def view_recap_document(request, docket_id=None, doc_num=None, att_num=None,
slug=''):
"""This view can either load an attachment or a regular document,
depending on the URL pattern that is matched.
"""
item = get_object_or_404(
RECAPDocument,
docket_entry__docket__id=docket_id,
document_number=doc_num,
attachment_number=att_num,
)
title = make_rd_title(item)
if is_og_bot(request):
make_thumb_if_needed(item)
item.refresh_from_db()
try:
fave = Favorite.objects.get(recap_doc_id=item.pk, user=request.user)
except (ObjectDoesNotExist, TypeError):
# Not favorited or anonymous user
favorite_form = FavoriteForm(initial={
'recap_doc_id': item.pk,
'name': trunc(title, 100, ellipsis='...'),
})
else:
favorite_form = FavoriteForm(instance=fave)
return render(request, 'recap_document.html', {
'document': item,
'title': title,
'favorite_form': favorite_form,
'private': True, # Always True for RECAP docs.
})
def view_recap_document(request, docket_id=None, doc_num=None, att_num=None,
slug=''):
"""This view can either load an attachment or a regular document,
depending on the URL pattern that is matched.
"""
item = get_object_or_404(
RECAPDocument,
docket_entry__docket__id=docket_id,
document_number=doc_num,
attachment_number=att_num,
)
title = make_rd_title(item)
if is_og_bot(request):
make_thumb_if_needed(item)
item.refresh_from_db()
try:
fave = Favorite.objects.get(recap_doc_id=item.pk, user=request.user)
except (ObjectDoesNotExist, TypeError):
# Not favorited or anonymous user
favorite_form = FavoriteForm(initial={
'recap_doc_id': item.pk,
'name': trunc(title, 100, ellipsis='...'),
})
else:
favorite_form = FavoriteForm(instance=fave)
return render(request, 'recap_document.html', {
'document': item,
'title': title,
'favorite_form': favorite_form,
'private': True, # Always True for RECAP docs.
})
def serve_static_file(request, file_path=''):
"""Sends a static file to a user.
This serves up the static case files such as the PDFs in a way that can be
blocked from search engines if necessary. We do four things:
- Look up the document or audio file associated with the filepath
- Check if it's blocked
- If blocked, we set the x-robots-tag HTTP header
- Serve up the file using Apache2's xsendfile
"""
response = HttpResponse()
file_loc = os.path.join(settings.MEDIA_ROOT, file_path.encode('utf-8'))
if file_path.startswith('mp3'):
item = get_object_or_404(Audio, local_path_mp3=file_path)
mimetype = 'audio/mpeg'
elif file_path.startswith('recap'):
# Either we serve up a special HTML file to make open graph crawlers
# happy, or we serve the PDF to make a human happy.
og_disabled = bool(request.GET.get('no-og'))
if is_og_bot(request) and not og_disabled:
# Serve up the regular HTML page, which has the twitter card info.
try:
rd = RECAPDocument.objects.get(filepath_local=file_path)
except (RECAPDocument.DoesNotExist,
RECAPDocument.MultipleObjectsReturned):
pass
else:
return view_recap_document(
request,
docket_id=rd.docket_entry.docket_id,
doc_num=rd.document_number,
att_num=rd.attachment_number
)
# A human or unable to find the item in the DB. Create an empty object,
# and set it to blocked. (All recap pdfs are blocked.)
item = RECAPDocument()
item.blocked = True
mimetype = 'application/pdf'
else:
item = get_object_or_404(Opinion, local_path=file_path)
item.blocked = item.cluster.blocked
try:
mimetype = magic.from_file(file_loc, mime=True)
except IOError:
raise Http404
if item.blocked:
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
if settings.DEVELOPMENT:
# X-Sendfile will only confuse you in a dev env.
response.content = open(file_loc, 'r').read()
else:
response['X-Sendfile'] = file_loc
file_name = file_path.split('/')[-1]
response['Content-Disposition'] = 'inline; filename="%s"' % \
file_name.encode('utf-8')
response['Content-Type'] = mimetype
return response
def serve_static_file(request, file_path=''):
"""Sends a static file to a user.
This serves up the static case files such as the PDFs in a way that can be
blocked from search engines if necessary. We do four things:
- Look up the document or audio file associated with the filepath
- Check if it's blocked
- If blocked, we set the x-robots-tag HTTP header
- Serve up the file using Apache2's xsendfile
"""
response = HttpResponse()
file_loc = os.path.join(settings.MEDIA_ROOT, file_path.encode('utf-8'))
if file_path.startswith('mp3'):
item = get_object_or_404(Audio, local_path_mp3=file_path)
mimetype = 'audio/mpeg'
elif file_path.startswith('recap'):
# Either we serve up a special HTML file to make open graph crawlers
# happy, or we serve the PDF to make a human happy.
og_disabled = bool(request.GET.get('no-og'))
if is_og_bot(request) and not og_disabled:
# Serve up the regular HTML page, which has the twitter card info.
try:
rd = RECAPDocument.objects.get(filepath_local=file_path)
except (RECAPDocument.DoesNotExist,
RECAPDocument.MultipleObjectsReturned):
pass
else:
return view_recap_document(
request,
docket_id=rd.docket_entry.docket_id,
doc_num=rd.document_number,
att_num=rd.attachment_number
)
# A human or unable to find the item in the DB. Create an empty object,
# and set it to blocked. (All recap pdfs are blocked.)
item = RECAPDocument()
item.blocked = True
mimetype = 'application/pdf'
else:
item = get_object_or_404(Opinion, local_path=file_path)
item.blocked = item.cluster.blocked
try:
mimetype = magic.from_file(file_loc, mime=True)
except IOError:
raise Http404
if item.blocked:
response['X-Robots-Tag'] = 'noindex, noodp, noarchive, noimageindex'
if settings.DEVELOPMENT:
# X-Sendfile will only confuse you in a dev env.
response.content = open(file_loc, 'r').read()
else:
response['X-Sendfile'] = file_loc
file_name = file_path.split('/')[-1]
response['Content-Disposition'] = 'inline; filename="%s"' % \
file_name.encode('utf-8')
response['Content-Type'] = mimetype
return response
def serve_static_file(request, file_path=""):
"""Serve a recap file or redirect to HTML if it's a bot.
Use nginx's X-Accel system to set headers without putting files in memory.
"""
# If it's a open graph crawler, serve the HTML page so they can get
# thumbnails instead of serving the PDF binary.
og_disabled = bool(request.GET.get("no-og"))
if is_og_bot(request) and not og_disabled:
# Serve up the regular HTML page, which has the twitter card info.
try:
rd = RECAPDocument.objects.get(filepath_local=file_path)
except (
RECAPDocument.DoesNotExist,
RECAPDocument.MultipleObjectsReturned,
):
# Fall through; serve it normally.
pass
else:
return view_recap_document(
request,
docket_id=rd.docket_entry.docket_id,
doc_num=rd.document_number,
att_num=rd.attachment_number,
)
response = HttpResponse()
response["Content-Type"] = "application/pdf"
file_name = file_path.split("/")[-1]
# HTTP headers didn't get encoding figured out until recently. As a result
# content disposition headers are a mess. Luckily, we can steal the junk
# below from Django.
try:
# Try with ascii. If it works, do it.
file_name.encode("ascii")
file_expr = 'filename="{}"'.format(file_name)
except UnicodeEncodeError:
# Ascii failed. Do utf-8 params.
file_expr = "filename*=utf-8''{}".format(quote(file_name))
response["Content-Disposition"] = "inline; %s" % file_expr
# Use microcache for RECAP PDFs. This should help with traffic bursts.
response["X-Accel-Expires"] = "5"
# Block all RECAP PDFs
response["X-Robots-Tag"] = "noindex, noodp, noarchive, noimageindex"
if settings.DEVELOPMENT:
# X-Accel-Redirect will only confuse you in a dev env.
file_loc = os.path.join(settings.MEDIA_ROOT, file_path)
with open(file_loc, "rb") as f:
response.content = f.read()
else:
file_loc = os.path.join("/protected/", file_path)
response["X-Accel-Redirect"] = file_loc
return response
def view_recap_document(request,
docket_id=None,
doc_num=None,
att_num=None,
slug=""):
"""This view can either load an attachment or a regular document,
depending on the URL pattern that is matched.
"""
item = get_object_or_404(
RECAPDocument,
docket_entry__docket__id=docket_id,
document_number=doc_num,
attachment_number=att_num,
)
# Check if the user has requested automatic redirection to the document
redirectToDownload = request.GET.get("redirectToDownload", False)
if redirectToDownload:
# Check if the document is available from Court Listener and
# if it is, redirect the user to that
# if it isn't, redirect the user to PACER
if item.is_available:
response = redirect(item.filepath_local)
else:
response = redirect(item.pacer_url)
return response
title = make_rd_title(item)
if is_og_bot(request):
make_thumb_if_needed(item)
item.refresh_from_db()
try:
fave = Favorite.objects.get(recap_doc_id=item.pk, user=request.user)
except (ObjectDoesNotExist, TypeError):
# Not favorited or anonymous user
favorite_form = FavoriteForm(initial={
"recap_doc_id": item.pk,
"name": trunc(title, 100, ellipsis="..."),
})
else:
favorite_form = FavoriteForm(instance=fave)
return render(
request,
"recap_document.html",
{
"document": item,
"title": title,
"favorite_form": favorite_form,
"private": True, # Always True for RECAP docs.
},
)
def make_thumb_if_needed(request, rd):
"""Make a thumbnail for a RECAP Document, if needed
If a thumbnail is needed, can be made and should be made, make one.
:param request: The request sent to the server
:param rd: A RECAPDocument object
"""
needs_thumb = rd.thumbnail_status != THUMBNAIL_STATUSES.COMPLETE
if all([needs_thumb, rd.has_valid_pdf, is_og_bot(request)]):
make_recap_document_thumbnail_from_pdf(rd.pk)
rd.refresh_from_db()
return rd
def make_thumb_if_needed(
request: HttpRequest,
rd: RECAPDocument,
) -> RECAPDocument:
"""Make a thumbnail for a RECAP Document, if needed
If a thumbnail is needed, can be made and should be made, make one.
:param request: The request sent to the server
:param rd: A RECAPDocument object
"""
needs_thumb = rd.thumbnail_status != THUMBNAIL_STATUSES.COMPLETE
if all([needs_thumb, rd.has_valid_pdf, is_og_bot(request)]):
make_png_thumbnail_for_instance(
pk=rd.pk,
klass=RECAPDocument,
file_attr="filepath_local",
max_dimension=1068,
)
rd.refresh_from_db()
return rd
