def load_templates():
for category_name in category_names:
category = pywikibot.Category(site, category_name)
for article in category.articles():
prefix = u'Шаблон:'
title = article.title()
if not title.startswith(prefix):
print title, '-', 'BAD!', '#' * 40
continue
title = title[len(prefix):]
content = article.get()
print title
edited = convert_wiki_date(article.editTime())
edited = make_aware(edited, pytz.UTC) # tmp
try:
template = TemplateLabel.objects.get(title=title)
template.edited = edited
template.category = category_name
template.content = content
template.save()
except ObjectDoesNotExist:
TemplateLabel.objects.create(
title=title,
edited=edited,
category=category_name,
content=content,
)
def load_redirects():
prefix = u'Шаблон:'
for category_name in category_names:
category = pywikibot.Category(site, category_name)
for article in category.articles():
redirect = article.title()[len(prefix):]
for page in article.backlinks(filterRedirects=True):
# print page.title(), '->', article.title()
title = page.title()
if not title.startswith(prefix):
print title, '-', 'BAD!', '#' * 40
continue
title = title[len(prefix):]
content = page.get(get_redirect=True)
print title
edited = convert_wiki_date(page.editTime())
edited = make_aware(edited, pytz.UTC) # tmp
try:
template = TemplateLabel.objects.get(title=title)
template.edited = edited
template.category = category_name
template.content = content
template.redirect = redirect
template.save()
except ObjectDoesNotExist:
TemplateLabel.objects.create(
title=title,
edited=edited,
category=category_name,
content=content,
redirect=redirect,
)
def get_data(self, item):
try:
content = item.get(get_redirect=True)
content_to_db = remove_utf8mb4(content)
edited = aware(convert_wiki_date(item.editTime()))
redirect = item.isRedirectPage()
return content, content_to_db, edited, redirect
except NoPage:
return None, None, None, None
def load_templates_contents():
# for template in TemplateInflection.objects.all():
for template in TemplateInflection.objects.filter(content__isnull=True):
title = template.title
print title
# continue
article = pywikibot.Page(site, u"Шаблон:%s" % title)
content = article.get()
edited = convert_wiki_date(article.editTime())
edited = make_aware(edited, pytz.UTC)
template.content = content
template.edited = edited
template.save()
def load_templates():
category_names = [
u"Категория:Шаблоны словоизменений/Глаголы/Возвратные глаголы",
u"Категория:Шаблоны словоизменений/Глаголы/Невозвратные глаголы",
u"Категория:Шаблоны словоизменений/Глаголы/Несовершенный вид",
u"Категория:Шаблоны словоизменений/Глаголы/Совершенный вид",
u"Категория:Шаблоны словоизменений/Существительные/Одушевлённые",
u"Категория:Шаблоны словоизменений/Существительные/Одушевлённые/Мужской род",
u"Категория:Шаблоны словоизменений/Существительные/Одушевлённые/Женский род",
u"Категория:Шаблоны словоизменений/Существительные/Одушевлённые/Средний род",
u"Категория:Шаблоны словоизменений/Существительные/Неодушевлённые",
u"Категория:Шаблоны словоизменений/Существительные/Неодушевлённые/Мужской род",
u"Категория:Шаблоны словоизменений/Существительные/Неодушевлённые/Женский род",
u"Категория:Шаблоны словоизменений/Существительные/Неодушевлённые/Средний род",
u"Категория:Шаблоны словоизменений/Прилагательные",
u"Категория:Шаблоны словоизменений/Причастия",
u"Категория:Шаблоны словоизменений/Числительные",
u"Категория:Шаблоны словоизменений/Фамилии",
u"Категория:Шаблоны словоизменений/Местоимения",
]
for category_name in category_names:
morph = category_name.split('/')[1]
category = pywikibot.Category(site, category_name)
for article in category.articles():
prefix = u'Шаблон:'
# Шаблон:прил ru
title = article.title()
if not title.startswith(prefix):
print title, '-', 'BAD!'
continue
title = title[len(prefix):]
content = article.get()
print title
edited = convert_wiki_date(article.editTime())
edited = make_aware(edited, pytz.UTC) # tmp
try:
TemplateInflection.objects.get(title=title,
content=content,
edited=edited,
morph=morph)
print '#' * 20, '-> ALREADY EXISTS'
except ObjectDoesNotExist:
TemplateInflection.objects.create(title=title,
content=content,
edited=edited,
category=category_name,
morph=morph)
def process_potentially_new_item(self, item, i):
try:
title = item.title()
except InvalidTitle:
print 'Wrong title', '#' * 120
return
if ':' in title:
return
# try:
# page = Page.objects.get(title=title)
# except (Page.DoesNotExist, _mysql_exceptions.Warning):
# return
content, content_to_db, edited, redirect = self.get_data(item)
if not edited:
return
if not self.output_interval or not i % self.output_interval:
print dt(),
if ':' in title:
print ':',
else:
print '#',
print edited, '-', i, '-',
if ':' in title:
print ':',
else:
print '#',
print transliterate(title), ' // pk=%s' % '?' # page.pk
# if item.previous_revision_id == -1 and u'{{-ru-' in content:
if u'{{-ru-' in content:
created_at = convert_wiki_date(item.oldest_revision.timestamp)
print transliterate(title), created_at
# if created_at + timedelta(hours=5) > datetime.now():
if created_at + timedelta(days=7) > datetime.now():
if title not in self.titles and item.oldest_revision.user != 'CinBot':
print
print '-' * 100
print '|', transliterate(item.oldest_revision.user), edited
print '|', transliterate(title)
print '-' * 100
print
self.items.append((title, item.oldest_revision.user, created_at))
self.titles.append(title)
if ':' in title:
return
return edited
def process_template(article, lang):
prefix = u'Шаблон:'
title = article.title()
if not title.startswith(prefix):
print title, '-', 'BAD!', 'BAD!', '#' * 100
return
title = title[len(prefix):]
print title
article = pywikibot.Page(site, u"Шаблон:%s" % title)
content = article.get()
edited = convert_wiki_date(article.editTime())
edited = make_aware(edited, pytz.UTC)
words = title.split(' ')
morph = words[0]
if morph not in [u'adv', u'conj', u'interj', u'гл', u'глагол',
u'мест', u'прил', u'сущ', u'числ', u'падежи',
u'prep', u'affix', u'intro', u'phrase', u'suffix',
u'predic', u'склонение', u'part',
u'артикль', u'article', u'арт', u'деепр', u'onomatop',
u'interj1', u'прич', u'герундий',
u'склон', u'степени',
u'междом',
u'спряжения', u'спряжение', u'словоизм', u'сущ2',
u'принад', u'palat', u'abbrev', u'measure',
u'morph', u'prefix', u'ein',
u'союз', u'словоформы', u'глаг', u'послел',
u'послелог', u'падежи-мест', u'нар', u'морфема',
u'межд', ]:
print u'm →', title
return
prefix = u"%s %s" % (morph, lang)
if not title.startswith(prefix):
print u'e →', title, '(%s)' % lang
return
info = title[len(prefix):].strip()
kind, gender, num = parse_template_title(title)
return TemplateInflection(
title=title, content=content, edited=edited, lang=lang,
morph=morph, info=info, kind=kind, gender=gender, num=num,
)
def load_templates():
import pywikibot
site = pywikibot.Site('ru')
category_names = [
u"Категория:Викисловарь:Шаблоны:Языки",
u"Категория:Шаблоны:Языковые_заголовки",
]
for category_name in category_names:
category = pywikibot.Category(site, category_name)
for article in category.articles():
title = article.title()
m = re.match(u'^Шаблон:-([^-].*)-$', title)
if m:
lang = m.group(1)
print lang
content = article.get(get_redirect=True)
edited = aware(convert_wiki_date(article.editTime()))
LanguageTemplate.objects.get_or_create(lang_code=lang,
edited=edited,
content=content)
def load_templates():
import pywikibot
site = pywikibot.Site('ru')
category_names = [
u"Категория:Шаблоны:Названия_языков",
# u"Категория:Шаблоны:Языковые_заголовки",
]
for category_name in category_names:
category = pywikibot.Category(site, category_name)
for article in category.articles():
title = article.title()
m = re.match(u'^Шаблон:(?P<case>[Ll])ang-(?P<lang>.+)$', title)
# if not m:
# print title
if m:
lang = m.group('lang')
case = m.group('case')
lower = case == 'l'
print lang
content = article.get(get_redirect=True)
edited = aware(convert_wiki_date(article.editTime()))
TemplateLangNames.objects.get_or_create(
lang=lang, lower=lower, edited=edited, content=content)
continue
title = orig_title[len(prefix) :]
print title
if title == u"сущ":
continue
if not re.match(u"^сущ/[fmn]\d/\d+$", title):
continue
# print 'ok'
# continue
old = WordInflectionMassEdit.objects.get(title=title)
old_content = old.content
new_content = remove_utf8mb4(item.get(get_redirect=True))
edited = aware(convert_wiki_date(item.editTime()))
print edited
update_content = new_content
if new_content != old_content:
if u"{{пишу}}" in new_content or u"{{message box" in new_content:
print u"×" * 20, u"{{пишу}} detected"
first_edited = None
break
changing_started = False
old_items, old_reports = parse_mass_edit(old_content)
new_items, new_reports = parse_mass_edit(new_content)
if len(old_items) != len(new_items) or set(old_items.keys()) != set(new_items.keys()):
print u"×" * 20, "WRONG DIFFERENCE IN TITLES (COUNTS OR VALUES)"
continue
for word in sorted(old_items.keys(), key=lambda x: x[::-1]):
def load_missed_pages():
print 'load_missed_pages()'
g = AllpagesPageGenerator(start=u"!")
i = 0
j = 0
redirects = []
edits = []
contents = []
# pages = []
print dt(), 'starting cycle?'
for item in g:
j += 1
if not j % 1000:
print dt(), 'j =', j
if j < 178000:
continue
#print item
title = item.title()
if ':' in title:
print dt(), ':' * 20, title
# continue
# print title
try:
# print dt(), title
if title != remove_utf8mb4(title):
print dt(), title, '#' * 30, 'UTF8-MB4'
continue
Page.objects.get(title=title)
# except MultipleObjectsReturned:
# print '#' * 10, 'MULTIPLE'
except ObjectDoesNotExist:
print dt(), title, '(ok)'
try:
content = item.get(get_redirect=True)
except NoPage:
print dt(), '#' * 30, 'NO_PAGE'
continue
content = remove_utf8mb4(content)
redirect = item.isRedirectPage()
# edited = item.editTime()
edited = convert_wiki_date(item.editTime())
# edited = make_naive(wiki_page.editTime(), pytz.UTC)
edited = make_aware(edited, pytz.UTC)
# edited = make_aware(item.editTime(), pytz.UTC)
# edited = make_aware(item.editTime(), None)
# print edited
page = Page.objects.create(title=title)
PageContent.objects.create(page=page, content=content)
PageRedirect.objects.create(page=page, redirect=redirect)
PageEdited.objects.create(page=page, edited=edited)
# contents.append(PageContent(page=page, content=content))
# if len(contents) > 100:
# PageContent.objects.bulk_create(contents)
# contents = []
# print '#' * 30, 'contents added'
#
# redirects.append(PageRedirect(page=page, redirect=redirect))
# if len(redirects) > 1000:
# PageRedirect.objects.bulk_create(redirects)
# redirects = []
# print '#' * 30, 'redirects added'
#
# edits.append(PageEdited(page=page, edited=edited))
# if len(edits) > 1000:
# PageEdited.objects.bulk_create(edits)
# edits = []
# print '#' * 30, 'edits added'
# break
i += 1
if not i % 100:
print dt(), '*' * 20, i
# PageContent.objects.bulk_create(pages)
# pages = []
sleep(1)
def update_missed_data():
print 'update_missed_data()'
i = 0
redirects = []
edits = []
for page in Page.objects.iterate():
i += 1
if i < 35000:
continue
if not i % 1000:
print dt(), 'i =', i
need_content = need_redirect = need_edited = False
try:
PageContent.objects.get(page=page)
except ObjectDoesNotExist:
need_content = True
try:
PageEdited.objects.get(page=page)
except ObjectDoesNotExist:
need_edited = True
try:
PageRedirect.objects.get(page=page)
except ObjectDoesNotExist:
need_redirect = True
if need_content or need_edited or need_redirect:
print dt(), page.title
site = pywikibot.Site('ru')
# wiki_page = pywikibot.Page(site, page.title.decode('utf-8'))
wiki_page = pywikibot.Page(site, page.title)
try:
if need_content:
content = wiki_page.get(get_redirect=True)
content = remove_utf8mb4(content)
PageContent.objects.create(page=page, content=content)
print '- contents added'
if need_redirect:
redirect = wiki_page.isRedirectPage()
PageRedirect.objects.create(page=page, redirect=redirect)
print '- redirect added'
# redirects.append(PageRedirect(page=page,
# redirect=redirect))
# if len(redirects) > 1000:
# PageRedirect.objects.bulk_create(redirects)
# redirects = []
# print '#' * 30, 'redirects added'
if need_edited:
edited = convert_wiki_date(wiki_page.editTime())
# edited = make_naive(wiki_page.editTime(), pytz.UTC)
# edited = make_aware(wiki_page.editTime(), None)
edited = make_aware(edited, pytz.UTC)
PageEdited.objects.create(page=page, edited=edited)
print '- edited added'
# edits.append(PageEdited(page=page, edited=edited))
# if len(edits) > 1000:
# PageEdited.objects.bulk_create(edits)
# edits = []
# print '#' * 30, 'edits added'
except NoPage:
print dt(), '#' * 20, page.title, 'DELETED?' # todo: remove?
content = page.content if not need_content else ''
redirect = page.page_redirect.redirect if not need_redirect else None
edited = page.page_edited.edited if not need_edited else None
PageDeleted.objects.create(page=page.pk, title=page.title,
content=content, redirect=redirect,
edited=edited)
# if not need_content:
# page.page_content.delete()
page.delete()
PageRedirect.objects.bulk_create(redirects)
PageEdited.objects.bulk_create(edits)
def process_recent():
print 'process_recent()'
i = 0
edited = PageEdited.objects.order_by('-edited')[0]
print 'updating untill:', edited.edited
end = datetime(edited.edited.year,
edited.edited.month,
edited.edited.day,
edited.edited.hour,
# edited.edited.minute,
)
g = RecentChangesPageGenerator(end=end)
for item in g:
title = item.title()
if ':' in title:
continue
i += 1
# print title
# print repr(title)
# print repr(title.encode('utf-8'))
# print repr(title.encode('utf-16be'))
# print repr(title.encode('utf-32be'))
# # print repr(title.encode('utf-16').decode('utf-16'))
# page = Page.objects.create(title=title)
# page = Page.objects.create(title=title.encode('utf-16be'))
# page = Page.objects.create(title=title.encode('utf-16'))
# page = Page.objects.create(title=title.encode('utf-8'))
# # page = Page.objects.create(title=title.encode('utf-8'))
# # page, created = Page.objects.get_or_create(title=title)
# continue
try:
page, created = Page.objects.get_or_create(title=title)
except Exception:
print dt(), title, '-', '@' * 30, 'bad title'
continue
try:
edited = convert_wiki_date(item.editTime())
edited = make_aware(edited, pytz.UTC)
except NoPage:
print dt(), title, '-', 'DELETED', '#' * 20
# todo: А если страница еще не была добавлена в БД? Т.е. создание/удаление в пределах одной проверки
content, created = PageContent.objects.get_or_create(page=page)
edited, created = PageEdited.objects.get_or_create(page=page)
redirect, created = PageRedirect.objects.get_or_create(page=page)
PageDeleted.objects.create(page=page.pk, title=page.title,
content=content.content,
edited=edited.edited,
redirect=redirect.redirect)
page.delete()
continue
content = item.get(get_redirect=True)
content = remove_utf8mb4(content)
redirect = item.isRedirectPage()
print dt(), edited, '-', i, '-', title, '| pk =', page.pk
if created:
print '- page CREATED: %s' % page.pk
# try:
# page = Page.objects.get(title__exact=title)
# except ObjectDoesNotExist:
# page = Page.objects.create(title=title)
data, created = PageContent.objects.get_or_create(page=page)
if data.content != content:
data.content = content
data.save()
print '- contents updated'
data, created = PageRedirect.objects.get_or_create(page=page)
if data.redirect != redirect:
data.redirect = redirect
data.save()
print '- redirect updated'
data, created = PageEdited.objects.get_or_create(page=page)
if data.edited != edited:
data.edited = edited
data.save()
print '- edited updated'
def process_item(self, item, i):
# todo: create external mechanism of pausing work (actual for big processors)
try:
title = item.title()
except InvalidTitle:
print 'Wrong title', '#' * 120
return
if ':' in title: # todo: we need this only for RecentProcessor
if title.startswith(u"Шаблон:"):
# if title.startswith(u"Категория:") or title.startswith(u"Шаблон:"):
# print '-' * 40
pass
else:
return
if self.readonly:
return self.process_item_readonly(item, title, i)
# if Page.objects.filter(title=title):
# print dt(), title, '- exists'
# return
page = self.get_page(title, i)
if not page:
return
content, content_to_db, edited, redirect = self.get_data(item)
if not edited:
# print dt(), title.encode('cp1251'), '-', 'DELETED', '#' * 20
print dt(), '& PAGE WAS DELETED - %d - & %s // pk=%d' \
% (i, transliterate(title), page.pk)
# print dt(), transliterate(title), '-', 'DELETED', '#' * 10
# print dt(), '-', 'DELETED', '#' * 20
page.delete_and_log()
return
# print dt(), edited, '-', i, '-', title.encode('cp1251'), '| pk =', page.pk
if not self.output_interval or not i % self.output_interval:
print dt(),
if ':' in title:
print ':',
else:
print '#',
print edited, '-', i, '-',
if ':' in title:
print ':',
else:
print '#',
print transliterate(title), ' // pk=%s' % page.pk
# print dt(), edited, '-', i, '-', '| pk =', page.pk
log = transliterate(title)
oldest = next(item.revisions(reverseOrder=True, total=1, content=True))
created_at = aware(convert_wiki_date(oldest.timestamp))
created_author = oldest.user
created_lang = '?'
if oldest.text is None:
created_lang = '??'
else:
# print oldest.text
# print repr(oldest.text)
# print
m = re.search(u'\{\{-([-\w]+|Праславянский)-(?:\|[^}]+)?\}\}',
oldest.text, flags=re.MULTILINE | re.UNICODE)
m2 = re.search(u'\{\{заголовок\|(en|tr|jbo|la|it|es|fo|da|de|pt|hr|pl|fi|lv|nl|sv|no|io|gd|az|ms|id|nv|nds|nah|hu|nrm|vls|fur|ga|hu|lb|hsb|li|tpi|gv|fr|cy|fy|sc|fo|zu|sw|mg|oc|ca|qu|ln|eo|so|cs|uz|et|vo|ku|su|sk|mi|kw|bar|br|an|sq|bs|af)\|add=\w*\}\}', oldest.text)
m_new = re.search(u'\{\{NEW\|lang=([-a-z]+)\|cat=', oldest.text, flags=re.UNICODE | re.IGNORECASE)
if m:
created_lang = m.group(1)
elif re.search(u'^= *Праславянский', oldest.text, flags=re.MULTILINE | re.UNICODE):
created_lang = u'Праславянский'
elif re.search(u'=<div style="background-color0?:#\{\{h1c\}\}">Эсперанто</div>=', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = u'eo'
elif m2:
created_lang = m2.group(1)
elif re.search(u'== *\[\[?(:w)?:en:[^|]+\|английский\]\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'en'
elif re.search(u'== *\[\[?(:w)?:de:[^|]+\|немецкий\]\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'de'
elif re.search(u'== *\[?\[?(английский|english)\]?\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'en'
elif re.search(u'== *\[?\[?(французский)\]?\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'fr'
elif re.search(u'== *\[?\[?(италь?янский)\]?\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'it'
elif re.search(u'== *\[?\[?(Нидерландский)\]?\]? *== *\n', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'nl'
elif re.search(u'\{\{(английский|en)\}\}', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'en'
elif re.search(u'\{\{(Нидерландский|nl)\}\}', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'nl'
elif re.search(u'\{\{(немецкий|de)\}\}', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'de'
elif re.search(u'\{\{(it)\}\}', oldest.text, flags=re.UNICODE | re.IGNORECASE):
created_lang = 'it'
elif m_new:
created_lang = m_new.group(1)
elif re.search(u'#(redirect|перенаправление)', oldest.text, flags=re.MULTILINE | re.UNICODE | re.IGNORECASE):
created_lang = u'-'
else:
save_file(settings.FILES_PATH + "/errors/created_lang/%s.txt" % page.pk, oldest.text.encode('utf-8'))
# print
# print transliterate(title), created_at
# print transliterate(oldest.user), transliterate(created_lang)
# print
self.update_data(page, content, content_to_db, edited, redirect, log,
created_at, created_author, created_lang)
if ':' in title:
return
return edited
