#ILIKE VERTICA UPDATE#
Update 5: I actually use regexp_ilike(x,y), but regexp_like() seems to be more common.Dbadmin => insert into t values ( 1 ) commit dbadmin => insert into t values ( 2 ) commit dbadmin => delete from t where col1 = 1 commit dbadmin => insert into t values ( 3 ) commit dbadmin => insert into t values ( 4 ) commit dbadmin => insert into t values ( 5 ) commit dbadmin => insert into t values ( 6 ) commit dbadmin => select current_epoch, ahm_epoch, last_good_epoch from system current_epoch | ahm_epoch | last_good_epoch -+-+- 800 | 791 | 799 ( 1 row ) dbadmin => select node_name, projection_name, storage_type, start_epoch, end_epoch dbadmin => from delete_vectors dbadmin => where projection_name ilike '%_b0' dbadmin => order by start_epoch node_name | projection_name | storage_type | start_epoch | end_epoch -+-+-+-+- v_test_db_node0002 | t_b0 | DVROS | 795 | 795 ( 1 row ) dbadmin => select make_ahm_now () make_ahm_now - AHM set ( New AHM Epoch : 800 ) ( 1 row ) dbadmin => insert into t values ( 7 ) commit dbadmin => insert into t values ( 8 ) commit dbadmin => insert into t values ( 9 ) commit dbadmin => insert into t values ( 10 ) commit dbadmin => insert into t values ( 11 ) commit dbadmin => delete from t where col1 = 10 commit dbadmin => select current_epoch, ahm_epoch, last_good_epoch from system current_epoch | ahm_epoch | last_good_epoch -+-+- 807 | 800 | 806 ( 1 row ) dbadmin => select node_name, projection_name, storage_type, start_epoch, end_epoch dbadmin => from delete_vectors dbadmin => where projection_name ilike '%_b0' dbadmin => order by start_epoch node_name | projection_name | storage_type | start_epoch | end_epoch -+-+-+-+- v_test_db_node0002 | t_b0 | DVROS | 795 | 795 v_test_db_node0003 | t_b0 | DVROS | 806 | 806 ( 2 rows ) dbadmin => -do_tm_task('mergeout') 호출해서 puerg수행 dbadmin => select do_tm_task ( 'mergeout' ) do_tm_task - Task : mergeout ( Table : public. 'Mammogram-bilateral' = 'bilateral mammogram scan' Update 2: Levenshtein distance might help for some of these, but i would need to find the shortest distance between those two, I'm not aware of a way to do that though. So Regexp_like doesn't quite work, but i'm trying to find a regexp_substr() to work.
i would basically need to do 1,2 = (1,1||1,2||1,3) and so on forever, some of these have over 100 whitespaces though. So that matches the first 3 words (also it just occurred to do a whitespace count as a filter, in this case if it has 3 whitespaces). Newbie to regular expressions, I have spent hours searching for help but can't find anything, sorry if I missed something The problem I'm having is in getting rows like that to match on just a regexp_like(x,y,'i') type join.
I could clean myself ( for instance 'removal of pacemaker' i need to be different from 'insertion of pacemaker', but I understand fixing that might be a big task). I really need help getting the base of this going, the rest I could clean myself. One item might be 'Mammogram-bilateral' the other may be 'bilateral mammogram scan' I really need help with matching to get 9 out of 10 words matched, if that doesn't work then 8 out of 10 and so on, or the most like characters (or words) in both tables. So I'm working in the perl sql flavor with a regexp_like and need to combine two tables based on pattern matching.
0 kommentar(er)
