在几个项目中使用Liquibase之后，上周我开始在Informix中使用它。对于执行的每个命令(更新或回滚)，我必须等待大约1分钟才能得到结果。在wireshark的帮助下，我发现Liquibase执行了以下SQL语句，这些语句需要花费很多时间。有可能避免这个语句或者使它更快吗?
SELECT trim(t.owner)::VARCHAR(128)， t.tabname::VARCHAR(128)， c.colname::VARCHAR(128)， informixx .get_data_type(128)。c.coltype,c.extended_id, 0)::SMALLINT, informixx .schema_coltypename(c.coltype,c.extended_id)::VARCHAR(128)， informixx .schema_precision(c。coltype, c.extended_id, c.collength)::INTEGER, 0::INTEGER,informix.schema_numscale(c. coltype, c.extended_id, c.collength)::INTEGER，coltype, c.c olength)::INTEGER, informixx .schema_numprecradix(c.coltype)::INTEGER, informixx .schema_isnullable(c.coltype)::INTEGER, NULL::VARCHAR(254)，CASE d.type WHEN 'L' THEN informixx .get_default_value(c.c olength)::INTEGER，coltype, c.d extended_id, c.c olength, d.k default:: VARCHAR(256))::VARCHAR(254) WHEN 'C' THEN 'current'::VARCHAR(254) WHEN 'S' THEN 'dbservername'::VARCHAR(254) WHEN 'U' THEN 'user'::VARCHAR(254) WHEN 'T' THEN 'today'::VARCHAR(254) ELSE NULL::VARCHAR(254) END，coltype::整数,informix.schema_datetype (c。coltype, c.collength)::INTEGER, informix.schema_charlen(c。c. coltype, c.extended_id, c.collength)::INTEGER，colno::INTEGER, informixx .schema_nullable(c.coltype)::VARCHAR(3)， NULL::VARCHAR(128)， NULL::VARCHAR(128)， NULL::VARCHAR(128)， NULL::smallint, informixx .schema_isautoincr(c.coltype)::VARCHAR(3)，c。extended_id::INTEGER FROM informix。systables 1、OUTER informix。Sysdefaults、informix。syscolumns c WHERE t.tabid = c.tabid and d.tabid = t.tabid and c.colno = d.colno and (t.tabtype not in ('P'， 'S')) and (d.class in ('T')) and t.tabname like '%' and t.tabname like 'informix' and c.colname like '%' union SELECT trim(st.owner)::VARCHAR(128)， st.tabname::VARCHAR(128)， c.colname::VARCHAR(128)， informinformix .get_data_type(c.coltype,c. colid))extended_id, 0)::SMALLINT, informixx .schema_coltypename(c.coltype, c.c extended_id)::VARCHAR(128)， informixx .schema_precision(128)。coltype, c.extended_id, c.collength)::INTEGER, 0::INTEGER, informix.schema_numscale(c. coltype, c.extended_id, c.collength)::INTEGER，coltype, c.c olength)::INTEGER, informixx .schema_numprecradix(c.coltype)::INTEGER, informixx .schema_isnullable(c.coltype)::INTEGER, NULL::VARCHAR(254)， CASE d.type WHEN 'L' THEN informixx .get_default_value(c.c olength)::INTEGER，C .coltype, C .extended_id, C .collength, d.default:: VARCHAR(256))::VARCHAR(254) WHEN 'C' THEN 'current'::VARCHAR(254) WHEN 'S' THEN 'dbservername'::VARCHAR(254) WHEN 'U' THEN 'user'::VARCHAR(254) WHEN 'T' THEN 'today'::VARCHAR(254) ELSE NULL::VARCHAR(254) END, C .coltype::INTEGER, informix.schema_datetype(256)coltype, c.collength)::INTEGER, informix.schema_charlen(c。c.coltype, c.extended_id, c.collength)::INTEGER, c.colno::INTEGER, informixx .schema_nullable(c.coltype)::VARCHAR(3)， NULL::VARCHAR(128)， NULL::VARCHAR(128)， NULL::VARCHAR(128)， NULL::smallint, informixx .schema_isautoincr(c.coltype)::VARCHAR(3)， c.extended_id::INTEGERSystables st, informix。Syssyntable、informix。systables 1、OUTER informix。Sysdefaults、informix。syscolumns c WHERE st.tabid = s.tabid AND st.tabid = t.tabid AND d.tabid = t.tabid AND c.tabid = t.tabid AND c.colno = d.colno AND (st.tabtype in ('P'， 'S')) AND (d.class in ('T')) AND st.tabname LIKE '%' AND st.tabname LIKE 'informix' AND c.colname LIKE '%'顺序为1,2,16
在我的实例(id 12.10uc5)上运行它:
dbaccess stores lq.sql
real 0m0.701s
user 0m0.036s
sys 0m0.032s