[ARVADOS] updated: cda964acdb8132d90b881e62db008c574fdd5cc4

[git at public.curoverse.com]

Summary of changes:
 services/api/lib/record_filters.rb | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

       via  cda964acdb8132d90b881e62db008c574fdd5cc4 (commit)
      from  2e5ac62b550f7dd608cf133ae66ef04f801be76b (commit)

Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.


commit cda964acdb8132d90b881e62db008c574fdd5cc4
Author: Radhika Chippada <radhika at curoverse.com>
Date:   Sun Feb 8 16:22:07 2015 -0500

    4523: update full text search to append ':*' for each word to support a collection file search such as dir_name_part space file_name_part.

diff --git a/services/api/lib/record_filters.rb b/services/api/lib/record_filters.rb
index 51153fe..0b65108 100644
--- a/services/api/lib/record_filters.rb
+++ b/services/api/lib/record_filters.rb
@@ -43,11 +43,9 @@ module RecordFilters
         end
         attrs = [] #  skip the generic per-column operator loop below
         # Use to_tsquery since plainto_tsquery does not support prefix search.
-        # Instead split operand and join the words with ' & ' and add ':*' to the last word
-        # Thus when searched for "some str", objects containing "some" and "str:*" are found.
+        # Instead split operand, add ':*' to each word and join the words with ' & '
         cond_out << model_class.full_text_tsvector+" @@ to_tsquery(?)"
-        operand << ':*'
-        param_out << operand.split.join(' & ')
+        param_out << operand.split.each {|s| s.concat(':*')}.join(' & ')
       else
        attrs.each do |attr|
         if !model_class.searchable_columns(operator).index attr.to_s

-----------------------------------------------------------------------


hooks/post-receive
--