diff options
| author | Dylan William Hardison <dylan@hardison.net> | 2017-03-21 04:08:26 +0100 |
|---|---|---|
| committer | Dylan William Hardison <dylan@hardison.net> | 2017-03-21 04:08:26 +0100 |
| commit | b921e3142e37977298571a4229faca4e1794017d (patch) | |
| tree | c965352b0db57f0d73f12ba2fd818845566dbe8b /Bugzilla/Elastic/Indexer.pm | |
| parent | e53de8c34e12e18ecea98ad97775893f083beff7 (diff) | |
| download | bugzilla-b921e3142e37977298571a4229faca4e1794017d.tar.gz bugzilla-b921e3142e37977298571a4229faca4e1794017d.tar.xz | |
Bug 1307485 - Add code to run a subset of buglist.cgi search queries against the ES backend
Skipping this for this week's push.
Diffstat (limited to 'Bugzilla/Elastic/Indexer.pm')
| -rw-r--r-- | Bugzilla/Elastic/Indexer.pm | 29 |
1 files changed, 17 insertions, 12 deletions
diff --git a/Bugzilla/Elastic/Indexer.pm b/Bugzilla/Elastic/Indexer.pm index dd71a7198..82f946af9 100644 --- a/Bugzilla/Elastic/Indexer.pm +++ b/Bugzilla/Elastic/Indexer.pm @@ -23,7 +23,7 @@ has 'mtime' => ( has 'shadow_dbh' => ( is => 'lazy' ); has 'debug_sql' => ( - is => 'ro', + is => 'ro', default => 0, ); @@ -40,24 +40,24 @@ sub create_index { index => $self->index_name, body => { settings => { - number_of_shards => 2, + number_of_shards => 1, analysis => { - filter => { - asciifolding_original => { - type => "asciifolding", - preserve_original => \1, - }, - }, analyzer => { folding => { + type => 'standard', tokenizer => 'standard', - filter => ['standard', 'lowercase', 'asciifolding_original'], + filter => [ 'lowercase', 'asciifolding' ] }, bz_text_analyzer => { type => 'standard', filter => ['lowercase', 'stop'], max_token_length => '20' }, + bz_substring_analyzer => { + type => 'custom', + filter => ['lowercase'], + tokenizer => 'bz_ngram_tokenizer', + }, bz_equals_analyzer => { type => 'custom', filter => ['lowercase'], @@ -71,20 +71,25 @@ sub create_index { whiteboard_shingle_words => { type => 'custom', tokenizer => 'whiteboard_words_pattern', - filter => ['stop', 'shingle', 'lowercase'] + filter => ['stop', 'shingle'] }, whiteboard_tokens => { type => 'custom', tokenizer => 'whiteboard_tokens_pattern', - filter => ['stop', 'lowercase'] + filter => ['stop'] }, whiteboard_shingle_tokens => { type => 'custom', tokenizer => 'whiteboard_tokens_pattern', - filter => ['stop', 'shingle', 'lowercase'] + filter => ['stop', 'shingle'] } }, tokenizer => { + bz_ngram_tokenizer => { + type => 'nGram', + min_ngram => 2, + max_ngram => 25, + }, whiteboard_tokens_pattern => { type => 'pattern', pattern => '\\s*([,;]*\\[|\\][\\s\\[]*|[;,])\\s*' |