ドキュメントで参照されている「FileIndex」は、haystack.indexes.SearchIndex の仮想サブクラスです。以下に例を示します。
from haystack import indexes
from myapp.models import MyFile
class FileIndex(indexes.SearchIndex, indexes.Indexable):
text = indexes.CharField(document=True, use_template=True)
title = indexes.CharField(model_attr='title')
owner = indexes.CharField(model_attr='owner__name')
def get_model(self):
return MyFile
def index_queryset(self, using=None):
return self.get_model().objects.all()
def prepare(self, obj):
data = super(FileIndex, self).prepare(obj)
# This could also be a regular Python open() call, a StringIO instance
# or the result of opening a URL. Note that due to a library limitation
# file_obj must have a .name attribute even if you need to set one
# manually before calling extract_file_contents:
file_obj = obj.the_file.open()
extracted_data = self.backend.extract_file_contents(file_obj)
# Now we'll finally perform the template processing to render the
# text field with *all* of our metadata visible for templating:
t = loader.select_template(('search/indexes/myapp/myfile_text.txt', ))
data['text'] = t.render(Context({'object': obj,
'extracted': extracted_data}))
return data
そのため、PDF/ DOCXextracted_data
コンテンツを抽出するために思いついたプロセスに置き換えられます。次に、テンプレートを更新してそのデータを含めます。