Replacing text - problem with $1

[Martin Kristensen]

I'm trying to replace some text in some txt-files with a perl call
like this...

perl -pi -e "s/brad(.+?)bill/brian$1bob/gm;" *.txt

This should replace "brad123bill" by "brian123bob" - but instead I
just get "brianbob". What's wrong? Probably something with the $1
variable, but what?

Regards
Martin

 

[Anno Siegel]

I'm trying to replace some text in some txt-files with a perl call
like this...

perl -pi -e "s/brad(.+?)bill/brian$1bob/gm;" *.txt

This should replace "brad123bill" by "brian123bob" - but instead I
just get "brianbob". What's wrong? Probably something with the $1
variable, but what?

Nothing is wrong with the regex, except for the /m modifier, which does
nothing.

Anno

 

[Martin Kristensen]

Nothing is wrong with the regex, except for the /m modifier, which does
nothing.

I solved the problem by replacing " by ' so now it works perfectly for
some reason.

Martin

 

[Hunter Johnson]

I'm trying to replace some text in some txt-files with a perl call
like this...

perl -pi -e "s/brad(.+?)bill/brian$1bob/gm;" *.txt

This should replace "brad123bill" by "brian123bob" - but instead I
just get "brianbob". What's wrong? Probably something with the $1
variable, but what?

Regexp is fine. By the time it gets to perl, though, the s command
looks like this:

s/brad(.+?)bill/brianbob/gm

Your shell has replaced $1 with its value (nothing) before passing it
to perl. Replacing the double quotes with single quotes should work.
If not, time to dig out the shell's documentation.

Hunter