SAS哈希合并宏 - 输出多个数据集

3 回答

• 0

  通过这种方式，您可以轻松获取 xtra 数据集 . 你可以通过类似的东西获得 miss 和 onto ，而不必过多地对你的 data 语句进行必要的修改：

  if rc = 0 then output &onto;
  else output miss&varnm;
  

  但是，要查找 from 数据集中尚未使用但通常会转到 xtra 数据集的所有值，您需要跟踪合并时使用的值，然后在最后数据步骤将其余部分复制到新的哈希对象中并输出该对象 .

  注：使用当前代码，所有行都输出到 &onto 数据集，即使它们缺少查找变量的值也是如此 .

  回复于 2024-04-29T08:08:07+08:00
• 0

  这样做有点难，但在这里你可以了解它是如何完成的 .

  创建另一个哈希哈希对象，就像第一个一样 . 从第二个对象，您将删除已经匹配的项目 . 最后，您只需将第二个对象输出到数据集 . 它需要散列哈希对象的内存，但它是第一次尝试 .

  （没有运行代码并抱歉任何错误）

  %macro hashmerge(varnm,onto,from,byvars);
  /* The inputs are the variables to merge,
     the merge onto data set, the merge from
     data set and the key(s). vanrnm and 
     byvars are set up to accept multiple inputs. */
  
  %let data_vars   = %trim   (&varnm);
  %let data_vars_a = %sysfunc(tranwrd(&data_vars.,%str( ),%str(",")));
  %let data_vars_b = %sysfunc(tranwrd(&data_vars.,%str( ), %str(,)));
  %let data_key    = %trim   (&byvars);
  %let data_key    = %sysfunc(tranwrd(&data_key.,%str( ), %str(",")));
  
  data &onto (drop=rc);
     if 0 set &onto &from(keep=&varnm. &byvars.);   
  
     declare hash h_merge    (dataset: "&from.");
     rc = h_merge.DefineKey  ("&data_key.");
     rc = h_merge.DefineData ("&data_vars_a.");
     rc = h_merge.DefineDone ();
  
     /*hash table that will hold data not yet founded by find method */
     declare hash h_merge_copy    (dataset: "&from.");
     rc = h_merge_copy.DefineKey  ("&data_key.");
     rc = h_merge_copy.DefineData ("&data_vars_a.");
     rc = h_merge_copy.DefineDone ();
  
     do until (eof);
  
        set &onto. end = eof;
  
        if h_merge.find () = 0 then do;
           /*removing data from h_merge_copy if founded and if not allready removed*/
           if h_merge_copy.check() = 0 then do;
              rc = h_merge_copy.remove();
           end;
        end;
        else
           output miss&varnm.; /*no find match - output to miss&varnm.*/
  
        output &onto.; /*find match in hash or not output to &onto. (left join)*/
     end;
  
     h_merge_copy.output(dataset: "&onto.");
  
     stop;
  run;
  
  %mend;
  

  我可以改进这个以消耗更少的内存，就像在第二个散列中没有存储数据一样，但今天它已经太晚了 . 希望这有点帮助 .

  回复于 2024-04-29T08:08:07+08:00
• 0

  感谢大家的帮助和建议 . 我不知道如何添加'xtra'数据集而不创建另一个哈希，这会破坏哈希合并的目的 . 所以这就是我必须创建缺失和合并的数据集 .

  %macro hashmerge(varnm,onto,from,byvars);
  
  
  %let data_vars   = %trim   (&varnm);
  %let data_vars_a = %sysfunc(tranwrd(&data_vars.,%str( ),%str(",")));
  %let data_vars_b = %sysfunc(tranwrd(&data_vars.,%str( ), %str(,)));
  %let data_key    = %trim   (&byvars);
  %let data_key    = %sysfunc(tranwrd(&data_key.,%str( ), %str(",")));
  
  data &onto(drop=rc) miss&varnm(drop=rc);
     if 0 then set &onto &from(keep=&varnm. &byvars.);
  
     declare hash h_merge (dataset: "&from.");
     rc = h_merge.DefineKey  ("&data_key.");
     rc = h_merge.DefineData ("&data_vars_a.");
     rc = h_merge.DefineDone ();
  
     do until (eof);
       set &onto end = eof;
       call missing(&data_vars_b.);
       rc = h_merge.find ();
       if rc = 0 then do;
         output &onto;
         from = "&from.";
       end;
       else do;
         output miss&varnm &onto;
         from = "&onto.";
       end;
     end;
  
  stop;
  run;
  
  %mend;
  

  如果您只想将非缺失值包含到新数据集中或删除“from”变量，那么修改代码就足够简单了 .

  如果有人知道更有效的方法，请发布建议 .

  回复于 2024-04-29T08:08:07+08:00