Fixed #2628 -- Added django.contrib.sitemap. Thanks for the patch, Dan Watson

git-svn-id: http://code.djangoproject.com/svn/django/trunk@3694 bcc190cf-cafb-0310-a4f2-bffc1f526a37

django/contrib/sitemap/__init__.py

@@ -0,0 +1,90 @@
+from django.core import urlresolvers
+import urllib
+
+PING_URL = "http://www.google.com/webmasters/sitemaps/ping"
+
+class SitemapNotFound(Exception):
+    pass
+
+def ping_google(sitemap_url=None, ping_url=PING_URL):
+    """
+    Alerts 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 urlresolvers.reverse().
+    """
+    if sitemap_url is None:
+        try:
+            # First, try to get the "index" sitemap URL.
+            sitemap_url = urlresolvers.reverse('django.contrib.sitemap.views.index')
+        except urlresolvers.NoReverseMatch:
+            try:
+                # Next, try for the "global" sitemap URL.
+                sitemap_url = urlresolvers.reverse('django.contrib.sitemap.views.sitemap')
+            except urlresolvers.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.")
+
+    from django.contrib.sites.models import Site
+    current_site = Site.objects.get_current()
+    url = "%s%s" % (current_site.domain, sitemap)
+    params = urllib.urlencode({'sitemap':url})
+    urllib.urlopen("%s?%s" % (ping_url, params))
+
+class Sitemap:
+    def __get(self, name, obj, default=None):
+        try:
+            attr = getattr(self, name)
+        except AttributeError:
+            return default
+        if callable(attr):
+            return attr(obj)
+        return attr
+
+    def items(self):
+        return []
+
+    def location(self, obj):
+        return obj.get_absolute_url()
+
+    def get_urls(self):
+        from django.contrib.sites.models import Site
+        current_site = Site.objects.get_current()
+        urls = []
+        for item in self.items():
+            loc = "http://%s%s" % (current_site.domain, self.__get('location', item))
+            url_info = {
+                'location':   loc,
+                'lastmod':    self.__get('lastmod', item, None),
+                'changefreq': self.__get('changefreq', item, None),
+                'priority':   self.__get('priority', item, None)
+            }
+            urls.append(url_info)
+        return urls
+
+class FlatpageSitemap(Sitemap):
+    def items(self):
+        from django.contrib.sites.models import Site
+        current_site = Site.objects.get_current()
+        return current_site.flatpage_set.all()
+
+class GenericSitemap(Sitemap):
+    priority = None
+    changefreq = None
+
+    def __init__(self, info_dict, priority=None, changefreq=None):
+        self.queryset = info_dict['queryset']
+        self.date_field = info_dict.get('date_field', None)
+        self.priority = priority
+        self.changefreq = changefreq
+
+    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

django/contrib/sitemap/templates/sitemap.xml

@@ -0,0 +1,11 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.google.com/schemas/sitemap/0.84">
+{% for url in urlset %}
+  <url>
+    <loc>{{ url.location|escape }}</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 %}
+   </url>
+{% endfor %}
+</urlset>

django/contrib/sitemap/templates/sitemap_index.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<sitemapindex xmlns="http://www.google.com/schemas/sitemap/0.84">
+{% for location in sitemaps %}
+	<sitemap>
+		<loc>{{ location|escape }}</loc>
+	</sitemap>
+{% endfor %}
+</sitemapindex>

django/contrib/sitemap/views.py

@@ -0,0 +1,30 @@
+from django.http import HttpResponse, Http404
+from django.template import loader
+from django.contrib.sites.models import Site
+from django.core import urlresolvers
+
+def index(request, sitemaps):
+    current_site = Site.objects.get_current()
+    sites = []
+    protocol = request.is_secure() and 'https' or 'http'
+    for section in sitemaps.keys():
+        sitemap_url = urlresolvers.reverse('django.contrib.sitemap.views.sitemap', kwargs={'section': section})
+        sites.append('%s://%s%s' % (protocol, current_site.domain, sitemap_url))
+    xml = loader.render_to_string('sitemap_index.xml', {'sitemaps': sites})
+    return HttpResponse(xml, mimetype='application/xml')
+
+def sitemap(request, sitemaps, section=None):
+    maps, urls = [], []
+    if section is not None:
+        if not sitemaps.has_key(section):
+            raise Http404("No sitemap available for section: %r" % section)
+        maps.append(sitemaps[section])
+    else:
+        maps = sitemaps.values()
+    for site in maps:
+        if callable(site):
+            urls.extend(site().get_urls())
+        else:
+            urls.extend(site.get_urls())
+    xml = loader.render_to_string('sitemap.xml', {'urlset': urls})
+    return HttpResponse(xml, mimetype='application/xml')