X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/ec4c9d6d4e2ce8c31123c48fb140b1d75e762155..4b2c05eedf0c6b3d213ce0b760d641c345e3c9b3:/src/search/index.py?ds=sidebyside diff --git a/src/search/index.py b/src/search/index.py index 2d84cb4d7..68a2b3b18 100644 --- a/src/search/index.py +++ b/src/search/index.py @@ -9,6 +9,9 @@ import os import re from django.conf import settings from librarian import dcparser +import librarian.meta.types.date +import librarian.meta.types.person +import librarian.meta.types.text from librarian.parser import WLDocument from lxml import etree import scorched @@ -95,7 +98,10 @@ class Snippets(object): of the snippet stored there. """ self.file.seek(pos[0], 0) - txt = self.file.read(pos[1]).decode('utf-8') + try: + txt = self.file.read(pos[1]).decode('utf-8') + except: + return '' return txt def close(self): @@ -315,21 +321,20 @@ class Index(SolrIndex): if hasattr(book_info, field.name): if not getattr(book_info, field.name): continue - # since no type information is available, we use validator - type_indicator = field.validator - if type_indicator == dcparser.as_unicode: + type_indicator = field.value_type + if issubclass(type_indicator, librarian.meta.types.text.TextValue): s = getattr(book_info, field.name) if field.multiple: s = ', '.join(s) fields[field.name] = s - elif type_indicator == dcparser.as_person: + elif issubclass(type_indicator, librarian.meta.types.person.Person): p = getattr(book_info, field.name) - if isinstance(p, dcparser.Person): + if isinstance(p, librarian.meta.types.person.Person): persons = str(p) else: persons = ', '.join(map(str, p)) fields[field.name] = persons - elif type_indicator == dcparser.as_date: + elif issubclass(type_indicator, librarian.meta.types.date.DateValue): dt = getattr(book_info, field.name) fields[field.name] = dt @@ -636,7 +641,7 @@ class SearchResult(object): if self._book is not None: return self._book try: - self._book = catalogue.models.Book.objects.get(id=self.book_id) + self._book = catalogue.models.Book.objects.get(id=self.book_id, findable=True) except catalogue.models.Book.DoesNotExist: self._book = None return self._book @@ -903,7 +908,7 @@ class Search(SolrIndex): def search_by_author(self, words): from catalogue.models import Book - books = Book.objects.filter(parent=None).order_by('-popularity__count') + books = Book.objects.filter(parent=None, findable=True).order_by('-popularity__count') for word in words: books = books.filter(cached_author__iregex='\m%s\M' % word).select_related('popularity__count') return [SearchResult.from_book(book, how_found='search_by_author', query_terms=words) for book in books[:30]] @@ -977,7 +982,7 @@ class Search(SolrIndex): idx += 1 except IOError as e: - book = catalogue.models.Book.objects.filter(id=book_id) + book = catalogue.models.Book.objects.filter(id=book_id, findable=True) if not book: log.error("Book does not exist for book id = %d" % book_id) elif not book.get().children.exists():