X-Git-Url: https://git.mdrn.pl/wolnelektury.git/blobdiff_plain/58554ad41d789d21c7477c01fd8c27ff1f590445..8885fd14f5f040eb3aa69f1d705856f99a1fa572:/src/search/views.py diff --git a/src/search/views.py b/src/search/views.py index da3a767f8..2f9adab36 100644 --- a/src/search/views.py +++ b/src/search/views.py @@ -1,30 +1,22 @@ -# This file is part of Wolnelektury, licensed under GNU Affero GPLv3 or later. -# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. +# This file is part of Wolne Lektury, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Wolne Lektury. See NOTICE for more information. # from django.conf import settings -from django.http.response import HttpResponseRedirect from django.shortcuts import render from django.views.decorators import cache from django.http import HttpResponse, JsonResponse +from sorl.thumbnail import get_thumbnail -from catalogue.models import Book, Tag -from pdcounter.models import Author -from picture.models import Picture -from search.index import Search, SearchResult, PictureResult -from suggest.forms import PublishingSuggestForm +import catalogue.models +import infopages.models +import picture.models +from .forms import SearchFilters import re import json from wolnelektury.utils import re_escape -def match_word_re(word): - if 'sqlite' in settings.DATABASES['default']['ENGINE']: - return r"\b%s\b" % word - elif 'mysql' in settings.DATABASES['default']['ENGINE']: - return "[[:<:]]%s[[:>:]]" % word - - query_syntax_chars = re.compile(r"[\\/*:(){}?.[\]+]") @@ -32,32 +24,6 @@ def remove_query_syntax_chars(query, replace=' '): return query_syntax_chars.sub(replace, query) -def did_you_mean(query, tokens): - return query - # change = {} - # for t in tokens: - # authors = Tag.objects.filter(category='author', name__iregex=match_word_re(t)) - # if len(authors) > 0: - # continue - - # if False: - # if not dictionary.check(t): - # try: - # change_to = dictionary.suggest(t)[0].lower() - # if change_to != t.lower(): - # change[t] = change_to - # except IndexError: - # pass - - # if change == {}: - # return None - - # for frm, to in change.items(): - # query = query.replace(frm, to) - - # return query - - @cache.never_cache def hint(request, mozhint=False, param='term'): prefix = request.GET.get(param, '') @@ -74,26 +40,91 @@ def hint(request, mozhint=False, param='term'): if limit < 1: limit = 20 - authors = Tag.objects.filter( - category='author', name_pl__iregex='\m' + prefix).only('name', 'id', 'slug', 'category') - data = [ - { - 'label': author.name, - 'id': author.id, - 'url': author.get_absolute_url(), - } - for author in authors[:limit] - ] + data = [] if len(data) < limit: - data += [ + authors = catalogue.models.Tag.objects.filter( + category='author', name_pl__iregex='\m' + prefix).only('name', 'id', 'slug', 'category') + data.extend([ { - 'label': b.title, - 'author': b.author_unicode(), - 'id': b.id, - 'url': b.get_absolute_url() + 'type': 'author', + 'label': author.name, + 'url': author.get_absolute_gallery_url() if author.for_pictures else author.get_absolute_url(), + 'img': get_thumbnail(author.photo, '72x72', crop='top').url if author.photo else '', } - for b in Book.objects.filter(findable=True, title__iregex='\m' + prefix)[:limit-len(data)] - ] + for author in authors[:limit - len(data)] + ]) + if request.user.is_authenticated and len(data) < limit: + tags = catalogue.models.Tag.objects.filter( + category='set', user=request.user, name_pl__iregex='\m' + prefix).only('name', 'id', 'slug', 'category') + data.extend([ + { + 'type': 'set', + 'label': tag.name, + 'url': tag.get_absolute_url(), + } + for tag in tags[:limit - len(data)] + ]) + if len(data) < limit: + tags = catalogue.models.Tag.objects.filter( + category__in=('theme', 'genre', 'epoch', 'kind'), name_pl__iregex='\m' + prefix).only('name', 'id', 'slug', 'category') + data.extend([ + { + 'type': tag.category, + 'label': tag.name, + 'url': tag.get_absolute_gallery_url() if tag.for_pictures else tag.get_absolute_url(), + } + for tag in tags[:limit - len(data)] + ]) + if len(data) < limit: + collections = catalogue.models.Collection.objects.filter( + title_pl__iregex='\m' + prefix).only('title', 'slug') + data.extend([ + { + 'type': 'collection', + 'label': collection.title, + 'url': collection.get_absolute_url(), + } + for collection in collections[:limit - len(data)] + ]) + if len(data) < limit: + for b in catalogue.models.Book.objects.filter(findable=True, title__iregex='\m' + prefix)[:limit-len(data)]: + author_str = b.author_unicode() + translator = b.translator() + if translator: + author_str += ' (tłum. ' + translator + ')' + data.append( + { + 'type': 'book', + 'label': b.title, + 'author': author_str, + 'url': b.get_absolute_url(), + 'img': get_thumbnail(b.cover_clean, '72x72').url if b.cover_clean else '', + } + ) + if len(data) < limit: + arts = picture.models.Picture.objects.filter( + title__iregex='\m' + prefix).only('title', 'id', 'slug') # img? + data.extend([ + { + 'type': 'art', + 'label': art.title, + 'author': art.author_unicode(), + 'url': art.get_absolute_url(), + 'img': get_thumbnail(art.image_file, '72x72').url if art.image_file else '', + } + for art in arts[:limit - len(data)] + ]) + if len(data) < limit: + infos = infopages.models.InfoPage.objects.filter( + title_pl__iregex='\m' + prefix).only('title', 'id', 'slug') + data.extend([ + { + 'type': 'info', + 'label': info.title, + 'url': info.get_absolute_url(), + } + for info in infos[:limit - len(data)] + ]) if mozhint: data = [ @@ -101,6 +132,14 @@ def hint(request, mozhint=False, param='term'): [ item['label'] for item in data + ], + [ + item.get('author', '') + for item in data + ], + [ + item['url'] + for item in data ] ] @@ -112,139 +151,19 @@ def hint(request, mozhint=False, param='term'): return JsonResponse(data, safe=False) + @cache.never_cache -def main(request): - query = request.GET.get('q', '') - if len(query) < 2: - return render( - request, 'catalogue/search_too_short.html', - {'prefix': query}) - elif len(query) > 256: - return render( - request, 'catalogue/search_too_long.html', - {'prefix': query}) - - query = prepare_query(query) - pd_authors = search_pd_authors(query) - books = search_books(query) - pictures = search_pictures(query) - suggestion = '' - - if not (books or pictures or pd_authors): - form = PublishingSuggestForm(initial={"books": query + ", "}) - return render( - request, - 'catalogue/search_no_hits.html', - { - 'form': form, - 'did_you_mean': suggestion - }) - - if not (books or pictures) and len(pd_authors) == 1: - return HttpResponseRedirect(pd_authors[0].get_absolute_url()) - - return render( - request, - 'catalogue/search_multiple_hits.html', - { - 'pd_authors': pd_authors, - 'books': books, - 'pictures': pictures, - 'did_you_mean': suggestion - }) - -def search_books(query): - search = Search() - results_parts = [] - search_fields = [] - words = query.split() - fieldsets = ( - (['authors', 'authors_nonstem'], True), - (['title', 'title_nonstem'], True), - (['metadata', 'metadata_nonstem'], True), - (['text', 'text_nonstem', 'themes_pl', 'themes_pl_nonstem'], False), - ) - for fields, is_book in fieldsets: - search_fields += fields - results_parts.append(search.search_words(words, search_fields, required=fields, book=is_book)) - results = [] - ids_results = {} - for results_part in results_parts: - for result in sorted(SearchResult.aggregate(results_part), reverse=True): - book_id = result.book_id - if book_id in ids_results: - ids_results[book_id].merge(result) - else: - results.append(result) - ids_results[book_id] = result - descendant_ids = set( - Book.objects.filter(id__in=ids_results, ancestor__in=ids_results).values_list('id', flat=True)) - results = [result for result in results if result.book_id not in descendant_ids] - for result in results: - search.get_snippets(result, query, num=3) - - def ensure_exists(r): - try: - return r.book - except Book.DoesNotExist: - return False - - results = [r for r in results if ensure_exists(r)] - return results - - -def search_pictures(query): - search = Search() - results_parts = [] - search_fields = [] - words = query.split() - fieldsets = ( - (['authors', 'authors_nonstem'], True), - (['title', 'title_nonstem'], True), - (['metadata', 'metadata_nonstem'], True), - (['themes_pl', 'themes_pl_nonstem'], False), - ) - for fields, is_book in fieldsets: - search_fields += fields - results_parts.append(search.search_words(words, search_fields, required=fields, book=is_book, picture=True)) - results = [] - ids_results = {} - for results_part in results_parts: - for result in sorted(PictureResult.aggregate(results_part), reverse=True): - picture_id = result.picture_id - if picture_id in ids_results: - ids_results[picture_id].merge(result) - else: - results.append(result) - ids_results[picture_id] = result - - def ensure_exists(r): - try: - return r.picture - except Picture.DoesNotExist: - return False - - results = [r for r in results if ensure_exists(r)] - return results - - -def search_pd_authors(query): - pd_authors = Author.objects.filter(name__icontains=query) - existing_slugs = Tag.objects.filter( - category='author', slug__in=list(pd_authors.values_list('slug', flat=True))) \ - .values_list('slug', flat=True) - pd_authors = pd_authors.exclude(slug__in=existing_slugs) - return pd_authors - - -def prepare_query(query): - query = ' '.join(query.split()) - # filter out private use characters - import unicodedata - query = ''.join(ch for ch in query if unicodedata.category(ch) != 'Co') - query = remove_query_syntax_chars(query) - - words = query.split() - if len(words) > 10: - query = ' '.join(words[:10]) - return query +def search(request): + filters = SearchFilters(request.GET) + ctx = { + 'title': 'Wynik wyszukiwania', + 'query': request.GET.get('q', ''), + 'filters': filters, + } + if filters.is_valid(): + ctx['results'] = filters.results() + for k, v in ctx['results'].items(): + if v: + ctx['hasresults'] = True + break + return render(request, 'search/results.html', ctx)