我有一个国家及其缩写表，其中有一列名为code缩写，另一列名为name包含国家名称。
我需要遍历一个外国地址并查找每个单词，直到在表中找到匹配项（即国家/地区）并检索案例陈述的缩写。
这不会在所有记录上完成，只有某些大于我正在构建的文件中 60 个字符的字段的记录。
所以我需要做的是影响：
SELECT
CASE WHEN address2 & foreign_address > 60
THEN split and iterate through '12345 MY SUPER LONG ADDRESS IN THE PHILIPPINES' and look up
each string until PHILIPPINES is matched in the country_codes table and 'PH' is returned
END
这是我能想到的处理这种情况的最好方法，而不是截断由于明显原因我不想做的地址。这也需要根据不同的地址和国家而动态变化。
在这一点上，我最大的挑战是分解字符串并查看每个字符串片段。
SELECT Addresses.foreign_address, Countries.Code
FROM Addresses, Countries
WHERE LEN(foreign_address) > 60
AND foreign_address LIKE '%' Countries.Name '%'
[代码] 将包含匹配国家的缩写代码。
这是相关的 SQLFiddle（以及下面的完整代码）：
CREATE TABLE Countries (Name varchar(128), Code varchar(2));
CREATE TABLE Addresses (foreign_address varchar(512));
INSERT INTO Countries(Name,Code) VALUES('PHILIPPINES', 'PH');
INSERT INTO Addresses(foreign_address)
VALUES('12345 MY SUPER LONG ADDRESS IN THE PHILIPPINES UNTIL PHILIPPINES IS MATCHED AND PH IS RETURNED');
SELECT Addresses.foreign_address, Countries.Code
FROM Addresses, Countries
WHERE LEN(foreign_address) > 60
AND foreign_address LIKE '%' + Countries.Name + '%'
像这样的东西，可能是？
select l.abbr, a.id
from
lookup_table l, address_table a
where
charindex(l.country_name, a.address2 + a.foreign_address) > 0
and len( a.address2 + a.foreign_address) > 60
未测试。
针对 SQL Server 2005 字符串连接运算符进行了编辑。
You can use LIKE instead of breaking up by word:
declare @address2 varchar(128)
declare @foreign_address varchar(128)
set @address2 = '12345 MY SUPER LONG ADDRESS IN THE PHILIPPINES'
set @foreign_address = '123456789012345678901234567890123456789012345678901234567890'
SELECT
CASE
WHEN len(@address2 + @foreign_address) > 60
AND (' ' + @address2 + ' ' + @foreign_address + ' ') like '% PHILIPPINES %'
THEN 'PH'
ELSE NULL
END
我不提倡在 SQL Server 中使用这种编程风格，但这就是我将如何做你要求做的事情。首先，将字符串标记为表变量中的单个单词。然后在表变量上打开一个光标并循环遍历单词，break如果我们从 dbo.countries 中找到结果则调用。请注意，循环在 SQL Server 中非常低效。UDF 表函数来自这里：How to split a string in T-SQL?
-- Create the UDF
CREATE FUNCTION Splitfn(@String varchar(8000), @Delimiter char(1))
returns @temptable TABLE (items varchar(8000))
as
begin
declare @idx int
declare @slice varchar(8000)
select @idx = 1
if len(@String)<1 or @String is null return
while @idx!= 0
begin
set @idx = charindex(@Delimiter,@String)
if @idx!=0
set @slice = left(@String,@idx - 1)
else
set @slice = @String
if(len(@slice)>0)
insert into @temptable(Items) values(@slice)
set @String = right(@String,len(@String) - @idx)
if len(@String) = 0 break
end
return
end
go
-- Create the dbo.countries table so we can test our code later
create table dbo.countries (code char(2), name varchar(100))
go
-- Insert one record in dbo.countries so we can test our code later
insert into dbo.countries (code, name)
select 'PH', 'PHILIPPINES'
go
-- for one @String input, this is what I would do
declare @String varchar(1000) = '12345 MY SUPER LONG ADDRESS IN THE PHILIPPINES'
declare @CountryCode char(2) = ''
declare @done bit = 0x0
declare @word varchar(1000)
declare @words table (word varchar(250) primary key)
-- Break apart your @String into a table of records, only returning the DISTINCT values.
-- Join on the domain list so we can only process the ones that will return data in the CURSOR (eliminating excess looping)
insert into @words (word)
--
select distinct items as word
from dbo.Splitfn(@String, ' ') s
join dbo.countries c
on lower(c.name) = lower(s.items)
declare word_cursor CURSOR for
select word
from @words w
open word_cursor
fetch next from word_cursor into @word
while @@FETCH_STATUS = 0
begin
select @CountryCode = code
from dbo.countries
where name = @word
if @@trancount > 0
begin
break
end
fetch next from word_cursor into @word
end
-- clean up the cursor
close word_cursor
deallocate word_cursor
-- return the found CountryCode
select @CountryCode