update tests
[wolnelektury.git] / apps / search / views.py
index 623b311..2945c42 100644 (file)
@@ -37,7 +37,9 @@ def did_you_mean(query, tokens):
 
         if not dictionary.check(t):
             try:
 
         if not dictionary.check(t):
             try:
-                change[t] = dictionary.suggest(t)[0]
+                change_to = dictionary.suggest(t)[0].lower()
+                if change_to != t.lower():
+                    change[t] = change_to
             except IndexError:
                 pass
 
             except IndexError:
                 pass
 
@@ -49,15 +51,16 @@ def did_you_mean(query, tokens):
 
     return query
 
 
     return query
 
+JVM.attachCurrentThread()
+search = Search()
 
 def hint(request):
     prefix = request.GET.get('term', '')
     if len(prefix) < 2:
         return JSONResponse([])
     JVM.attachCurrentThread()
 
 def hint(request):
     prefix = request.GET.get('term', '')
     if len(prefix) < 2:
         return JSONResponse([])
     JVM.attachCurrentThread()
-    s = Search()
 
 
-    hint = s.hint()
+    hint = search.hint()
     try:
         tags = request.GET.get('tags', '')
         hint.tags(Tag.get_tag_list(tags))
     try:
         tags = request.GET.get('tags', '')
         hint.tags(Tag.get_tag_list(tags))
@@ -69,14 +72,17 @@ def hint(request):
     # 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.
 
     # 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, pdcounter=True)
-    books = s.hint_books(prefix)
+    tags = search.hint_tags(prefix, pdcounter=True)
+    books = search.hint_books(prefix)
 
 
-    # TODO DODAC TU HINTY
+    def category_name(c):
+        if c.startswith('pd_'):
+            c=c[len('pd_'):]
+        return _(c)
 
     return JSONResponse(
         [{'label': t.name,
 
     return JSONResponse(
         [{'label': t.name,
-          'category': _(t.category),
+          'category': category_name(t.category),
           'id': t.id,
           'url': t.get_absolute_url()}
           for t in tags] + \
           'id': t.id,
           'url': t.get_absolute_url()}
           for t in tags] + \
@@ -90,11 +96,10 @@ def hint(request):
 def main(request):
     results = {}
     JVM.attachCurrentThread()  # where to put this?
 def main(request):
     results = {}
     JVM.attachCurrentThread()  # where to put this?
-    srch = Search()
 
     results = None
     query = None
 
     results = None
     query = None
-    fuzzy = False
+    fuzzy = False #0.8
 
     if 'q' in request.GET:
         # tags = request.GET.get('tags', '')
 
     if 'q' in request.GET:
         # tags = request.GET.get('tags', '')
@@ -104,7 +109,7 @@ def main(request):
         # if book_id is not None:
         #     book = get_object_or_404(Book, id=book_id)
 
         # if book_id is not None:
         #     book = get_object_or_404(Book, id=book_id)
 
-        # hint = srch.hint()
+        # hint = search.hint()
         # try:
         #     tag_list = Tag.get_tag_list(tags)
         # except:
         # try:
         #     tag_list = Tag.get_tag_list(tags)
         # except:
@@ -117,21 +122,18 @@ def main(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 = search.hint_tags(query, pdcounter=True, prefix=False, fuzzy=fuzzy)
         tags = split_tags(tags)
 
         toks = StringReader(query)
         tokens_cache = {}
         tags = split_tags(tags)
 
         toks = StringReader(query)
         tokens_cache = {}
-        fuzzy = 'fuzzy' in request.GET
-        if fuzzy:
-            fuzzy = 0.7
 
 
-        author_results = srch.search_phrase(toks, 'authors', fuzzy=fuzzy, tokens_cache=tokens_cache)
-        title_results = srch.search_phrase(toks, 'title', fuzzy=fuzzy, tokens_cache=tokens_cache)
+        author_results = search.search_phrase(toks, 'authors', fuzzy=fuzzy, tokens_cache=tokens_cache)
+        title_results = search.search_phrase(toks, 'title', fuzzy=fuzzy, tokens_cache=tokens_cache)
 
         # Boost main author/title results with mixed search, and save some of its results for end of list.
         # boost author, title results
 
         # 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 = srch.search_some(toks, ['authors', 'title', 'tags'], fuzzy=fuzzy, tokens_cache=tokens_cache)
+        author_title_mixed = search.search_some(toks, ['authors', 'title', 'tags'], fuzzy=fuzzy, tokens_cache=tokens_cache)
         author_title_rest = []
         for b in author_title_mixed:
             bks = filter(lambda ba: ba.book_id == b.book_id, author_results + title_results)
         author_title_rest = []
         for b in author_title_mixed:
             bks = filter(lambda ba: ba.book_id == b.book_id, author_results + title_results)
@@ -143,10 +145,10 @@ def main(request):
         # Do a phrase search but a term search as well - this can give us better snippets then search_everywhere,
         # Because the query is using only one field.
         text_phrase = SearchResult.aggregate(
         # Do a phrase search but a term search as well - this can give us better snippets then search_everywhere,
         # Because the query is using only one field.
         text_phrase = SearchResult.aggregate(
-            srch.search_phrase(toks, 'content', fuzzy=fuzzy, tokens_cache=tokens_cache, snippets=True, book=False, slop=4),
-            srch.search_some(toks, ['content'], tokens_cache=tokens_cache, snippets=True, book=False))
+            search.search_phrase(toks, 'content', fuzzy=fuzzy, tokens_cache=tokens_cache, snippets=True, book=False, slop=4),
+            search.search_some(toks, ['content'], tokens_cache=tokens_cache, snippets=True, book=False))
 
 
-        everywhere = srch.search_everywhere(toks, fuzzy=fuzzy, tokens_cache=tokens_cache)
+        everywhere = search.search_everywhere(toks, fuzzy=fuzzy, tokens_cache=tokens_cache)
 
         def already_found(results):
             def f(e):
 
         def already_found(results):
             def f(e):
@@ -173,7 +175,7 @@ def main(request):
                                         re.subn(r"(^[ \t\n]+|[ \t\n]+$)", u"",
                                                 re.subn(r"[ \t\n]*\n[ \t\n]*", u"\n", s)[0])[0], h['snippets'])
 
                                         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"))
+        suggestion = did_you_mean(query, search.get_tokens(toks, field="SIMPLE"))
         print "dym? %s" % repr(suggestion).encode('utf-8')
 
         results = author_results + title_results + text_phrase + everywhere
         print "dym? %s" % repr(suggestion).encode('utf-8')
 
         results = author_results + title_results + text_phrase + everywhere
@@ -182,9 +184,9 @@ def main(request):
         if len(results) == 1:
             fragment_hits = filter(lambda h: 'fragment' in h, results[0].hits)
             if len(fragment_hits) == 1:
         if len(results) == 1:
             fragment_hits = filter(lambda h: 'fragment' in h, results[0].hits)
             if len(fragment_hits) == 1:
-                anchor = fragment_hits[0]['fragment']
-                frag = Fragment.objects.get(anchor=anchor)
-                return HttpResponseRedirect(frag.get_absolute_url())
+                #anchor = fragment_hits[0]['fragment']
+                #frag = Fragment.objects.get(anchor=anchor)
+                return HttpResponseRedirect(fragment_hits[0]['fragment'].get_absolute_url())
             return HttpResponseRedirect(results[0].book.get_absolute_url())
         elif len(results) == 0:
             form = PublishingSuggestForm(initial={"books": query + ", "})
             return HttpResponseRedirect(results[0].book.get_absolute_url())
         elif len(results) == 0:
             form = PublishingSuggestForm(initial={"books": query + ", "})