Q: RegEx - string substitution

[Troll]

Hi,

I'd like to change 'Jones & Bros.' to 'Jones and Company' when it appears as
part of a h1 heading in a HTML doc. I have tried the following with no luck:

s/<h1>.*Jones & Bros..*</h1>/<h1>.*Jones and Company.*</h1>/
s/<h1>[\\s|S]*Jones & Bros.[\\s|S]*</h1>/<h1>[\\s|S]*Jones and
Company[\\s|S]*</h1>/

Where am I going wrong? We cannot assume that Jones & Bros. is alone in the
heading. There may be other characters/words including spaces around it.

 

[Glenn Jackman]

Hi,

I'd like to change 'Jones & Bros.' to 'Jones and Company' when it appears as
part of a h1 heading in a HTML doc. I have tried the following with no luck:

s/<h1>.*Jones & Bros..*</h1>/<h1>.*Jones and Company.*</h1>/
s/<h1>[\\s|S]*Jones & Bros.[\\s|S]*</h1>/<h1>[\\s|S]*Jones and
Company[\\s|S]*</h1>/

Where am I going wrong? We cannot assume that Jones & Bros. is alone in the
heading. There may be other characters/words including spaces around it.

Your error is leaving regex metacharacters in the substitution.
s{(<h1>.*Jones) & Bros\.}{$1 and Company}

 

[Chris Mattern]

Hi,

I'd like to change 'Jones & Bros.' to 'Jones and Company' when it appears as
part of a h1 heading in a HTML doc. I have tried the following with no luck:

s/<h1>.*Jones & Bros..*</h1>/<h1>.*Jones and Company.*</h1>/
s/<h1>[\\s|S]*Jones & Bros.[\\s|S]*</h1>/<h1>[\\s|S]*Jones and
Company[\\s|S]*</h1>/

Where am I going wrong? We cannot assume that Jones & Bros. is alone in the
heading. There may be other characters/words including spaces around it.

Your error is leaving regex metacharacters in the substitution.
s{(<h1>.*Jones) & Bros\.}{$1 and Company}

This will, of course, break in the first h1 heading that doesn't have
"Jones & Bros." on the same line as the h1 tag. It will also break
if "Jones & Bros." appears on the same line as the h1 tag but after
the heading has been closed. Finally, it will break if "Jones & Bros."
appears more than once in a heading.

Don't use regexs to parse HTML. Parsing HTML is *not* trivial and
beyond the capabilities of regexs. Use a perl package for parsing
HTML, of which several are available on CPAN.

Chris Mattern

 

[Troll]

Hi,

I'd like to change 'Jones & Bros.' to 'Jones and Company' when it appears as
part of a h1 heading in a HTML doc. I have tried the following with no luck:

s/<h1>.*Jones & Bros..*</h1>/<h1>.*Jones and Company.*</h1>/
s/<h1>[\\s|S]*Jones & Bros.[\\s|S]*</h1>/<h1>[\\s|S]*Jones and
Company[\\s|S]*</h1>/

Where am I going wrong? We cannot assume that Jones & Bros. is alone in the
heading. There may be other characters/words including spaces around

it.


Your error is leaving regex metacharacters in the substitution.
s{(<h1>.*Jones) & Bros\.}{$1 and Company}

This will, of course, break in the first h1 heading that doesn't have
"Jones & Bros." on the same line as the h1 tag. It will also break
if "Jones & Bros." appears on the same line as the h1 tag but after
the heading has been closed. Finally, it will break if "Jones & Bros."
appears more than once in a heading.

Don't use regexs to parse HTML. Parsing HTML is *not* trivial and
beyond the capabilities of regexs. Use a perl package for parsing
HTML, of which several are available on CPAN.

Chris Mattern

Thanks guys. If I have no option to get a CPAN module will something along
the lines come close:

replace {
s{(.*Jones) & Bros.}{$1 and Company};
}

if ($_ =~ "<h1>") {
until ($_ =~ "</h1>" ) {
replace ();
}
}

Tho it's failing somewhere in the UNTIL part an the function doesn't get
called. Cheers.