I have a fairly complex regexp, consisting of 9 match groups one of which has to match. I'm using copies of the exact same regexp in 4 different spots. All of them should match the same text. However, in one spot I want to access the match parameters as in $1 to $9, while in the other spots only one parameter ($3) and $`, $&, and $' are used.
How do I abstract a regexp so I only need to define and update it in one spot, but still be able to access the capture match groups anyway?
I tried storing the regexp pattern in a string variable then use that variable in the search, but I don't know how to correctly set the variable to a multi-line string (using a here document?) and use that for the pattern. Below is a dump of the working code, then the failing code and the error messages:
Working:
while (not $second and $linking =~
/(\bINT\s?[0-9A-Fa-f]{2}[Hh]?
(?:\/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-F
+a-f]{2,}[Hh]?)+
(?:\"[^"]+\")?
)
|(\bINT\s?[0-9A-Fa-f]{2}[Hh]?
(?:\"[^"]+\")?
)
|(\b(?:E?A[XHL])=[0-9A-Fa-f]{2,}[Hh]?
(?:\/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-F
+a-f]{2,}[Hh]?)*
(?:\"[^"]+\")?
)
|(\#[0-9A-Z][0-9]{4}\b)
|(\bMEM\s?[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,4}[Hh]
+?
(?:\"[^"]+\")?
)
|(\bMEM\s?[0-9A-Fa-fXx]{1,8}[Hh]?
(?:\"[^"]+\")?
)
|(\@[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,4}[Hh]?
(?:\"[^"]+\")?
)
|(\bPORT\s?[0-9A-Fa-fXx]{1,4}[Hh]?-[0-9A-Fa-fXx]{1,4}[Hh
+]?
(?:\"[^"]+\")?
)
|(\bPORT\s?[0-9A-Fa-fXx]{1,4}[Hh]?
(?:\"[^"]+\")?
)
/x) {
my $intplus = $1;
my $intonly = $2;
my $regonly = $3;
if (defined $regonly and not defined $int) {
if (defined $link) {
print_or_errorline("Entered reg only link without im
+plicit INT");
return;
}
$maskhighlight .= " " x (length($`) + length($&));
$linking = $';
next;
}
my $table = $4;
my $mem_16_16 = $5;
my $mem_32 = $6;
my $call = $7;
my $portrange = $8;
my $portsingle = $9;
$maskhighlight .= " " x length $`;
Failing:
my $linkpattern = <<PATTERNEND;
(\bINT\s?[0-9A-Fa-f]{2}[Hh]?
(?:\/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-F
+a-f]{2,}[Hh]?)+
(?:\"[^\"]+\")?
)
|(\bINT\s?[0-9A-Fa-f]{2}[Hh]?
(?:\"[^\"]+\")?
)
|(\b(?:E?A[XHL])=[0-9A-Fa-f]{2,}[Hh]?
(?:\/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-F
+a-f]{2,}[Hh]?)*
(?:\"[^\"]+\")?
)
|(\#[0-9A-Z][0-9]{4}\b)
|(\bMEM\s?[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,4}[Hh]
+?
(?:\"[^\"]+\")?
)
|(\bMEM\s?[0-9A-Fa-fXx]{1,8}[Hh]?
(?:\"[^\"]+\")?
)
|(\@[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,4}[Hh]?
(?:\"[^\"]+\")?
)
|(\bPORT\s?[0-9A-Fa-fXx]{1,4}[Hh]?-[0-9A-Fa-fXx]{1,4}[Hh
+]?
(?:\"[^\"]+\")?
)
|(\bPORT\s?[0-9A-Fa-fXx]{1,4}[Hh]?
(?:\"[^\"]+\")?
)
PATTERNEND
while (not $second and $linking =~
/$linkpattern/x) {
my $intplus = $1;
my $intonly = $2;
my $regonly = $3;
if (defined $regonly and not defined $int) {
if (defined $link) {
print_or_errorline("Entered reg only link without im
+plicit INT");
return;
}
$maskhighlight .= " " x (length($`) + length($&));
$linking = $';
next;
}
my $table = $4;
my $mem_16_16 = $5;
my $mem_32 = $6;
my $call = $7;
my $portrange = $8;
my $portsingle = $9;
$maskhighlight .= " " x length $`;
My program uses curses for a TUI, so I redirected stderr to a file to capture perl error messages:
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unrecognized escape \s passed through at /home/[user]/proj/tractest/in
+tlist.pl line 419.
Unmatched ( in regex; marked by <-- HERE in m/ INTs?[0-9A
+-Fa-f]{2}[Hh]?
(?:/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-Fa
+-f]{2,}[Hh]?)+
(?:"[^"]+")?
)
|INTs?[0-9A-Fa-f]{2}[Hh]?
(?:"[^"]+")?
)
|(?:E?A[XHL])=[0-9A-Fa-f]{2,}[Hh]?
(?:/(?:E?[ABCD][XHL]|E?[SD]I|E?[SB]P|[DESC]S)=[0-9A-Fa
+-f]{2,}[Hh]?)*
(?:"[^"]+")?
)
|(#[0-9A-Z][0-9]{4)
<-- HERE |MEMs?[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,
+4}[Hh]?
(?:"[^"]+")?
)
|MEMs?[0-9A-Fa-fXx]{1,8}[Hh]?
(?:"[^"]+")?
)
|(@[0-9A-Fa-fXx]{1,4}[Hh]?:[0-9A-Fa-fXx]{1,4}[Hh]?
(?:"[^"]+")?
)
|PORTs?[0-9A-Fa-fXx]{1,4}[Hh]?-[0-9A-Fa-fXx]{1,4}[Hh]?
(?:"[^"]+")?
)
|PORTs?[0-9A-Fa-fXx]{1,4}[Hh]?
(?:"[^"]+")?
)
/ at /home/[user]/proj/tractest/intlist.pl line 447, <$array_lstff[...
+]> line 197305.
|
I have an NFS file system mounted on many machines. I am attempting to use file locking to get exclusive access by one system to a file.
The critical code section shown here:
# loop through file names looking for one we can lock
open ( my $file_handle, '+<', $file_name ) or next;
if ( !flock( $file_handle, LOCK_EX | LOCK_NB ) ) {
close( $file_handle );
next;
}
# do some work here based on the file contents
unlink $file_name;
close( $file_handle );
This almost always works, but occasionally there is a failure where two machines get a lock on the same file. It appears there is a race between unlinking (deleting) the file and releasing the lock. It seems like there is a small window where the lock is actually released before the file is deleted allowing another machine to lock and read the file before it is deleted.
I am wondering if this is a known problem and if there is a preferred way to work around this.
|