X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/35b64fd8bec183054b63234aebf8782b87cf5cc5..c3545c19bd62da303fe0f8c49f82d8c340d965ea:/apps/search/views.py diff --git a/apps/search/views.py b/apps/search/views.py index 36dd52cd2..dc9e27b51 100644 --- a/apps/search/views.py +++ b/apps/search/views.py @@ -1,9 +1,10 @@ # -*- coding: utf-8 -*- - +# This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. +# from django.conf import settings from django.shortcuts import render_to_response, get_object_or_404 from django.template import RequestContext -from django.contrib.auth.decorators import login_required from django.views.decorators import cache from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect from django.utils.translation import ugettext as _ @@ -13,9 +14,7 @@ from catalogue.models import Book, Tag, Fragment from pdcounter.models import Author as PDCounterAuthor, BookStub as PDCounterBook from catalogue.views import JSONResponse from search import Search, SearchResult -from lucene import StringReader from suggest.forms import PublishingSuggestForm -from time import sleep import re #import enchant import json @@ -28,6 +27,13 @@ def match_word_re(word): return "[[:<:]]%s[[:>:]]" % word +query_syntax_chars = re.compile(r"[\\/*:(){}]") + + +def remove_query_syntax_chars(query, replace=' '): + return query_syntax_chars.sub(' ', query) + + def did_you_mean(query, tokens): return query # change = {} @@ -59,6 +65,8 @@ def hint(request): if len(prefix) < 2: return JSONResponse([]) + prefix = remove_query_syntax_chars(prefix) + search = Search() # tagi beda ograniczac tutaj # ale tagi moga byc na ksiazce i na fragmentach @@ -114,6 +122,9 @@ def main(request): return render_to_response('catalogue/search_too_short.html', {'prefix': query}, context_instance=RequestContext(request)) + + query = remove_query_syntax_chars(query) + search = Search() theme_terms = search.index.analyze(text=query, field="themes_pl") \ @@ -124,15 +135,17 @@ def main(request): tags = split_tags(tags) author_results = search.search_phrase(query, 'authors', book=True) + translator_results = search.search_phrase(query, 'translators', book=True) + title_results = search.search_phrase(query, 'title', book=True) # Boost main author/title results with mixed search, and save some of its results for end of list. # boost author, title results - author_title_mixed = search.search_some(query, ['authors', 'title', 'tags'], query_terms=theme_terms) + author_title_mixed = search.search_some(query, ['authors', 'translators', 'title', 'tags'], query_terms=theme_terms) author_title_rest = [] for b in author_title_mixed: - also_in_mixed = filter(lambda ba: ba.book_id == b.book_id, author_results + title_results) + also_in_mixed = filter(lambda ba: ba.book_id == b.book_id, author_results + translator_results + title_results) for b2 in also_in_mixed: b2.boost *= 1.1 if also_in_mixed is []: @@ -155,15 +168,17 @@ def main(request): return True return False return f - f = already_found(author_results + title_results + text_phrase) + f = already_found(author_results + translator_results + title_results + text_phrase) everywhere = filter(lambda x: not f(x), everywhere) author_results = SearchResult.aggregate(author_results) + translator_results = SearchResult.aggregate(translator_results) title_results = SearchResult.aggregate(title_results) everywhere = SearchResult.aggregate(everywhere, author_title_rest) for field, res in [('authors', author_results), + ('translators', translator_results), ('title', title_results), ('text', text_phrase), ('text', everywhere)]: @@ -180,13 +195,15 @@ def main(request): return False author_results = filter(ensure_exists, author_results) + translator_results = filter(ensure_exists, translator_results) title_results = filter(ensure_exists, title_results) text_phrase = filter(ensure_exists, text_phrase) everywhere = filter(ensure_exists, everywhere) - results = author_results + title_results + text_phrase + everywhere + results = author_results + translator_results + title_results + text_phrase + everywhere # ensure books do exists & sort them - results.sort(reverse=True) + for res in (author_results, translator_results, title_results, text_phrase, everywhere): + res.sort(reverse=True) # We don't want to redirect to book text, but rather display result page even with one result. # if len(results) == 1: @@ -209,6 +226,7 @@ def main(request): {'tags': tags, 'prefix': query, 'results': {'author': author_results, + 'translator': translator_results, 'title': title_results, 'content': text_phrase, 'other': everywhere},