\d{2})(\.\d{3,6}Z?)?$')
-LOCALS = threading.local()
-LOCALS.RAM_STORE = None
-
-
-class WhooshHtmlFormatter(HtmlFormatter):
- """
- This is a HtmlFormatter simpler than the whoosh.HtmlFormatter.
- We use it to have consistent results across backends. Specifically,
- Solr, Xapian and Elasticsearch are using this formatting.
- """
- template = '<%(tag)s>%(t)s'
-
-
-class WhooshSearchBackend(BaseSearchBackend):
- # Word reserved by Whoosh for special use.
- RESERVED_WORDS = (
- 'AND',
- 'NOT',
- 'OR',
- 'TO',
- )
-
- # Characters reserved by Whoosh for special use.
- # The '\\' must come first, so as not to overwrite the other slash
- # replacements.
- RESERVED_CHARACTERS = (
- '\\', '+', '-', '&&', '||', '!', '(', ')', '{', '}',
- '[', ']', '^', '"', '~', '*', '?', ':', '.',
- )
-
- def __init__(self, connection_alias, **connection_options):
- super(
- WhooshSearchBackend,
- self).__init__(
- connection_alias,
- **connection_options)
- self.setup_complete = False
- self.use_file_storage = True
- self.post_limit = getattr(
- connection_options,
- 'POST_LIMIT',
- 128 * 1024 * 1024)
- self.path = connection_options.get('PATH')
-
- if connection_options.get('STORAGE', 'file') != 'file':
- self.use_file_storage = False
-
- if self.use_file_storage and not self.path:
- raise ImproperlyConfigured(
- "You must specify a 'PATH' in your settings for connection '%s'." %
- connection_alias)
-
- self.log = logging.getLogger('haystack')
-
- def setup(self):
- """
- Defers loading until needed.
- """
- from haystack import connections
- new_index = False
-
- # Make sure the index is there.
- if self.use_file_storage and not os.path.exists(self.path):
- os.makedirs(self.path)
- new_index = True
-
- if self.use_file_storage and not os.access(self.path, os.W_OK):
- raise IOError(
- "The path to your Whoosh index '%s' is not writable for the current user/group." %
- self.path)
-
- if self.use_file_storage:
- self.storage = FileStorage(self.path)
- else:
- global LOCALS
-
- if getattr(LOCALS, 'RAM_STORE', None) is None:
- LOCALS.RAM_STORE = RamStorage()
-
- self.storage = LOCALS.RAM_STORE
-
- self.content_field_name, self.schema = self.build_schema(
- connections[self.connection_alias].get_unified_index().all_searchfields())
- self.parser = QueryParser(self.content_field_name, schema=self.schema)
-
- if new_index is True:
- self.index = self.storage.create_index(self.schema)
- else:
- try:
- self.index = self.storage.open_index(schema=self.schema)
- except index.EmptyIndexError:
- self.index = self.storage.create_index(self.schema)
-
- self.setup_complete = True
-
- def build_schema(self, fields):
- schema_fields = {
- ID: WHOOSH_ID(stored=True, unique=True),
- DJANGO_CT: WHOOSH_ID(stored=True),
- DJANGO_ID: WHOOSH_ID(stored=True),
- }
- # Grab the number of keys that are hard-coded into Haystack.
- # We'll use this to (possibly) fail slightly more gracefully later.
- initial_key_count = len(schema_fields)
- content_field_name = ''
-
- for field_name, field_class in fields.items():
- if field_class.is_multivalued:
- if field_class.indexed is False:
- schema_fields[field_class.index_fieldname] = IDLIST(
- stored=True, field_boost=field_class.boost)
- else:
- schema_fields[field_class.index_fieldname] = KEYWORD(
- stored=True, commas=True, scorable=True, field_boost=field_class.boost)
- elif field_class.field_type in ['date', 'datetime']:
- schema_fields[field_class.index_fieldname] = DATETIME(
- stored=field_class.stored, sortable=True)
- elif field_class.field_type == 'integer':
- schema_fields[field_class.index_fieldname] = NUMERIC(
- stored=field_class.stored, numtype=int, field_boost=field_class.boost)
- elif field_class.field_type == 'float':
- schema_fields[field_class.index_fieldname] = NUMERIC(
- stored=field_class.stored, numtype=float, field_boost=field_class.boost)
- elif field_class.field_type == 'boolean':
- # Field boost isn't supported on BOOLEAN as of 1.8.2.
- schema_fields[field_class.index_fieldname] = BOOLEAN(
- stored=field_class.stored)
- elif field_class.field_type == 'ngram':
- schema_fields[field_class.index_fieldname] = NGRAM(
- minsize=3, maxsize=15, stored=field_class.stored, field_boost=field_class.boost)
- elif field_class.field_type == 'edge_ngram':
- schema_fields[field_class.index_fieldname] = NGRAMWORDS(minsize=2, maxsize=15, at='start',
- stored=field_class.stored,
- field_boost=field_class.boost)
- else:
- # schema_fields[field_class.index_fieldname] = TEXT(stored=True, analyzer=StemmingAnalyzer(), field_boost=field_class.boost, sortable=True)
- schema_fields[field_class.index_fieldname] = TEXT(
- stored=True, analyzer=ChineseAnalyzer(), field_boost=field_class.boost, sortable=True)
- if field_class.document is True:
- content_field_name = field_class.index_fieldname
- schema_fields[field_class.index_fieldname].spelling = True
-
- # Fail more gracefully than relying on the backend to die if no fields
- # are found.
- if len(schema_fields) <= initial_key_count:
- raise SearchBackendError(
- "No fields were found in any search_indexes. Please correct this before attempting to search.")
-
- return (content_field_name, Schema(**schema_fields))
-
- def update(self, index, iterable, commit=True):
- if not self.setup_complete:
- self.setup()
-
- self.index = self.index.refresh()
- writer = AsyncWriter(self.index)
-
- for obj in iterable:
- try:
- doc = index.full_prepare(obj)
- except SkipDocument:
- self.log.debug(u"Indexing for object `%s` skipped", obj)
- else:
- # Really make sure it's unicode, because Whoosh won't have it any
- # other way.
- for key in doc:
- doc[key] = self._from_python(doc[key])
-
- # Document boosts aren't supported in Whoosh 2.5.0+.
- if 'boost' in doc:
- del doc['boost']
-
- try:
- writer.update_document(**doc)
- except Exception as e:
- if not self.silently_fail:
- raise
-
- # We'll log the object identifier but won't include the actual object
- # to avoid the possibility of that generating encoding errors while
- # processing the log message:
- self.log.error(
- u"%s while preparing object for update" %
- e.__class__.__name__,
- exc_info=True,
- extra={
- "data": {
- "index": index,
- "object": get_identifier(obj)}})
-
- if len(iterable) > 0:
- # For now, commit no matter what, as we run into locking issues
- # otherwise.
- writer.commit()
-
- def remove(self, obj_or_string, commit=True):
- if not self.setup_complete:
- self.setup()
-
- self.index = self.index.refresh()
- whoosh_id = get_identifier(obj_or_string)
-
- try:
- self.index.delete_by_query(
- q=self.parser.parse(
- u'%s:"%s"' %
- (ID, whoosh_id)))
- except Exception as e:
- if not self.silently_fail:
- raise
-
- self.log.error(
- "Failed to remove document '%s' from Whoosh: %s",
- whoosh_id,
- e,
- exc_info=True)
-
- def clear(self, models=None, commit=True):
- if not self.setup_complete:
- self.setup()
-
- self.index = self.index.refresh()
-
- if models is not None:
- assert isinstance(models, (list, tuple))
-
- try:
- if models is None:
- self.delete_index()
- else:
- models_to_delete = []
-
- for model in models:
- models_to_delete.append(
- u"%s:%s" %
- (DJANGO_CT, get_model_ct(model)))
-
- self.index.delete_by_query(
- q=self.parser.parse(
- u" OR ".join(models_to_delete)))
- except Exception as e:
- if not self.silently_fail:
- raise
-
- if models is not None:
- self.log.error(
- "Failed to clear Whoosh index of models '%s': %s",
- ','.join(models_to_delete),
- e,
- exc_info=True)
- else:
- self.log.error(
- "Failed to clear Whoosh index: %s", e, exc_info=True)
-
- def delete_index(self):
- # Per the Whoosh mailing list, if wiping out everything from the index,
- # it's much more efficient to simply delete the index files.
- if self.use_file_storage and os.path.exists(self.path):
- shutil.rmtree(self.path)
- elif not self.use_file_storage:
- self.storage.clean()
-
- # Recreate everything.
- self.setup()
-
- def optimize(self):
- if not self.setup_complete:
- self.setup()
-
- self.index = self.index.refresh()
- self.index.optimize()
-
- def calculate_page(self, start_offset=0, end_offset=None):
- # Prevent against Whoosh throwing an error. Requires an end_offset
- # greater than 0.
- if end_offset is not None and end_offset <= 0:
- end_offset = 1
-
- # Determine the page.
- page_num = 0
-
- if end_offset is None:
- end_offset = 1000000
-
- if start_offset is None:
- start_offset = 0
-
- page_length = end_offset - start_offset
-
- if page_length and page_length > 0:
- page_num = int(start_offset / page_length)
-
- # Increment because Whoosh uses 1-based page numbers.
- page_num += 1
- return page_num, page_length
-
- @log_query
- def search(
- self,
- query_string,
- sort_by=None,
- start_offset=0,
- end_offset=None,
- fields='',
- highlight=False,
- facets=None,
- date_facets=None,
- query_facets=None,
- narrow_queries=None,
- spelling_query=None,
- within=None,
- dwithin=None,
- distance_point=None,
- models=None,
- limit_to_registered_models=None,
- result_class=None,
- **kwargs):
- if not self.setup_complete:
- self.setup()
-
- # A zero length query should return no results.
- if len(query_string) == 0:
- return {
- 'results': [],
- 'hits': 0,
- }
-
- query_string = force_str(query_string)
-
- # A one-character query (non-wildcard) gets nabbed by a stopwords
- # filter and should yield zero results.
- if len(query_string) <= 1 and query_string != u'*':
- return {
- 'results': [],
- 'hits': 0,
- }
-
- reverse = False
-
- if sort_by is not None:
- # Determine if we need to reverse the results and if Whoosh can
- # handle what it's being asked to sort by. Reversing is an
- # all-or-nothing action, unfortunately.
- sort_by_list = []
- reverse_counter = 0
-
- for order_by in sort_by:
- if order_by.startswith('-'):
- reverse_counter += 1
-
- if reverse_counter and reverse_counter != len(sort_by):
- raise SearchBackendError("Whoosh requires all order_by fields"
- " to use the same sort direction")
-
- for order_by in sort_by:
- if order_by.startswith('-'):
- sort_by_list.append(order_by[1:])
-
- if len(sort_by_list) == 1:
- reverse = True
- else:
- sort_by_list.append(order_by)
-
- if len(sort_by_list) == 1:
- reverse = False
-
- sort_by = sort_by_list[0]
-
- if facets is not None:
- warnings.warn(
- "Whoosh does not handle faceting.",
- Warning,
- stacklevel=2)
-
- if date_facets is not None:
- warnings.warn(
- "Whoosh does not handle date faceting.",
- Warning,
- stacklevel=2)
-
- if query_facets is not None:
- warnings.warn(
- "Whoosh does not handle query faceting.",
- Warning,
- stacklevel=2)
-
- narrowed_results = None
- self.index = self.index.refresh()
-
- if limit_to_registered_models is None:
- limit_to_registered_models = getattr(
- settings, 'HAYSTACK_LIMIT_TO_REGISTERED_MODELS', True)
-
- if models and len(models):
- model_choices = sorted(get_model_ct(model) for model in models)
- elif limit_to_registered_models:
- # Using narrow queries, limit the results to only models handled
- # with the current routers.
- model_choices = self.build_models_list()
- else:
- model_choices = []
-
- if len(model_choices) > 0:
- if narrow_queries is None:
- narrow_queries = set()
-
- narrow_queries.add(' OR '.join(
- ['%s:%s' % (DJANGO_CT, rm) for rm in model_choices]))
-
- narrow_searcher = None
-
- if narrow_queries is not None:
- # Potentially expensive? I don't see another way to do it in
- # Whoosh...
- narrow_searcher = self.index.searcher()
-
- for nq in narrow_queries:
- recent_narrowed_results = narrow_searcher.search(
- self.parser.parse(force_str(nq)), limit=None)
-
- if len(recent_narrowed_results) <= 0:
- return {
- 'results': [],
- 'hits': 0,
- }
-
- if narrowed_results:
- narrowed_results.filter(recent_narrowed_results)
- else:
- narrowed_results = recent_narrowed_results
-
- self.index = self.index.refresh()
-
- if self.index.doc_count():
- searcher = self.index.searcher()
- parsed_query = self.parser.parse(query_string)
-
- # In the event of an invalid/stopworded query, recover gracefully.
- if parsed_query is None:
- return {
- 'results': [],
- 'hits': 0,
- }
-
- page_num, page_length = self.calculate_page(
- start_offset, end_offset)
-
- search_kwargs = {
- 'pagelen': page_length,
- 'sortedby': sort_by,
- 'reverse': reverse,
- }
-
- # Handle the case where the results have been narrowed.
- if narrowed_results is not None:
- search_kwargs['filter'] = narrowed_results
-
- try:
- raw_page = searcher.search_page(
- parsed_query,
- page_num,
- **search_kwargs
- )
- except ValueError:
- if not self.silently_fail:
- raise
-
- return {
- 'results': [],
- 'hits': 0,
- 'spelling_suggestion': None,
- }
-
- # Because as of Whoosh 2.5.1, it will return the wrong page of
- # results if you request something too high. :(
- if raw_page.pagenum < page_num:
- return {
- 'results': [],
- 'hits': 0,
- 'spelling_suggestion': None,
- }
-
- results = self._process_results(
- raw_page,
- highlight=highlight,
- query_string=query_string,
- spelling_query=spelling_query,
- result_class=result_class)
- searcher.close()
-
- if hasattr(narrow_searcher, 'close'):
- narrow_searcher.close()
-
- return results
- else:
- if self.include_spelling:
- if spelling_query:
- spelling_suggestion = self.create_spelling_suggestion(
- spelling_query)
- else:
- spelling_suggestion = self.create_spelling_suggestion(
- query_string)
- else:
- spelling_suggestion = None
-
- return {
- 'results': [],
- 'hits': 0,
- 'spelling_suggestion': spelling_suggestion,
- }
-
- def more_like_this(
- self,
- model_instance,
- additional_query_string=None,
- start_offset=0,
- end_offset=None,
- models=None,
- limit_to_registered_models=None,
- result_class=None,
- **kwargs):
- if not self.setup_complete:
- self.setup()
-
- # Deferred models will have a different class ("RealClass_Deferred_fieldname")
- # which won't be in our registry:
- model_klass = model_instance._meta.concrete_model
-
- field_name = self.content_field_name
- narrow_queries = set()
- narrowed_results = None
- self.index = self.index.refresh()
-
- if limit_to_registered_models is None:
- limit_to_registered_models = getattr(
- settings, 'HAYSTACK_LIMIT_TO_REGISTERED_MODELS', True)
-
- if models and len(models):
- model_choices = sorted(get_model_ct(model) for model in models)
- elif limit_to_registered_models:
- # Using narrow queries, limit the results to only models handled
- # with the current routers.
- model_choices = self.build_models_list()
- else:
- model_choices = []
-
- if len(model_choices) > 0:
- if narrow_queries is None:
- narrow_queries = set()
-
- narrow_queries.add(' OR '.join(
- ['%s:%s' % (DJANGO_CT, rm) for rm in model_choices]))
-
- if additional_query_string and additional_query_string != '*':
- narrow_queries.add(additional_query_string)
-
- narrow_searcher = None
-
- if narrow_queries is not None:
- # Potentially expensive? I don't see another way to do it in
- # Whoosh...
- narrow_searcher = self.index.searcher()
-
- for nq in narrow_queries:
- recent_narrowed_results = narrow_searcher.search(
- self.parser.parse(force_str(nq)), limit=None)
-
- if len(recent_narrowed_results) <= 0:
- return {
- 'results': [],
- 'hits': 0,
- }
-
- if narrowed_results:
- narrowed_results.filter(recent_narrowed_results)
- else:
- narrowed_results = recent_narrowed_results
-
- page_num, page_length = self.calculate_page(start_offset, end_offset)
-
- self.index = self.index.refresh()
- raw_results = EmptyResults()
-
- if self.index.doc_count():
- query = "%s:%s" % (ID, get_identifier(model_instance))
- searcher = self.index.searcher()
- parsed_query = self.parser.parse(query)
- results = searcher.search(parsed_query)
-
- if len(results):
- raw_results = results[0].more_like_this(
- field_name, top=end_offset)
-
- # Handle the case where the results have been narrowed.
- if narrowed_results is not None and hasattr(raw_results, 'filter'):
- raw_results.filter(narrowed_results)
-
- try:
- raw_page = ResultsPage(raw_results, page_num, page_length)
- except ValueError:
- if not self.silently_fail:
- raise
-
- return {
- 'results': [],
- 'hits': 0,
- 'spelling_suggestion': None,
- }
-
- # Because as of Whoosh 2.5.1, it will return the wrong page of
- # results if you request something too high. :(
- if raw_page.pagenum < page_num:
- return {
- 'results': [],
- 'hits': 0,
- 'spelling_suggestion': None,
- }
-
- results = self._process_results(raw_page, result_class=result_class)
- searcher.close()
-
- if hasattr(narrow_searcher, 'close'):
- narrow_searcher.close()
-
- return results
-
- def _process_results(
- self,
- raw_page,
- highlight=False,
- query_string='',
- spelling_query=None,
- result_class=None):
- from haystack import connections
- results = []
-
- # It's important to grab the hits first before slicing. Otherwise, this
- # can cause pagination failures.
- hits = len(raw_page)
-
- if result_class is None:
- result_class = SearchResult
-
- facets = {}
- spelling_suggestion = None
- unified_index = connections[self.connection_alias].get_unified_index()
- indexed_models = unified_index.get_indexed_models()
-
- for doc_offset, raw_result in enumerate(raw_page):
- score = raw_page.score(doc_offset) or 0
- app_label, model_name = raw_result[DJANGO_CT].split('.')
- additional_fields = {}
- model = haystack_get_model(app_label, model_name)
-
- if model and model in indexed_models:
- for key, value in raw_result.items():
- index = unified_index.get_index(model)
- string_key = str(key)
-
- if string_key in index.fields and hasattr(
- index.fields[string_key], 'convert'):
- # Special-cased due to the nature of KEYWORD fields.
- if index.fields[string_key].is_multivalued:
- if value is None or len(value) == 0:
- additional_fields[string_key] = []
- else:
- additional_fields[string_key] = value.split(
- ',')
- else:
- additional_fields[string_key] = index.fields[string_key].convert(
- value)
- else:
- additional_fields[string_key] = self._to_python(value)
-
- del (additional_fields[DJANGO_CT])
- del (additional_fields[DJANGO_ID])
-
- if highlight:
- sa = StemmingAnalyzer()
- formatter = WhooshHtmlFormatter('em')
- terms = [token.text for token in sa(query_string)]
-
- whoosh_result = whoosh_highlight(
- additional_fields.get(self.content_field_name),
- terms,
- sa,
- ContextFragmenter(),
- formatter
- )
- additional_fields['highlighted'] = {
- self.content_field_name: [whoosh_result],
- }
-
- result = result_class(
- app_label,
- model_name,
- raw_result[DJANGO_ID],
- score,
- **additional_fields)
- results.append(result)
- else:
- hits -= 1
-
- if self.include_spelling:
- if spelling_query:
- spelling_suggestion = self.create_spelling_suggestion(
- spelling_query)
- else:
- spelling_suggestion = self.create_spelling_suggestion(
- query_string)
-
- return {
- 'results': results,
- 'hits': hits,
- 'facets': facets,
- 'spelling_suggestion': spelling_suggestion,
- }
-
- def create_spelling_suggestion(self, query_string):
- spelling_suggestion = None
- reader = self.index.reader()
- corrector = reader.corrector(self.content_field_name)
- cleaned_query = force_str(query_string)
-
- if not query_string:
- return spelling_suggestion
-
- # Clean the string.
- for rev_word in self.RESERVED_WORDS:
- cleaned_query = cleaned_query.replace(rev_word, '')
-
- for rev_char in self.RESERVED_CHARACTERS:
- cleaned_query = cleaned_query.replace(rev_char, '')
-
- # Break it down.
- query_words = cleaned_query.split()
- suggested_words = []
-
- for word in query_words:
- suggestions = corrector.suggest(word, limit=1)
-
- if len(suggestions) > 0:
- suggested_words.append(suggestions[0])
-
- spelling_suggestion = ' '.join(suggested_words)
- return spelling_suggestion
-
- def _from_python(self, value):
- """
- Converts Python values to a string for Whoosh.
-
- Code courtesy of pysolr.
- """
- if hasattr(value, 'strftime'):
- if not hasattr(value, 'hour'):
- value = datetime(value.year, value.month, value.day, 0, 0, 0)
- elif isinstance(value, bool):
- if value:
- value = 'true'
- else:
- value = 'false'
- elif isinstance(value, (list, tuple)):
- value = u','.join([force_str(v) for v in value])
- elif isinstance(value, (six.integer_types, float)):
- # Leave it alone.
- pass
- else:
- value = force_str(value)
- return value
-
- def _to_python(self, value):
- """
- Converts values from Whoosh to native Python values.
-
- A port of the same method in pysolr, as they deal with data the same way.
- """
- if value == 'true':
- return True
- elif value == 'false':
- return False
-
- if value and isinstance(value, six.string_types):
- possible_datetime = DATETIME_REGEX.search(value)
-
- if possible_datetime:
- date_values = possible_datetime.groupdict()
-
- for dk, dv in date_values.items():
- date_values[dk] = int(dv)
-
- return datetime(
- date_values['year'],
- date_values['month'],
- date_values['day'],
- date_values['hour'],
- date_values['minute'],
- date_values['second'])
-
- try:
- # Attempt to use json to load the values.
- converted_value = json.loads(value)
-
- # Try to handle most built-in types.
- if isinstance(
- converted_value,
- (list,
- tuple,
- set,
- dict,
- six.integer_types,
- float,
- complex)):
- return converted_value
- except BaseException:
- # If it fails (SyntaxError or its ilk) or we don't trust it,
- # continue on.
- pass
-
- return value
-
-
-class WhooshSearchQuery(BaseSearchQuery):
- def _convert_datetime(self, date):
- if hasattr(date, 'hour'):
- return force_str(date.strftime('%Y%m%d%H%M%S'))
- else:
- return force_str(date.strftime('%Y%m%d000000'))
-
- def clean(self, query_fragment):
- """
- Provides a mechanism for sanitizing user input before presenting the
- value to the backend.
-
- Whoosh 1.X differs here in that you can no longer use a backslash
- to escape reserved characters. Instead, the whole word should be
- quoted.
- """
- words = query_fragment.split()
- cleaned_words = []
-
- for word in words:
- if word in self.backend.RESERVED_WORDS:
- word = word.replace(word, word.lower())
-
- for char in self.backend.RESERVED_CHARACTERS:
- if char in word:
- word = "'%s'" % word
- break
-
- cleaned_words.append(word)
-
- return ' '.join(cleaned_words)
-
- def build_query_fragment(self, field, filter_type, value):
- from haystack import connections
- query_frag = ''
- is_datetime = False
-
- if not hasattr(value, 'input_type_name'):
- # Handle when we've got a ``ValuesListQuerySet``...
- if hasattr(value, 'values_list'):
- value = list(value)
-
- if hasattr(value, 'strftime'):
- is_datetime = True
-
- if isinstance(value, six.string_types) and value != ' ':
- # It's not an ``InputType``. Assume ``Clean``.
- value = Clean(value)
- else:
- value = PythonData(value)
-
- # Prepare the query using the InputType.
- prepared_value = value.prepare(self)
-
- if not isinstance(prepared_value, (set, list, tuple)):
- # Then convert whatever we get back to what pysolr wants if needed.
- prepared_value = self.backend._from_python(prepared_value)
-
- # 'content' is a special reserved word, much like 'pk' in
- # Django's ORM layer. It indicates 'no special field'.
- if field == 'content':
- index_fieldname = ''
- else:
- index_fieldname = u'%s:' % connections[self._using].get_unified_index(
- ).get_index_fieldname(field)
-
- filter_types = {
- 'content': '%s',
- 'contains': '*%s*',
- 'endswith': "*%s",
- 'startswith': "%s*",
- 'exact': '%s',
- 'gt': "{%s to}",
- 'gte': "[%s to]",
- 'lt': "{to %s}",
- 'lte': "[to %s]",
- 'fuzzy': u'%s~',
- }
-
- if value.post_process is False:
- query_frag = prepared_value
- else:
- if filter_type in [
- 'content',
- 'contains',
- 'startswith',
- 'endswith',
- 'fuzzy']:
- if value.input_type_name == 'exact':
- query_frag = prepared_value
- else:
- # Iterate over terms & incorportate the converted form of
- # each into the query.
- terms = []
-
- if isinstance(prepared_value, six.string_types):
- possible_values = prepared_value.split(' ')
- else:
- if is_datetime is True:
- prepared_value = self._convert_datetime(
- prepared_value)
-
- possible_values = [prepared_value]
-
- for possible_value in possible_values:
- terms.append(
- filter_types[filter_type] %
- self.backend._from_python(possible_value))
-
- if len(terms) == 1:
- query_frag = terms[0]
- else:
- query_frag = u"(%s)" % " AND ".join(terms)
- elif filter_type == 'in':
- in_options = []
-
- for possible_value in prepared_value:
- is_datetime = False
-
- if hasattr(possible_value, 'strftime'):
- is_datetime = True
-
- pv = self.backend._from_python(possible_value)
-
- if is_datetime is True:
- pv = self._convert_datetime(pv)
-
- if isinstance(pv, six.string_types) and not is_datetime:
- in_options.append('"%s"' % pv)
- else:
- in_options.append('%s' % pv)
-
- query_frag = "(%s)" % " OR ".join(in_options)
- elif filter_type == 'range':
- start = self.backend._from_python(prepared_value[0])
- end = self.backend._from_python(prepared_value[1])
-
- if hasattr(prepared_value[0], 'strftime'):
- start = self._convert_datetime(start)
-
- if hasattr(prepared_value[1], 'strftime'):
- end = self._convert_datetime(end)
-
- query_frag = u"[%s to %s]" % (start, end)
- elif filter_type == 'exact':
- if value.input_type_name == 'exact':
- query_frag = prepared_value
- else:
- prepared_value = Exact(prepared_value).prepare(self)
- query_frag = filter_types[filter_type] % prepared_value
- else:
- if is_datetime is True:
- prepared_value = self._convert_datetime(prepared_value)
-
- query_frag = filter_types[filter_type] % prepared_value
-
- if len(query_frag) and not isinstance(value, Raw):
- if not query_frag.startswith('(') and not query_frag.endswith(')'):
- query_frag = "(%s)" % query_frag
-
- return u"%s%s" % (index_fieldname, query_frag)
-
- # if not filter_type in ('in', 'range'):
- # # 'in' is a bit of a special case, as we don't want to
- # # convert a valid list/tuple to string. Defer handling it
- # # until later...
- # value = self.backend._from_python(value)
-
-
-class WhooshEngine(BaseEngine):
- backend = WhooshSearchBackend
- query = WhooshSearchQuery
diff --git a/src/DjangoBlog/djangoblog/wsgi.py b/src/DjangoBlog/djangoblog/wsgi.py
deleted file mode 100644
index 2295efd..0000000
--- a/src/DjangoBlog/djangoblog/wsgi.py
+++ /dev/null
@@ -1,16 +0,0 @@
-"""
-WSGI config for djangoblog project.
-
-It exposes the WSGI callable as a module-level variable named ``application``.
-
-For more information on this file, see
-https://docs.djangoproject.com/en/1.10/howto/deployment/wsgi/
-"""
-
-import os
-
-from django.core.wsgi import get_wsgi_application
-
-os.environ.setdefault("DJANGO_SETTINGS_MODULE", "djangoblog.settings")
-
-application = get_wsgi_application()
diff --git a/src/DjangoBlog/doc/LICENSE b/src/DjangoBlog/doc/LICENSE
deleted file mode 100644
index 3b08474..0000000
--- a/src/DjangoBlog/doc/LICENSE
+++ /dev/null
@@ -1,20 +0,0 @@
-The MIT License (MIT)
-
-Copyright (c) 2025 车亮亮
-
-Permission is hereby granted, free of charge, to any person obtaining a copy of
-this software and associated documentation files (the "Software"), to deal in
-the Software without restriction, including without limitation the rights to
-use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
-the Software, and to permit persons to whom the Software is furnished to do so,
-subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
-FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
-COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
-IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
-CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
\ No newline at end of file
diff --git a/src/DjangoBlog/doc/README.md b/src/DjangoBlog/doc/README.md
deleted file mode 100644
index 56aa4cc..0000000
--- a/src/DjangoBlog/doc/README.md
+++ /dev/null
@@ -1,158 +0,0 @@
-# DjangoBlog
-
-
- 
- 
- 
- 
-
-
-
- 一款功能强大、设计优雅的现代化博客系统
-
- English • 简体中文
-
-
----
-
-DjangoBlog 是一款基于 Python 3.10 和 Django 4.0 构建的高性能博客平台。它不仅提供了传统博客的所有核心功能，还通过一个灵活的插件系统，让您可以轻松扩展和定制您的网站。无论您是个人博主、技术爱好者还是内容创作者，DjangoBlog 都旨在为您提供一个稳定、高效且易于维护的写作和发布环境。
-
-## ✨ 特性亮点
-
-- **强大的内容管理**: 支持文章、独立页面、分类和标签的完整管理。内置强大的 Markdown 编辑器，支持代码语法高亮。
-- **全文搜索**: 集成搜索引擎，提供快速、精准的文章内容搜索。
-- **互动评论系统**: 支持回复、邮件提醒等功能，评论内容同样支持 Markdown。
-- **灵活的侧边栏**: 可自定义展示最新文章、最多阅读、标签云等模块。
-- **社交化登录**: 内置 OAuth 支持，已集成 Google, GitHub, Facebook, 微博, QQ 等主流平台。
-- **高性能缓存**: 原生支持 Redis 缓存，并提供自动刷新机制，确保网站高速响应。
-- **SEO 友好**: 具备基础 SEO 功能，新内容发布后可自动通知 Google 和百度。
-- **便捷的插件系统**: 通过创建独立的插件来扩展博客功能，代码解耦，易于维护。我们已经通过插件实现了文章浏览计数、SEO 优化等功能！
-- **集成图床**: 内置简单的图床功能，方便图片上传和管理。
-- **自动化前端**: 集成 `django-compressor`，自动压缩和优化 CSS 及 JavaScript 文件。
-- **健壮的运维**: 内置网站异常邮件提醒和微信公众号管理功能。
-
-## 🛠️ 技术栈
-
-- **后端**: Python 3.10, Django 4.0
-- **数据库**: MySQL, SQLite (可配置)
-- **缓存**: Redis
-- **前端**: HTML5, CSS3, JavaScript
-- **搜索**: Whoosh, Elasticsearch (可配置)
-- **编辑器**: Markdown (mdeditor)
-
-## 🚀 快速开始
-
-### 1. 环境准备
-
-确保您的系统中已安装 Python 3.10+ 和 MySQL/MariaDB。
-
-### 2. 克隆与安装
-
-```bash
-# 克隆项目到本地
-git clone https://github.com/liangliangyy/DjangoBlog.git
-cd DjangoBlog
-
-# 安装依赖
-pip install -r requirements.txt
-```
-
-### 3. 项目配置
-
-- **数据库**:
- 打开 `djangoblog/settings.py` 文件，找到 `DATABASES` 配置项，修改为您的 MySQL 连接信息。
-
- ```python
- DATABASES = {
- 'default': {
- 'ENGINE': 'django.db.backends.mysql',
- 'NAME': 'djangoblog',
- 'USER': 'root',
- 'PASSWORD': 'your_password',
- 'HOST': '127.0.0.1',
- 'PORT': 3306,
- }
- }
- ```
- 在 MySQL 中创建数据库:
- ```sql
- CREATE DATABASE `djangoblog` DEFAULT CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
- ```
-
-- **更多配置**:
- 关于邮件发送、OAuth 登录、缓存等更多高级配置，请参阅我们的 [详细配置文档](/docs/config.md)。
-
-### 4. 初始化数据库
-
-```bash
-python manage.py makemigrations
-python manage.py migrate
-
-# 创建一个超级管理员账户
-python manage.py createsuperuser
-```
-
-### 5. 运行项目
-
-```bash
-# (可选) 生成一些测试数据
-python manage.py create_testdata
-
-# (可选) 收集和压缩静态文件
-python manage.py collectstatic --noinput
-python manage.py compress --force
-
-# 启动开发服务器
-python manage.py runserver
-```
-
-现在，在您的浏览器中访问 `http://127.0.0.1:8000/`，您应该能看到 DjangoBlog 的首页了！
-
-## 部署
-
-- **传统部署**: 我们为您准备了非常详细的 [服务器部署教程](https://www.lylinux.net/article/2019/8/5/58.html)。
-- **Docker 部署**: 项目已全面支持 Docker。如果您熟悉容器化技术，请参考 [Docker 部署文档](/docs/docker.md) 来快速启动。
-- **Kubernetes 部署**: 我们也提供了完整的 [Kubernetes 部署指南](/docs/k8s.md)，助您轻松上云。
-
-## 🧩 插件系统
-
-插件系统是 DjangoBlog 的核心特色之一。它允许您在不修改核心代码的情况下，通过编写独立的插件来为您的博客添加新功能。
-
-- **工作原理**: 插件通过在预定义的“钩子”上注册回调函数来工作。例如，当一篇文章被渲染时，`after_article_body_get` 钩子会被触发，所有注册到此钩子的函数都会被执行。
-- **现有插件**: `view_count`（浏览计数）, `seo_optimizer`（SEO优化）等都是通过插件系统实现的。
-- **开发您自己的插件**: 只需在 `plugins` 目录下创建一个新的文件夹，并编写您的 `plugin.py`。欢迎探索并为 DjangoBlog 社区贡献您的创意！
-
-## 🤝 贡献指南
-
-我们热烈欢迎任何形式的贡献！如果您有好的想法或发现了 Bug，请随时提交 Issue 或 Pull Request。
-
-## 📄 许可证
-
-本项目基于 [MIT License](LICENSE) 开源。
-
----
-
-## ❤️ 支持与赞助
-
-如果您觉得这个项目对您有帮助，并且希望支持我继续维护和开发新功能，欢迎请我喝杯咖啡！您的每一份支持都是我前进的最大动力。
-
-
- 
- 
-
-
- (左) 支付宝 / (右) 微信
-
-
-## 🙏 鸣谢
-
-特别感谢 **JetBrains** 为本项目提供的免费开源许可证。
-
-
-
- 
-
-
-
----
-> 如果本项目帮助到了你，请在[这里](https://github.com/liangliangyy/DjangoBlog/issues/214)留下你的网址，让更多的人看到。您的回复将会是我继续更新维护下去的动力。
diff --git a/src/DjangoBlog/docs/README-en.md b/src/DjangoBlog/docs/README-en.md
deleted file mode 100644
index 37ea069..0000000
--- a/src/DjangoBlog/docs/README-en.md
+++ /dev/null
@@ -1,158 +0,0 @@
-# DjangoBlog
-
-
-
-
-
-
-
-
-
- A powerful, elegant, and modern blog system.
-
- English • 简体中文
-
-
----
-
-DjangoBlog is a high-performance blog platform built with Python 3.10 and Django 4.0. It not only provides all the core functionalities of a traditional blog but also features a flexible plugin system, allowing you to easily extend and customize your website. Whether you are a personal blogger, a tech enthusiast, or a content creator, DjangoBlog aims to provide a stable, efficient, and easy-to-maintain environment for writing and publishing.
-
-## ✨ Features
-
-- **Powerful Content Management**: Full support for managing articles, standalone pages, categories, and tags. Comes with a powerful built-in Markdown editor with syntax highlighting.
-- **Full-Text Search**: Integrated search engine for fast and accurate content searching.
-- **Interactive Comment System**: Supports replies, email notifications, and Markdown formatting in comments.
-- **Flexible Sidebar**: Customizable modules for displaying recent articles, most viewed posts, tag cloud, and more.
-- **Social Login**: Built-in OAuth support, with integrations for Google, GitHub, Facebook, Weibo, QQ, and other major platforms.
-- **High-Performance Caching**: Native support for Redis caching with an automatic refresh mechanism to ensure high-speed website responses.
-- **SEO Friendly**: Basic SEO features are included, with automatic notifications to Google and Baidu upon new content publication.
-- **Extensible Plugin System**: Extend blog functionalities by creating standalone plugins, ensuring decoupled and maintainable code. We have already implemented features like view counting and SEO optimization through plugins!
-- **Integrated Image Hosting**: A simple, built-in image hosting feature for easy uploads and management.
-- **Automated Frontend**: Integrated with `django-compressor` to automatically compress and optimize CSS and JavaScript files.
-- **Robust Operations**: Built-in email notifications for website exceptions and management capabilities through a WeChat Official Account.
-
-## 🛠️ Tech Stack
-
-- **Backend**: Python 3.10, Django 4.0
-- **Database**: MySQL, SQLite (configurable)
-- **Cache**: Redis
-- **Frontend**: HTML5, CSS3, JavaScript
-- **Search**: Whoosh, Elasticsearch (configurable)
-- **Editor**: Markdown (mdeditor)
-
-## 🚀 Getting Started
-
-### 1. Prerequisites
-
-Ensure you have Python 3.10+ and MySQL/MariaDB installed on your system.
-
-### 2. Clone & Installation
-
-```bash
-# Clone the project to your local machine
-git clone https://github.com/liangliangyy/DjangoBlog.git
-cd DjangoBlog
-
-# Install dependencies
-pip install -r requirements.txt
-```
-
-### 3. Project Configuration
-
-- **Database**:
- Open `djangoblog/settings.py`, locate the `DATABASES` section, and update it with your MySQL connection details.
-
- ```python
- DATABASES = {
- 'default': {
- 'ENGINE': 'django.db.backends.mysql',
- 'NAME': 'djangoblog',
- 'USER': 'root',
- 'PASSWORD': 'your_password',
- 'HOST': '127.0.0.1',
- 'PORT': 3306,
- }
- }
- ```
- Create the database in MySQL:
- ```sql
- CREATE DATABASE `djangoblog` DEFAULT CHARACTER SET utf8mb4 COLLATE utf8mb4_unicode_ci;
- ```
-
-- **More Configurations**:
- For advanced settings such as email, OAuth, caching, and more, please refer to our [Detailed Configuration Guide](/docs/config-en.md).
-
-### 4. Database Initialization
-
-```bash
-python manage.py makemigrations
-python manage.py migrate
-
-# Create a superuser account
-python manage.py createsuperuser
-```
-
-### 5. Running the Project
-
-```bash
-# (Optional) Generate some test data
-python manage.py create_testdata
-
-# (Optional) Collect and compress static files
-python manage.py collectstatic --noinput
-python manage.py compress --force
-
-# Start the development server
-python manage.py runserver
-```
-
-Now, open your browser and navigate to `http://127.0.0.1:8000/`. You should see the DjangoBlog homepage!
-
-## Deployment
-
-- **Traditional Deployment**: A detailed guide for server deployment is available here: [Deployment Tutorial](https://www.lylinux.net/article/2019/8/5/58.html) (in Chinese).
-- **Docker Deployment**: This project fully supports Docker. If you are familiar with containerization, please refer to the [Docker Deployment Guide](/docs/docker-en.md) for a quick start.
-- **Kubernetes Deployment**: We also provide a complete [Kubernetes Deployment Guide](/docs/k8s-en.md) to help you go cloud-native easily.
-
-## 🧩 Plugin System
-
-The plugin system is a core feature of DjangoBlog. It allows you to add new functionalities to your blog without modifying the core codebase by writing standalone plugins.
-
-- **How it Works**: Plugins operate by registering callback functions to predefined "hooks". For instance, when an article is rendered, the `after_article_body_get` hook is triggered, and all functions registered to this hook are executed.
-- **Existing Plugins**: Features like `view_count` and `seo_optimizer` are implemented through this plugin system.
-- **Develop Your Own Plugin**: Simply create a new folder under the `plugins` directory and write your `plugin.py`. We welcome you to explore and contribute your creative ideas to the DjangoBlog community!
-
-## 🤝 Contributing
-
-We warmly welcome contributions of any kind! If you have great ideas or have found a bug, please feel free to open an issue or submit a pull request.
-
-## 📄 License
-
-This project is open-sourced under the [MIT License](LICENSE).
-
----
-
-## ❤️ Support & Sponsorship
-
-If you find this project helpful and wish to support its continued maintenance and development, please consider buying me a coffee! Your support is my greatest motivation.
-
-
-
-
-
-
- (Left) Alipay / (Right) WeChat
-
-
-## 🙏 Acknowledgements
-
-A special thanks to **JetBrains** for providing a free open-source license for this project.
-
-
-
-
-
-
-
----
-> If this project has helped you, please leave your website URL [here](https://github.com/liangliangyy/DjangoBlog/issues/214) to let more people see it. Your feedback is the driving force for my continued updates and maintenance.
diff --git a/src/DjangoBlog/docs/config-en.md b/src/DjangoBlog/docs/config-en.md
deleted file mode 100644
index b877efb..0000000
--- a/src/DjangoBlog/docs/config-en.md
+++ /dev/null
@@ -1,64 +0,0 @@
-# Introduction to main features settings
-
-## Cache:
-Cache using `memcache` for default. If you don't have `memcache` environment, you can remove the `default` setting in `CACHES` and change `locmemcache` to `default`.
-```python
-CACHES = {
- 'default': {
- 'BACKEND': 'django.core.cache.backends.memcached.MemcachedCache',
- 'LOCATION': '127.0.0.1:11211',
- 'KEY_PREFIX': 'django_test' if TESTING else 'djangoblog',
- 'TIMEOUT': 60 * 60 * 10
- },
- 'locmemcache': {
- 'BACKEND': 'django.core.cache.backends.locmem.LocMemCache',
- 'TIMEOUT': 10800,
- 'LOCATION': 'unique-snowflake',
- }
-}
-```
-
-## OAuth Login:
-QQ, Weibo, Google, GitHub and Facebook are now supported for OAuth login. Fetch OAuth login permissions from the corresponding open platform, and save them with `appkey`, `appsecret` and callback address in **Backend->OAuth** configuration.
-
-### Callback address examples:
-QQ: http://your-domain-name/oauth/authorize?type=qq
-Weibo: http://your-domain-name/oauth/authorize?type=weibo
-type is in the type field of `oauthmanager`.
-
-## owntracks:
-owntracks is a location tracking application. It will send your locaiton to the server by timing.Simple support owntracks features. Just install owntracks app and set api address as `your-domain-name/owntracks/logtracks`. Visit `your-domain-name/owntracks/show_dates` and you will see the date with latitude and langitude, click it and see the motion track. The map is drawn by AMap.
-
-## Email feature:
-Same as before, Configure your own error msg recvie email information with`ADMINS = [('liangliang', 'liangliangyy@gmail.com')]` in `settings.py`. And modify:
-```python
-EMAIL_HOST = 'smtp.zoho.com'
-EMAIL_PORT = 587
-EMAIL_HOST_USER = os.environ.get('DJANGO_EMAIL_USER')
-EMAIL_HOST_PASSWORD = os.environ.get('DJANGO_EMAIL_PASSWORD')
-DEFAULT_FROM_EMAIL = EMAIL_HOST_USER
-SERVER_EMAIL = os.environ.get('DJANGO_EMAIL_USER')
-```
-with your email account information.
-
-## WeChat Official Account
-Simple wechat official account features integrated. Set token as `your-domain-name/robot` in wechat backend. Default token is `lylinux`, you can change it to your own in `servermanager/robot.py`. Add a new command in `Backend->Servermanager->command`, in this way, you can manage the system through wechat official account.
-
-## Introduction to website configuration
-You can add website configuration in **Backend->BLOG->WebSiteConfiguration**. Such as: keywords, description, Google Ad, website stats code, case number, etc.
-OAuth user avatar path is saved in *StaticFileSavedAddress*. Please input absolute path, code directory for default.
-
-## Source code highlighting
-If the code block in your article didn't show hightlight, please write the code blocks as following:
-
-
-
-That is, you should add the corresponding language name before the code block.
-
-## Update
-If you get errors as following while executing database migrations:
-```python
-django.db.migrations.exceptions.MigrationSchemaMissing: Unable to create the django_migrations table ((1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '(6) NOT NULL)' at line 1"))
-```
-This problem may cause by the mysql version under 5.6, a new version( >= 5.6 ) mysql is needed.
-
diff --git a/src/DjangoBlog/docs/config.md b/src/DjangoBlog/docs/config.md
deleted file mode 100644
index 24673a3..0000000
--- a/src/DjangoBlog/docs/config.md
+++ /dev/null
@@ -1,58 +0,0 @@
-# 主要功能配置介绍:
-
-## 缓存：
-缓存默认使用`localmem`缓存，如果你有`redis`环境，可以设置`DJANGO_REDIS_URL`环境变量，则会自动使用该redis来作为缓存，或者你也可以直接修改如下代码来使用。
-https://github.com/liangliangyy/DjangoBlog/blob/ffcb2c3711de805f2067dd3c1c57449cd24d84ee/djangoblog/settings.py#L185-L199
-
-
-## oauth登录:
-
-现在已经支持QQ，微博，Google，GitHub，Facebook登录，需要在其对应的开放平台申请oauth登录权限，然后在
-**后台->Oauth** 配置中新增配置，填写对应的`appkey`和`appsecret`以及回调地址。
-### 回调地址示例：
-qq：http://你的域名/oauth/authorize?type=qq
-微博：http://你的域名/oauth/authorize?type=weibo
-type对应在`oauthmanager`中的type字段。
-
-## owntracks：
-owntracks是一个位置追踪软件，可以定时的将你的坐标提交到你的服务器上，现在简单的支持owntracks功能，需要安装owntracks的app，然后将api地址设置为:
-`你的域名/owntracks/logtracks`就可以了。然后访问`你的域名/owntracks/show_dates`就可以看到有经纬度记录的日期，点击之后就可以看到运动轨迹了。地图是使用高德地图绘制。
-
-## 邮件功能：
-同样，将`settings.py`中的`ADMINS = [('liangliang', 'liangliangyy@gmail.com')]`配置为你自己的错误接收邮箱，另外修改:
-```python
-EMAIL_HOST = 'smtp.zoho.com'
-EMAIL_PORT = 587
-EMAIL_HOST_USER = os.environ.get('DJANGO_EMAIL_USER')
-EMAIL_HOST_PASSWORD = os.environ.get('DJANGO_EMAIL_PASSWORD')
-DEFAULT_FROM_EMAIL = EMAIL_HOST_USER
-SERVER_EMAIL = os.environ.get('DJANGO_EMAIL_USER')
-```
-为你自己的邮箱配置。
-
-## 微信公众号
-集成了简单的微信公众号功能，在微信后台将token地址设置为:`你的域名/robot` 即可，默认token为`lylinux`，当然你可以修改为你自己的，在`servermanager/robot.py`中。
-然后在**后台->Servermanager->命令**中新增命令，这样就可以使用微信公众号来管理了。
-## 网站配置介绍
-在**后台->BLOG->网站配置**中,可以新增网站配置，比如关键字，描述等，以及谷歌广告，网站统计代码及备案号等等。
-其中的*静态文件保存地址*是保存oauth用户登录的头像路径，填写绝对路径，默认是代码目录。
-## 代码高亮
-如果你发现你文章的代码没有高亮，请这样书写代码块:
-
-
-
-
-也就是说，需要在代码块开始位置加入这段代码对应的语言。
-
-## update
-如果你发现执行数据库迁移的时候出现如下报错：
-```python
-django.db.migrations.exceptions.MigrationSchemaMissing: Unable to create the django_migrations table ((1064, "You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '(6) NOT NULL)' at line 1"))
-```
-可能是因为你的mysql版本低于5.6，需要升级mysql版本>=5.6即可。
-
-
-django 4.0登录可能会报错CSRF，需要配置下`settings.py`中的`CSRF_TRUSTED_ORIGINS`
-
-https://github.com/liangliangyy/DjangoBlog/blob/master/djangoblog/settings.py#L39
-
diff --git a/src/DjangoBlog/docs/docker-en.md b/src/DjangoBlog/docs/docker-en.md
deleted file mode 100644
index 8d5d59e..0000000
--- a/src/DjangoBlog/docs/docker-en.md
+++ /dev/null
@@ -1,114 +0,0 @@
-# Deploying DjangoBlog with Docker
-
-
-
-
-
-This project fully supports containerized deployment using Docker, providing you with a fast, consistent, and isolated runtime environment. We recommend using `docker-compose` to launch the entire blog service stack with a single command.
-
-## 1. Prerequisites
-
-Before you begin, please ensure you have the following software installed on your system:
-- [Docker Engine](https://docs.docker.com/engine/install/)
-- [Docker Compose](https://docs.docker.com/compose/install/) (Included with Docker Desktop for Mac and Windows)
-
-## 2. Recommended Method: Using `docker-compose` (One-Click Deployment)
-
-This is the simplest and most recommended way to deploy. It automatically creates and manages the Django application, a MySQL database, and an optional Elasticsearch service for you.
-
-### Step 1: Start the Basic Services
-
-From the project's root directory, run the following command:
-
-```bash
-# Build and start the containers in detached mode (includes Django app and MySQL)
-docker-compose up -d --build
-```
-
-`docker-compose` will read the `docker-compose.yml` file, pull the necessary images, build the project image, and start all services.
-
-- **Access Your Blog**: Once the services are up, you can access the blog by navigating to `http://127.0.0.1` in your browser.
-- **Data Persistence**: MySQL data files will be stored in the `data/mysql` directory within the project root, ensuring that your data persists across container restarts.
-
-### Step 2: (Optional) Enable Elasticsearch for Full-Text Search
-
-If you want to use Elasticsearch for more powerful full-text search capabilities, you can include the `docker-compose.es.yml` configuration file:
-
-```bash
-# Build and start all services in detached mode (Django, MySQL, Elasticsearch)
-docker-compose -f docker-compose.yml -f deploy/docker-compose/docker-compose.es.yml up -d --build
-```
-- **Data Persistence**: Elasticsearch data will be stored in the `data/elasticsearch` directory.
-
-### Step 3: First-Time Initialization
-
-After the containers start for the first time, you'll need to execute some initialization commands inside the application container.
-
-```bash
-# Get a shell inside the djangoblog application container (named 'web')
-docker-compose exec web bash
-
-# Inside the container, run the following commands:
-# Create a superuser account (follow the prompts to set username, email, and password)
-python manage.py createsuperuser
-
-# (Optional) Create some test data
-python manage.py create_testdata
-
-# (Optional, if ES is enabled) Create the search index
-python manage.py rebuild_index
-
-# Exit the container
-exit
-```
-
-## 3. Alternative Method: Using the Standalone Docker Image
-
-If you already have an external MySQL database running, you can run the DjangoBlog application image by itself.
-
-```bash
-# Pull the latest image from Docker Hub
-docker pull liangliangyy/djangoblog:latest
-
-# Run the container and connect it to your external database
-docker run -d \
- -p 8000:8000 \
- -e DJANGO_SECRET_KEY='your-strong-secret-key' \
- -e DJANGO_MYSQL_HOST='your-mysql-host' \
- -e DJANGO_MYSQL_USER='your-mysql-user' \
- -e DJANGO_MYSQL_PASSWORD='your-mysql-password' \
- -e DJANGO_MYSQL_DATABASE='djangoblog' \
- --name djangoblog \
- liangliangyy/djangoblog:latest
-```
-
-- **Access Your Blog**: After startup, visit `http://127.0.0.1:8000`.
-- **Create Superuser**: `docker exec -it djangoblog python manage.py createsuperuser`
-
-## 4. Configuration (Environment Variables)
-
-Most of the project's configuration is managed through environment variables. You can modify them in the `docker-compose.yml` file or pass them using the `-e` flag with the `docker run` command.
-
-| Environment Variable | Default/Example Value | Notes |
-|---------------------------|--------------------------------------------------------------------------|---------------------------------------------------------------------|
-| `DJANGO_SECRET_KEY` | `your-strong-secret-key` | **Must be changed to a random, complex string!** |
-| `DJANGO_DEBUG` | `False` | Toggles Django's debug mode. |
-| `DJANGO_MYSQL_HOST` | `mysql` | Database hostname. |
-| `DJANGO_MYSQL_PORT` | `3306` | Database port. |
-| `DJANGO_MYSQL_DATABASE` | `djangoblog` | Database name. |
-| `DJANGO_MYSQL_USER` | `root` | Database username. |
-| `DJANGO_MYSQL_PASSWORD` | `djangoblog_123` | Database password. |
-| `DJANGO_REDIS_URL` | `redis:6379/0` | Redis connection URL (for caching). |
-| `DJANGO_ELASTICSEARCH_HOST`| `elasticsearch:9200` | Elasticsearch host address. |
-| `DJANGO_EMAIL_HOST` | `smtp.example.org` | Email server address. |
-| `DJANGO_EMAIL_PORT` | `465` | Email server port. |
-| `DJANGO_EMAIL_USER` | `user@example.org` | Email account username. |
-| `DJANGO_EMAIL_PASSWORD` | `your-email-password` | Email account password. |
-| `DJANGO_EMAIL_USE_SSL` | `True` | Whether to use SSL. |
-| `DJANGO_EMAIL_USE_TLS` | `False` | Whether to use TLS. |
-| `DJANGO_ADMIN_EMAIL` | `admin@example.org` | Admin email for receiving error reports. |
-| `DJANGO_BAIDU_NOTIFY_URL` | `http://data.zz.baidu.com/...` | Push API from [Baidu Webmaster Tools](https://ziyuan.baidu.com/linksubmit/index). |
-
----
-
-After deployment, please review and adjust these environment variables according to your needs, especially `DJANGO_SECRET_KEY` and the database and email settings.
\ No newline at end of file
diff --git a/src/DjangoBlog/docs/docker.md b/src/DjangoBlog/docs/docker.md
deleted file mode 100644
index e7c255a..0000000
--- a/src/DjangoBlog/docs/docker.md
+++ /dev/null
@@ -1,114 +0,0 @@
-# 使用 Docker 部署 DjangoBlog
-
-
-
-
-
-本项目全面支持使用 Docker 进行容器化部署，为您提供了快速、一致且隔离的运行环境。我们推荐使用 `docker-compose` 来一键启动整个博客服务栈。
-
-## 1. 环境准备
-
-在开始之前，请确保您的系统中已经安装了以下软件：
-- [Docker Engine](https://docs.docker.com/engine/install/)
-- [Docker Compose](https://docs.docker.com/compose/install/) (对于 Docker Desktop 用户，它已内置)
-
-## 2. 推荐方式：使用 `docker-compose` (一键部署)
-
-这是最简单、最推荐的部署方式。它会自动为您创建并管理 Django 应用、MySQL 数据库，以及可选的 Elasticsearch 服务。
-
-### 步骤 1: 启动基础服务
-
-在项目根目录下，执行以下命令：
-
-```bash
-# 构建并以后台模式启动容器 (包含 Django 应用和 MySQL)
-docker-compose up -d --build
-```
-
-`docker-compose` 会读取 `docker-compose.yml` 文件，自动拉取所需镜像、构建项目镜像，并启动所有服务。
-
-- **访问您的博客**: 服务启动后，在浏览器中访问 `http://127.0.0.1` 即可看到博客首页。
-- **数据持久化**: MySQL 的数据文件将存储在项目根目录下的 `data/mysql` 文件夹中，确保数据在容器重启后不丢失。
-
-### 步骤 2: (可选) 启用 Elasticsearch 全文搜索
-
-如果您希望使用 Elasticsearch 提供更强大的全文搜索功能，可以额外加载 `docker-compose.es.yml` 配置文件：
-
-```bash
-# 构建并以后台模式启动所有服务 (Django, MySQL, Elasticsearch)
-docker-compose -f docker-compose.yml -f deploy/docker-compose/docker-compose.es.yml up -d --build
-```
-- **数据持久化**: Elasticsearch 的数据将存储在 `data/elasticsearch` 文件夹中。
-
-### 步骤 3: 首次运行的初始化操作
-
-当容器首次启动后，您需要进入容器来执行一些初始化命令。
-
-```bash
-# 进入 djangoblog 应用容器
-docker-compose exec web bash
-
-# 在容器内执行以下命令:
-# 创建超级管理员账户 (请按照提示设置用户名、邮箱和密码)
-python manage.py createsuperuser
-
-# (可选) 创建一些测试数据
-python manage.py create_testdata
-
-# (可选，如果启用了 ES) 创建索引
-python manage.py rebuild_index
-
-# 退出容器
-exit
-```
-
-## 3. 备选方式：使用独立的 Docker 镜像
-
-如果您已经拥有一个正在运行的外部 MySQL 数据库，您也可以只运行 DjangoBlog 的应用镜像。
-
-```bash
-# 从 Docker Hub 拉取最新镜像
-docker pull liangliangyy/djangoblog:latest
-
-# 运行容器，并链接到您的外部数据库
-docker run -d \
- -p 8000:8000 \
- -e DJANGO_SECRET_KEY='your-strong-secret-key' \
- -e DJANGO_MYSQL_HOST='your-mysql-host' \
- -e DJANGO_MYSQL_USER='your-mysql-user' \
- -e DJANGO_MYSQL_PASSWORD='your-mysql-password' \
- -e DJANGO_MYSQL_DATABASE='djangoblog' \
- --name djangoblog \
- liangliangyy/djangoblog:latest
-```
-
-- **访问您的博客**: 启动完成后，访问 `http://127.0.0.1:8000`。
-- **创建管理员**: `docker exec -it djangoblog python manage.py createsuperuser`
-
-## 4. 配置说明 (环境变量)
-
-本项目的大部分配置都通过环境变量来管理。您可以在 `docker-compose.yml` 文件中修改它们，或者在使用 `docker run` 命令时通过 `-e` 参数传入。
-
-| 环境变量名称 | 默认值/示例 | 备注 |
-|-------------------------|--------------------------------------------------------------------------|---------------------------------------------------------------------|
-| `DJANGO_SECRET_KEY` | `your-strong-secret-key` | **请务必修改为一个随机且复杂的字符串！** |
-| `DJANGO_DEBUG` | `False` | 是否开启 Django 的调试模式 |
-| `DJANGO_MYSQL_HOST` | `mysql` | 数据库主机名 |
-| `DJANGO_MYSQL_PORT` | `3306` | 数据库端口 |
-| `DJANGO_MYSQL_DATABASE` | `djangoblog` | 数据库名称 |
-| `DJANGO_MYSQL_USER` | `root` | 数据库用户名 |
-| `DJANGO_MYSQL_PASSWORD` | `djangoblog_123` | 数据库密码 |
-| `DJANGO_REDIS_URL` | `redis:6379/0` | Redis 连接地址 (用于缓存) |
-| `DJANGO_ELASTICSEARCH_HOST` | `elasticsearch:9200` | Elasticsearch 主机地址 |
-| `DJANGO_EMAIL_HOST` | `smtp.example.org` | 邮件服务器地址 |
-| `DJANGO_EMAIL_PORT` | `465` | 邮件服务器端口 |
-| `DJANGO_EMAIL_USER` | `user@example.org` | 邮件账户 |
-| `DJANGO_EMAIL_PASSWORD` | `your-email-password` | 邮件密码 |
-| `DJANGO_EMAIL_USE_SSL` | `True` | 是否使用 SSL |
-| `DJANGO_EMAIL_USE_TLS` | `False` | 是否使用 TLS |
-| `DJANGO_ADMIN_EMAIL` | `admin@example.org` | 接收异常报告的管理员邮箱 |
-| `DJANGO_BAIDU_NOTIFY_URL` | `http://data.zz.baidu.com/...` | [百度站长平台](https://ziyuan.baidu.com/linksubmit/index) 的推送接口 |
-
----
-
-部署完成后，请务必检查并根据您的实际需求调整这些环境变量，特别是 `DJANGO_SECRET_KEY` 和数据库、邮件相关的配置。
diff --git a/src/DjangoBlog/docs/es.md b/src/DjangoBlog/docs/es.md
deleted file mode 100644
index 97226c5..0000000
--- a/src/DjangoBlog/docs/es.md
+++ /dev/null
@@ -1,28 +0,0 @@
-# 集成Elasticsearch
-如果你已经有了`Elasticsearch`环境，那么可以将搜索从`Whoosh`换成`Elasticsearch`，集成方式也很简单，
-首先需要注意如下几点:
-1. 你的`Elasticsearch`支持`ik`中文分词
-2. 你的`Elasticsearch`版本>=7.3.0
-
-接下来在`settings.py`做如下改动即可：
-- 增加es链接，如下所示：
-```python
-ELASTICSEARCH_DSL = {
- 'default': {
- 'hosts': '127.0.0.1:9200'
- },
-}
-```
-- 修改`HAYSTACK`配置：
-```python
-HAYSTACK_CONNECTIONS = {
- 'default': {
- 'ENGINE': 'djangoblog.elasticsearch_backend.ElasticSearchEngine',
- },
-}
-```
-然后终端执行:
-```shell script
-./manage.py build_index
-```
-这将会在你的es中创建两个索引，分别是`blog`和`performance`，其中`blog`索引就是搜索所使用的，而`performance`会记录每个请求的响应时间，以供将来优化使用。
\ No newline at end of file
diff --git a/src/DjangoBlog/docs/imgs/alipay.jpg b/src/DjangoBlog/docs/imgs/alipay.jpg
deleted file mode 100644
index 424d70a2ffbb629b481e0c27d72d6076727e8041..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001
literal 17961
zcmcJ%1z1$w*Y`hk4B$wLgu@LetspQo3`$A2q)LZ$NQk5;NOyN5ASECmq9ENJ(p`cQ
zD)pb4GkAO7&+~hq>-oRe3+I|^=Ir5|9qa79)@OY;=i}#dAjmywIcX3E1_*?K_ywI$
zf$o43WvM?$PLN;@xT5#{{Xs-kI{uOfQ9iF2=g)q
z)@6+I7SL4?2onog+rNKsu&{A4k!5fQ2m=cPlMoLN3mX#)goSJpCKfi%Wl{e36aR6Wg
z2=J#H4*&7Lt&InbnnzMz%c1mDyZ3W;?fm)p*bIn6y{af(4(q`cOXe|8ol0^lNCdh?
zO=jBiI)l!k51Zd@U9QH$zDXhxvG^|X9EAVBZ0_(rf6f7)V$fCy@2uB{*$5kndHj3q
zHgfAl>0~t2*0S$*Jzun9IWztHYf2D&jcvC4L5Z=?SLIx0Mjy9wp!@t%|3xLyeU7!y
zgzokbpwA^v32=dK&jtprUtAK79=rc?ksTVdEV+b_?LJOEFsvS|7G$^JW0%ej7=Qmo
zKEPhHfB)p{IVL#+tfFAPAdf4uCgm>}HX)>{wlrDWeknv-vAEU|o+LTA@Xp5IQxvxW
zf9ZS&t|eGLc)j`wiEOL$Wc>>!>Kslk_Pmj4K{l-tCf?!Zfj5S$29j$zYLuJu=gs-0`X+gm94OHxcN|W1iINFu+
zdk+p{Dp>xNWv}b(UhSp3MyNUv4N}Ha3|Z~aMgn!XEvu`Ks-~O`B@bK8=g-qL(NC7~
z-YPdRj_=@D#z|5?AoAUU3Ah#u4Cd>(ulOrGPy8G!G}Xszd@t2zHHnx<$963Q(HoUl
zH8$-;MvAB(JYdf*fK``#tQRLIR8Pj@J=WVO1#X`I;2ae8VPg@G-+jV}s%wNu)F5LvHIN>&@_N(_(VxD!u*Q1Ly5c{b5%R
zuEsvRXr@`mQGh4ihY}u~xSf}M;+e8!$4~B>r}yXJV{_v0r7Hb$o7Rb`B8zy!>hOOa
z#NQSEmkp>5S*p6XQCN)x3b{|VPlB4?pM#E!3gvU9R%unNBLbk104Qm_w^VrFn)+R|
zf#ItkgYP^^*(JdAeq>?$Jq-hks{6k@1c%D4xm&Vu(EW|K&f=GL$JbYmhw|eaRCqMXXA+S
zF+<_l`s;4%8?@`hAKq@yYrh^``CcESk}(K$hJm$4s~RJBEiFy!>`qGwK1K@X%bt_h
zZoQNr%+6Fi0vIsCPZaDT9N1DxQt`U&C>*~G(Jay4%fFKk|4!42TUh5;V$rf*F!LEa
zExYi#=9Xbb_rkkWgEqv2i>!@hgf9gS(XM8N%$w<(xO1M89sy9a@%xI`I7_>q5=!`^
zpVd6JtD>aW7~OE#E6K_CQu*KShFB_=G&t#Gr{eS04kqilt)NfjK!^AT@}2y37v#36
zC>%^#oY%yR-AdG*u<>oaaKDGQ6!A`K7&^2X);NFuiAP$1vLo`V!~bC_I{VoFWmN=I
zB7Iz!reW0w?)ck4bl+{`ddKa6S+^ZV{fdM2(>Y0$85tu0mkElsjyZxLa-
zwP+}ngS2UEWDSD@$gozuf4%?d=RLR`nWF16+L46olzKwcwoEBEkJgf5w`76FWz>ab
z>@0Q|>RlrBpVIp>Xk!PdWLLEIe5Y8lKxvALE-Ee3Qm)S$
zcZ+qw6x>S8N&Sk)n4)a{Y~(X@deaNuQ+$Y#&hZZ?>EE9$1NDA5X>+$Q3Q(O8EF}ta
z-(O6{O^d3oHXSF1!34Ditx>a7N`|(j&NodByf;h4SixmDvApU$3O`72w^D{$pu1*<
znfN%;Swq2T=6C1ptQso$YZ^=^1+a1IqCz+OZ7
zW2sUcv`hZ92i$6WNesk+UW9n>ZJ$z#i)TtaM?ZHFHx4NSzY5Ml^h^8Af7;Gfg!O8@
z%?#AFWV9J;%)yQsABs+Y!3%~YF8tkBV2Z9AGncoXVCJ_}9__B=XIGl#7n$Q01C3c>XR}&qI@kegI0gia4aYW2WaI*E|-6jR-3{_TFOLyl?rE7*b4^+wDB<1k2Pi
zu|?I06_**St~>KZ@%(kQU*_Qs_->50PSmibm{v-dm@j6w`t9efKS>li$@@tAWvNCPAB{9PKcIBxw2cCf&Y)3KbSSjehi~3PL?(q>+
zW;AbXnq0}j=~YleN=}|YlIxt_iW&*Hd`VAPExVTUH+el8
z!_Ck7J_h6U|8+Rf)`ydKkNtIr6LiB_&P({B
zk_t9W+dC(vS=HSY75r}u7eEcqRA7=A4a^6l;Tj1t{@xYDN6?1>$@JcKPnNl}Q-;-p
zWI@6_V0;zLzLCN&oMT^}kk_CJao7O|BQNEE!}uJuH0%E&UVQJ%t~BG>$~ovGuCoV|
znDj28N%B<}yS0b&&d+*&h)xKP1hd~A>G8P&!u;kN3@Owea}l{@cEaAT?`l-uCgc?M
zwZk`k{--o^6BorS5&J3Rx@6-6NtRD`4$Pgwc);UL##;f*_)Q|IM3faQsIQ}6{JmGb0vTfr_!5YNIc`yZ!{U>!+^4N;*gf_u
z+)Gq^rS*LCxXPLFpS&l-0v`~cgRDlKx!oQgtM<3fS8l_j*GzY353^C_)u}{=G-AW3
zVil8L)K6ij1*qZ@^E*_5ebscRVrntqGD67mVja|FWKiY*<84Z$=wDGa``D_XZlNs!
z;=-KsK@HG&J#rzz@+9bam`w1&)YhpQc3aBfEL&_j%6fWsaDksDG
zXWXxH)eo`3hWA;(Xjd1s9GoZU2b>J&#Bgwf2c1l^BFk$9TS7u%DVw7GUGq)bT3cg6
zy5m!fuae;Hi*DUjqNwJRZ*^ShWSLy$E2Wwts13>0U(ed3KH@mdXPnSu)LrGvc;y4b
zh+y;Leb0EdM5fuwP!#iW8twq~O=qhyD(^0KrIEKY{kZ4yQ_@n7+$`=dV@u|YDCwuc
zuAe_APk#2_#TS;_9u-`nrHqc82yB^mT;ZE7%-F`~F`~A+Hsn27&6)PsP11_Rmi{Mz!YOo3%p63Zbg)S=VstJ2w`3CY0gGouNm!5AnftgC-X`F8>lA80Oi&~R~q=e
z3Q*-49s9^DsiMktdr{W`ig~<{ml+0bREfGw*gtL~r%3LX_m}aLO}can;9_M&+?O4K
z;@u?)5YEHD4yv1O#_}n);9MoZwT32_G?&$8BVS%8L9B(QuO0_ShFebs47uGuKtJhl
z)^070K}GMi8OjoTo|}HMtzT_ES%sJr^faXrhwaNnShecIo;9|5j?RruYboOMrP>mE
zwKBamoM6*Egh=Dly63hdo@Tghx08;3o4_=e2x0$0E#DDEv?MNNfO1lW93w{0jq4A5
zG(B>DC4clbIey3aB1l_BJugjQebIUvP7Hb#?wDGAS+u#tWof0ok|Ife=7|H__Ve7P
zjtz*oB*gzy!t27Eq;t?f1E(&8rQGEt_$|
zYlmKcZDdzzY_bC3g;jC_G)!(WWcoG-
zor4DLD@UD4rSxc0rULb|ZyNPyPzVFQbiT`71qbU?hif=${c(bbNJoH*lxu4no~q3;
zX*Qy?BC>+q`$6vN1h8$++qiAN7_8wmefS#MQ-yD7WgV&)RxYx
zl@4hH*zGC@1ujPv?E19*E6L5>Dq6)^riLWcr?8^WpZ&H@-1DqIC1kVDpk$yk{g`R(rjXHWGKF2nS{|#3mvsc7~L2C@%^54rm3iUmidV-jaIICiIDpP
z1IDDTX2zO12$y7Xk#b=JBk~1Wu58(*3^ul);*$@{-8V>7{z=)u10qfD<~
z<$kiKirnyBApd7$5oS&9QtNU=PJs5v$+t*{WoW`=1TYhHL;3g^_B`)k6CKfy7{;aD
z-^lVzu-DyHobXh+P034TzS!fyN$yO<$DMAO$YatM{r2Fg<{kUPBMRMrjL022tPf-W5{54<&F&|33lqgqHodXps
z-?P!lE=Vm4{SZP^`yFnG4v>hNP1X&^5xpAPT!wwB<)iC7cFEdp#czWG_$z4?hsZ$0
zjGqm%eWZY?U43LoPDf>Z{)&Rjn2G2ZR{@K;UrCf`;Jh0QYZ4-QK|_l8t4KA}tIJ=t
zyR5FMnDryD)!*Si4*x3H=(aaIKSyVTOl9piU4EKL;uf-T73b&J2de51_rvPJJTIt+
z5YVOHKts{=4kQrMA
zNhlJHr=l{auZQqU=7zFX*MwOBM2AYHNX7!r;Tysfoq$RB)k+O8X{uB|-^T*b7Id-C0(1;!-o(K@D4Q14QUZXu)b)
z_;GFO8Ho!&svK$#a3*xfSfJ85iLanZ-}UNMWw=4GPH{#WUQ3&&8gG^B>bSp{x2o0_
zFt_ec#4^qL*M8Gt`DedgV7DQTo|3QBu3;D$CdVH|!{)RS%^m}PIRu4drMaz7(lOMM
zIhFyf=E_w26J>!5^wGUIh=I=*cRO!v24e8<9x4Q|*S@aC4vaTyF#;rQu*QYvS^w|sa(H+qzzI;oW2P}O?c%}7M*JrGlw)O
z_zQl6l;$Y(IS|U`Qj=!E+pd#hsb13XU7ELWbjCenh{5c6p}P0RjC%=>+=P35cE%HTcpxT5MVeZb!wY(<4N+0K;14l@{kP_&gs;peAZD|#x#s7f
z@@sc`FuTxUN?Kfh6Gt_XXShwX`+!awyO@6zwq@$`+>WIK!agZ8x7WW&|
z|1u*^1+|JsRapuq+jkB!5_^%fbf6=CG{P@&+fZ=Emz1nGuxC$apElN9WuUGlc$ELF
zo7Ufh;585m$LmxzYUgl9R+KO4ubkXefHy$lz{;ndIh64;bfFX-Gy*YZ1_FW@meC}z
zjfRv4hJJli_aTgh^TL9zpeS@klDX0e?91~>KTq`h*LC~Cv@n6tFo)n2y?|rxn-iS=
zgW)PBD&Wi`&^r~b-LgmJ{mu+CmsxPffgu`cvsY4t366A|BS%q_);>iE?(KL9lACL6
zQWO+MRow?dOglO;W0%bBm;8Bi^N@=-I~>62ls3YO+t>nTP@?1r14)Pv@*wT9+zCsA
z-SXLDFRQOuBI>X3vi0(67Al`vXH7YZG6;~K^Nj-H-*D5gnb*FsO9jB~}z%)6>
zDwD5T)C!=kSi_X+&}Ive8HPqfsdmx2pwj@xIS5k6ed~}lIWoV|QCgEZH_~AEwB#3y
zyey1%)rU^@KAn`yX02{FqS6!uQXRJGo?BR&+uT)*%$nO;tUY`^uf0aVOA*?%
zUxRN``wje900=7!0lug&cYC>>cx<{JFr}<&#cm3Dz}qQ7AY9SR?PmK1XdAlM`X(Rp
zRBu~Q@z3Gy3TB{PL&WrS!A-yb7mwh48u?ZNYsyXEMb-rtg9C6-1iIS`v8M6s2N@!K
zz%apn|5W~R29qp({#geqJBeJdmgS?`W17w`k$8bo+)DDxZjDCd
z*yEOiZ;Y-YAX0?Jy`PKLDfmu4P~UhqVoAUZvMtrcimr3}EV&#lX-D|tB0(mf5yA7g
z-ax&e8{pGl*Vl3C^I+`)F;t1xQcR)7dhQOj;ByecOZGp!%>#T%BJ{jV3B*2>0I>wf
zFZoqd98Gyk<4bPsik;4A=1b&)_k{l536FDm^&EuurVhnLc))kWoUlUjwd5rF2GSkK
zFt!|R)4HPL#@Te^xSVTSa^D>P$^wcW8H#np@eA&_^jI_`H?d3H_GY_^0?+t7&xPKs
z*n%_8K@p8(Cp=80z3Revir!r!l7!`NW(W@-{3yeEr5*0>(j|58GfHZ1~nX4|P|@0fR4T1ei%jLJ}3Z1bs%(}>LE&~BaNL}TVt6FwIaZBuZ<^iSe(
zDxAK(froKlA_~$r!8&nh
zoh!X77`3?-=^H{j`=ia`$LF9kFKxTFYl9m_+G-8lJBKgdFH;D&XmYo2?il;BodP3J
zE+tOc4KkZ?XNCUez*-J}Lv<;aL{2I7Wy|)(H8N+Y3
zy79~P(LFGazr3OaPmw+@jB{*DTMZ1BkM7tY+XpXKXNJk#3MpZ_cDUU9_EGK9V#oE1
z)W`bRBU&=IEvD;-xR!@CBiEH4qvl*Z5@gy&j|>%v)MoQ^wogUW`rOpsLn6|m7{G%=
zBs_>34cH3}uN;~FbsXb7sUh~C@(c|4IOGvB42|0jD~a>~gu2UR>^bL`SHiJVX*cwX^i%G(F9wBOE2w%zUNVPg(*3&2pYh7{Uqh}B
zf?c-z5@RiUu|sGx9giLoy-KpDjvy*dyva0E_fBN}i%VD35k)(-il)p%S`7bG(%Jr*
zd8a_BPo|OK3?&U*R%!AQf}L{PHv0wYo>=FgrgKnVky`S76ATq4JM)nz_UP49E!)Dy
z1ko@mADM4uQ=B{+)dN+HwIIXktdpWtM*7a$loi1e>t?1`rG&(goWS`lJN5xZu;P58
zOsn(t;Q3Q&udW%?(2$IJ3>C!24)T?@<8S@IK1K~|A%WT`0@3F163K3nwwouQG*kG-|!%R
z0BLYx8U4&$m0uiLYEIg>g^FweJ&fd^@JpHqqfyT&1q|!*)SuwF-UhOfSC@0y2WQbz
zqwpUr8d5kmk|fO|Uxxj~Wj2Ni5^davfWXDWeA1;bacm5)cBYd4{GaB2-PE}a=OEQ{
zkmJx8%9+=ZF=r88&)vIr#F!V~;yqm}N%G6`smy6R>KqsOaD+=zV5`^IY%?Am3Zyv?
z{s=#gek^m^a}EkH65p{qfvOM5FNp8x>dI#GmXp!Qabic^Uee&cb_6@S94b_^f6M-z
zxY-H)R_YoJKpP)jN73&q0C;V1HHuzDQrD0t&lHHxL6+#8P|NUO18zy1sPyIF=T{|hC34B;lDL7AHt}gYt9(6q1BQCP9)WR`g&q#~<%!Q)y
zIqr|vGrN1t)p-ng7;U&d^PedizP7PN*xo*(m3Xc7Tu%`S3dA@ns`0;xJyNiM`tEtnK(Yn`65H1EyAbDKYcTEv%@%Ae)S={$inbWS`qz9J~gYx*XeBb8m4>VxM5k>a+az-Vi@mjW1(wKn1`(NK2W3>_A
zbn>3(&h{CS`0M)abUQ;;(s)6RvkkF5BfC_aP#>z_;=(Rc(0>jnYk{T5bg3D|82{;d
zAR6OqcrZPn^siHjHdZP*C8FgDpJ>9Xzu^JoE2gr--}0Z5zH%se_>vdb-)qjK3C6c0
zayIj|Q*>{_e_*N4)w|3Qdl6;5a(m1Z_TQg+G&e2!KPkf1d`Er`;@H*k=dDTK
z5Z_UB$k#mw?d^&m9@9+>1l{a|}PhK<52nyU#D%8*uf>__Hjq7dHS
zSweNH$-xeqgu|IJ@XMy%cv2t9@;hKFwq&w0kbg6*<;|156<^Fcz}r?Ak;n6@3N_`R25uwzIZIQ_6a&M3dQTUdd8VNiWffn(4M$wnEdu52(|pKV!+adX0m`P
z`y&$Cx8FFAom93khy8Xb@MZOn;&U$NG7R=@G`_*-Hu76oMFmuPyq;wxs
z6$vHYdbIZSRA(r?aIDqnBgL_Zft10wQ^GUiG}Zq>uF70yrQ({zJbWe?mOb5jMuL)0(afa+?<+blAg
z3M^Ms85DIRbft{H5j}ik{q8P*m$~9)0VM&k#+oCxAINTwP(%9y`LF-kwQJ-32I=9v
z_+$W_j1k(pX`(NTY4^}9-P%h<;Exmr_Wo~&pcC+_bu$y7=uCtv%CguFQ0`8wayGoHK{XGa)$1&vZe;Kxx~N!Y7TmZ@I7Pm52@dU
zoS2AU5-J*S-6p+cbSoZqf4|gL>L@wTqHpy#
z?6q4zL8)*yX1I$e+%DsO`~YI}546v%?wS%k`jSvpWMFX)>Zm!<@`$VMQ}eoqVK7~s
zRJu}1Cp^1*4JZBTSjznuYF8ZYSEsB$=wOlRNCjK-vt(}!e1;P53
zGC}#bnxh>>l5f5>c8=*L!ajbo!+K34UENK$1?m@dRt)KKv%ozKMxecs`EC{CFlV4U
za}Zfaw0Q^ycP7b|ozi?9$;34SIA-Ej^;~krVg5o9&5N4aEVQyDWf==>x*e5U_6Yl-
zZ(C9>O46iZrHe`j`|Bd2P2eaTL4aOa`tJ_!mL2fP`%LeUUy2q3%hliH4OwE>HpuAv
z_NCdBE+qp!wfxD5x+eiE;d2~43aOU?6=sm{$8zOHG$zVXu|Ry&gwheg{7ULh6s~BC
z;T(x_@ZXA+m-5`abmt(wpLrG%Nf|~d^e=A8rC0J#9S`G9ByCN#F^d8OCc+uvVPejY
z3&*oaVu+RVGR_9NVP9w|{MaQ^2O7FR$kWA~5M~Edh1r~gdcA{H)tRx?-w8hQG@SlH
z9rxHuqpF+-|895}W93tdLn2DwDM#(Gf}6PO6;H{3EJPotTgmFarz(EQY>J3&CN=_Kq^#>$8&UsxPZm9rX>VNBQ(o|F|NCmWkMAjZ^cx9zsoB>13J
zXpvha*4}Jx9h@L%iuDY;&B{_n6X%ZnLu_1quD_P()CQ&QbtUi|)~>R-Y&Ua6jpjJn
zir(DilDe(INh6{5usqR3Q27I!lO7C)o5V9aeUi)WIekPW5Ve>Rb0>j;GF!;96r1*1
z{%IjM2i=K#7LZ#u^EG4kZwlvYsZz)Y8UE~*_vERg>r676YfXL5ns2G`J`T6hpVeh5SO>Kz=%e$^z|b+z15lb%k|YI@
zbkrBda|9+SH50%r5h-T|Y>G*Q$HO^IsXEa)f_BKXpJn-`CY!fuD;a)|5eOq(eEc$D
zZl(4S@61iSq;KfDqZkatM}FIKR+-2yWLh2(cV0+1!2nwWpi*HmM72kyTVTrV{bs-H
z=Do~qPA6YH0a%e0Ku;MdhBtOZBD4avKS4c-d+CZsc?-v8+{*xU*pZzYz1PQ
zBa#C>6yqH!$RWKYSTpU}?}`wzSp$MAqIvP^b<@JNcl@*dUdK@bW*Yv_NESP=e9hFY
zZ1DkosWIE5)G@k@UuAk0Lv_t14orvO;7f?R=FNAgl*SH|;jf>@#l0{WOD1c{8p%$|
z$`Rt9xdm$a$-)BQFh9#w;oVtRDywu<`y+!lT%OR&3RGr29tjCuIE-++k$`Gs+@M;Q
zoXW(qO}EWa%0%mz!Z?7$mtR2Bds%^Iz1{o|{Ks?QXrDlfBZ0A?z{!LPsl{PC08)Of
z(IU^@7eQ)5h)Ote@R@sQC%osVwF&I*9?Z`?^b_0?pIf}fgrF}*sIck4PGmGR
zYsJ@{$u(^xh0;`(abnt?lZJR3cY$Dwh2YA>l@$TQF>r7(3QgWyfZV|mW^KMagR!L!
zhzI{WDf7SdgAzU9Ad|~?f6~S98rsV5bu(7FFZm`>;ByMB#+H}Y)Xh?O;hLc*<$RY>
zlWGd(*nGzD3{4bh$haTyr!wn)bqq=dL%@$x=B*@#h2e64Fzm-w_9KW0fKt8S)-2l8
zk@F9{`LR)}?N&-W;(5a8%||n0Blj>B#aW18X`)$|!eQ?MSI7&RGkkgkE6z<=!HC+L
za%b0wRV;esM6>1Zvc{1~7%G)dA<`T4VP=?4cV;Kr+UIvHh6nK@H{iKw
zd1+B>^v;jB^vnhqW~owG%v;y6v!KC(&7hN$3w
zzYXuCNV<1pWd0$|8`}e|a+Lh64sB@npb0vU*Gy}p?3-q@6631WWJmdG;%R)Ec721<
zn3v*%rpKD16#If@ZInPpXXb5`%&So2gXJ=J-G|8ZIX`k%%Xyc(ANiNAA9$i9ZkQhj
zm%e!foHF_0>Db-A=M_&x|25q7wDYr&e|NFWD=!vJ{yg2}F7vX<*oB{@65C
zs+Ap?J)EpbRrxi_#bzQv2RB5cXoIZkx(y@&;fjP`5=P(Gh1xghyK!DZsu1Bg=poma
zp)tnht|1wjBuX`lhow&6dv1-!6y)#60|g3o?@tA}w5_QyqRE;3eJhJG7qc!A|D0##?
zZAYu3cab-HQPNjd%f%Eq2r=C3GXh^Y({Z9<9Ij-Ft#U|&3=u@HrMX^r%Pg(aBM$Ym
zByLLEH%JmA#Ls0~>PR`^^*E03n9kKGdE0sq++GwKWX40{6L>Q4bRnoYxu_OGZ>fY{
zE)j_f4I4J|k-ML7(T2T03H}CwqO3DSUfi3+^4GsU-S}H6dION9s#g`JYGU2*o*&y`
zTR?w1Wr%y=!)F2TB&cIoodLtE!%Wb>e$$(&VtvT-tO#2
z-hNY6Cz6(lfx6!nR0Z~-5*(IgeV>v%rgo(vphF*RS-4hdbxeYQ(Fyy=W7JJ(&<1NK
za(y^DFL)Q!d=7e6+|E&$l^SKn|8&GLlpSGkSCeob0cS%Y}7>XECBda#?AYw#(t#=+-AnYLjn@;v7-xc%|pZB
z-@25Zf6@E?iqT-1>QI}V|2}Pi)D#-7z5T*_tVDNm_fWm2J7~E{0EUH
zaTS1=8oR}i)Yu5IzhuaJUQpk;{71-fgn99r)D@c@raD#Jg%rK8OP47l=598G)R@Wj
z3sTPhZiV`Q96~rJBXt8Q=)!u8yhE%_RpnofgHAR5AyUT@l4pNCRG}~Y=Dj!tKLCypG0l6F-03=k8_xdFVtKtN$L;0@
zdllSZ8bb4G@{$AlXnPw?(qZ&2i;za~=DRe!%4ZZC8;W+TFN_-@m{VN{HK)4kjPA{M
zXH#(!=3rgs@D?*K6N=(Bj}?rM^|%Mhm(NZSlyM4$ZhUriJrnEe)
zRTrlvJqpsdr3&x#o3PF)w9dA4vIjNRNjnw|N*`)|T;=Y;n2VBrT)?}?qB7Dv%KC-k
z=)n_;TisLzWGH;eJQ6`@!No#iqD9T-fedVB88xA
zo;JC|uZgi!#Co3ph0}|BB{faa&67DZQZ{0o-XKATf{#ZZ%+&fF-U+AAOtAa?aLb}0l5WnG!>`?^H
zx;`p{5TRCD<&pDIgyuBDu1=!sVf*15{)I(*0r6PfNrn+cCs{@i`14?)I&y!`f8JMd
z*kb%Z+CGVHJ7sZ>lnL!6;R51P0M#nlTO&kwSc0l3X%OB1-+EPj7eFzDkPZFU(}W57
z5|=uiOyIL9nIqTRgi$pZPG-%!Mf-Or&w2l|+#ZpQ?4UM5&}dG?LIg_IU1eTafvAC3jwBAh%
z9S0LF>sJsKN_hHwa+N=0)+^Uv9jPa19pl+^f4`NfU6>zvKq?rDO+|{&*r;1RAH`Nf
zRpp~+Uw_j=Vh`;lbgJ95{8YG4e0|Of!f3Ex;VS#J=xb3=VFOFb;234CzO=o6DZq?{
zkhs2UeEwu79L22C4qkbyps>(LLEx{S&C9!
zqFx9jK>9e%B;6VYNcEhXRc7t6XE0G_Z}PaJb<4OS9=lU$p?WhSV2HkccLqhdGHg(S
zWrF|~p<%J>3qJo&Pz-2lNFnq^)@59eu$a4eoMPzfcAjiF2;bX5(Yk`+D%2LD|MixU
zd_mo!|NBit2-hAGKj~7Iv$l)#L*f@3=0wuTv>^0Z2EcD~O@A=r&v+`CF1;pg`O#~R
zdFh)!bZZ&CjU`0!k~h6o!*?3ig)3#y5Dr9yl!a%FcI?y_3y*6ZwT!Y&Gp`gzGp6qf
z3!mXrO#mC4@-(282uWdko4($zc0XG5OPDmMp%ba0uuiMEql&`da93afQHT^9j7r>j
zo_L1@zagwUwX-$#0jB}nxD*8LIW4%Fi4R{y%0fola}$zC&Os$7{dZ8LOg&aDDkvt*
z6{|v6K&v164{d@r;P~b4$r{B*5p6*OuFfai)OxCSo1~XEHuz7)CLS2wBwbVKtnT0}
zbJ+zuN1{ox6C_`^KtuQc_7gu!Q;3NTkdj=Idhi4l$%o`|>io%B4U#{JDTAC+Gv{50
z`r9cOO06j7F5xS$h}?jv{u+hxM<*_GH}}|ov}`aD>E{8|E9h-Sis-$GDh$e7|8%#h
z*sG||^lCY>D`Sg)=Ma!M`(m@9#t2%1?XU1Z?;%RWm&DZEc742@DRG(@p4~zzl*2hQqqGixEB@_XUwvGfowJb6dIkeE}_PtEnQ>*cqZ@
z`u%SM63p&Hl*CkhPZP<5Rnat6d4q+|3pvTSk7jPUpIy~b1;$&&8bRd{H~AkooZ@Wc
z3Ek7VKj7|!e@!Z#-&AoGi5S=*W)7$wkpT0mYKJGbx(L;WGn@GZy0J|#j@wfC+lyw3
zN@Ae*m6)Li9uo&GC(+@%(3S8&0wBto6#SQDH{VpUDCR3jjq4$J$7~--n+MaP0Njak
zIHjlIG2_4!VxnDz4#Qc~FcQ(kw6`mi`Svqf62mZRuzHr__N
zX|O0bI!3Bi@jSTgKn>Bg&Q53II7XEA`NhoOY$25L(eK?<1>~>hujoRs}yUHVX
zQTt}~MN1Zw0!xe7-aUbnm3fxd*KjwLLobx-SSZdGT`vUtB-=hS^~#Rs9K<#Wao{Zp
z{kmB)_tVwwglj-W2qla4$Dx
zz53|611jp`Ro8}#TnY+1OdA}SN;lGY+!=)=QY;q0b}yA~i0O^|s1#r34{5DET&Qh`
z+yAF+h#H($?p6{{X
zz@F@jlhUjuyW%|!{Q!~$q>$)y1Hsl$FMYS;#r=c7U7SN-8Uv-lv<@mZOXKOlXLy4C
z3v?;Yh1KbtsO!ynC%cyr-`4MR8{Y~e28GXK7yRZ9{3d5(xoP=XYdX*Uy?#pU$a5>&
z`JUOV_
zRl}`OH8`y;U%XD+MFBD!HxL@B`JSH_G!;0V9-A+_4Hhf%=wJ%txE{qNaHkfkLsS+h
zfe@JrYreKrFKhU2`1kD)K_#LCd!v$sI$is2ee^7Am=@kq2Z-`cU(l{<8*ON_J?@@&
zOR+4E3%sN-vIK-%81R}RdPK$bCrBoGF
zRu$>N^0s*R^!6R919|j@|cY|A}(o`$`Onha=;(7(EQNg;0
z^sK+j)DQ>i-wCKpkXH^lH{QNw|FWJ%bUle^xtPfLdv#(gNpA7lv59GyIOXp*E+vz_!y|AkM=lB7mYXSWGfxr6#WlV5UW{v7tVn1!e
zwg=v}kG!o+0qXfY-FyUozk}%&PS$ZRL^NoDuz%nVw(p)FVMPG6Ph*mgs4z0~C)P`9
zWsg_X!-)T2FkH=VqaRBu8Zkq=AX6oLSWpg(yQ^NIftdmxjx
diff --git a/src/DjangoBlog/docs/imgs/pycharm_logo.png b/src/DjangoBlog/docs/imgs/pycharm_logo.png
deleted file mode 100644
index 7f2a4b0ea66469bd218774de8cb3027a9c18b84d..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001
literal 132045
zcmZTw2|U#4|DU$qY=^9^R*2bBDb|r>=Jvqn&-a$U8~l#LJ&jyHfAI(S>t8P)K52l%`RvBw08PF#ZVCPOerZ-T{F^v^o!^bU+DE!L
z0m)bXT*H)y=^3
z$a;HC%g)N@X_wZwuj+_)rQ{FZbv>!$q^EP&B_JU1(w`32c}ac6W;C{jW_0vdjuV4q
zzz>^PEu>|?gD}v^9cwc;@bI*IRO3D$y|f31i|s96XOP7(w&SKQ{S5yi4mY>2tgtuT
zA~$UxU0v1Ctzo0@bJaVE(b?w+3tmgHI9zpy$$FvkubGSQOUl|AFJ1?r^>Klc4;mVL
z928pj$>&I)@BNS1jOl1i(j{s7iYU8qUZJ2;P$B7Y7d_vjM|)k
zL7SL|13NY6W=77^zDJ)n4zm2(ACcWq&#~B{s;c@RjT0^1;@RMzw)6#u%eee5IWF1j
zazJn4j4Fjq2&RqM*i4xtAHzMc$~MSrcsM?GFPS+u>zZ03v0=h%@sFyJk&GJ#Q-
ze#E`{z?3pmTZD~14w0vXfU)uxZ>>U~+ai&i>}~U`r1znYr=ByTf)g6XOeyo7dFVWm
zl*JMH4!4zPE^~f9N_0+iT;=VAT4xKx%iPvVtgn0<`uOoOtp^5f4gOci>_?|=j79Km
zBqj+g1uj~Z>wi`0?!#}uU0}LT6Batc7sA=ypBYH2WQ7vv3adItmL9B4`2@H8>hB-nmfv5xJX{t$rH!Xc
z`ZopWW1r1;cMs2PsPnglkNRHQpAh#}pLLr_nF{Ap+Ojx&YjDa6*e!1n6~f!8XK1dpL*Fft>(IsO9d^wlCNU|T8XfkdZ?g}t>MXhLAh3H4
z(^liQoi~~IG69j~pdK~hhu7v_Lx)%#=6!m{Q)^vqvcqZLRNqhI6tyIino#1vjfFi=
zt!QhkI5F
z44m>@M<%g%yW}O>XbtZ%hb$N1d3ovWl;eowOV1~jQqj$u@
z;Zqd+A^QeTmF`c)yXln6#V3=`7?`wLnH%6V53EB$eW_U}#6vmYQPJY9)#$6Q+4wnD
zKE2UdyL7Do>SIQ$m4gnxF|aJ_+@_+4f!U$&a1Zqdf^S8XJymwrZ~YcFkdjkA!MB*h
z`)ulVg7lpab;Y~m6Wv==HVmv`t%_u_TP+9t`jhyR_2z6=Fvz8{&H^1sx|2XYuPCSkD_3FH*0>PH88SE;-y%^5iPFH6axOrBt_nTOSiyh)J
z@{`)1=S;-;Mh;Xcrfa;ktNfvoBZSt#t=Z{57Z$yVb4I4B^HCjG@EiN;4-#CbFs7^c
zF}DWJm+R+;_mzpP3OwRs^pt}y=86`MI;!D^1Q#;PZfu&D#uf?LnVig-H+ImqR3J=s
zR|*X1vM!1@;`gemo}(r&Tr!WxY3_QFtvr!ox?tc59*?iyUYeskF}k=5eLQd>;kK>%
z?pW${w}nH0!;D6PXrrE;&NwAcm9Cy8^ZauBu{$dry457xA+M
z*WlgzcZ-Lxuh+YNJaL3FT?}7t00szILrfB->$$-OrktMn&aj%5+$ewF62|RWFju#=
z0sotCy0c;6ntHnDsLlM10AaMq1D9T3Rl_454fpQV*42u&Aus4zMJAmaR;rt`G;hOc
zW)F2b(do8a(YTHMUc@~T+}!u(bFC;-21><#FreNZp&TT{Cwe#hy%F5t=dE;Y`0h@W
z?vF
zd&8MG#ty1(;zR^Y#|q$L{u&Nwb{R@oe9vzi?4+jAw0$AR^O#yk`DV0$@e43s-ln2>
zDyjeK$dUf5*Yy&F!a#d7An81=%3r^5>G1qVy7v
z1B)dAGH8(pI@?<^r}mpif2jK-xwg*h+js@1TbM3APi6+l5wbMmsjd`9$l5O+k2(6Z
z!A=-NXEiyx;DYPoJM{Vcf2m0jJda5crh2nG!wlxc>-d?ceAX?{ubQ8WOZ2}vz8QBx
z`BB?P*4+aO;Q?I?FXDL#Zpy|%NJx*r(XcSy@@v=rsaj(P?js9__f#ktqVybJ9qQCD
zcYb7Ppkxhh>yezfL}FQHxikCq`>CH*!~Jgs97JEf$HewroL&~c-Z%z!U(d6_x|*
zKY}dMso_;ow5s$WDMjNf*`dDmWKmJWWcV7~wp`={c~?z(m$nUh5t~ekJzLs
zf)iQxOttOtRPT!rgaxnhb>U21R^{}Z4!cu81pk{xUI2d;JS`Oz4>s-fMM?7kjQ9>2=3ugtSY3GJ99Qw
zPTdi23bD|UQ(DLia7KICYM0_}EJau_sG4Uvf@k5ylX^M>$2N0+jAV9UZfR=MJd)=Z
zy;{?KmfK|#W@>6T-~H$NsUs@kWtpE@`<8~#GV+IiPRvw05TEFwuRtjEbkcK`C>AWe
zuFD!L3II?vXI$6*Db25SwPxQ?yYoceX*QwTV^6R5yD;Xv8^XLI<8(c$$Xf15f2NFK
zf98U_RS34k&NVFcg=Y@UBLkf8ae>v4m-=cyE*Pl73PFEvU|tEx@O?U1zA%ic+^dUQ
z86D1c3WVu}s?HYos?Pjxvti8c5VHdk+|@TaUp7$lQl_k^>u|?~i@B0ze(nLy|B-}OG3QqR`@}43{T4A_>0cmuyEjhge6JJ2rnhn8qCow9NNbgpkTR4
z_)JGIZ*+EsL=_d83&V~d3Gn0gXymLnpq@9iYcDPSyji^U*A0Vxim8lm!}@#8F$(fT
zfWM=i67|8r7R)oVtsU|{A=PB`)R^x9Nt3w5sgu-SamU5A9%PrMGaibsWZW
z{NCU+2RWFhG&ANGv_%?nn2Ox+lHK)_ACYx6KR)bGBcGQdOy2_QnhT6!F(gXIkM`(z
z7*=MBc92*%R^u{yij$)$WbLPe1)?_%dV>Gl$c!bXj@xi^X+NQTJUsg2^K830+vuWf
z0=TWXNO4bRD5OJfPUn)D!`O_4!(#a??z_M{?h9h!mR-l~+KVjJ@Ku7PIy5)NP!5((
zA39wnDJsQNeVi2t&sF0SMHajBFF+u=^%R_|hX_7m=)f_8I#*cIyhFqgK*{)GV*#Z9ZP7a!$>$*;jL2mJEkb$5s
ztIFDh+dvh)S?5@u^4X2?=X-!q6x+^o^{g$(do{0|>F_ar=G(3cXVU0qV;2l*!3)yD
zSW(T-iT&yy`F_D|cr3cEZj<94C3U|&y_}1Y%qwc?8qS*Lc{zhJWj4q(4}Q!(-+!iR
z%$j->JOz(H3L&+mn*2Q}@Q{1dtDxHhD^Z=vEtB?FH{gc`gP1Qc;wx0x1wx(r5ntPDVXrylE$
z8f1llFW;GC)Fvj`h)_Vi_raiaPdv4F;)5=hZF61Zf&|q8I1(TTJUMQw110jF6%M8k
zuQ$-x8aDCr2+eKj4s8e8Nwu!F+Gv=u!%(e-!w{a&Q5hx5aON(Qui2fV&28~n-2Byn
zf0j55q+8NtmAZ-ddcC)p4VtlH(G3GTHcc}snkme1F)G(kOhc7sOp5)pNF%VbF~{g4
z@1+x)v>pp+;*=G0JPo;(;A*OX*J+v3UFF`Dha_ijBFb_9YuI>czi{@8g=%B}v*uD7
z2ht}i_EALXr2^y^dTv>b0*!cOA=AUpI?do#hn{*$y<1QSJ9Qf#xF*p5=L}Zw3
zLC4A1e!bn;zr7}kGc$aWGc#wznR#NVM!1iK&>KN@0y}cL$Q$n=B8Hn84Zbwh0oU9`
zIN$4i>inRYonIF1H?)dwi?!%k(&2suynbKD()b;^-*~17SVfmLM7$yXbJx^j*7nx;
zZ`NyZ0fi4gw|J4C4)zS+GU(w7Zf)cnSMTJJVT&Hm>dqU{rBOk=GJHn
z8;YDK4v>xxbjXqVItuYztFk4*u0Rw$rNF9uH?>|B0i}?SuRF#8Iz-Iu|=n`&Bbul4mym161t>=nXp$`t@JSo^Lr9>4%l%aON7;Y7cV-)q%-`
z+*~@5yy}4t<|n7#WtROJG4N42Rp8U@vSO!+qkes8BTjEmPQ(~XF}m;87BbXlE0DvP
ztsB>dv*=(9(baYb!SXNCC7e{%G+BmLT=tnYS0UD5I_U4^zD0DY!UT^t6C#-(apav5
z<6iai0Et+~052$p5PtVngB)};{topIAbX5iHncT*=L=Zks=FRuYF5pKl5Q{m-qoa7
z;K`-SIW$eQ&YGQ1t*ecUIN8zBoSE(((YjDIeeP+Y#_J6$R`>kWcR8PYUcV9`JM4${pTH
z?Zn00`qZd0#J%LIXk1Yw-6b%wj%^xV&ieQxEb+?U;KOtA=&+6@ILi{}Lu5bHz+!fa
zD4p-PheH8Tb?zOpYSkWhluAx-Ni=~C1*xC<5^A>Fq9Fx_$&)Zsv+>HtN|
zHsE~zpCx<_do)meuQo(I^;f{jCxN4sxt@AD=(t*}Az2syNpNAec{bX|mfSK)6`5FD
zM}xkO@dqz^I`_gYp(Oyv+5#UWh@=2ponc1@&hqKzH&tJ(lFgL7>~BM}N;8#$XblyO_vJx+0NslH@%RX!Mb
zZH+anyde0~i82>tTB*13>h#aX#HlJ{e;ZHtuzp?Eq!^u_X_yk%2ohB*wYGkapm7Md
z^%!H&Ud7|e(?RZQade(`!G)X!1NSl?v`Y1P1<{4e3avy{a6DJ{jRV~wyv*SLR+6s0
zgW!0`)L~+;|3jQp`aeFy-{&=^$Zb47*!__)D_8axwoapMYk_!dSXPXncDl(3FJcp0obkt?7YgjaZM=FarSAYN6Y$3wyLG+~-wKw#
zB&Fe-LKh9p>(FAmhLH4}**g=N@nyzb_-&~`7k1lZ9bMLKFfGm4VJvUn_`3@Mr>6;J
zzF^+P(n{d^MHF4?=U!5(hLn%xOdxZ?X+GJ{C
zXLsfd*O%cpoH;oa&h1bthruTqY0soPZaSJyb$Sg$_h|<%6E#iy!sp4n)c84#H
zI*Y0f%{P8`|3H33aqj2O{`Y%2tyw3_o5&j4J}HB4<~`UN7JlOg8V$peQn1*xRN)o(3hvm^RK*67w|;=;dE6489_l*35kOo-ilH2TN=<1WmzZRL-Ek
z!z3Mwi0F$Io0N`#O+M7m{VqkJPHOYtFaE7#-{9BScbU&a_ESu_q5kBcFeYJhVgfk>
z!MF|AiodXW0dxptO^u3fXg7pBhidc5ZS(>D{PLY80|t*kgq)9A{lHIpE^+AeFNx3e
z6euN4r7W78xshE$Z;lu&bK%Eb1+`X(-;a($M2~N1L;iv;T$;I=dpV0fu#_%K@O(-2
z!o?1AK^}uuZeBZpR=G|}6A~>SkQPUaJPG=AqROxTB{)n6tW)vc7E8Pcv3~=&F;1KA
za+Pd#a}S@<{^|a+^X7Kk0XUe(h8Zu&mr-JL&)J8Vadx^F4WdL>bF!#b>GE@S3NKlj
zw1MZzf%?^&q2x~KliZ%$Fk3u^vyN0#~g2T&XZan*{fHR4hK#?e^zeYRDG$r0+1FQD_@tDYfezF^n_J#
zuW%WNK-ZHNG^H5+w@$Q!)6lvF$rn~bSuuIHmsM^)T$X~7#v5yoo1N7jelW~U%rqs0
zA*j1cF;yt>&5&uG<5V(Q%_+2rlU-#jyMI@FL#KVHQfgd3ARzGPy{srya9S$5yPx!;`Bx4*=Q
zbIfa}cK(XBMCk%%&wRVzTlm+v`QQI8CijZ!?qomHr`DtH@A{vx1Bp@E#nN4RS9O3M
zj)c|q_;`r_^9$H(jS3M
z&z_ZMj}>2Xow+mNN6Ofj3N?LRg3JFdb7|TXUBZ(%H_wjsQ5>s_Qq&^PrInJN>WhbJ
z8gi;kmN1av-~L@p@Rp$d0W%XjS)#uhW9fETlQqjmIlQdK6DbhOJn&sSJHs_TQNSL<
zODN*98pD|(U1Qvl&TRMLO#HZ{{HqT|xypmjTeqBBe6cj~GT$(b<|0D=3>OL-Q?DvI
z=Yu1?1YLLV)EY={^%bJ^{6O1z;SmV`7OuL?GV
z9+JChiZ{0>5&*5uL7+a3Zn5KA(IS_2x!7+y$*Fw_HIR~m4x|fcWfe-Z++SmoOD=$x
zukY`(dP6SqCz`Y(?cgLT>al968cOLx3AxjzD4Q6^xGK4cC)Z!;POi3G^kgDk1tBM_
zfN<}heo#|Kj;8kGu3c`6(3dnfv0G%k7KPkK`%Aj3b*>&=x0Dui9NK`qtk3ql+WHMq
zIdn(70wkYU{@YXQZleV^K4m|`NO8LVm8aRhR_|T2=+d)_Tt_k+R>1X1UvGEwAF@5k
zu`*DgdOKOH@B|)NzVD$&cIZ9_^?mD$*mv&-9`8|N)3!EEbcUK`OZb>LPwp+d1*-%;
zQ`p|BvOD$`@lLg+Qf$2y1X*m!sxDBQz_Q%ZVTAUlixzkGowUs
z;r^C{#RgB*evwZ#>}d4SQlWCJAm~dKK)B~IBxJCV%R1DhO0K~aaYRw&c=+jQ@?>`F
zmc#&;7lOItK
z#9N653F%BDH6q?5uEJ`QdR%rvCu@m<2Q0g9sRq&9jL*V)I3GHV#gwB6Iu+U
zg}%Cdcpx?4NZ
z;XJdwRd;dxwDH^94mxrl1_y1+_k3yogCCWO8PXupJnGqT}^oz0Yh%!+|u%jLG;A{@yi+
zqhsm=AvLZOq@BDeM}x2Z$G+WKe_ZzN=!?H!z_tkDR>)A+tnJ&HmOc%gg0TwRE6tX^
zOlnkE+EL&wr!T({+wtRPNL+6#R
zO80Ach+C8{?Vuwt-NtF-jupvGoD-05I?W<1);x!NB{qmWG>gt>t5Xi@TC?YFSm?OX
zCM|mvd?^vKm8TT7&AwA9wPu-LnN3J=$BM^(u+J8&iI}?ell;%8H=kSVg-xx-5M5U&
zajQ0W=u&SX*%MBvx#egVvG=Pnu{o+*FEJc|p)ghrK*@c`vzCsBKS`nF+ZKQ6?}Cs1
zc;Gq2jV%6V4-}VNH~e+M(pJyCuqERZH`3gcQPD9=NC>Tg_BwH3a7V#vgdfa0er?ws5KT
zg1u%OBt3wdD8*1a|9b4+#-UB0+a$jk^DM~iiNUTs&#Dk^GJ%dyg(uRtjGq+OHCOJM
zqQ^n`r15{$bvwAs%)^;{TN~Mm>_PRT@3}JTw*GeORy65MH)B2z+~JAe_iIoFA!|zb
z1T>d(a({q|a|F!(Owz|xC9Vhnl$u2nn0=*~uOj~{;bU6hYPqB2?`i++?;={hsP5h!
zCzLSd`+n-qHacYEBW!IE`kbibTs>NBoiipEHyk_GxBuYf{1wl_M430y%+x@u;PUtr_2^ot`phU86m%Xue%c8TJfoBSk`5@b9
z3aeOZ56}nCvvQ4zc{6BkMmQ~;>GDtR;=7{BaD6;?v$*q%(fluKw$)BvYMzS|O2jM2
zAjkpSM3io_C6Vg~lT3)J_S5`5r0V^Mv5_95zL?~C1ye?V5d{*NS&+XhO+6Mn7IfE)
zApfelRcErnLK#b@8Oo`iTO{%0s)!;UnbSA6SBdj>m%=|b>`&tpfHo;04+Zdtqqd+S
zK^WPfF#5rk;@;6@IlGJha~d>HSg^ej31MK+V6|$H*Ba-~Gd?1#{RJLYneJkbC-tR;
zhd}EAINpEE_9Tqy+3ML4;ArsUbAQ`wL!VA1`OSmjTGw1g^{|Wuln>xK+H
zn~k_p1rF==0HG`(`G0A1e`fT+Lh@RMjL(a;7P3?E#>64f1Zc!LU}1@|kx=4qpygNy
z`T2IfKjiLTajm-WQ9#prxX1CWLr9zguQ&>`1diPJaTl~LV6?k?XX3JYt*%JnyEO6<
z&<~WwGCs-}tR7MNaNhbw<5u;h%Z6#%eP@ev2F+o)CAIUtpy7pFJ#0Rx$o+E-OO>1j
zj`2z!GQsd#aM+8moRS{~jAd7#q#*eZKGt!n=}$w&Li6(ObMr9_T*|ZGWD4d7;0CVM
z+^oCU_TvTs6P_7!PT@>x-$cVnF#*`7_E%PC%*5b%MVZ)0e+?J3$xvCW%8iksT=s}z
zn>@+rI`|Zc(W2X93V*V(iQ96HSZb-vP?dlhCp~e5eTrN+Fk25~HSl-xo0bE`+v}64
zH`B)Nz6X>^-~~M9Pbq5dL7tIjr|l?tG}ORwGbskTl#ZkYa@icz
z-wQv5-J*G)m11}tlET}|l_EFcB{m)e4-IDqJZ9wo9fbs9|I61W>B#At!09$BLY4$x
z+9Hne{v_?HF@B
z;qK_Yn6szt#)O$(fN4ZZ|1wNt48g1>)ZP(i+PJ&l694;v<#e|DJTrXTBAK4&
zxo*&^#o@kvf5X6tk`UG*tKXm2T`hv0cBLfLT-Kq@LU^^`>-_9pz#v}y`rEMM;w$8G
z*A7xldPBz)efrK1`d8%ko6K6kYJqyYT#d|P>h*2`V5G&W|C5nWtd4E=se+pJ(NdKr
zj{33;-HWquxRU&UrdaKLNhM8yf}Rn>q2g?(qAsNI^NCWi%H6JX=;ThAJvz0Dxpz(o{4-Vf-@K-kzl2m)Z|0YG2GsK9_da!G8{fhy{UmK6KG(
zN6C|3;i96;^GfUW9O(7$SxsekDeAqP?(ny_F?e*;OQ)Ow+CCK(3gh(P7o*GXs`y%`^wmQn37r?lN51CD7>W->
zLz)V_>5Qgr0sWO{P0~plz98QHa#`O5gi`={{k(fuaxxmDn(UC>P#dMyEDGlxJJ{&u
z|M&LNHdJ17#*eTU`&&6I*xmh+k8GW+3>SOzV|{7nqOVc3t@vMvC)6>4iM=fKEyF-K
zzY@%q6_0Yy5?|yuPX3+3ixJiE#m|BMcSy)%s*Rv+?Nx0X?d*103a@T43z!lEB%;tv
z1i|Sh433i7e_g?5W#cY>W>EE^z^;>_3NxL+ZYVn~9&Tk545sxF7Cdt>Pf{eYR!=Bv
z@sR(Z8n?i2gR&g4ttW8=qFQKwRjER)C>#>IC63fP#0fPrV
zyIrfhbg$mBKYfx;Yq4Fd!zB{3Ks~u$<49
z2jQg3n$%}tcmStauJ#o{k%xH`r15x2eBtW`cXS
z^I&t_FK0gq%!|8iO7H5WlpJH4aB
z+eoE^XFOX
z00yDJuC$Q=cj4xMn+IXsT&~slT864*a%p;_%RrRY(LWS6tj0otmLT1=KHQPD7L~-%
zgYK8-ZVW2Bo{)<0zw!v7OH^R7(jTjoDliP=nT{zxQbATwpwZkcjV1@
z^4;NMvxkyy&bmsqw8+XFsmRrtpXp%`K2tt0n_KuQol{XBIiq5AR77f1aP2F^N
z^Vu?gWdu(UI>dhpEh~Xxl`s-&o31&8NH@QFe5B%A`cg*ko{_3=cK_PYv=645p#Gq}
z?tZrZT#|0qdAAks!&f2JhUg?8bkzd%de0F5ge8@C+>kL6X>!g~7t<^604xk-`3eTf
zAk{&#_qh=c1+wH@f3uS8!Dh#am}eQ}Xi?Z?!3Gd26fuXN#LbUsFmIN|2kD`Dldp^-+M8JdX2;|uDkF8O+dbAbouz@;B=zCH_pw|59ILc>cWA;(M6vhp`}pim)U-O(=bph1QHcVwW)4VoXekiAa;t
zR$pcZytI$mlPJM`)8*c^zT>3oU9kb$nl&9w(CWDCoYHr9VY$LV(^xf?Hst76NRon7
zCKcV?IrEp>Hko6Ubn}q^I;9PfiP@RnYJ$CZPc5u
zLUMz)oVtIC9Z?I=vvUy%xP0I`r7{A{A9j=_MWiues{g++&l{$UXUmp2*_wLLZ5M)G
z_xy1(iPirOTytCfN<@Z`b(V@qGO!$?^V+g1ge`jJ29{CNjHwd9Oi%H
z8?&Rk4y{IZB&Wfn?6vu?+YSTWy>&~5iVs6_ieVaXg_qjghh-$<$1ITjHHB5x8H)h6
z_RERsv16K7VjFQTUdK$JJCyinHJ9>BmV>#+v+~1Ush=8sTE3Ls4T527gk(_fIg$xY
z{-v+cP!sOg@;dvZ{w{Dy{i29vi1{zMubJJC6gp+cx?zhDUmNrGeK@zCuh5DIQ539&))wnh-kb
zevUhwY5AM1B0~j8A*h_-Hv(*%wO~3Ml2_&;eZxLf%E;T+ypm1Zr_cDHBOwI;4s0m{?h>2r2
z)}gVvRe++9(8p`A#ZC}elB^Ktu3H*kf#AUFAP?B+dOlHpowRcTLy2KlK$WuI4npj2MN(*l+|x&3pE_y7^|gy(zv^Tq=Lsgs{X&PJYXX?t8|k*w_AF8Ndcl7YZsK
zJEU=}i>T2@%b463Fnp!qys;JM-5rFoF(-Of)U=R8d5r`UuxN5#A`h3a)Sn=vt<;zno
zS->T-nw;~SzhOe@e@*9$=l-tCIv!`riUxsgVlhFgZDOAK%a*faA~z>GQNRErODCl<
zT89FhpuQ^m4c5^$GI|Sao1IyN%gYLN((+^$Lbmw7Ll;}Q2J7Xr(Fiv!r-@$V+Egy{
zTYhJW(~rDX3MgY^p=Dnp?p`NpL1J9>K-y4)xW!WTa3x(Xyo2W||0+~|%Lz*e{YWkK>)Nsd&2
z+>`u!N1!Yw4}fDs(+;_iz#G)JUvzGJXYvX4?aj?A6`p6w2@vhQiIda?pPI(zCP!Gb)&^hp*xiC0v@Jy&LbU$6#fwPla;!fXQ+e9XC0G(
z@df@XaDg!eOTR~e4`c@+v*`AxBgz4#{FHa^s(Rsi3dF|{j+E|#c%f&}@#jU)Mo#Ll
z1M&0Vf3P?nm(kX?Ez
zFujVgz$iV|t;)H4IP^npE3w#mJsN3X|59>9T34_YN@L(e%gtg`A8M|ka@ej8o-|}F
zszUl!+0}FAbF0fDELytUtpLU4irHXU>SsR+39#Dz)C@Gj@KCZDj9C|Kxf)w^ne+nU
zmFopT?uyX#C&%+Fn}9)nCpEbOKf8+;o5q5I^e^s#dPVC~pB^qykuxBTD8
zlU`|C`f^WXpiJNH(l+b_GxS$Xn%-NcG0Pd`Jj)JBgr^5qhYv?b1`{9%-5j?A~3&|a|OJKLyEciPA4zKf6
z)<+kpJF98rBS?US&k?rX*=~aOUy@FMg9Zf#iVD^aQmhCK6>Z^lq*_w22j|JY#oF1`VD^0*U!AnguwCwh(RvZRH)>UJ0tHXGiNKid^M&%SJF
zUQ?;MKH5V73wKn&X}N+6^1-Zk6f|=oAqVpkvQtBDqL
z9_INxsF5w(n{4W4b7f)e8KVx^W=4fDkM5u~H4bY`+=o4G-M7i*@#T}w9hJ?_`Bk&h
z$d%T(d1J(hH<`k{9nLJzsgdP2B>ZGkvoQXlo?l3;>x|VIm95ZzqC!Q(8g}qtvwY?)
zTG3;W!mM@MR(GV*K0sl%8Y{J?|21R?ONw23cCxP6+*CF9{Lb
zGuNocA~Sv2rIrIik~KZ!)jo5=l^Yq
zfMXK+v4bPWL}oiAb|QO^O*`tYJM69cIpe5R9h%}>}ZCi4X#!9i5Wd@A3#
zZ1g4&3He%VQS{bo#gH%*BlnHS(jG1=9*L^4{Zt@+ASuAn1G>dbPLO_$fFdPshKG$H
z)c0z0pD%2p_V4`_4zJ0dnDug}m|hJAj~=0Wuldgy7fd}o1#^Ck!qkinm5ijLV`Y~{
z{(6(XbPyYh#uVY$F>TfRB%h)zP;p*GW6$kwrU_*9*2;y^6nF&~L3Ey0r)Q&MuJ{-D
zCf`trkhwdOk3)tW8;VqA13sa3MVY*oJ}=96hjcS0e^4*0lpvu|E6&bfBy-rWzEG^*T*LI?^!8mbU(noBJIvUWR^^b!~!
z{1^>-W<*(hL|O>{V{~$?a3C^)>TjvQ!di7@hvq0bgeg4a7dMzTY7F+RUl;AKx|N61
zq0uQwK2CFzQ>0Bi0m=EH@ulXNETy;hap}DeTAxUv(Wgj4r1gaR|nUb)~C|+xk$IamX!&9K&;U^1(8@0PKcL}s!iK|z0@4dN+&GxiL
zC7xn(&u!ERd*#6^p4rP(!R^$@?hQ{tUiC5i7_U%
zfMOzE@xVfp01W~RF3w>%46t#(pn6!1EQQUY?7D1gV`I}l_G13b)Ge99QFzN(zv5UY
z#pyKZ0t+>3csA)GUf)|VaNuqH50Fr9{ceny!4=K8oKB=Gg9H_777ybi($RvY4$Vzv
zg)@e8ALieri_IZ2L`&KZdo)Zl`DV>pnt3Y-*rxV6ywrzrBu2A3oy31Xil-m^m%+&W
z$xeZ(ouhARCi5y-Wk)N}V>iGusB?bQPp83~3lBrCq&iHoocWzQgdoyw)3PiuVRacA
zOL?MM-ngc{EZ1J)iyn-x@Q@JUK#YO!^2?Z8Fzx9|Q@;VHJGn*6UWzi=?zXM#9)!(?
z?+e^LZSd&T&f?r@m}d`
z+ASmJ7W>W^tl%{EK@%D(wkfEe5GW^0s~1+6n&4-gptztC4XpT!HLEWuli;NSsC?HB
z_z{;Rx1{*!h%^mRMeqr9S{=*$x9LSw>6N>ZcOm0{
z2ZP#qptrj6JD)43pInYo@OiA?fjGyu?hyae?7zMRiBZL$p?rj>+G^n@=t=5Ga6Q$F
zz_L9*YuQ)y7nmG)O9w?1bc4N!2*s^}Xj~cfjfH`a(_iNr&2J})W{$j6tCc>sZYua9
z;8!veO(dW@V+`Cb`=||EVC$tU2Rn^@VwSRiV0GbVcz9@Yr8|T%j8Q3eT%YQj{Ia8X
zQE=x_OwIuI1EX$Zx!`vcE3nZ|R-v?mY1v&LCRqtMcax#w8-
z07#y^jX>8+g_LGqnA3NRq_hgg&RD{^m|8aPhe-UK7nnBORGle(BpI78_oyEWs-qKnZ;^!kAf}p20PR0W+bz
zu@eZT3|~$pQ$}O*dr+x2!b8{E!9nGO1W;;4H}Conxxl1LS~@$08k>ke>*?pyh*rR2
zNV==W>#&^9&(Noz+f=Me_}a1fcsrgf37}W>XUzM-^N_P(J1K24OevQqNal~e^lL;q
z2G5{z!=c^iv89(hEaNMTa0N`}c&kJdh;H14G%qMu%d>L816y)B&F@mTWU@_FERZpH
zNug9tk9BvzR|{4+aNq#HI{lXDmx3!`&xSwtRKX=Ix4(EAC(Kw=JNC=CZQdk{M^EZ^i{ajBTg;oQHwg3o7_7n{6@wD-)xkHj|3EMn=KGb*>@K$T!Hp
zZ$T|7xEFXP*0)e+;$yQ8rwP-j@E}l|)B=5(lJLKxkb4QQBo)+{vGKygq}_>G&2f`A
zoo~`B6HGvfL3TkN>;HM0AUDl|-3tzVOw_!hrfW=L=1p8feQHk*34}kq(wE#oEXuQkJ-lT}y
zUHa?!vW2`kv43&7VcNK9jicI$A8_|~WVpl&W%{Nbf?AUEl=Pou%bR+z*+yU5N09To
zSeYfDf;4U3SsZkhe3Lt5eQsvuK>57B79RVFd?{Eu{Epth@5kE*&o0O;gYz;9{?+a7p&powzH%82^baaG$3*OZ+GB`WWo0lEx(eXM0r-X`Mt^~P
zq2pw)>Kn8Nq=Q0%eIrPY&3Xdcc-cHP+F<`dZwl(6(6!{97*x9t+-Vt4LfTy--j{r~
zzJkxUeDdprUpcC&ko?hX8{ANUrXwdK>nW_1j=dTfv-TyaMEGC2AN~IlEwtg|s%3xN
zROo+kX+_tE-shFWG>UYW3Cy2v)iQXTl?O|OaVgCFU}YO*Q=1xCF~30b9Tn;Nc~_N)VleUjeTHCSv=-9n
z5D|eGWdt+~Qbfz1EHq!NhuL0Kk4`zEqruo}m>`)0@2Y(+tcRw%G7LR*G&avvxox`=
z?uhvgfW_6wZxH>!`>6$-$dx#s*Lzbw%W?8@_}dl5=c8YS?-i=9xtZ1C6UjnQv2(uO
zW#XHi6;+{(D80~M#@j0~wT7iFN;jua1v5uqjzCPoV|!F-OW%;t(1|sx!v0VSM}*`A
zzgOdejR0zIE0x${>2Gl%+CnpKMX^WcgQrOn-13=Y{CQm2wgLD-m1P^Dv(cWw6HJmT
zETS+t@-ZvZSHUEs_kq0OA8;3(f{y$8
zp!b2w#Y9NITbphr2CPPbq94_RsaHIsUsfio}MQN&@WSM$X6DRM;sqcl#p8sk}dRD6J1kpwsa7E-}Z9;reL^??8OTV_>};%
zbwF>q*AQI#D=5{VSv-dRv9GQI+!PuSf{|sUccM4IuZ8K`Oj&mh
zhMQ3&xX%{*ttsEH#ytsGek;h4TnSm5H#`7n9=fedXGiojUx
zSoCsG0F)^*DHcj7?@5uz;YcSF3VgL0BbH-!8uCcEtWsq|?)nE@$;qT}Y_s=}p?IspaIU
z82J)MmwW29*h5T=CqrcT!0#^!wPBtpP)ckHsGrWJF{6`aSiT)7WQbcx1a|4CUsACfw9rCcp4}o
zNMt4-%h<#0h1c)^Rgn_;6+t~-LL;PVOo*`G6MTJOM~D71j?i##7s@~o3&;U@6H0iB
zAlZ-I>BK2$81qzB$ZuHk?gJSi2PkpiePl#3
zBHsoMu8Td1?uF>f0B2q|309BLFa(Ee%m!@bPx-4!BXfMkOc-MG*ab8
zZG9pn8(->Q}Sf4`h=a=K!CKDXQpWg;<#0Ck37X!EUCq+Napp0lAE}9^-4{
zzPGFYt$JRNi)rLDppIrZc7EcP)@obfjD0Z53#2|Jeo{)+3Eot5BfeC@vu
zpt%3(gx?EYR@*5w5wh%t6$ewz&)J=2zE;F(-atRob5bQjq?xnF@%-}f?t_TJbDoTY
zsiY6?*Bi!XTcXj;qs&Ung;-@(!>=Sg`aEAwIldYP^A8l;#M{-D{lGWGuSM|^^h)zR
z)T~&7+r|;@Nl1gg-(M>JO8!!aG~b_Wqxtq6Xd>?%+7rVBxA)>!QX@A>aw6KirS
zozb`{NSGYA0V)M+1&^AQO8C5^=#EoaTBIkdaKYL?J|Fb!ux)~8fzbX}{6(La2OHtK
z5Eg_VdZ7v~h@HzNK0@OXysQq4xiz%xCA<$A$Q=sCj10rEI_kYy=v3@xX3v(rP|Mfgdm2@MlZEZ9T}@Vo)JSYqo>u
zU~}X=2nT%7QPYu05#8-D*Li=rVH`X5(H+5yKA2#WQ8XPMRDx!&H#WS;nJLWjO~s1@
z?nLP(<)3u37ySxB!-u-HCU#4;x5JLJW=Ih)EZwoi
z@-%qNs>_kQ%l-%AvL^BLhWdsdTSs2{eg2o#!I3iHNV0~+cdMOnf!c2E!Jjq(z0?D_
zX7s%nFQ$0P1R7B=sSjMN_Bw0D@V35p;JbfBMC8MA$=k`+Px~?`*WougdB4ce$?MSH
zm%tJZ?|A|?_G|JfbTOE+=h%43&f1~G`4?eWSpe`a;j<-zxlo7yrn3$A_pc{8%qzj$
zczgJFtfO>Nm*r_!#MQ6hz=YoakaZp4RKH)GR2m|wY3Mpd+x@=Z^PYV^=e+Ml9*=~JEN(%M
z>Q13Yzf&!&2=|>K2e;Kwn;c9p-6k#O_#W>8w^i{uK~8yb)nyj
zY^}Ac=zvZ)*TSRDgvojKFv(G{RvcWRx_BKR6HYX1gX7JvcS0v7Rbl&|s1{mwh^=~7
z($ofwDcGEn<&ovOom1+l8`qJkcB$<-{I+;#la)IOGtn2xZ
z&0mSly?q9$pwj#~{5NFje3&y}uaiDlC6B5Xhy{!22%I;^(N>ZWD$ukz$$dhvpir-(
zNgkcW2Iq^WFu{N3@pbv{4J!Qk8ARZWTKu=(zaYCz61)yoq6~5mrpP_GTOmK?ltD^r
z9h$tcO?E8#u^qlZK7lFBDCkBsJPm^&=$IR+Nl~GPTml2QgzInG#3U-&9#F2jT%@(9
zJ+Ut6+M^Tf)D*4mZr3f7UMQ>veJ=ZXFCd=EekSB!SuQatD3^tOyapWxNu6c#kir8`
ztjENgH#kc+iW{JcDmZ5yXC`>Mt_Y2$~~Adeu{Nets@<_gzds2#=#yodt7%n8Oi`
zhj_f!zZ-16%H*bf&s%EZR5_GaG(*$iI2Mum3bEELY8F6JO7W>+GcFYCuE0zV=xf8}
zK<{!jzpeyKj*$ol-6J~XhgTs*g7?<$%p9N*`djUl^yE~mYqoL@6Z&oT{osxx&a3UL|*E#%oNuGbGA3p_Ul(MByyMavsNCA>)4&6
zg_{-)kC@L6eFbu2u62l95y~PvX=2C3
z>2{C14tq>3YmR~Y(+(M%uPC-%h@xA3qEkwQDJqD3wpyX~$>H?yfLIiKp{d`gYcl8}m
zsidSp9!;|U*;)olM7udVyR!!>z$0+`i0Uy}prDAPjKt=AfNNo@;OG4l$EYddbmTi{
zIotBus2Fo+3guPh?IYw<>=$l2J=$;;97OUQ$1QNKcilYO9ke
zgs%!XVT;tFfezXGwZtPNL=`7`(r=LV8tDwdvp3`>o?c~AT~%??^e9{bf~~2l*%H?A
zE0%~?64L?F&hteSaX+=B$&}N19{y8IoKMlUHTHMCv4Yhq+jYC`=5LaQ+;rI0b5yKJ
z@RwlNJ_6mo#zuQXS_;*9R}OtISc8ZTqVXT-GdMwV6mKuIuidVV-fQr+Xr%7tCbVPJ
zZ;S6ghd(Fl1rYmN6?`Kb#rs)Gf~h=NY^&f0!lzipRJLAO6ksPE3Wl-`p;M5qmp>`6
zG??Z)Aqrw^MySGWomL!kjH=nF^cj(r1ZrK9WpSXEa2kvcK>#?-n4
z;0ZEk$kY}EXar?_d+bOJRiL`$c|lrh{Q&WC828np;ep~JlvrF4JE=G-8wJy=S;mX6
zZ?==>kWu1D?rONZoC?G7sZ>+C8gkUiVnkkmk`tQ}VC2iA7m44o{4e2wT*Na9NV0(hJ5^y$!0
zDhf5Y|7L`Sd#xTLrc5jOHm37a`{)Lnjv@~xvhY=BNT!XX4xmPEz0(~xoCYyR5`m)rF75?OUE%l;$k_>gh@q^`Nnw>_VnYCD
zDEqkkdC;DNrn$pU&}0vhUG>b6A0Z`^~2++8-cPL@)~eME+=-|1jpE6cvuaF^#a`JSDA&99QElZ%Xv$<$!v(z
z#6BxIXrbC7zY}ijY8T7?e8mzbJfr&#CyPG&bJI?9inw~FT3t4k9muGkuSIK8ep@jk
zvH*wiSfx*cX&A}XWahUVBGXN(igdhUYIVC99%qF3(-NChNWIBUrfTfCe~+C01$_*
zc73B^vm)V&tX}`1&FXu?d1F>(n4fb=8&jyhXEo+0>6
zg&kNMW;uj(kj4ffQHP9x6=$_ap7B1{D}IW-ZwPnXoi)E469dGniA#rCc-N61CQTG1r_mFX
zgOX{-9*V#*7!>W+GIukLk-MPMdswDSRxp!*5v|SQ*>ihgCx!|2;dGESLtaO=e|u3l
zq_K^k5c&?-yf4H@K*Sa~98@bM4Wn(o77CBKpIYRe>FoS<&21p10_;K*rhU;NDBLL!
z@DgP{&qBh-KetJr-w7~(c0h!AhZ<K&vc4z8?aB9iV@IB{~q%>Dt-FooOz
z9o8J<9kFh+f`jyggeA{=ZL@sxi`7b;?6rc4&-Rg1K}GS%H6_(RaoR2ut+R&)Ya+%h
zK@>T3@QC-DB!S5L@Ak%xg>3_b%;|cQ5OKPVJXeBp%6(37kWD!BJuu>ExS|*5VLnaIITf!o5|3?=+v!>8tF^a6uer1u8HTgWUVcNj^GK#&=aaa
z>H!aFh!UhLeZEd-C&p(lf*9@?%%0NoVgXxOJNtm1)Yd8xK^G{JJs{~HX#a>s){HFv
zxFW3que~boSjDWc1KB8qDFF}JH>Z}CouHl@u)$v+Q4b&%)^M~~RHkZBml&oC}
z>^s4-U`%xL>f^+qli!Hs&wR5lBi?%rC?Wb_(1L{*JBE~=`F2ly1C7mwoU+5glbLAa
zDBeNb5gb6GPeY=ohN}=r@DAA%E8vRRP>qR24}ZKYeJORHeErF?f>F_kaR>D?EYYO|
zAKn<1vn(%U2y+Ee*R>_iJ7L}|j2^XWqPJ?s%diXRI9!@*T+tDac3R+ZSor$EYGBJu
z06jLfxK>TR+dZstFz$T~oxCnngeG;n3chI*)vyOwqrY30p!y|Ui94JuFOxX>k|Z9<
z;H&V?sqfxl@@QM@nyD!cwNOe|pFc)>PV$;nPwxK4#>O)*51fU+lD7ScbQBlgLGN3?
z^dzJO=);#_p|ligaA>SYl=hrZxDx5Rn=L7hhL_bPdp-o{iS*G+N_Le0TE>`Pg!Gs6
znNT>~%?}q{Lr<$i{1DhWX-Vc<%J4YuX?59-qw7BWP_-Qd!9qst(;-Roe?&Iw|NXTZMOk5*
zzxQ`3`R2O4xKc{P%=r4Ry*ES?(4POT<3n8cQ1;<|7^?V_IdM8@V2OFtUpujc19S@O
zcd{8L-f(;O|7|4c__lQvzPFEc?AR5xj$N*CEzf-l^wvKNH3iK7Ui$yN9Q|uSQ<%m@
zBZGg@m~J^cd}YV_-OylzI&^OATu0=^nak1vV$2)Y?U1RqzjD*&^&3=6-eP2Ln!Xp2w&EJ?~^NsV$lB)XpS>j|EO!6HTpR$uLpn0dzI@YCf>&+l}Uv*vs)BllZ^7oQ7*!c`WR;s`F;1F5z*McDE1;1l|v!UAf
z=;n2TMy4k47lFfjVeFj_j7ge&KcZ}|J+F+s7vKO5}mG2Qr=cCy&F%iR6@Lv#|x({-Ij0Bh1}x2)SSXqZ94
z&GpZMbBWSrp*)bVqJW_L
ze;?+=jLH7>Aq_xm1Ai?E{0_uW{KbmP=9t%%f62<|Nfak5#k%c)=E#))^%-mti~s8#
zYW(R@GvfVWe|m1$IJgE^nPmk6{hyv&HHq-HRCb+R0Ro$m$-MrHOn{oHFs=U#+`hP>
z=rSbA>%vgLroYxh;YwHR@9zypX?eN)E$2pqB{UJU7BE0qsL&()y<(z(7-4tu-~0Qj
z?Bv~x=FEPFyC$i8gYrVA<-c|f@rvNKe%FXff0p%M1m2#{UB4mViYp4ad{!!I@
zRW>WuzdwWdv+7Bn^^q41=r?26e>fZx8gkVPx1JX>EXP&sp4R=fK_&btGtq_r6M~1Q
z^E&3n>A*#8-G9GE^heaZ!g@X+{Jrr19+vcOW+kj+BJk!;%wNP%L)0{KnE`>r9X!
zfTzuH<2#pV_@J=wj^8Y*e?o`#u^jn%S~R=!*j^e6CdwhlSI7RKmRl&iCq=?DKYh@p
zteZu)>KP?Pu{87w`DlJGa(U|xtWWT%9!cx2pQ%Tf;SPNWj!r$_f6MlDtLyR^7sqmb
zEgiEe7@KCTP2K#P{jmqAaYtwUs_4A3^`}pHY^I>8adpo8{9E9>ZIsfU`T9!YLw?;7
zqDCzo`fYHY$^xB3@p4)d_vIlvEU;u(X=lp7B2Q02|LyI7g+4%+B{DDm=I#_zEhoRs
z-LB?eS7|AjZlh7yKNr#1fxS7(ujMR;ahfpVXQw#&hRQxzqNnwb&2+hPly;{KeEyg@
zt8-@zE>wF7$Gi&pTrhgE@Ybe*#&?LIgGsTVKQ^Xdc0*lW6R?|Ha
z;qD(tGM#4{y;xO5oq`XK2b(HL<;ZtLn8Vh
zkruq$$B~R~^LoH=zDRDMSbXtJc*Pj|n_F==;7HIrUtq1}9P?vv
z`Pi?qmjN{s&X(0#$p68uvs4S86mY)HX8ggnBeRUxlu=<}(j$!f&5$Oka3AeLQBUo{
zF+tc=L6gM>xz}{qI)y*i8KYWgK59*9b}vinu|EyhEbiS+5oaa0;uY~*n}xzBhQ7>b
z`8PSn;IHv*qfi?lUXf5%SL`P9M?flwIX05%c5EbL*|Ji_A?&6yWw>D@+_S|zzqauM
zR13bFaWh)QPe(FMIHx_jx9;5oI0~H2?)ds!PRzA<(Qw)JunX5SXOyq*q2Sel`*lKJ
z{7De!4e%AsW?j-4Li0~JA(_aB>QtGW+I7p&35Cnn^J_6S8n@m{d%VfDoDHI54a+)u
z_U*uusm4ZIm0U?ULZR9XX>qETkmq$E72MpVC<>B#WRx9VW_+e^V~EN7PcZQbjr
z#lzi!1qA2HM6H#Q_6>5Ic3|tO6#nFqGZSces&mIfy=D_aR>>yUXV1$OwbsAuDEx$k
zW~K?Isy$s%Lx=d=P4D`#I$Xb{e_KvAlG2Hg0fNhzIAK+s5yNzY;*b(T2w?L1ck$$O
zbTj-Kh4g^G7E8iWidP{p2FOXX`UT$XOHm4suUwGe_T-e6$@{@AxYf+In6)UkdnV~#
z?_)+cCBb}=w2Tb5+1c5nF3U?F9OtJ#V5$Nqz04fBmcCQVZjmJnsTA{PcJ4AQB+Ndq
z!pIBwZc4kx6FPUeo-U=6lmsV(y$Pb7j4Y
zJuA+0rD6s2rg^_gKin#LR@>IbjHwQ0PRjlnc&|xK{qZ+n9(NySTvTt*3ZbXiZZL#V
zhcK+}J}A2~zc1jto3DzDL}PRF0{+zjD*J32?9Ivr6C0m=tHpT!V-&Am!--u(Dl*%m
zEk>h#FKnx^tG-_}qPwzi@AD?cM=CB04G~dMTnQSf2NSe&kN=pSzt^H|^SCS5#{Jx6
ztvuc19>TcIP1ws#!op0iR?ouKYXy#WZzQb@vzDeRirr;!5%;?71|O%^$Vdg-569lb
zVe1MmK=Cj0k|byjZb%q0e{9YNm%_H5VKE~d{mBm7NMMXv8s~vh4yC2tE{^94;&8Y}6VCYDo7KS*H|wL%_AbGtl?EYJSH*w6(R1`V
z_ma}-;$|GE`eU&66})is&(docibG?NawVm8M@}
zW=hu#C=O}Btgq18hCXvtl=*OSb}x40-Oe9maGopXbn983&D=Y*Ek&=)d$P*RS=e!a
zPj~BtU5!JL>{7ApBT)MJ1%GfvpEu)7;Z}+lLoP%k=0;BI_H*a}B8nl~k
zFwDhy4!F!#utJz{mtE?m+xwH(vPK&JD9EfpqB2mF&+$7I=N90N-z6l1MV4CRC04Ba
za!e*t>nSeU1O8>1`+)4b>LaddX|Yy|_j^vh@NqnFL@VQnaa&4>MWr*m%w?PG@-Lra
z(U$8B-Jk_-1}FNN3t&Rbbs2BkHd_+7Q$fRs;!p?zQ$&y&ja;;^GSXq_#=R3eE9u!i
z=_dBbh0x-%BiW&Aq$pmOFg#iBUmz0fG~0Hw=+V|*T!RXom6cWRt1B0_mT@>3zfVk@
z3E(=*oBer7qQ>sR+z%dE=H>(q24}*EuGV9T{aP8uTS4)+GwRshWRi8x>v1^{8;D%f
zmDFBuh9k!*JoC_n?~1uZmm~Qq!?bJoEGNxFcTQ*VCA1p?fv_*E#$C%ao7)ke+q1){
zIbrMey+^iOd3S!R|8ONLPSgz@P;U2A`{})nfEqv-p
zG&bFKF`TtDu~HLx!U;P0l8${0VMpI935BZkJR&r)q255rwDxCTy;KDEzBYvbtp^dnWf!M
zOTo22JA16ER2rf#bM1qj9NCi2i?d_B9X-p|UvCTb%5yGU6dw$}AFSygz0JXYe!K2m
z(PW9B^RcC2^}?xQpA;OZ@IJ)EJWyvko-oIh36ry%apjP0z+f7ox~B2bMWb(&H?9Fr
z!FhZfl&r#)mhG!No%HTbXZ!bXDFDUUQzuoCM~_pBFM8{*!%{7c@0@%UbE0H0^1$n}
zQHnma$zXBCp3K5U)ge-Z8&Q4ED)Ya5T-XAGg5>wVn3NWTjh&=gcrzna)H)-T9Iuhv
zFK$FjQ7i^%&Fj^z6@Rpw{`tXTpuD!TlZ~h!Q9d&525%3%J`<(jwGF+3=m|l9OSO@P
zpg@IN!WJ51{*{-1ShE4Pv<;Hxu|Z