awk小写替换的问题

kernel69

哪个群，我也加一下，
回复 5# yestreenstars


   

yestreenstars

回复 11# kernel69


    shell awk sed

Shell_HAT

回复 10# V_客


    不是这个问题 @blackold

V_客

回复 13# Shell_HAT


    我想到就是编码是en 或者zh啥的 这个影响到a-z 顺序的，请教下别的情况 谢谢

blackold

回复 13# Shell_HAT


    我还没碰到过这种问题，只是按照逻辑，应该是locale的原因。

http://www.gnu.org/software/gawk ... #Ranges-and-Locales

在你的环境下 LC_ALL=C 也不行？

blackold

回复 13# Shell_HAT
@Shell_HAT@V_客
grep man

   
       Within  a  bracket  expression,  a  range  expression  consists  of two
       characters separated by a hyphen.  It matches any single character that
       sorts  between  the  two  characters,  inclusive,  using  the  locale's
       collating sequence and character set.  For example, in  the  default  C
       locale, [a-d] is equivalent to [abcd].  Many locales sort characters in
       dictionary  order,  and  in  these  locales  [a-d]  is  typically   not
       equivalent to [abcd]; it might be equivalent to [aBbCcDd], for example.
       To obtain the traditional interpretation of  bracket  expressions,  you
       can  use the C locale by setting the LC_ALL environment variable to the
       value C.

blackold

awk man:

Within a character list, a "range expression" consists of two
characters separated by a hyphen.  It matches any single character that
sorts between the two characters, using the locale's collating sequence
and character set.  For example, in the default C locale, `[a-dx-z]' is
equivalent to `[abcdxyz]'.  Many locales sort characters in dictionary
order, and in these locales, `[a-dx-z]' is typically not equivalent to
`[abcdxyz]'; instead it might be equivalent to `[aBbCcDdxXyYz]', for
example.  To obtain the traditional interpretation of bracket
expressions, you can use the C locale by setting the `LC_ALL'
environment variable to the value `C'.

seesea2517

回复 17# blackold

啊啊啊……疯狂的 awk

1. [seesea@UC ~]$ LC_ALL=C; echo abcABC | awk '{print IGNORECASE ":" gensub(/([a-z])/, "x", "g");}'
   
2. 0:xxxxxx
   
3. [seesea@UC ~]$ echo $LC_ALL
   
4. C

复制代码

blackold

回复 18# seesea2517
locale 输出是啥？


什么系统？

seesea2517

嘿嘿没事了……
这个 export 还真是必须的好像：

1. [seesea@UC ~]$ LC_ALL=C; echo abcABC | awk '{print IGNORECASE ":" gensub(/([a-z])/, "x", "g");}'
   
2. 0:xxxxxx
   
3. [seesea@UC ~]$ echo $LC_ALL
   
4. C
   
5. [seesea@UC ~]$ locale
   
6. LANG=zh_CN
   
7. LC_CTYPE="zh_CN"
   
8. LC_NUMERIC="zh_CN"
   
9. LC_TIME="zh_CN"
   
10. LC_COLLATE="zh_CN"
    
11. LC_MONETARY="zh_CN"
    
12. LC_MESSAGES="zh_CN"
    
13. LC_PAPER="zh_CN"
    
14. LC_NAME="zh_CN"
    
15. LC_ADDRESS="zh_CN"
    
16. LC_TELEPHONE="zh_CN"
    
17. LC_MEASUREMENT="zh_CN"
    
18. LC_IDENTIFICATION="zh_CN"
    
19. LC_ALL=
    
20. [seesea@UC ~]$ uname -a
    
21. Linux p01coding 2.6.18-128.2.1.el5 #1 SMP Tue Jul 14 06:39:56 EDT 2009 i686 i686 i386 GNU/Linux
    
22. [seesea@UC ~]$ export LC_ALL=C
    
23. [seesea@UC ~]$ locale         
    
24. LANG=zh_CN
    
25. LC_CTYPE="C"
    
26. LC_NUMERIC="C"
    
27. LC_TIME="C"
    
28. LC_COLLATE="C"
    
29. LC_MONETARY="C"
    
30. LC_MESSAGES="C"
    
31. LC_PAPER="C"
    
32. LC_NAME="C"
    
33. LC_ADDRESS="C"
    
34. LC_TELEPHONE="C"
    
35. LC_MEASUREMENT="C"
    
36. LC_IDENTIFICATION="C"
    
37. LC_ALL=C
    
38. [seesea@UC ~]$ echo abcABC | awk '{print IGNORECASE ":" gensub(/([a-z])/, "x", "g");}'         
    
39. 0:xxxABC

复制代码