minor fixes in search

diff --git a/doc/schema.xml b/doc/schema.xml
index 33886be..17868e1 100644 (file)
--- a/doc/schema.xml
+++ b/doc/schema.xml
@@ -135,10 +135,12 @@
    <field name="authors" type="text_general" stored="false" indexed="true" multiValued="true" termPositions="true" termVectors="true"/>
    <field name="translators" type="text_general" stored="false" indexed="true" multiValued="true" termPositions="true" termVectors="true" />
    <field name="title" type="text_pl" stored="false" indexed="true"/>
-   <field name="title_orig" type="lowercase" stored="true" indexed="true"/>
+   <field name="title_orig" type="lowercase" stored="false" indexed="true"/>
 <!--   <field name="published_date" type="tdate" stored="false" indexed="true"/>-->
    <field name="published_date" type="string" stored="true" indexed="true"/>
 
+   <field name="metadata" type="lowercase" stored="false" indexed="true"/>
+
    <field name="themes" type="lowercase" stored="true" indexed="true" termVectors="true" termPositions="true" multiValued="true" />
    <field name="themes_pl" type="text_pl" stored="true" indexed="true" termVectors="true" termPositions="true" multiValued="true" />
    <field name="header_index" type="int" stored="true" indexed="true"/>

diff --git a/src/catalogue/templates/catalogue/book_searched.html b/src/catalogue/templates/catalogue/book_searched.html
index 26b9f61..23d4736 100644 (file)
--- a/src/catalogue/templates/catalogue/book_searched.html
+++ b/src/catalogue/templates/catalogue/book_searched.html
@@ -21,7 +21,7 @@
               {% if hit.snippet %}
                 {{hit.snippet|safe}}
               {% else %}
-                {{hit.fragment.text|truncatewords_html:15|safe}}
+                {{hit.fragment.text|truncatewords_html:15|safe}} {# bad, now impossible #}
               {% endif %}
             </a>
           </div>

diff --git a/src/search/custom.py b/src/search/custom.py
index dfface9..da21e01 100644 (file)
--- a/src/search/custom.py
+++ b/src/search/custom.py
@@ -184,5 +184,6 @@ class CustomSolrInterface(sunburnt.SolrInterface):
             off = -start
             snip = snip[:e + off] + mark[1] + snip[e + off:]
             snip = snip[:s + off] + mark[0] + snip[s + off:]
+        snip = re.sub('%s[ \t\n]+%s' % (mark[1], mark[0]), " ", snip)
 
         return snip

diff --git a/src/search/index.py b/src/search/index.py
index 1cac82e..f9fb4b2 100644 (file)
--- a/src/search/index.py
+++ b/src/search/index.py
@@ -272,7 +272,7 @@ class Index(SolrIndex):
         ]
 
     ignore_content_tags = [
-        'uwaga', 'extra',
+        'uwaga', 'extra', 'nota_red',
         'zastepnik_tekstu', 'sekcja_asterysk', 'separator_linia', 'zastepnik_wersu',
         'didaskalia',
         'naglowek_aktu', 'naglowek_sceny', 'naglowek_czesc',
@@ -365,8 +365,8 @@ class Index(SolrIndex):
         if master is None:
             return []
 
-        def walker(node, ignore_tags=()):
-            if node.tag not in ignore_tags:
+        def walker(node):
+            if node.tag not in self.ignore_content_tags:
                 yield node, None, None
                 if node.text is not None:
                     yield None, node.text, None
@@ -421,12 +421,6 @@ class Index(SolrIndex):
                 book.id, doc['header_index'], doc['header_span'], doc.get('fragment_anchor', ''))
             return doc
 
-        def give_me_utf8(s):
-            if isinstance(s, unicode):
-                return s.encode('utf-8')
-            else:
-                return s
-
         fragments = {}
         snippets = Snippets(book.id).open('w')
         try:
@@ -447,7 +441,7 @@ class Index(SolrIndex):
                     content.append(text)
                 handle_text = [all_content]
 
-                for start, text, end in walker(header, ignore_tags=self.ignore_content_tags):
+                for start, text, end in walker(header):
                     # handle footnotes
                     if start is not None and start.tag in self.footnote_tags:
                         footnote = []
@@ -819,9 +813,10 @@ class Search(SolrIndex):
                 text = snippets.get((int(position),
                                      int(length)))
                 snip = self.index.highlight(text=text, field=field, q=query)
-                snips[idx] = snip
-                if snip:
-                    num -= 1
+                if snip not in snips:
+                    snips[idx] = snip
+                    if snip:
+                        num -= 1
                 idx += 1
 
         except IOError, e:

diff --git a/src/search/templatetags/search_tags.py b/src/search/templatetags/search_tags.py
index c135b80..517f18f 100644 (file)
--- a/src/search/templatetags/search_tags.py
+++ b/src/search/templatetags/search_tags.py
@@ -32,7 +32,6 @@ def book_searched(context, result):
     # We don't need hits which lead to sections but do not have
     # snippets.
     hits = filter(lambda (idx, h):
-                  'fragment' in h or
                   result.snippets[idx] is not None,
                   enumerate(result.hits))
     # print "[tmpl: from %d hits selected %d]" % (len(result.hits), len(hits))
@@ -45,8 +44,8 @@ def book_searched(context, result):
             continue
         snip = result.snippets[idx]
         # fix some formattting
-        snip = re.subn(r"(^[ \t\n]+|[ \t\n]+$)", u"",
-                       re.subn(r"[ \t\n]*\n[ \t\n]*", u"\n", snip)[0])[0]
+        snip = re.sub(r"[ \t\n]*\n[ \t\n]*", u"\n", snip)
+        snip = re.sub(r"(^[ \t\n]+|[ \t\n]+$)", u"", snip)
 
         snip = snip.replace("\n", "<br />").replace('---', '&mdash;')
         hit['snippet'] = snip
@@ -54,5 +53,5 @@ def book_searched(context, result):
     return {
         'request': context['request'],
         'book': book,
-        'hits':  hits and zip(*hits)[1] or []
+        'hits':  zip(*hits)[1] if hits else []
     }

diff --git a/src/search/views.py b/src/search/views.py
index d105be6..14da6f6 100644 (file)
--- a/src/search/views.py
+++ b/src/search/views.py
@@ -222,7 +222,6 @@ def main(request):
     text_phrase = filter(ensure_exists, text_phrase)
     everywhere = filter(ensure_exists, everywhere)
 
-    results = author_results + translator_results + title_results + text_phrase + everywhere
     # ensure books do exists & sort them
     for res in (author_results, translator_results, title_results, text_phrase, everywhere):
         res.sort(reverse=True)
@@ -235,7 +234,7 @@ def main(request):
     #         #frag = Fragment.objects.get(anchor=anchor)
     #         return HttpResponseRedirect(fragment_hits[0]['fragment'].get_absolute_url())
     #     return HttpResponseRedirect(results[0].book.get_absolute_url())
-    if len(results) == 0:
+    if not (author_results or translator_results or title_results or text_phrase or everywhere):
         form = PublishingSuggestForm(initial={"books": query + ", "})
         return render_to_response(
             'catalogue/search_no_hits.html',