from librarian.parser import WLDocument
from lxml import etree
import catalogue.models
+from pdcounter.models import Author as PDCounterAuthor
from multiprocessing.pool import ThreadPool
from threading import current_thread
import atexit
doc.add(Field("tag_category", tag.category, Field.Store.NO, Field.Index.NOT_ANALYZED))
self.index.addDocument(doc)
+ for pdtag in PDCounterAuthor.objects.all():
+ doc = Document()
+ doc.add(NumericField("tag_id", Field.Store.YES, True).setIntValue(int(pdtag.id)))
+ doc.add(Field("tag_name", pdtag.name, Field.Store.NO, Field.Index.ANALYZED))
+ doc.add(Field("tag_name_pl", pdtag.name, Field.Store.NO, Field.Index.ANALYZED))
+ doc.add(Field("tag_category", 'pdcounter', Field.Store.NO, Field.Index.NOT_ANALYZED))
+ doc.add(Field("is_pdcounter", 'true', Field.Store.YES, Field.Index.NOT_ANALYZED))
+ self.index.addDocument(doc)
+
def create_book_doc(self, book):
"""
Create a lucene document referring book id.
stored = search.searcher.doc(scoreDocs.doc)
self.book_id = int(stored.get("book_id"))
- header_type = stored.get("header_type")
- if not header_type:
- return
-
- sec = (header_type, int(stored.get("header_index")))
- header_span = stored.get('header_span')
- header_span = header_span is not None and int(header_span) or 1
-
- fragment = stored.get("fragment_anchor")
-
pd = stored.get("published_date")
if pd is None:
pd = 0
self.published_date = int(pd)
- if snippets:
- snippets = snippets.replace("/\n", "\n")
- hit = (sec + (header_span,), fragment, scoreDocs.score, {'how_found': how_found, 'snippets': snippets and [snippets] or []})
+ header_type = stored.get("header_type")
+ # we have a content hit in some header of fragment
+ if header_type is not None:
+ sec = (header_type, int(stored.get("header_index")))
+ header_span = stored.get('header_span')
+ header_span = header_span is not None and int(header_span) or 1
+
+ fragment = stored.get("fragment_anchor")
- self._hits.append(hit)
+ if snippets:
+ snippets = snippets.replace("/\n", "\n")
+ hit = (sec + (header_span,), fragment, scoreDocs.score, {'how_found': how_found, 'snippets': snippets and [snippets] or []})
+
+ self._hits.append(hit)
self.search = search
self.searched = searched
def __cmp__(self, other):
c = cmp(self.score, other.score)
if c == 0:
+ if not hasattr(other,'published_date') or not hasattr(self, 'published_date'):
+ import pdb; pdb.set_trace()
# this is inverted, because earlier date is better
return cmp(other.published_date, self.published_date)
else:
if terms:
return JArray('object')(terms, Term)
- def search_tags(self, query, filter=None, max_results=40):
+ def search_tags(self, query, filters=None, max_results=40, pdcounter=False):
"""
Search for Tag objects using query.
"""
- tops = self.searcher.search(query, filter, max_results)
+ if not pdcounter:
+ filters = self.chain_filters([filter, self.term_filter(Term('is_pdcounter', 'true'), inverse=True)])
+ tops = self.searcher.search(query, filters, max_results)
tags = []
for found in tops.scoreDocs:
doc = self.searcher.doc(found.doc)
- tag = catalogue.models.Tag.objects.get(id=doc.get("tag_id"))
- tags.append(tag)
- print "%s (%d) -> %f" % (tag, tag.id, found.score)
-
+ is_pdcounter = doc.get('is_pdcounter')
+ if is_pdcounter:
+ tag = PDCounterAuthor.objects.get(id=doc.get('tag_id'))
+ else:
+ tag = catalogue.models.Tag.objects.get(id=doc.get("tag_id"))
+ # don't add the pdcounter tag if same tag already exists
+ if not (is_pdcounter and filter(lambda t: tag.slug == t.slug, tags)):
+ tags.append(tag)
+ # print "%s (%d) -> %f" % (tag, tag.id, found.score)
+ print 'returning %s' % tags
return tags
def search_books(self, query, filter=None, max_results=10):
bks.append(catalogue.models.Book.objects.get(id=doc.get("book_id")))
return bks
- def create_prefix_phrase(self, toks, field):
+ def make_prefix_phrase(self, toks, field):
q = MultiPhraseQuery()
for i in range(len(toks)):
t = Term(field, toks[i])
return only_term
- def hint_tags(self, string, max_results=50):
+ def hint_tags(self, string, max_results=50, pdcounter=True, prefix=True):
"""
Return auto-complete hints for tags
using prefix search.
top = BooleanQuery()
for field in ['tag_name', 'tag_name_pl']:
- q = self.create_prefix_phrase(toks, field)
+ if prefix:
+ q = self.make_prefix_phrase(toks, field)
+ else:
+ q = self.make_term_query(toks, field)
top.add(BooleanClause(q, BooleanClause.Occur.SHOULD))
no_book_cat = self.term_filter(Term("tag_category", "book"), inverse=True)
- return self.search_tags(top, no_book_cat, max_results=max_results)
+ return self.search_tags(top, no_book_cat, max_results=max_results, pdcounter=pdcounter)
- def hint_books(self, string, max_results=50):
+ def hint_books(self, string, max_results=50, prefix=True):
"""
Returns auto-complete hints for book titles
Because we do not index 'pseudo' title-tags.
"""
toks = self.get_tokens(string, field='SIMPLE')
- q = self.create_prefix_phrase(toks, 'title')
+ if prefix:
+ q = self.make_prefix_phrase(toks, 'title')
+ else:
+ q = self.make_term_query(toks, 'title')
return self.search_books(q, self.term_filter(Term("is_book", "true")), max_results=max_results)
from django.http import HttpResponse, HttpResponseRedirect, Http404, HttpResponsePermanentRedirect
from django.utils.translation import ugettext as _
-from catalogue.utils import get_random_hash
+from catalogue.utils import split_tags
from catalogue.models import Book, Tag, Fragment
from catalogue.fields import dumps
from catalogue.views import JSONResponse
authors = Tag.objects.filter(category='author', name__iregex=match_word_re(t))
if len(authors) > 0:
continue
-
+
if not dictionary.check(t):
try:
change[t] = dictionary.suggest(t)[0]
# jezeli tagi dot tylko ksiazki, to wazne zeby te nowe byly w tej samej ksiazce
# jesli zas dotycza themes, to wazne, zeby byly w tym samym fragmencie.
- tags = s.hint_tags(prefix)
+ tags = s.hint_tags(prefix, pdcounter=True)
books = s.hint_books(prefix)
# TODO DODAC TU HINTY
fuzzy = False
if 'q' in request.GET:
- tags = request.GET.get('tags', '')
+ # tags = request.GET.get('tags', '')
query = request.GET['q']
- book_id = request.GET.get('book', None)
- book = None
- if book_id is not None:
- book = get_object_or_404(Book, id=book_id)
+ # book_id = request.GET.get('book', None)
+ # book = None
+ # if book_id is not None:
+ # book = get_object_or_404(Book, id=book_id)
- hint = srch.hint()
- try:
- tag_list = Tag.get_tag_list(tags)
- except:
- tag_list = []
+ # hint = srch.hint()
+ # try:
+ # tag_list = Tag.get_tag_list(tags)
+ # except:
+ # tag_list = []
if len(query) < 2:
- return render_to_response('catalogue/search_too_short.html', {'tags': tag_list, 'prefix': query},
+ return render_to_response('catalogue/search_too_short.html', {'prefix': query},
context_instance=RequestContext(request))
- hint.tags(tag_list)
- if book:
- hint.books(book)
+ # hint.tags(tag_list)
+ # if book:
+ # hint.books(book)
+ tags = srch.hint_tags(query, pdcounter=True, prefix=False)
+ tags = split_tags(tags)
toks = StringReader(query)
tokens_cache = {}
author_results = SearchResult.aggregate(author_results)
title_results = SearchResult.aggregate(title_results)
-
+
everywhere = SearchResult.aggregate(everywhere, author_title_rest)
for res in [author_results, title_results, text_phrase, everywhere]:
for r in res:
for h in r.hits:
h['snippets'] = map(lambda s:
- re.subn(r"(^[ \t\n]+|[ \t\n]+$)", u"",
+ re.subn(r"(^[ \t\n]+|[ \t\n]+$)", u"",
re.subn(r"[ \t\n]*\n[ \t\n]*", u"\n", s)[0])[0], h['snippets'])
-
+
suggestion = did_you_mean(query, srch.get_tokens(toks, field="SIMPLE"))
print "dym? %s" % repr(suggestion).encode('utf-8')
-
+
results = author_results + title_results + text_phrase + everywhere
results.sort(reverse=True)
-
+
if len(results) == 1:
fragment_hits = filter(lambda h: 'fragment' in h, results[0].hits)
if len(fragment_hits) == 1:
elif len(results) == 0:
form = PublishingSuggestForm(initial={"books": query + ", "})
return render_to_response('catalogue/search_no_hits.html',
- {'tags': tag_list,
+ {'tags': tags,
'prefix': query,
"form": form,
'did_you_mean': suggestion},
context_instance=RequestContext(request))
+ print "TAGS: %s" % tags
return render_to_response('catalogue/search_multiple_hits.html',
- {'tags': tag_list,
+ {'tags': tags,
'prefix': query,
'results': { 'author': author_results,
'title': title_results,
padding-top:3em;
background: #fff;
}
+
+/* just on search page */
+.top-tag-list {
+ margin-top: 2.2em;
+ margin-bottom: 1.6em;
+}
\ No newline at end of file
return false;
});
+ $(function(){
+ $("#search").search();});
});
})(jQuery)
},
-
- });
-
- $(".search-result .see-more-snippets").click(function() {
- $(this).closest('.search-result').find('.snippets').removeClass('ui-helper-hidden');
- });
});
<span class="did_you_mean">{% trans "Did you mean" %} <a href="{% url search %}?q={{did_you_mean|urlencode}}">{{did_you_mean|lower}}</a>?</span>
{% endif %}
<!-- tu pójdą trafienia w tagi: Autorzy - z description oraz motywy i rodzaje (z book_count) -->
+ <div class="inline-tag-lists top-tag-list">
+ {% if tags.author %}
+ <div>
+ <div class="mono inline-header">{% trans "Authors" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list tags.author %}
+ </div>
+ </div>
+ {% endif %}
+ {% if tags.kind %}
+ <div>
+ <div class="mono inline-header">{% trans "Kinds" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list tags.kind %}
+ </div>
+ </div>
+ {% endif %}
+ {% if tags.genre %}
+ <div>
+ <div class="mono inline-header">{% trans "Genres" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list tags.genre %}
+ </div>
+ </div>
+ {% endif %}
+ {% if tags.epoch %}
+ <div class="inline-tag-list">
+ <div class="mono inline-header">{% trans "Epochs" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list tags.epoch %}
+ </div>
+ </div>
+ {% endif %}
+ </div>
+
+
{% if results.author %}
<div class="book-list-header">
<div class="book-box-inner">
</div>
<div>
<ol class="work-list">
- {% for author in results.author %}
- <li class="Book-item">
- {% book_short author.book %}
- </li>
- {% endfor %}
+ {% for author in results.author %}<li class="Book-item">{% book_short author.book %}</li>{% endfor %}
</ol>
</div>
{% endif %}
</div>
<div>
<ol class="work-list">
- {% for result in results.title %}
- <li class="Book-item">
+ {% for result in results.title %}<li class="Book-item">
{% book_short result.book %}
- </li>
- {% endfor %}
+ </li>{% endfor %}
</ol>
</div>
{% endif %}
{% block bodyid %}tagged-object-list{% endblock %}
{% block body %}
-
<div class="left-column">
-
<div class="page-desc">
+<div class="left-column">
+ <div class="page-desc">
<h1>{% html_title_from_tags tags %}</h1>
{% with tags|last as last_tag %}
{% if last_tag.has_description %}
-
<div id="description" class="normal-text">
-
<div id='description-long' style="display:none">{{ last_tag.description|safe }}</div>
-
<div id='description-short'>{{ last_tag.description|safe|truncatewords_html:40 }}</div>
- </div>
+ <div id="description" class="normal-text">
+ <div id='description-long' style="display:none">{{ last_tag.description|safe }}</div>
+ <div id='description-short'>{{ last_tag.description|safe|truncatewords_html:40 }}</div>
+ </div>
{% endif %}
-
+ <div class="clearboth"></div>
<div class="inline-tag-lists">
- {% if categories.author %}
- <div>
- <div class="mono inline-header">{% trans "Authors" %}:</div>
- <div class="inline-body">
- {% inline_tag_list categories.author tags %}
- </div>
+ {% if categories.author %}
+ <div>
+ <div class="mono inline-header">{% trans "Authors" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list categories.author tags %}
</div>
- {% endif %}
- {% if categories.kind %}
- <div>
- <div class="mono inline-header">{% trans "Kinds" %}:</div>
- <div class="inline-body">
- {% inline_tag_list categories.kind tags %}
- </div>
+ </div>
+ {% endif %}
+ {% if categories.kind %}
+ <div>
+ <div class="mono inline-header">{% trans "Kinds" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list categories.kind tags %}
</div>
- {% endif %}
- {% if categories.genre %}
- <div>
- <div class="mono inline-header">{% trans "Genres" %}:</div>
- <div class="inline-body">
- {% inline_tag_list categories.genre tags %}
- </div>
+ </div>
+ {% endif %}
+ {% if categories.genre %}
+ <div>
+ <div class="mono inline-header">{% trans "Genres" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list categories.genre tags %}
</div>
- {% endif %}
- {% if categories.epoch %}
- <div class="inline-tag-list">
- <div class="mono inline-header">{% trans "Epochs" %}:</div>
- <div class="inline-body">
- {% inline_tag_list categories.epoch tags %}
- </div>
+ </div>
+ {% endif %}
+ {% if categories.epoch %}
+ <div class="inline-tag-list">
+ <div class="mono inline-header">{% trans "Epochs" %}:</div>
+ <div class="inline-body">
+ {% inline_tag_list categories.epoch tags %}
</div>
- {% endif %}
+ </div>
+ {% endif %}
+
+ {% if categories.theme %}
+ <div class="hidden-box-wrapper">
+ <p><a href="#" class="hidden-box-trigger theme-list-link mono">
+ {% trans "Motifs and themes" %}</a></p>
+ <div class="hidden-box">
+ {% tag_list categories.theme tags %}
+ </div>
+ </div>
+ {% endif %}
</div>
- {% if categories.theme %}
- <div class="hidden-box-wrapper">
- <p><a href="#" class="hidden-box-trigger theme-list-link mono">
- {% trans "Motifs and themes" %}</a></p>
- <div class="hidden-box">
- {% tag_list categories.theme tags %}
- </div>
- </div>
- {% endif %}
+ <div class="clearboth"></div>
{% if theme_is_set %}
fnp / wolnelektury.git / commitdiff