程序是用SAS寫的，本來也打算用python寫的，碰到困難，就用SAS了（SAS也困難重重，寫了一天，才搞定，╮(╯_╰)╭）
首先獲取數(shù)據(jù)，本程序利用了宏變量，宏程序；宏變量寫在程序開頭，方便修改
libname aa "d:\english_word";
/*是否合格的定義：大于候選個(gè)數(shù)限制且長度大于已輸入單詞，的為不合格*/
%let wait_max=5;/*候選個(gè)數(shù)限制*/
%let infile="d:\word.txt";
%let outfile="d:\output.txt";
/*讀取txt*/
data aa.allword_temp;
infile &infile.;
attrib word length=$20.;
input word $;
/*input word $20.;*/
/*是錯(cuò)誤寫法，$20.指連續(xù)的20位，效果是每隔一行讀一次，語法錯(cuò)誤會導(dǎo)致各種奇怪的錯(cuò)誤*/
run;
/*去重*/
proc sql noprint;
create table aa.allword as
select distinct word from aa.allword_temp;
quit;
;
接下來的工作就麻煩了。
/*0-3位縮寫不合格的*/
proc sql noprint;
create table aa.temp2 as
select distinct word,substr(word,1,3) as suox from aa.allword
group by suox
having count(word)>&wait_max. and length(word)>3;
/*0-3位縮寫合格的*/
create table aa.yes2 as
select distinct word,substr(word,1,3) as suox from aa.allword
group by suox
having count(word)<= &wait_max. or length(word)<=3;
quit;
;
測試一下，縮寫成3位,檢查，能正確運(yùn)行之后，就可以寫宏程序（弄一個(gè)循環(huán)）
我想弄一個(gè)sql里面的循環(huán)的，發(fā)現(xiàn)sql里面沒有循環(huán)結(jié)構(gòu)，反正我搞不懂
/*3位縮寫，不合格的*/
proc sql noprint;
create table aa.temp3 as
select word,substr(word,1,3) as suox from aa.temp2
group by suox
having count(word)>&wait_max. and length(word)>3
;
/*3位縮寫，合格的*/
/*不宜從allword里選,而是從上一步不合格（0-3位縮寫)的數(shù)中選*/
create table aa.yes3 as
select word,substr(word,1,3) as suox from aa.temp2
group by suox
having count(word)<=&wait_max. or length(word)<=3
;
quit;
下面就是完整代碼鏈接：
