def recall_document(title, url, uuid, text):
doc = None
if url:
try:
doc = SearchDocument.objects.lookup_by_url(url)
except SearchDocument.DoesNotExist:
pass
if uuid and not doc:
try:
doc = SearchDocument.objects.get(uuid=uuid)
except SearchDocument.DoesNotExist:
pass
if text and not doc:
doc = SearchDocument.objects.filter(text=text)
if len(doc) > 0:
doc = doc[0]
if not doc:
raise SearchDocument.DoesNotExist()
# If the extractor fails to extract a title, this will copy a title from
# a later submission from an improved extractor.
if title and not doc.title:
doc.title = title
doc.save()
return doc
def fetch_and_store(url):
try:
(title, text) = fetch_and_clean(url)
except Exception:
raise Http404(url)
doc = SearchDocument()
doc.url = url
doc.title = title
doc.text = text
doc.save()
return doc
def fetch_and_store(url):
try:
(title, text) = fetch_and_clean(url)
except Exception:
raise Http404(url)
doc = SearchDocument()
doc.url = url
doc.title = title
doc.text = text
doc.save()
return doc
url = doc.url
text = doc.text
except UnicodeDecodeError:
raise
except SearchDocument.DoesNotExist:
pass
except Exception, e:
return HttpResponseServerError(str(e))
if not doc:
if not text:
return HttpResponseNotFound(str(url or uuid))
else:
doc = SearchDocument()
doc.text = text
if title:
doc.title = title
if url:
doc.url = url
ua_string = request.META.get('HTTP_USER_AGENT')
if ua_string is not None:
doc.user_agent = ua_string[:255]
doc.save()
# The actual proxying:
response = execute_search(doc, doctype)
if isinstance(response, HttpResponse):
return response
url = doc.url
text = doc.text
except UnicodeDecodeError:
raise
except SearchDocument.DoesNotExist:
pass
except Exception, e:
return HttpResponseServerError(str(e))
if not doc:
if not text:
return HttpResponseNotFound(str(url or uuid))
else:
doc = SearchDocument()
doc.text = text
if title:
doc.title = title
if url:
doc.url = url
ua_string = request.META.get('HTTP_USER_AGENT')
if ua_string is not None:
doc.user_agent = ua_string[:255]
doc.save()
# The actual proxying:
response = execute_search(doc, doctype)
if isinstance(response, HttpResponse):
return response