1}',
f'{self.title}|{self.lang_key}|{i}')
for i in range(self.homonyms_count)
]
for chunk in chunks(values, 6):
buttons.append([
telegram.InlineKeyboardButton(text, callback_data=data)
for text, data in chunk
])
if len(self.lang_keys) == 1 and self.homonyms_count == 1:
buttons.append([
telegram.InlineKeyboardButton(
'Обновить',
callback_data=f'{self.title}|{self.lang_keys[0]}|0'
)
])
return telegram.InlineKeyboardMarkup(buttons)
def _parse(self):
parts = self.parse_pattern.split(self.content)
if len(parts) == 1:
self._top = ''
self._sub_sections = {
'': self.child_section_type(self, '', '', parts[0],
self.silent),
}
return
self._top = parts.pop(0)
self._sub_sections = dict()
if self.copy_top_to_sub_sections:
self._sub_sections[''] = \
self.child_section_type(self, '', '', self._top, self.silent)
for full_header, header, content in chunks(parts, 3):
if header in self._sub_sections and not self.silent:
raise Exception(f'Duplicated section `{header}` on the page '
f'"{self.title}"')
child_section = \
self.child_section_type(self, full_header, header, content,
self.silent)
key = child_section.key
if key in self._sub_sections and not self.silent:
raise Exception(f'Duplicated header key `{key}` on the page '
f'"{self.title}"')
self._sub_sections[key] = child_section
def check_page(self, page) -> list:
values = []
parts = TR.lang_header.split(self.content)
headers = set()
if len(parts) == 1:
return [] # наверное, редирект
parts.pop(0)
for full_header, header, content in chunks(parts, 3):
if header in headers:
values.append(header)
headers.add(header)
return values
def save_pages():
endings = json.loads(read('endings_ru_new.json'))
for ending, entries in endings.items():
entries.sort(key=lambda x: x[0][::-1])
values = [entry[1] for entry in entries]
for i, chunk in enumerate(chunks(values, 3700)):
# content = f"= Глаголы на '''-{ending}''' =\n" \
# f'<div class="reverseindex">\n' + \
# f''.join(chunk) + \
# f'</div>'
content = f"= Глаголы на '''-{ending}''' =\n" + f''.join(chunk)
save_page(f'User:Vitalik/reports/verbs/-{ending}/{i+1}', content,
'Обратный список глаголов по окончаниям')
print(ending, len(chunk))
def check_page(self, page) -> list:
values = []
for lang, language_obj in page.languages.last_dict(
unique=True).items():
# print(page.title, lang, language_obj)
parts = R.second_header.split(language_obj.content)
if len(parts) == 1:
continue # наверное, редирект
headers = set()
# print(parts)
parts.pop(0)
for full_header, header, content in chunks(parts, 3):
if header in headers:
values.append((lang, header))
headers.add(header)
return values
def group_entries(self):
grouped = defaultdict(list)
for title, details in self.entries.items():
aspect, stress, impersonal = details # unpack
if impersonal:
res_key = '(безличные)'
elif '-' in title:
res_key = '(через дефис)'
else:
res_key = '-???'
for key in self.keys:
if key.startswith('-') and title.endswith(key[1:]):
res_key = key
break
grouped[res_key].append(title)
for key, entries in grouped.copy().items():
max_size = 2000
# todo: implement хитрый алгоритм, разделения на подразделы по удобным разделяющим буквам...
if len(entries) > max_size:
for i, chunk in enumerate(chunks(entries, max_size)):
new_key = f'{key} @{i+1}'
grouped[new_key] = chunk
del grouped[key]
def sort_key(pair):
check_key, value = pair
if check_key in self.keys:
return self.keys.index(check_key), 0
m = re.search('(.*) @(\d+)$', check_key)
if m:
check_key, page = m.groups()
if check_key in self.keys:
return self.keys.index(check_key), page
return float('inf'), check_key
return dict(sorted(grouped.items(), key=sort_key))