档案合并加总问题 ?

lucas528

我有两各档案内容如下:

file 1.txt
00:00,0003,0000453
00:05,0006,0000681
00:10,0006,0000580
00:15,0005,0000497
00:20,0008,0000595
00:25,0003,0000461
00:30,0004,0000396
00:35,0007,0000507
00:40,0009,0000521

file 2.txt
00:00,0005,0000542
00:05,0007,0000493
00:10,0005,0000222
00:15,0004,0000227
00:20,0006,0000420
00:25,0004,0000398
00:30,0002,0000250
00:35,0003,0000279
00:40,0005,0000373

需要把$2加总，$3加总后/2 产生file 3.txt该如何写呢?

随风漂

awk -F "," '{sum2+=$2;sum3+=$3}END{print sum2,sum3/2}' 1.txt 2.txt >3.txt

chzht001

原帖由 随风漂 于 2006-11-17 15:44 发表
awk -F "," '{sum2+=$2;sum3+=$3}END{print sum2,sum3/2}' 1.txt 2.txt >3.txt

哎，被抢了先

lucas528

原帖由 随风漂 于 2006-11-17 15:44 发表
awk -F "," '{sum2+=$2;sum3+=$3}END{print sum2,sum3/2}' 1.txt 2.txt >3.txt

File 3.txt 的内容也要是 file 1.txt的format一样只是把$2，$3做计算，
$1时间的栏位不动，不是要纯粹计算出答案，请问有方法吗?

计算后File 3.txt内容:
00:00,0008,000497
00:05,0013,000587
00:10,0011,000401
00:15,0009,000362
00:20,0014,000507
00:25,0007,000429
00:30,0006,000323
00:35,0010,000393
00:40,0014,000447

[ 本帖最后由 lucas528 于 2006-11-17 18:32 编辑 ]

lucas528

PUSH ~

Edengundam

1. 
   
2. #!/bin/awk -f
   
3. BEGIN {
   
4.     FS = ","
   
5. }
   
6. FILENAME == "1.txt" {
   
7.     hash[$1] = $0
   
8. }
   
9. FILENAME == "2.txt" {
   
10.     if ($1 in hash)
    
11.     {
    
12.         split(hash[$1], tmp);
    
13.         printf("%s,%.4d,%.7d\n", $1, $2 + tmp[2], ($3 + tmp[3]) / 2);
    
14.     }
    
15. }
    

复制代码

./脚本名 1.txt 2.txt

我这里的结果

1. 
   
2. Administrator@EDENSXY ~
   
3. $ ./test.awk 1.txt 2.txt
   
4. 00:00,0008,0000497
   
5. 00:05,0013,0000587
   
6. 00:10,0011,0000401
   
7. 00:15,0009,0000362
   
8. 00:20,0014,0000507
   
9. 00:25,0007,0000429
   
10. 00:30,0006,0000323
    
11. 00:35,0010,0000393
    
12. 00:40,0014,0000447
    

复制代码

如有错误请指出

[ 本帖最后由 Edengundam 于 2006-11-19 21:28 编辑 ]

lucas528

感谢!!

在请教一下，如果一次有六各档案要合并呢?

Edengundam

1. 
   
2. #!/bin/awk -f
   
3. BEGIN {
   
4.     FS = ","
   
5. }
   
6. 
   
7. FNR == NR {
   
8.     hash[$1] = $0
   
9. }
   
10. NR > FNR {
    
11.     if ($1 in hash)
    
12.     {
    
13.         split(hash[$1], tmp);
    
14.         hash[$1] = sprintf("%s,%.4d,%.7d", $1, $2 + tmp[2], $3 + tmp[3]);
    
15.     }
    
16. }
    
17. 
    
18. END {
    
19.     for (item in hash)
    
20.     {
    
21.         split(hash[item], tmp);
    
22.         printf("%s,%.4d,%.7d\n", tmp[1], tmp[2], tmp[3] / 6);
    
23.     }
    
24. }
    

复制代码

使用脚本名awk.script

./awk.script file1 file2 file3 file4 file5 file6 > output.txt

sort output.txt

不然结果是没有顺序的>_<我水平差, 就只能这样搞了

lucas528

再请教一各白痴问题，就是我在linux上可以跑，但是在aix 5.3的环境上都会出现错误:
+ BEGIN {
xxx.awk[2]: BEGIN:  not found
+ FS = ,
xxx.awk[3]: FS:  not found
xxx.awk[4]: syntax error at line 5 : `}' unexpected

Edengundam

aix默认是ksh嘛???

我的脚本第一行 #!/bin/awk -f
aix上你是不是写错了?