replace-match: end of the replacement text for strings

replace-match: end of the replacement text for strings

[Roland Winkler]

Hi

When I use replace-match to operate on a string, is there an easy
way to find out the end of the replacement text in the string that
is returned by replace-match (like point in a buffer)? I have in
mind the situation when the optional arg LITERAL is nil, i.e., the
replacement text is `derived' from the string that is passed to
replace-match.

Thanks a lot,

Roland

Re: replace-match: end of the replacement text for strings

[Stefan Monnier <foo@acm.com>]

>>>>> "Roland" == Roland Winkler <Roland.Winkler@physik.uni-erlangen.de> writes:
> When I use replace-match to operate on a string, is there an easy
> way to find out the end of the replacement text in the string that
> is returned by replace-match (like point in a buffer)? I have in
> mind the situation when the optional arg LITERAL is nil, i.e., the
> replacement text is `derived' from the string that is passed to
> replace-match.

You should be able to get it by looking at the size of the string
before and after replacement as well as the size of the replaced text.
It's a bit round-about, but I can't think of any other way.


        Stefan

Re: replace-match: end of the replacement text for strings

[Roland Winkler]

"Stefan Monnier <foo@acm.com>" <monnier+gnu.emacs.help/news/@flint.cs.yale.edu> writes:

> >>>>> "Roland" == Roland Winkler <Roland.Winkler@physik.uni-erlangen.de> writes:
> > When I use replace-match to operate on a string, is there an easy
> > way to find out the end of the replacement text in the string that
> > is returned by replace-match (like point in a buffer)? I have in
> > mind the situation when the optional arg LITERAL is nil, i.e., the
> > replacement text is `derived' from the string that is passed to
> > replace-match.
> 
> You should be able to get it by looking at the size of the string
> before and after replacement as well as the size of the replaced
> text. It's a bit round-about, but I can't think of any other way.

In general, doesn't this work only when I know the replacement text
-- which I don't know when I have LITERAL set to nil and I am using
something like "\1" in the replacement?

Would it make sense to have something similar to match-data, i.e., a
lisp variable where the end of the replacement text is stored? Or is
the above situation too rare to justify such a thing? When
replace-match is operating on a buffer, it leaves point at the end
of the replacement text. So I guess that the function `knows' the
end of the replacement text anyway, even when operating on strings.

Roland

Re: replace-match: end of the replacement text for strings

[Stefan Monnier <foo@acm.com>]

>> You should be able to get it by looking at the size of the string
>> before and after replacement as well as the size of the replaced
>> text. It's a bit round-about, but I can't think of any other way.

> In general, doesn't this work only when I know the replacement text
> -- which I don't know when I have LITERAL set to nil and I am using
> something like "\1" in the replacement?

Reread what I wrote: I said "replaced" not "replacement".


        Stefan

Re: replace-match: end of the replacement text for strings

[Kai Großjohann]

Roland Winkler <Roland.Winkler@physik.uni-erlangen.de> writes:

> "Stefan Monnier <foo@acm.com>" <monnier+gnu.emacs.help/news/@flint.cs.yale.edu> writes:
>
>> >>>>> "Roland" == Roland Winkler <Roland.Winkler@physik.uni-erlangen.de> writes:
>> > When I use replace-match to operate on a string, is there an easy
>> > way to find out the end of the replacement text in the string that
>> > is returned by replace-match (like point in a buffer)? I have in
>> > mind the situation when the optional arg LITERAL is nil, i.e., the
>> > replacement text is `derived' from the string that is passed to
>> > replace-match.
>> 
>> You should be able to get it by looking at the size of the string
>> before and after replacement as well as the size of the replaced
>> text. It's a bit round-about, but I can't think of any other way.
>
> In general, doesn't this work only when I know the replacement text
> -- which I don't know when I have LITERAL set to nil and I am using
> something like "\1" in the replacement?

You need to do some arithmetic to find out the value you want.

length of replaced text = new length of string
                        - old length of string 
                        + length of text to be replaced

kai
-- 
~/.signature is: umop 3p!sdn    (Frank Nobis)

Re: replace-match: end of the replacement text for strings

[Roland Winkler]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> You need to do some arithmetic to find out the value you want.
> 
> length of replaced text = new length of string
>                         - old length of string 
>                         + length of text to be replaced

got you.

Thanks a lot,

Roland