以下两个查询组件的性能如何比较？
LOWER LIKE
... LOWER(description) LIKE '%abcde%' ...
ILIKE
... description iLIKE '%abcde%' ...
答案 0 :(得分：25)
根据我的测试（每个查询 10 ），LOWER LIKE比17%快iLIKE。
<强>解释
我创建了一百万行包含一些随机混合文本数据：
require 'securerandom'
inserts = []
1000000.times do |i|
inserts << "(1, 'fake', '#{SecureRandom.urlsafe_base64(64)}')"
end
sql = "insert into books (user_id, title, description) values #{inserts.join(', ')}"
ActiveRecord::Base.connection.execute(sql)
验证行数：
my_test_db=# select count(id) from books ;
count
---------
1000009
（是的，我从其他测试中增加了9行 - 不是问题。）
示例查询和结果：
my_test_db=# SELECT "books".* FROM "books" WHERE "books"."published" = 'f'
my_test_db=# and (LOWER(description) LIKE '%abcde%') ;
id | user_id | title | description | published
---------+---------+-------+----------------------------------------------------------------------------------------+------
1232322 | 1 | fake | 5WRGr7oCKABcdehqPKsUqV8ji61rsNGS1TX6pW5LJKrspOI_ttLNbaSyRz1BwTGQxp3OaxW7Xl6fzVpCu9y3fA | f
1487103 | 1 | fake | J6q0VkZ8-UlxIMZ_MFU_wsz_8MP3ZBQvkUo8-2INiDIp7yCZYoXqRyp1Lg7JyOwfsIVdpPIKNt1uLeaBCdelPQ | f
1817819 | 1 | fake | YubxlSkJOvmQo1hkk5pA1q2mMK6T7cOdcU3ADUKZO8s3otEAbCdEcmm72IOxiBdaXSrw20Nq2Lb383lq230wYg | f
LOWER LIKE的结果
my_test_db=# EXPLAIN ANALYZE SELECT "books".* FROM "books" WHERE "books"."published" = 'f' and (LOWER(description) LIKE '%abcde%') ;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------
Seq Scan on books (cost=0.00..32420.14 rows=1600 width=117) (actual time=938.627..4114.038 rows=3 loops=1)
Filter: ((NOT published) AND (lower(description) ~~ '%abcde%'::text))
Rows Removed by Filter: 1000006
Total runtime: 4114.098 ms
iLIKE的结果
my_test_db=# EXPLAIN ANALYZE SELECT "books".* FROM "books" WHERE "books"."published" = 'f' and (description iLIKE '%abcde%') ;
QUERY PLAN
----------------------------------------------------------------------------------------------------------------
Seq Scan on books (cost=0.00..29920.11 rows=100 width=117) (actual time=1147.612..4986.771 rows=3 loops=1)
Filter: ((NOT published) AND (description ~~* '%abcde%'::text))
Rows Removed by Filter: 1000006
Total runtime: 4986.831 ms
数据库信息披露
Postgres版本：
my_test_db=# select version();
version
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------
PostgreSQL 9.2.4 on x86_64-apple-darwin12.4.0, compiled by i686-apple-darwin11-llvm-gcc-4.2 (GCC) 4.2.1 (Based on Apple Inc. build 5658) (LLVM build 2336.11.00), 64-bit
整理设置：
my_test_db=# select datcollate from pg_database where datname = 'my_test_db';
datcollate
-------------
en_CA.UTF-8
表格定义：
my_test_db=# \d books
Table "public.books"
Column | Type | Modifiers
-------------+-----------------------------+-------------------------------------------------------
id | integer | not null default nextval('books_id_seq'::regclass)
user_id | integer | not null
title | character varying(255) | not null
description | text | not null default ''::text
published | boolean | not null default false
Indexes:
"books_pkey" PRIMARY KEY, btree (id)
答案 1 :(得分：18)
答案取决于许多因素，如Postgres版本，编码和区域设置 - 尤其是LC_COLLATE。
裸表达式lower(description) LIKE '%abc%'通常比description ILIKE '%abc%'快一点，并且要比同等正则表达式description ~* 'abc'快一点。这对于顺序扫描至关重要，其中必须针对每个测试行评估表达式。
但是对于你在答案中展示的大表，肯定会使用索引。对于任意模式（不仅是左锚定）我建议使用附加模块pg_trgm的trigram索引。然后我们讨论毫秒而不是秒，并且上述表达式之间的差异无效。
GIN和GiST索引（使用gin_trgm_ops或gist_trgm_ops运算符类）支持LIKE（~~），ILIKE（~~*） ，~，~*（以及更多变体）都一样。如果description上的三元组GIN索引（通常大于GiST，但读取速度更快），则查询将使用description ILIKE 'case_insensitive_pattern'。
相关：
Postgres中模式匹配的基础知识：
使用上述trigram索引时，通常更实用：
description ILIKE '%abc%'
或者使用不区分大小写的regexp运算符（没有%通配符）：
description ~* 'abc'
(description)上的索引不支持lower(description)上的查询，例如：
lower(description) LIKE '%abc%'
反之亦然。
对于lower(description) 上的谓词，表达式索引是稍微好一点的选择。
在所有其他情况下，(description)上的索引更可取，因为它支持 区分大小写和不敏感的谓词。
答案 2 :(得分：0)
在我的Rails项目中。 ILIKE比LOWER LIKE快10倍，我在GIN列上添加了一个entities.name索引
> Entity.where("LOWER(name) LIKE ?", name.strip.downcase).limit(1).first
Entity Load (2443.9ms) SELECT "entities".* FROM "entities" WHERE (lower(name) like 'baidu') ORDER BY "entities"."id" ASC LIMIT $1 [["LIMIT", 1]]
> Entity.where("name ILIKE ?", name.strip).limit(1).first
Entity Load (285.0ms) SELECT "entities".* FROM "entities" WHERE (name ilike 'Baidu') ORDER BY "entities"."id" ASC LIMIT $1 [["LIMIT", 1]]
# explain analyze SELECT "entities".* FROM "entities" WHERE (name ilike 'Baidu') ORDER BY "entities"."id" ASC LIMIT 1;
QUERY PLAN
------------------------------------------------------------------------------------------------------------------------------------------------
Limit (cost=3186.03..3186.04 rows=1 width=1588) (actual time=7.812..7.812 rows=1 loops=1)
-> Sort (cost=3186.03..3187.07 rows=414 width=1588) (actual time=7.811..7.811 rows=1 loops=1)
Sort Key: id
Sort Method: quicksort Memory: 26kB
-> Bitmap Heap Scan on entities (cost=1543.21..3183.96 rows=414 width=1588) (actual time=7.797..7.805 rows=1 loops=1)
Recheck Cond: ((name)::text ~~* 'Baidu'::text)
Rows Removed by Index Recheck: 6
Heap Blocks: exact=7
-> Bitmap Index Scan on index_entities_on_name (cost=0.00..1543.11 rows=414 width=0) (actual time=7.787..7.787 rows=7 loops=1)
Index Cond: ((name)::text ~~* 'Baidu'::text)
Planning Time: 6.375 ms
Execution Time: 7.874 ms
(12 rows)
GIN索引确实有助于提高ILIKE的性能