def setUp(self):
# Set up testing cores in Solr and swap them in
self.core_name_opinion = settings.SOLR_OPINION_TEST_CORE_NAME
self.core_name_audio = settings.SOLR_AUDIO_TEST_CORE_NAME
self.core_name_people = settings.SOLR_PEOPLE_TEST_CORE_NAME
self.core_name_recap = settings.SOLR_RECAP_TEST_CORE_NAME
self.si_opinion = sunburnt.SolrInterface(
settings.SOLR_OPINION_URL, mode="rw"
)
self.si_audio = sunburnt.SolrInterface(
settings.SOLR_AUDIO_URL, mode="rw"
)
self.si_people = sunburnt.SolrInterface(
settings.SOLR_PEOPLE_URL, mode="rw"
)
# This will cause headaches, but it follows in the mission to slowly
# migrate off of sunburnt. This was added after the items above, and so
# uses scorched, not sunburnt.
self.si_recap = scorched.SolrInterface(
settings.SOLR_RECAP_URL, mode="rw"
)
self.all_sis = [
self.si_opinion,
self.si_audio,
self.si_people,
self.si_recap,
]
def setUp(self):
# Set up testing cores in Solr and swap them in
self.core_name_opinion = settings.SOLR_OPINION_TEST_CORE_NAME
self.core_name_audio = settings.SOLR_AUDIO_TEST_CORE_NAME
self.core_name_people = settings.SOLR_PEOPLE_TEST_CORE_NAME
self.core_name_recap = settings.SOLR_RECAP_TEST_CORE_NAME
root = settings.INSTALL_ROOT
create_temp_solr_core(
self.core_name_opinion,
os.path.join(root, 'Solr', 'conf', 'schema.xml'),
)
create_temp_solr_core(
self.core_name_audio,
os.path.join(root, 'Solr', 'conf', 'audio_schema.xml'),
)
create_temp_solr_core(
self.core_name_people,
os.path.join(root, 'Solr', 'conf', 'person_schema.xml'),
)
create_temp_solr_core(
self.core_name_recap,
os.path.join(root, 'Solr', 'conf', 'recap_schema.xml'))
self.si_opinion = sunburnt.SolrInterface(settings.SOLR_OPINION_URL,
mode='rw')
self.si_audio = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL,
mode='rw')
self.si_people = sunburnt.SolrInterface(settings.SOLR_PEOPLE_URL,
mode='rw')
# This will cause headaches, but it follows in the mission to slowly
# migrate off of sunburnt. This was added after the items above, and so
# uses scorched, not sunburnt.
self.si_recap = scorched.SolrInterface(settings.SOLR_RECAP_URL,
mode='rw')
def __init__(self, *args, **kwargs):
super(Command, self).__init__(*args, **kwargs)
self.connections = {
'o': sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r'),
'oa': sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='r'),
}
self.options = {}
self.valid_ids = {}
def index_sitemap_maker(request):
"""Generate a sitemap index page
Counts the number of cases in the site, divides by `items_per_sitemap` and
provides links items.
"""
connection_string_obj_type_pairs = (
(settings.SOLR_OPINION_URL, 'opinions'),
(settings.SOLR_AUDIO_URL, 'oral-arguments'),
)
sites = []
for connection_string, obj_type in connection_string_obj_type_pairs:
conn = sunburnt.SolrInterface(connection_string, mode='r')
search_results_object = conn.raw_query(**index_solr_params).execute()
count = search_results_object.result.numFound
num_pages = count / items_per_sitemap + 1
for i in range(1, num_pages + 1):
sites.append('https://www.courtlistener.com/sitemap-%s.xml?p=%s' %
(obj_type, i))
# Random additional sitemaps.
sites.extend([
'https://www.courtlistener.com/sitemap-simple-pages.xml',
'https://www.courtlistener.com/sitemap-visualizations.xml',
])
xml = loader.render_to_string('sitemap_index.xml', {'sitemaps': sites})
# These links contain case names, so they should get crawled but not
# indexed
response = HttpResponse(xml, content_type='application/xml')
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
return response
def people_sitemap_maker(request):
conn = sunburnt.SolrInterface(settings.SOLR_PEOPLE_URL, mode='r')
page = int(request.GET.get("p", 1))
start = (page - 1) * items_per_sitemap
params = {
'q': '*:*',
'rows': items_per_sitemap,
'start': start,
'fl': ','.join([
'absolute_url',
'timestamp',
]),
'sort': 'dob asc',
'caller': 'people_sitemap_maker',
}
results = conn.raw_query(**params).execute()
# Translate Solr object into something Django's template can use
urls = []
for result in results:
urls.append({
'location':
'https://www.courtlistener.com%s' % result['absolute_url'],
'changefreq':
'monthly',
'lastmod':
result['timestamp'],
'priority':
'0.5',
})
xml = smart_str(loader.render_to_string('sitemap.xml', {'urlset': urls}))
response = HttpResponse(xml, content_type='application/xml')
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
return response
def coverage_data(request, version, court):
"""Provides coverage data for a court.
Responds to either AJAX or regular requests.
"""
if court != 'all':
court_str = get_object_or_404(Court, pk=court).pk
else:
court_str = 'all'
q = request.GET.get('q')
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
response = conn.raw_query(**build_coverage_query(court_str, q)).execute()
counts = response.facet_counts.facet_ranges[0][1][0][1]
counts = strip_zero_years(counts)
# Calculate the totals
annual_counts = {}
total_docs = 0
for date_string, count in counts:
annual_counts[date_string[:4]] = count
total_docs += count
return JsonResponse({
'annual_counts': annual_counts,
'total': total_docs,
},
safe=True)
def make_court_variable():
courts = Court.objects.exclude(jurisdiction=Court.TESTING_COURT)
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
response = conn.raw_query(**build_court_count_query()).execute()
court_count_tuples = response.facet_counts.facet_fields['court_exact']
courts = annotate_courts_with_counts(courts, court_count_tuples)
return courts
def add_or_update_items(items, solr_url=settings.SOLR_OPINION_URL):
"""Adds an item to a solr index.
This function is for use with the update_index command. It's slightly
different than the commands below because it expects a Django object,
rather than a primary key. This rejects the standard Celery advice about
not passing objects around, but thread safety shouldn't be an issue since
this is only used by the update_index command, and we want to query and
build the SearchDocument objects in the task, not in its caller.
"""
si = sunburnt.SolrInterface(solr_url, mode='w')
if hasattr(items, "items") or not hasattr(items, "__iter__"):
# If it's a dict or a single item make it a list
items = [items]
search_item_list = []
for item in items:
try:
if type(item) == Audio:
search_item_list.append(SearchAudioFile(item))
elif type(item) == Opinion:
search_item_list.append(SearchDocument(item))
except AttributeError as e:
print "AttributeError trying to add: %s\n %s" % (item, e)
except ValueError as e:
print "ValueError trying to add: %s\n %s" % (item, e)
except InvalidDocumentError:
print "Unable to parse: %s" % item
try:
si.add(search_item_list)
except socket.error, exc:
add_or_update_items.retry(exc=exc, countdown=120)
def view_opinion(request, pk, _):
"""Using the cluster ID, return the cluster of opinions.
We also test if the cluster ID is a favorite for the user, and send data
if needed. If it's a favorite, we send the bound form for the favorite so
it can populate the form on the page. If it is not a favorite, we send the
unbound form.
"""
# Look up the court, cluster, title and favorite information
cluster = get_object_or_404(OpinionCluster, pk=pk)
title = '%s, %s' % (
trunc(best_case_name(cluster), 100),
cluster.citation_string,
)
get_string = search_utils.make_get_string(request)
try:
fave = Favorite.objects.get(
cluster_id=cluster.pk,
user=request.user,
)
favorite_form = FavoriteForm(instance=fave)
except (ObjectDoesNotExist, TypeError):
# Not favorited or anonymous user
favorite_form = FavoriteForm(
initial={
'cluster_id': cluster.pk,
'name': trunc(best_case_name(cluster), 100, ellipsis='...'),
}
)
# Get the citing results from Solr for speed.
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
q = {
'q': 'cites:({ids})'.format(
ids=' OR '.join([str(pk) for pk in
(cluster.sub_opinions
.values_list('pk', flat=True))])
),
'rows': 5,
'start': 0,
'sort': 'citeCount desc',
'caller': 'view_opinion',
}
citing_clusters = conn.raw_query(**q).execute()
return render_to_response(
'view_opinion.html',
{
'title': title,
'cluster': cluster,
'favorite_form': favorite_form,
'get_string': get_string,
'private': cluster.blocked,
'citing_clusters': citing_clusters,
'top_authorities': cluster.authorities[:5],
},
RequestContext(request)
)
def add_or_update_cluster(pk, force_commit=True):
si = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='w')
try:
si.add([SearchDocument(item) for item in
OpinionCluster.objects.get(pk=pk).sub_opinions.all()])
if force_commit:
si.commit()
except SolrError, exc:
add_or_update_cluster.retry(exc=exc, countdown=30)
def add_or_update_people(item_pks, force_commit=True):
si = sunburnt.SolrInterface(settings.SOLR_PEOPLE_URL, mode='w')
try:
si.add([SearchPerson(item) for item in
Person.objects.filter(pk__in=item_pks)])
if force_commit:
si.commit()
except SolrError, exc:
add_or_update_people.retry(exc=exc, countdown=30)
def add_or_update_opinions(item_pks, force_commit=True):
si = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='w')
try:
si.add([SearchDocument(item) for item in
Opinion.objects.filter(pk__in=item_pks)])
if force_commit:
si.commit()
except SolrError, exc:
add_or_update_opinions.retry(exc=exc, countdown=30)
def add_or_update_audio_files(item_pks, force_commit=True):
si = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='w')
try:
si.add([SearchAudioFile(item) for item in
Audio.objects.filter(pk__in=item_pks)])
if force_commit:
si.commit()
except SolrError, exc:
add_or_update_audio_files.retry(exc=exc, countdown=30)
def items(self, obj):
conn = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='r')
params = {
'q': '*:*',
'sort': 'dateArgued desc',
'rows': '20',
'start': '0',
'caller': 'AllJurisdictionsPodcast',
}
return conn.raw_query(**params).execute()
def place_facet_queries(cd,
conn=sunburnt.SolrInterface(settings.SOLR_OPINION_URL,
mode='r')):
"""Get facet values for the status filters
Using the search form, query Solr and get the values for the status filters.
"""
# Build up all the queries needed
facet_params = {
'rows': '0',
'facet': 'true',
'facet.mincount': 0,
'facet.field': '{!ex=dt}status_exact',
'q': cd['q'] or '*:*',
'caller': 'facet_parameters',
}
fq = []
# Case Name and judges
if cd['case_name']:
fq.append(make_fq(cd, 'caseName', 'case_name'))
if cd['judge']:
fq.append(make_fq(cd, 'judge', 'judge'))
# Citations
if cd['citation']:
fq.append(make_fq_proximity_query(cd, 'citation', 'citation'))
if cd['docket_number']:
fq.append(make_fq(cd, 'docketNumber', 'docket_number'))
if cd['neutral_cite']:
fq.append(make_fq(cd, 'neutralCite', 'neutral_cite'))
fq.append(
make_date_query('dateFiled', cd['filed_before'], cd['filed_after']))
fq.append(make_cite_count_query(cd))
# Faceting
selected_courts_string = get_selected_field_string(
cd, 'court_') # Status facets depend on court checkboxes
selected_stats_string = get_selected_field_string(cd, 'stat_')
if len(selected_stats_string) > 0:
fq.extend([
'{!tag=dt}status_exact:(%s)' % selected_stats_string,
'court_exact:(%s)' % selected_courts_string
])
# If a param has been added to the fq variables, then we add them to the
# main_params var. Otherwise, we don't, as doing so throws an error.
if len(fq) > 0:
facet_params['fq'] = fq
stat_facet_fields = conn.raw_query(
**facet_params).execute().facet_counts.facet_fields
return stat_facet_fields
def setUp(self):
# Set up testing cores in Solr and swap them in
self.core_name_opinion = settings.SOLR_OPINION_TEST_CORE_NAME
self.core_name_audio = settings.SOLR_AUDIO_TEST_CORE_NAME
create_solr_core(self.core_name_opinion)
create_solr_core(
self.core_name_audio,
schema=os.path.join(settings.INSTALL_ROOT, 'Solr', 'conf',
'audio_schema.xml'),
instance_dir='/usr/local/solr/example/solr/audio',
)
# swap_solr_core('collection1', self.core_name_opinion)
# swap_solr_core('audio', self.core_name_audio)
self.si_opinion = sunburnt.SolrInterface(settings.SOLR_OPINION_URL,
mode='rw')
self.si_audio = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL,
mode='rw')
self.si_opinion.commit()
self.si_audio.commit()
def items(self, obj):
"""Do a Solr query here. Return the first 20 results"""
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
params = {
'q': '*:*',
'sort': 'dateFiled desc',
'rows': '20',
'start': '0',
'caller': 'AllJurisdictionsFeed',
}
return conn.raw_query(**params).execute()
def oral_argument_sitemap_maker(request):
conn = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='r')
page = int(request.GET.get("p", 1))
start = (page - 1) * items_per_sitemap
params = {
'q':
'*:*',
'rows':
items_per_sitemap,
'start':
start,
'fl':
','.join([
'absolute_url',
'dateArgued',
'local_path',
'citeCount',
'timestamp',
]),
'sort':
'dateArgued asc',
'caller':
'oral_argument_sitemap_maker',
}
search_results_object = conn.raw_query(**params).execute()
# Translate Solr object into something Django's template can use
urls = []
for result in search_results_object:
url_strs = ['https://www.courtlistener.com%s' % result['absolute_url']]
if result.get('local_path') and result.get('local_path') != '':
url_strs.append('https://www.courtlistener.com/%s' %
result['local_path'])
sitemap_item = {}
for url_str in url_strs:
sitemap_item['location'] = url_str
sitemap_item['changefreq'] = 'yearly'
sitemap_item['lastmod'] = result['timestamp']
if any(s in url_str for s in ['mp3']):
sitemap_item['priority'] = '0.3'
else:
sitemap_item['priority'] = '0.5'
urls.append(dict(sitemap_item))
xml = smart_str(loader.render_to_string('sitemap.xml', {'urlset': urls}))
# These links contain case names, so they should get crawled but not
# indexed
response = HttpResponse(xml, content_type='application/xml')
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
return response
def items(self, obj):
"""
Returns a list of items to publish in this feed.
"""
conn = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='r')
params = {
'q': '*:*',
'fq': 'court_exact:%s' % obj.pk,
'sort': 'dateArgued desc',
'rows': '20',
'start': '0',
'caller': 'JurisdictionPodcast',
}
return conn.raw_query(**params).execute()
def do_search(request, rows=20, order_by=None, type=None):
# Bind the search form.
search_form = SearchForm(request.GET)
if search_form.is_valid():
cd = search_form.cleaned_data
# Allows an override by calling methods.
if order_by:
cd['order_by'] = order_by
if type:
cd['type'] = type
search_form = _clean_form(request, cd)
try:
if cd['type'] == 'o':
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL,
mode='r')
stat_facet_fields = search_utils.place_facet_queries(cd, conn)
status_facets = search_utils.make_stats_variable(
stat_facet_fields, search_form)
elif cd['type'] == 'oa':
conn = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL,
mode='r')
status_facets = None
results_si = conn.raw_query(**search_utils.build_main_query(cd))
courts = Court.objects.filter(in_use=True).values(
'pk', 'short_name', 'jurisdiction',
'has_oral_argument_scraper')
courts, court_count_human, court_count = search_utils\
.merge_form_with_courts(courts, search_form)
except Exception, e:
logger.warning("Error loading search with request: %s" %
request.GET)
logger.warning("Error was %s" % e)
return {'error': True}
def items(self, obj):
search_form = SearchForm(obj.GET)
if search_form.is_valid():
cd = search_form.cleaned_data
conn = sunburnt.SolrInterface(settings.SOLR_AUDIO_URL, mode='r')
main_params = search_utils.build_main_query(cd, highlight=False)
main_params.update({
'sort': 'dateArgued desc',
'rows': '20',
'start': '0',
'caller': 'SearchFeed',
})
return conn.raw_query(**main_params).execute()
else:
return []
def setUp(self):
# Set up testing cores in Solr and swap them in
self.core_name_opinion = settings.SOLR_OPINION_TEST_CORE_NAME
self.core_name_audio = settings.SOLR_AUDIO_TEST_CORE_NAME
self.core_name_people = settings.SOLR_PEOPLE_TEST_CORE_NAME
root = settings.INSTALL_ROOT
create_temp_solr_core(
self.core_name_opinion,
os.path.join(root, 'Solr', 'conf', 'schema.xml'),
)
create_temp_solr_core(
self.core_name_audio,
os.path.join(root, 'Solr', 'conf', 'audio_schema.xml'),
)
create_temp_solr_core(
self.core_name_people,
os.path.join(root, 'Solr', 'conf', 'person_schema.xml'),
)
self.si_opinion = sunburnt.SolrInterface(
settings.SOLR_OPINION_URL, mode='rw')
self.si_audio = sunburnt.SolrInterface(
settings.SOLR_AUDIO_URL, mode='rw')
self.si_people = sunburnt.SolrInterface(
settings.SOLR_PEOPLE_URL, mode='rw')
def match_citation(citation, citing_doc=None):
"""For a citation object, try to match it to an item in the database using
a variety of heuristics.
Returns:
- a Solr Result object with the results, or an empty list if no hits
"""
# TODO: Create shared solr connection for all queries
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
main_params = {
'q': '*',
'fq': [
'status:Precedential', # Non-precedential documents aren't cited
],
'caller': 'citation.match_citations.match_citation',
}
if citing_doc is not None:
# Eliminate self-cites.
main_params['fq'].append('-id:%s' % citing_doc.pk)
# Set up filter parameters
if citation.year:
start_year = end_year = citation.year
else:
start_year, end_year = get_years_from_reporter(citation)
if citing_doc is not None and citing_doc.cluster.date_filed:
end_year = min(end_year, citing_doc.cluster.date_filed.year)
main_params['fq'].append(
'dateFiled:%s' % build_date_range(start_year, end_year)
)
if citation.court:
main_params['fq'].append('court_exact:%s' % citation.court)
# Take 1: Use a phrase query to search the citation field.
main_params['fq'].append(
'citation:("%s")' % citation.base_citation()
)
results = conn.raw_query(**main_params).execute()
if len(results) == 1:
return results
if len(results) > 1:
if citing_doc is not None and citation.defendant:
# Refine using defendant, if there is one
results = case_name_query(conn, main_params, citation, citing_doc)
return results
# Give up.
return []
def handle(self, *args, **options):
self.verbosity = int(options.get('verbosity', 1))
self.solr_url = options['solr_url']
self.si = sunburnt.SolrInterface(self.solr_url, mode='rw')
self.options = options
self.type = options['type']
self.noinput = options['noinput']
if options['update']:
if self.verbosity >= 1:
self.stdout.write('Running in update mode...\n')
if options.get('everything'):
self.add_or_update_all()
elif options.get('datetime'):
self.add_or_update_by_datetime(options['datetime'])
elif options.get('query'):
self.stderr.write("Updating by query not implemented.")
sys.exit(1)
elif options.get('items'):
self.add_or_update(*options['items'])
elif options.get('delete'):
if self.verbosity >= 1:
self.stdout.write('Running in deletion mode...\n')
if options.get('everything'):
self.delete_all()
elif options.get('datetime'):
self.delete_by_datetime(options['datetime'])
elif options.get('query'):
self.delete_by_query(options['query'])
elif options.get('items'):
self.delete(*options['items'])
elif options.get('do_commit'):
self.commit()
elif options.get('optimize'):
self.optimize()
else:
self.stderr.write('Error: You must specify whether you wish to '
'update, delete, commit, or optimize your '
'index.\n')
sys.exit(1)
def get_expected_item_count(self):
# OpinionsSitemap uses the solr index to generate the page, so the only
# accurate count comes from the index itself which will also be based on
# the fixtures.
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
opinion_solr_params['rows'] = 1000
search_results_object = conn.raw_query(**opinion_solr_params).execute()
# the underlying SitemapTest relies on counting url elements in the xml
# response...this logic mimics the creation of the xml, so we at least
# know what we *should* get getting for a count if the SiteMapTest's
# HTTP client-based test gets an HTTP 200
count = 0
for result in search_results_object:
if result.get('local_path'):
count += 3
else:
count += 2
return count
def handle(self, *args, **options):
super(Command, self).handle(*args, **options)
both_list_and_endpoints = options.get("doc_id") is not None and (
options.get("start_id") is not None or options.get("end_id")
is not None or options.get("filed_after") is not None)
no_option = not any([
options.get("doc_id") is None,
options.get("start_id") is None,
options.get("end_id") is None,
options.get("filed_after") is None,
options.get("all") is False,
])
if both_list_and_endpoints or no_option:
raise CommandError("Please specify either a list of documents, a "
"range of ids, a range of dates, or "
"everything.")
self.index = options["index"]
self.si = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode="rw")
# Use query chaining to build the query
query = Opinion.objects.all()
if options.get("doc_id"):
query = query.filter(pk__in=options.get("doc_id"))
if options.get("end_id"):
query = query.filter(pk__lte=options.get("end_id"))
if options.get("start_id"):
query = query.filter(pk__gte=options.get("start_id"))
if options.get("filed_after"):
query = query.filter(
cluster__date_filed__gte=options["filed_after"])
if options.get("all"):
query = Opinion.objects.all()
self.count = query.count()
self.average_per_s = 0
self.timings = []
count = query.count()
opinion_pks = query.values_list("pk", flat=True).iterator()
self.update_documents(opinion_pks, count)
self.add_to_solr()
def handle(self, *args, **options):
super(Command, self).handle(*args, **options)
both_list_and_endpoints = (options.get('doc_id') is not None and
(options.get('start_id') is not None or
options.get('end_id') is not None or
options.get('filed_after') is not None))
no_option = (not any([options.get('doc_id') is None,
options.get('start_id') is None,
options.get('end_id') is None,
options.get('filed_after') is None,
options.get('all') is False]))
if both_list_and_endpoints or no_option:
raise CommandError('Please specify either a list of documents, a '
'range of ids, a range of dates, or '
'everything.')
self.index = options['index']
self.si = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='rw')
# Use query chaining to build the query
query = Opinion.objects.all()
if options.get('doc_id'):
query = query.filter(pk__in=options.get('doc_id'))
if options.get('end_id'):
query = query.filter(pk__lte=options.get('end_id'))
if options.get('start_id'):
query = query.filter(pk__gte=options.get('start_id'))
if options.get('filed_after'):
query = query.filter(
cluster__date_filed__gte=options['filed_after'])
if options.get('all'):
query = Opinion.objects.all()
self.count = query.count()
self.average_per_s = 0
self.timings = []
count = query.count()
opinion_pks = query.values_list('pk', flat=True).iterator()
self.update_documents(opinion_pks, count)
self.add_to_solr()
def opinion_sitemap_maker(request):
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='r')
page = request.GET.get("p", '1')
opinion_solr_params['start'] = (int(page) - 1) * items_per_sitemap
search_results_object = conn.raw_query(**opinion_solr_params).execute()
# Translate Solr object into something Django's template can use
urls = []
for result in search_results_object:
url_strs = [
'https://www.courtlistener.com%s' % result['absolute_url'],
'https://www.courtlistener.com%sauthorities/' %
result['absolute_url']
]
if result.get('local_path') and result.get('local_path') != '':
url_strs.append(
'https://www.courtlistener.com/%s' % result['local_path'])
sitemap_item = {}
for url_str in url_strs:
sitemap_item['location'] = url_str
sitemap_item['changefreq'] = 'yearly'
sitemap_item['lastmod'] = result['timestamp']
if any(s in url_str for s in
['authorities', 'pdf', 'doc', 'wpd']):
sitemap_item['priority'] = '0.3'
else:
sitemap_item['priority'] = '0.5'
urls.append(dict(sitemap_item))
xml = smart_str(loader.render_to_string('sitemap.xml', {'urlset': urls}))
# These links contain case names, so they should get crawled but not
# indexed
response = HttpResponse(xml, content_type='application/xml')
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
return response
def get_citing_clusters_with_cache(cluster, is_bot):
"""Use Solr to get clusters citing the one we're looking at
If it's not a bot, cache the results for a long time. If it is a bot, load
those results if they exist. Otherwise, return None.
:param cluster: The cluster we're targeting
:type cluster: OpinionCluster
:param is_bot: Whether the page running this was loaded by a bot
:type is_bot: bool
:return: A search result of the top five citing clusters or None
:rtype: SolrSearch or None
"""
cache_key = "citing:%s" % cluster.pk
cache = caches["db_cache"]
if is_bot:
# If the cache was set by a real user, bots can access it. But if no
# user set the cache, this will just return None.
return cache.get(cache_key)
# Get the citing results from Solr for speed. Only do this for humans
# to save on disk usage.
sub_opinion_pks = cluster.sub_opinions.values_list("pk", flat=True)
ids_str = " OR ".join([str(pk) for pk in sub_opinion_pks])
q = {
"q": "cites:(%s)" % ids_str,
"rows": 5,
"start": 0,
"sort": "citeCount desc",
"caller": "view_opinion",
}
conn = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode="r")
citing_clusters = conn.raw_query(**q).execute()
a_month = 60 * 60 * 24 * 30
cache.set(cache_key, citing_clusters, a_month)
return citing_clusters
def handle(self, *args, **options):
both_list_and_endpoints = (options.get('doc_id') is not None and
(options.get('start_id') is not None
or options.get('end_id') is not None
or options.get('filed_after') is not None))
no_option = (not any([
options.get('doc_id') is None,
options.get('start_id') is None,
options.get('end_id') is None,
options.get('filed_after') is None,
options.get('all') is False
]))
if both_list_and_endpoints or no_option:
raise CommandError('Please specify either a list of documents, a '
'range of ids, a range of dates, or '
'everything.')
self.index = options['index']
self.si = sunburnt.SolrInterface(settings.SOLR_OPINION_URL, mode='rw')
# Use query chaining to build the query
query = Opinion.objects.all()
if options.get('doc_id'):
query = query.filter(pk__in=options.get('doc_id'))
if options.get('end_id'):
query = query.filter(pk__lte=options.get('end_id'))
if options.get('start_id'):
query = query.filter(pk__gte=options.get('start_id'))
if options.get('filed_after'):
query = query.filter(
cluster__date_filed__gte=options['filed_after'])
if options.get('all'):
query = Opinion.objects.all()
count = query.count()
docs = queryset_generator(query, chunksize=10000)
self.update_documents(docs, count)
