help extracting tag with boost:regex

[MCH]

hi there,
I am working with a HTML-like text with boost:regex. For example,
the following pattern might occur in my text

<abc efg>
[TAG] <p>EFG</p> [/TAG] 12<3>

In this case, I would like to extract everything between [TAG] [/TAG]
and replace [TAG] with <pre>, [/TAG] with </pre>. Meanwhile,
everything outside [TAG][/TAG] should be unchaged except that < is
replaced by &lt; and > is replaced by &gt;

In far more complicated case, a nested [TAG] might occur as follow

<abc efg>
[TAG] <p>EF [TAG] eee [/TAG] G</p> [/TAG] 12<3>

in this case, the program just tackle the outermost TAG and left the
inside TAG there. I try to implement the program with boost::regex;
however, it seems never succeed in extracing the TAG even for the
simple case.

 

[David Harmon]

<abc efg>
[TAG] <p>EFG</p> [/TAG] 12<3>

In this case, I would like to extract everything between [TAG] [/TAG]
and replace [TAG] with <pre>, [/TAG] with </pre>. Meanwhile,
everything outside [TAG][/TAG] should be unchaged except that < is
replaced by &lt; and > is replaced by &gt;

This exceeds what I know how to do with regex. Too much context.
I would do it with std::string, std::find_first_of(),
and a couple of if statements.

Maybe you could pretend you are writing a Perl program and get some help
from the real regex experts over on comp.lang.perl.misc

 

[Fei Liu]

hi there,
I am working with a HTML-like text with boost:regex. For example,
the following pattern might occur in my text

<abc efg>
[TAG] <p>EFG</p> [/TAG] 12<3>

In this case, I would like to extract everything between [TAG] [/TAG]
and replace [TAG] with <pre>, [/TAG] with </pre>. Meanwhile,
everything outside [TAG][/TAG] should be unchaged except that < is
replaced by &lt; and > is replaced by &gt;

In far more complicated case, a nested [TAG] might occur as follow

<abc efg>
[TAG] <p>EF [TAG] eee [/TAG] G</p> [/TAG] 12<3>

in this case, the program just tackle the outermost TAG and left the
inside TAG there. I try to implement the program with boost::regex;
however, it seems never succeed in extracing the TAG even for the
simple case.

try
const boost::regex expression("\[TAG\].*?\[\/TAG\]");

The trick is to use the '?' after .* to turn off greedy pattern match,
instead of matching the last occurance of [/TAG], it will match the
first [/TAG] which may or may not be what you want. It seems your
problem is not so much as boost::regex but utilizing regular expression
pattern match in general. I would recommend you to consult regular
expression documents first and experiment with simpler string pattern
with boost::regex.

Fei

 

[Pete Becker]

hi there,
I am working with a HTML-like text with boost:regex. For example,
the following pattern might occur in my text

<abc efg>
[TAG] <p>EFG</p> [/TAG] 12<3>

In this case, I would like to extract everything between [TAG] [/TAG]
and replace [TAG] with <pre>, [/TAG] with </pre>. Meanwhile,
everything outside [TAG][/TAG] should be unchaged except that < is
replaced by &lt; and > is replaced by &gt;

In far more complicated case, a nested [TAG] might occur as follow

<abc efg>
[TAG] <p>EF [TAG] eee [/TAG] G</p> [/TAG] 12<3>

in this case, the program just tackle the outermost TAG and left the
inside TAG there. I try to implement the program with boost::regex;
however, it seems never succeed in extracing the TAG even for the
simple case.

try
const boost::regex expression("\[TAG\].*?\[\/TAG\]");

The trick is to use the '?' after .* to turn off greedy pattern match,
instead of matching the last occurance of [/TAG], it will match the
first [/TAG] which may or may not be what you want.

It's definitely not what's needed, since it will match the [TAG] of
the outer block to the [/TAG] of the inner block. See the "far more
complicated case" above.

It seems your
problem is not so much as boost::regex but utilizing regular expression
pattern match in general. I would recommend you to consult regular
expression documents first and experiment with simpler string pattern
with boost::regex.

Right. Regular expressions don't deal well with recursive patterns.

--

-- Pete
Roundhouse Consulting, Ltd. (www.versatilecoding.com)
Author of "The Standard C++ Library Extensions: a Tutorial and
Reference." (www.petebecker.com/tr1book)