How does one avoid matching \n when matching all but whitespace?

How does one avoid matching \n when matching all but whitespace?

[ld1976]

Hello all. This is a rookie's question.
 I am trying to a simple text regular expression replacement.
The text goes as follows (white spaces may change:
tcl_cmd bla > kuku
I want to change it to:
tcl_cmd bla > kuku.rev1
When trying to use replace-regexp \(>[ ]+[^ ]+\)  to \1.rev1 I get:
tcl_cmd bla > kuku
.rev1
This is quiet different than using the \W+ with perl...
I think that emacs matches within [^ ]+ \n too. Trying to use [^ \n]+ did
not solve the issue either.
Any thoughts?
-Lior
-- 
View this message in context: http://www.nabble.com/How-does-one-avoid-matching-%5Cn-when-matching-all-but-whitespace--tf4041646.html#a11481693
Sent from the Emacs - Help mailing list archive at Nabble.com.

Re: How does one avoid matching \n when matching all but whitespace?

[Stephen Berman]

On Sat, 7 Jul 2007 12:59:31 -0700 (PDT) ld1976 <lior_dagan@yahoo.com> wrote:

> Hello all. This is a rookie's question.
>  I am trying to a simple text regular expression replacement.
> The text goes as follows (white spaces may change:
> tcl_cmd bla > kuku
> I want to change it to:
> tcl_cmd bla > kuku.rev1
> When trying to use replace-regexp \(>[ ]+[^ ]+\)  to \1.rev1 I get:
> tcl_cmd bla > kuku
> .rev1
> This is quiet different than using the \W+ with perl...
> I think that emacs matches within [^ ]+ \n too. Trying to use [^ \n]+ did
> not solve the issue either.

With replace-regexp you need to type C-q C-j instead of \n to match a
newline.  (Emacs 22.1 gives you a hint to that effect when you type
`\n' in the regexp.)

Steve Berman

Re: How does one avoid matching \n when matching all but whitespace?

[Tyler Smith]

On 2007-07-07, ld1976 <lior_dagan@yahoo.com> wrote:
>
> Hello all. This is a rookie's question.
>  I am trying to a simple text regular expression replacement.
> The text goes as follows (white spaces may change:
> tcl_cmd bla > kuku
> I want to change it to:
> tcl_cmd bla > kuku.rev1
> When trying to use replace-regexp \(>[ ]+[^ ]+\)  to \1.rev1 I get:
> tcl_cmd bla > kuku
> .rev1
> This is quiet different than using the \W+ with perl...

You could do it (almost) the same way you'd do it in perl:

'\(> +\w+\)'  to '\1.rev1'

You can also use '$' to explicitly signal the end of line, which is
helpful if you have other incidences of '> ' that occur earlier in the
line:

'\(> +\w+\)$'  to '\1.rev1'

HTH,

Tyler