docker setup
This commit is contained in:
AdrienLSH
@ -0,0 +1,263 @@
|
||||
import warnings
|
||||
from urllib.parse import urlencode
|
||||
from urllib.request import urlopen
|
||||
|
||||
from django.apps import apps as django_apps
|
||||
from django.conf import settings
|
||||
from django.core import paginator
|
||||
from django.core.exceptions import ImproperlyConfigured
|
||||
from django.urls import NoReverseMatch, reverse
|
||||
from django.utils import translation
|
||||
from django.utils.deprecation import RemovedInDjango50Warning
|
||||
|
||||
PING_URL = "https://www.google.com/webmasters/tools/ping"
|
||||
|
||||
|
||||
class SitemapNotFound(Exception):
|
||||
pass
|
||||
|
||||
|
||||
def ping_google(sitemap_url=None, ping_url=PING_URL, sitemap_uses_https=True):
|
||||
"""
|
||||
Alert Google that the sitemap for the current site has been updated.
|
||||
If sitemap_url is provided, it should be an absolute path to the sitemap
|
||||
for this site -- e.g., '/sitemap.xml'. If sitemap_url is not provided, this
|
||||
function will attempt to deduce it by using urls.reverse().
|
||||
"""
|
||||
sitemap_full_url = _get_sitemap_full_url(sitemap_url, sitemap_uses_https)
|
||||
params = urlencode({"sitemap": sitemap_full_url})
|
||||
urlopen("%s?%s" % (ping_url, params))
|
||||
|
||||
|
||||
def _get_sitemap_full_url(sitemap_url, sitemap_uses_https=True):
|
||||
if not django_apps.is_installed("django.contrib.sites"):
|
||||
raise ImproperlyConfigured(
|
||||
"ping_google requires django.contrib.sites, which isn't installed."
|
||||
)
|
||||
|
||||
if sitemap_url is None:
|
||||
try:
|
||||
# First, try to get the "index" sitemap URL.
|
||||
sitemap_url = reverse("django.contrib.sitemaps.views.index")
|
||||
except NoReverseMatch:
|
||||
try:
|
||||
# Next, try for the "global" sitemap URL.
|
||||
sitemap_url = reverse("django.contrib.sitemaps.views.sitemap")
|
||||
except NoReverseMatch:
|
||||
pass
|
||||
|
||||
if sitemap_url is None:
|
||||
raise SitemapNotFound(
|
||||
"You didn't provide a sitemap_url, and the sitemap URL couldn't be "
|
||||
"auto-detected."
|
||||
)
|
||||
|
||||
Site = django_apps.get_model("sites.Site")
|
||||
current_site = Site.objects.get_current()
|
||||
scheme = "https" if sitemap_uses_https else "http"
|
||||
return "%s://%s%s" % (scheme, current_site.domain, sitemap_url)
|
||||
|
||||
|
||||
class Sitemap:
|
||||
# This limit is defined by Google. See the index documentation at
|
||||
# https://www.sitemaps.org/protocol.html#index.
|
||||
limit = 50000
|
||||
|
||||
# If protocol is None, the URLs in the sitemap will use the protocol
|
||||
# with which the sitemap was requested.
|
||||
protocol = None
|
||||
|
||||
# Enables generating URLs for all languages.
|
||||
i18n = False
|
||||
|
||||
# Override list of languages to use.
|
||||
languages = None
|
||||
|
||||
# Enables generating alternate/hreflang links.
|
||||
alternates = False
|
||||
|
||||
# Add an alternate/hreflang link with value 'x-default'.
|
||||
x_default = False
|
||||
|
||||
def _get(self, name, item, default=None):
|
||||
try:
|
||||
attr = getattr(self, name)
|
||||
except AttributeError:
|
||||
return default
|
||||
if callable(attr):
|
||||
if self.i18n:
|
||||
# Split the (item, lang_code) tuples again for the location,
|
||||
# priority, lastmod and changefreq method calls.
|
||||
item, lang_code = item
|
||||
return attr(item)
|
||||
return attr
|
||||
|
||||
def get_languages_for_item(self, item):
|
||||
"""Languages for which this item is displayed."""
|
||||
return self._languages()
|
||||
|
||||
def _languages(self):
|
||||
if self.languages is not None:
|
||||
return self.languages
|
||||
return [lang_code for lang_code, _ in settings.LANGUAGES]
|
||||
|
||||
def _items(self):
|
||||
if self.i18n:
|
||||
# Create (item, lang_code) tuples for all items and languages.
|
||||
# This is necessary to paginate with all languages already considered.
|
||||
items = [
|
||||
(item, lang_code)
|
||||
for item in self.items()
|
||||
for lang_code in self.get_languages_for_item(item)
|
||||
]
|
||||
return items
|
||||
return self.items()
|
||||
|
||||
def _location(self, item, force_lang_code=None):
|
||||
if self.i18n:
|
||||
obj, lang_code = item
|
||||
# Activate language from item-tuple or forced one before calling location.
|
||||
with translation.override(force_lang_code or lang_code):
|
||||
return self._get("location", item)
|
||||
return self._get("location", item)
|
||||
|
||||
@property
|
||||
def paginator(self):
|
||||
return paginator.Paginator(self._items(), self.limit)
|
||||
|
||||
def items(self):
|
||||
return []
|
||||
|
||||
def location(self, item):
|
||||
return item.get_absolute_url()
|
||||
|
||||
def get_protocol(self, protocol=None):
|
||||
# Determine protocol
|
||||
if self.protocol is None and protocol is None:
|
||||
warnings.warn(
|
||||
"The default sitemap protocol will be changed from 'http' to "
|
||||
"'https' in Django 5.0. Set Sitemap.protocol to silence this "
|
||||
"warning.",
|
||||
category=RemovedInDjango50Warning,
|
||||
stacklevel=2,
|
||||
)
|
||||
# RemovedInDjango50Warning: when the deprecation ends, replace 'http'
|
||||
# with 'https'.
|
||||
return self.protocol or protocol or "http"
|
||||
|
||||
def get_domain(self, site=None):
|
||||
# Determine domain
|
||||
if site is None:
|
||||
if django_apps.is_installed("django.contrib.sites"):
|
||||
Site = django_apps.get_model("sites.Site")
|
||||
try:
|
||||
site = Site.objects.get_current()
|
||||
except Site.DoesNotExist:
|
||||
pass
|
||||
if site is None:
|
||||
raise ImproperlyConfigured(
|
||||
"To use sitemaps, either enable the sites framework or pass "
|
||||
"a Site/RequestSite object in your view."
|
||||
)
|
||||
return site.domain
|
||||
|
||||
def get_urls(self, page=1, site=None, protocol=None):
|
||||
protocol = self.get_protocol(protocol)
|
||||
domain = self.get_domain(site)
|
||||
return self._urls(page, protocol, domain)
|
||||
|
||||
def get_latest_lastmod(self):
|
||||
if not hasattr(self, "lastmod"):
|
||||
return None
|
||||
if callable(self.lastmod):
|
||||
try:
|
||||
return max([self.lastmod(item) for item in self.items()], default=None)
|
||||
except TypeError:
|
||||
return None
|
||||
else:
|
||||
return self.lastmod
|
||||
|
||||
def _urls(self, page, protocol, domain):
|
||||
urls = []
|
||||
latest_lastmod = None
|
||||
all_items_lastmod = True # track if all items have a lastmod
|
||||
|
||||
paginator_page = self.paginator.page(page)
|
||||
for item in paginator_page.object_list:
|
||||
loc = f"{protocol}://{domain}{self._location(item)}"
|
||||
priority = self._get("priority", item)
|
||||
lastmod = self._get("lastmod", item)
|
||||
|
||||
if all_items_lastmod:
|
||||
all_items_lastmod = lastmod is not None
|
||||
if all_items_lastmod and (
|
||||
latest_lastmod is None or lastmod > latest_lastmod
|
||||
):
|
||||
latest_lastmod = lastmod
|
||||
|
||||
url_info = {
|
||||
"item": item,
|
||||
"location": loc,
|
||||
"lastmod": lastmod,
|
||||
"changefreq": self._get("changefreq", item),
|
||||
"priority": str(priority if priority is not None else ""),
|
||||
"alternates": [],
|
||||
}
|
||||
|
||||
if self.i18n and self.alternates:
|
||||
item_languages = self.get_languages_for_item(item[0])
|
||||
for lang_code in item_languages:
|
||||
loc = f"{protocol}://{domain}{self._location(item, lang_code)}"
|
||||
url_info["alternates"].append(
|
||||
{
|
||||
"location": loc,
|
||||
"lang_code": lang_code,
|
||||
}
|
||||
)
|
||||
if self.x_default and settings.LANGUAGE_CODE in item_languages:
|
||||
lang_code = settings.LANGUAGE_CODE
|
||||
loc = f"{protocol}://{domain}{self._location(item, lang_code)}"
|
||||
loc = loc.replace(f"/{lang_code}/", "/", 1)
|
||||
url_info["alternates"].append(
|
||||
{
|
||||
"location": loc,
|
||||
"lang_code": "x-default",
|
||||
}
|
||||
)
|
||||
|
||||
urls.append(url_info)
|
||||
|
||||
if all_items_lastmod and latest_lastmod:
|
||||
self.latest_lastmod = latest_lastmod
|
||||
|
||||
return urls
|
||||
|
||||
|
||||
class GenericSitemap(Sitemap):
|
||||
priority = None
|
||||
changefreq = None
|
||||
|
||||
def __init__(self, info_dict, priority=None, changefreq=None, protocol=None):
|
||||
self.queryset = info_dict["queryset"]
|
||||
self.date_field = info_dict.get("date_field")
|
||||
self.priority = self.priority or priority
|
||||
self.changefreq = self.changefreq or changefreq
|
||||
self.protocol = self.protocol or protocol
|
||||
|
||||
def items(self):
|
||||
# Make sure to return a clone; we don't want premature evaluation.
|
||||
return self.queryset.filter()
|
||||
|
||||
def lastmod(self, item):
|
||||
if self.date_field is not None:
|
||||
return getattr(item, self.date_field)
|
||||
return None
|
||||
|
||||
def get_latest_lastmod(self):
|
||||
if self.date_field is not None:
|
||||
return (
|
||||
self.queryset.order_by("-" + self.date_field)
|
||||
.values_list(self.date_field, flat=True)
|
||||
.first()
|
||||
)
|
||||
return None
|
||||
@ -0,0 +1,8 @@
|
||||
from django.apps import AppConfig
|
||||
from django.utils.translation import gettext_lazy as _
|
||||
|
||||
|
||||
class SiteMapsConfig(AppConfig):
|
||||
default_auto_field = "django.db.models.AutoField"
|
||||
name = "django.contrib.sitemaps"
|
||||
verbose_name = _("Site Maps")
|
||||
@ -0,0 +1,16 @@
|
||||
from django.contrib.sitemaps import ping_google
|
||||
from django.core.management.base import BaseCommand
|
||||
|
||||
|
||||
class Command(BaseCommand):
|
||||
help = "Ping Google with an updated sitemap, pass optional url of sitemap"
|
||||
|
||||
def add_arguments(self, parser):
|
||||
parser.add_argument("sitemap_url", nargs="?")
|
||||
parser.add_argument("--sitemap-uses-http", action="store_true")
|
||||
|
||||
def handle(self, *args, **options):
|
||||
ping_google(
|
||||
sitemap_url=options["sitemap_url"],
|
||||
sitemap_uses_https=not options["sitemap_uses_http"],
|
||||
)
|
||||
@ -0,0 +1,16 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:xhtml="http://www.w3.org/1999/xhtml">
|
||||
{% spaceless %}
|
||||
{% for url in urlset %}
|
||||
<url>
|
||||
<loc>{{ url.location }}</loc>
|
||||
{% if url.lastmod %}<lastmod>{{ url.lastmod|date:"Y-m-d" }}</lastmod>{% endif %}
|
||||
{% if url.changefreq %}<changefreq>{{ url.changefreq }}</changefreq>{% endif %}
|
||||
{% if url.priority %}<priority>{{ url.priority }}</priority>{% endif %}
|
||||
{% for alternate in url.alternates %}
|
||||
<xhtml:link rel="alternate" hreflang="{{ alternate.lang_code }}" href="{{ alternate.location }}"/>
|
||||
{% endfor %}
|
||||
</url>
|
||||
{% endfor %}
|
||||
{% endspaceless %}
|
||||
</urlset>
|
||||
@ -0,0 +1,13 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<sitemapindex xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
|
||||
{% spaceless %}
|
||||
{% for site in sitemaps %}
|
||||
<sitemap>
|
||||
<loc>{{ site.location }}</loc>
|
||||
{% if site.last_mod %}
|
||||
<lastmod>{{ site.last_mod|date:"c" }}</lastmod>
|
||||
{% endif %}
|
||||
</sitemap>
|
||||
{% endfor %}
|
||||
{% endspaceless %}
|
||||
</sitemapindex>
|
||||
@ -0,0 +1,151 @@
|
||||
import datetime
|
||||
import warnings
|
||||
from dataclasses import dataclass
|
||||
from functools import wraps
|
||||
|
||||
from django.contrib.sites.shortcuts import get_current_site
|
||||
from django.core.paginator import EmptyPage, PageNotAnInteger
|
||||
from django.http import Http404
|
||||
from django.template.response import TemplateResponse
|
||||
from django.urls import reverse
|
||||
from django.utils import timezone
|
||||
from django.utils.deprecation import RemovedInDjango50Warning
|
||||
from django.utils.http import http_date
|
||||
|
||||
|
||||
@dataclass
|
||||
class SitemapIndexItem:
|
||||
location: str
|
||||
last_mod: bool = None
|
||||
|
||||
# RemovedInDjango50Warning
|
||||
def __str__(self):
|
||||
msg = (
|
||||
"Calling `__str__` on SitemapIndexItem is deprecated, use the `location` "
|
||||
"attribute instead."
|
||||
)
|
||||
warnings.warn(msg, RemovedInDjango50Warning, stacklevel=2)
|
||||
return self.location
|
||||
|
||||
|
||||
def x_robots_tag(func):
|
||||
@wraps(func)
|
||||
def inner(request, *args, **kwargs):
|
||||
response = func(request, *args, **kwargs)
|
||||
response.headers["X-Robots-Tag"] = "noindex, noodp, noarchive"
|
||||
return response
|
||||
|
||||
return inner
|
||||
|
||||
|
||||
def _get_latest_lastmod(current_lastmod, new_lastmod):
|
||||
"""
|
||||
Returns the latest `lastmod` where `lastmod` can be either a date or a
|
||||
datetime.
|
||||
"""
|
||||
if not isinstance(new_lastmod, datetime.datetime):
|
||||
new_lastmod = datetime.datetime.combine(new_lastmod, datetime.time.min)
|
||||
if timezone.is_naive(new_lastmod):
|
||||
new_lastmod = timezone.make_aware(new_lastmod, datetime.timezone.utc)
|
||||
return new_lastmod if current_lastmod is None else max(current_lastmod, new_lastmod)
|
||||
|
||||
|
||||
@x_robots_tag
|
||||
def index(
|
||||
request,
|
||||
sitemaps,
|
||||
template_name="sitemap_index.xml",
|
||||
content_type="application/xml",
|
||||
sitemap_url_name="django.contrib.sitemaps.views.sitemap",
|
||||
):
|
||||
req_protocol = request.scheme
|
||||
req_site = get_current_site(request)
|
||||
|
||||
sites = [] # all sections' sitemap URLs
|
||||
all_indexes_lastmod = True
|
||||
latest_lastmod = None
|
||||
for section, site in sitemaps.items():
|
||||
# For each section label, add links of all pages of its sitemap
|
||||
# (usually generated by the `sitemap` view).
|
||||
if callable(site):
|
||||
site = site()
|
||||
protocol = req_protocol if site.protocol is None else site.protocol
|
||||
sitemap_url = reverse(sitemap_url_name, kwargs={"section": section})
|
||||
absolute_url = "%s://%s%s" % (protocol, req_site.domain, sitemap_url)
|
||||
site_lastmod = site.get_latest_lastmod()
|
||||
if all_indexes_lastmod:
|
||||
if site_lastmod is not None:
|
||||
latest_lastmod = _get_latest_lastmod(latest_lastmod, site_lastmod)
|
||||
else:
|
||||
all_indexes_lastmod = False
|
||||
sites.append(SitemapIndexItem(absolute_url, site_lastmod))
|
||||
# Add links to all pages of the sitemap.
|
||||
for page in range(2, site.paginator.num_pages + 1):
|
||||
sites.append(
|
||||
SitemapIndexItem("%s?p=%s" % (absolute_url, page), site_lastmod)
|
||||
)
|
||||
# If lastmod is defined for all sites, set header so as
|
||||
# ConditionalGetMiddleware is able to send 304 NOT MODIFIED
|
||||
if all_indexes_lastmod and latest_lastmod:
|
||||
headers = {"Last-Modified": http_date(latest_lastmod.timestamp())}
|
||||
else:
|
||||
headers = None
|
||||
return TemplateResponse(
|
||||
request,
|
||||
template_name,
|
||||
{"sitemaps": sites},
|
||||
content_type=content_type,
|
||||
headers=headers,
|
||||
)
|
||||
|
||||
|
||||
@x_robots_tag
|
||||
def sitemap(
|
||||
request,
|
||||
sitemaps,
|
||||
section=None,
|
||||
template_name="sitemap.xml",
|
||||
content_type="application/xml",
|
||||
):
|
||||
req_protocol = request.scheme
|
||||
req_site = get_current_site(request)
|
||||
|
||||
if section is not None:
|
||||
if section not in sitemaps:
|
||||
raise Http404("No sitemap available for section: %r" % section)
|
||||
maps = [sitemaps[section]]
|
||||
else:
|
||||
maps = sitemaps.values()
|
||||
page = request.GET.get("p", 1)
|
||||
|
||||
lastmod = None
|
||||
all_sites_lastmod = True
|
||||
urls = []
|
||||
for site in maps:
|
||||
try:
|
||||
if callable(site):
|
||||
site = site()
|
||||
urls.extend(site.get_urls(page=page, site=req_site, protocol=req_protocol))
|
||||
if all_sites_lastmod:
|
||||
site_lastmod = getattr(site, "latest_lastmod", None)
|
||||
if site_lastmod is not None:
|
||||
lastmod = _get_latest_lastmod(lastmod, site_lastmod)
|
||||
else:
|
||||
all_sites_lastmod = False
|
||||
except EmptyPage:
|
||||
raise Http404("Page %s empty" % page)
|
||||
except PageNotAnInteger:
|
||||
raise Http404("No page '%s'" % page)
|
||||
# If lastmod is defined for all sites, set header so as
|
||||
# ConditionalGetMiddleware is able to send 304 NOT MODIFIED
|
||||
if all_sites_lastmod:
|
||||
headers = {"Last-Modified": http_date(lastmod.timestamp())} if lastmod else None
|
||||
else:
|
||||
headers = None
|
||||
return TemplateResponse(
|
||||
request,
|
||||
template_name,
|
||||
{"urlset": urls},
|
||||
content_type=content_type,
|
||||
headers=headers,
|
||||
)
|
||||
Reference in New Issue
Block a user