processing multiline pattern in perl - regex

Question

i am working on adding new languages support for my mobile platform.I have to add entry for each language in several files,so i thought to do it using perl.To automate this process,i am feeling problem in how to match multi-line patterns in perl.
Here is my scenario :
const mmi_imeres_mode_details_struct g_ime_mode_array_int[] =
{
{
INPUT_MODE_NONE,
0,
0,
0,
0,
0,
0
},
{
INPUT_MODE_MULTITAP_LOWERCASE_ABC,
STR_INPUT_METHOD_MENU_MULTITAP_abc,
WGUI_IME_MULTITAP_LOWERCASE_ABC_IMG,
INPUT_MODE_DEFAULT_ALTERNATE_METHOD,
MMI_IME_ALL_EDITORS | MMI_IME_ENGLISH_ONLY_MODE | MMI_IME_ALPHABETIC | MMI_IME_LOWERCASE,
MMI_IMM_WRITING_LANGUAGE_ENGLISH,
"en-US"
},
}
First i had problem because in perl file is read one line at a time.so i first converted my file stream in to single variable.
my $newstr = '';
open (FH, "$filename") || die "Could not open file.\n";
while(<FH>)
{
$newstr = $newstr.$_;
}
No can someone help me how to search for text within { } , if it is a multi-line pattern.please reply soon...:)

Answer 1

First, there's a better idiom for slurping a file:
my $newstr;
{
open my $fh, '<', $filename or die "Could not open file $filename.\n$!\n";
local $/ = undef;
$newstr = <$fh>;
}
Next, you can set the /s modifier on your regexp, which treats the string as a single line by allowing '.' (dot) to match anything including newlines. But even that's not really necessary since you won't be using 'dot' in your regexp anyway.....
while(
$newstr =~ m/
{ # Match the opening bracket.
([^}]*) # Capture any number of characters that exclude '}'
} # Match the closing bracket.
/gx # Use /g for multiple matches, and /x for readability.
) {
print "$1\n";
}
Another solution would be to set your input record separator, $/, to '}'. That way you're reading the file in as chunks that end with a closing bracket. Nifty trick.