SAS字符型函数

528866wmYSJ*

```sql
data compress_1;
x1="  1 2 a 3";
x2="Abacabbad";
x3="134-7596-5241";
x4="123***14567";
y1=compress(x1," ");
y2=compress(x2,"Ab");/*消除的是A 或b不是且的关系*/
y3=compress(x3,"0123456789","k");
/*消除除了数字外的所有其它字符*/
y4=compress(x4,"0123456789","k");
run;

在这里插入代码片`在这里插入代码片`

```sql
data compress_1;
x1="  1 2 a 3";
x2="Abacabbad";
x3="134-7596-5241";
x4="123***14567";
y1=compress(x1," ");
y2=compress(x2,"Ab");/*消除的是A 或b不是且的关系*/
y3=compress(x3,"0123456789","k");
y4=compress(x4,"0123456789","k");
y5=compress(x1);/*什么都不写，默认是清楚空格*/
run;

data compress_mob;
length Mob_Phe$ 14;
input Mob_Phe$ @@;
Cards;
13915625877 013145672561 134-7596-5241
1364524124 01045125421
;
run;
/*要消除不是数字的其它字符*/
data mob1;
set compress_mob;
if length(compress(Mob_Phe,"0123456789","k"))=11 then do;
mob_phone=substr(compress(Mob_Phe,"0123456789","k"),1,11);
   if substr(compress(Mob_Phe),1,3) in("130","131","132","133","150","151","152","153") then do;
   flag="联通";output;END;
   else if substr(compress(Mob_Phe),1,3) in("134","135","136","137","138","139",
   "154","155","156","157","158","159") then do;
   flag="移动";output;END;
end;
else if length(compress(Mob_Phe,"0123456789","k"))=12 
and substr(compress(Mob_Phe),1,1)="0" then do;
mob_phone=substr(compress(Mob_Phe,"0123456789","k"),2,11);
   if substr(compress(Mob_Phe),2,3) in("130","131","132","133","150","151","152","153") then do;
   flag="联通";output;END;
   else if substr(compress(Mob_Phe),2,3) in("134","135","136","137","138","139",
   "154","155","156","157","158","159") then do;
   flag="移动";output;END;
end;
run;

if then 是满足if条件并执行后面一个语句
if then do…end;满足if语句并执行do和end之间的所有语句，通常用于多个语句的时候可以用do end
另外，substr(str,3,11)指的是从第3位开始取，共取11个，且没有第0位，都是第一个

/*put函数，把数值型和字符型变量转为字符型变量，put(source,format),format是输出的格式*/
/*sas中数据类型只有两种，一种数值型一种字符型*/
/*"01Jan2007"D这实际是数值型的变量，可以参与数值变化，如何转化为字符型*/
data test;
X="01Jan2007"D;
y=put(X,yymmdd10.);
Z=compress(y,"-");
run;

发现对X没有规定格式，输出的是数字

data test;
X="01Jan2007"D;
format x yymmdd10.;规定了x的输出格式，但仍然是数值型的
y=put(X,yymmdd10.);
Z=compress(y,"-");
run;

/*substr(),从第几位开始取，共取几位*/
data test3;
x="abcd e";
y=substr(x,2);*从第二位开始取直到结束;
y2=substr(x,2,1);
z=substr(x,length(x)-3,4);/*从后面开始取，取四位，那么length()-3*/
/*就是取几位，那么length（）减去取几位再减一*/
put x=;
put y=;
put y2=;
put z=;
run;

x=abcd e
y=bcd e
y2=b
z=cd e

/*tranwrd替代字符，tranwrd(str,需要替代的字符，替换成哪个字符)*/
data test4;
x="abcdaba";
x1=compress(x,"ab");
y=tranwrd(x,"ab","k");
z=tranwrd(x,"ab","");
lz=length(z);
z1=tranwrd(x,"ab"," ");
lz1=length(z1);
m=compress(z);
m1=compress(z1);
put x=;
put x1=;
put y=;
put z=;
put lz=;
put z1=;
put lz1=;
put m=;
put m1=;
run;

x=abcdaba
x1=cd
y=kcdka
z=cd a
lz=5
z1=cd a
lz1=5
m=cda
m1=cda