May the source be with you, but remember the KISS principle ;-)
Home Switchboard Unix Administration Red Hat TCP/IP Networks Neoliberalism Toxic Managers
(slightly skeptical) Educational society promoting "Back to basics" movement against IT overcomplexity and  bastardization of classic Unix

Introduction to Perl 5.10 for Unix System Administrators

(Perl 5.10 without excessive complexity)

by Dr Nikolai Bezroukov

Contents : Foreword : Ch01 : Ch02 : Ch03 : Ch04 : Ch05 : Ch06 : Ch07 : Ch08 :

Prev | Up | Contents | Down | Next

4.2. Determining the Type and Attributes of a File

Version 0.91

There are several dozens operators in Perl that can help to test the status of the file. They are derived from and are similar to used in Unix shells.

They should be viewed as shortcuts for stat function. That's why in Perl documentation they are considered to be built-in functions, not operators (see -X -  The most widely used are just five listed in the table below:

Operator Build-in function
-d Is this a directory?
-e Does this file exist?
-r Is this file readable by the person running the script?
-s Returns the size of the file
-w Is this file writeable by the person running the script?
-x Is this file executable by the person running the script?

First we will discuss probably the most popular test among listed above the -e test. If you want to open the file for writing if the file does not already exist, you can first test to see if a file exists using the -e operator.

All other file-test operators have the same syntax as the -e operator used below. This is a unary operator that accept string as its only operand. The value of the string should contain the name of the file to be tested. It can be fully qualified name or relative name.

unless (-e "/home/nnb/.profile") {
   die ("file /home/nnb/.profile does not exist");
open (SYSPROF, "/home/nnb/.profile");

If the file exists, the -e operator returns true; otherwise, it returns false. Similar tests exist to test other file conditions. Here is another example:

unless (open(SYSIN, "</etc/hosts") {
   if (-e "/etc/hosts") {                 
      die ("File /etc/hosts exists, but cannot be opened for reading (wrong permissions?).\n"));
   } else {
      die ("File /etc/hosts does not exist.\n");

If you use Windows and need specify full path using backslashes, please remember that they need to be doubles as backslash serves as a escape character in Perl literals. Actually you can use regular slash, but this is convenient solution only for people who get used to Unix.

You can even extend previous trick with || operator, but it becomes somewhat less comprehensible and although this is a Perl idiom this is probably a bad Perl idiom -- I recommend against using it:

open(SYSIN, "infile") && !(-e "infile") || die("Cannot open infile\n");

In Unix and Windows NT (unless you are always using root and administrator IDs to login ;-) before you can open a file for reading, you must have permission to read the file. The -r file-test operator tests whether you have permission to read a file.

unless (open(SYSIN, "$fname")) {
  if (!(-e "file1")) {die ("File $fname does not exist.\n");}
  unless (-r "file1")) {die ("You are not allowed to read $fname.\n");
  die ("File1 cannot be opened. Reason unknown\n");

To check whether you have write permission on a file, use the -w file-test operator.

unless (-w "$fname") {
   print STDERR ("Can't write to $fname.\n");

The -x file-test operator is used mainly in Unix environment and it checks whether you have execute permission on the file (in other words, whether the system thinks this is an executable script, and whether you have permission to run it if it is), as illustrated here:

unless (-x "$fname") {
   print STDERR ("I can run $fname.\n");

The -s file-test operator returns the size of the file in bytes. This provides a more refined test for whether or not to open a file for writing: if the file exists but is empty, no information is lost if you overwrite the existing file.

$size = -s "outfile";
if ($size == 0) {
   print ("The file is empty.\n");
} else {
   print ("The file is $size bytes long.\n");

The file-test operators provide a way of retrieving information on a particular file. The most common file-test operators are

Full list of file test operators

Note that Perl allows comparing too file timestamps via -M imitating what is available in shell as -nt operator

-M How long since filename modified? Start time minus file modification time, in days.

if ( ( -m $file1 ) > ( -M file 2 ) ) { 
     # file1 is newer then file 2

Full list of file test built-in functions contain more than a dozen entries as shown in the table below:

Build-in function Description
-b Is filename a block device?
-c Is filename a character device?
-d Is name a directory?
-e Does filename exist?
-f Is filename an ordinary file?
-g Does filename have its setgid bit set?
-k Does filename have its "sticky bit" set?
-l Is filename a symbolic link?
-o Is filename owned by the user?
-p Is name a named pipe?
-r Is filename a readable file?
-s Returns the size of the file
-t Does name represent a terminal?
-u Does filename have its setuid bit set? The effective userid in this case will be different fromt the user login ID
-w Is filename a writable file?
-x Is filename an executable file?
-z Is filename an empty file?
-A How long since filename accessed? Similar to -M
-B Is filename a binary file?
-C How long since filename's inode accessed?
-M How long since filename modified? Start time minus file modification time, in days.
-O Is filename owned by the "real user" only? (See -u (setuid bit) comment)
-R Is filename readable by the "real user" only? (See -u (setuid bit) comment)
-S Is name a socket?
-T Is filename a text file?
-W Is filename writable by the "real user" only? (See -u (setuid bit) comment)
-X Is filename executable by the "real user" only? (See -u (setuid bit) comment)

stat function

If built-in functions does not provide what you want you can use stat function. The stat function accepts a filehandle or the name of the file. It returns a 13-element list giving the status info for a file. If name of the file is specified it can be either fully qualified or relative name. If no arguments are supplied $_ is used as a default argument. The function returns an empty list in case of failure. Typically used as follows:
($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,$atime,$mtime,$ctime,$blksize,$blocks)= stat($filename);

Not all fields are supported on all filesystem types as this is a Unix-oriented structure.

Here are the meaning of the fields:

  0 dev      device number of filesystem
  1 ino      inode number
  2 mode     file mode  (type and permissions).
             The result is in octal form and contains five digits
  3 nlink    number of (hard) links to the file
  4 uid      numeric user ID of file's owner
  5 gid      numeric group ID of file's owner
  6 rdev     the device identifier (special files only)
  7 size     total size of file, in bytes
  8 atime    last access time since the epoch
  9 mtime    last modify time since the epoch
 10 ctime    inode change time (NOT creation time!) since the epoch
 11 blksize  preferred block size for file system I/O
 12 blocks   actual number of blocks allocated 

If stat is passed the special filehandle consisting of an underline, no stat is done, but the current contents of the stat structure from the last stat or filetest are returned. Example:

if (-x $file && (($d) = stat(_)) && $d < 0) {
   print "$file is executable NFS file\n";

(This works on machines only for which the device number is negative under NFS.)

In scalar context, stat() returns a Boolean value indicating success or failure, and, if successful, sets the information associated with the special filehandle

If you want to retrieve the time at which the file was last read, written, or had its meta-data (owner, etc) changed, you use the -M, -A, or -C filetest operations as documented in the perlfunc manpage. These retrieve the age of the file (measured against the start-time of your program) in days as a floating point number. To retrieve the ``raw'' time in seconds since the epoch, you would call the stat function, then use

Here's an example:

    $write_secs = (stat($file))[9];
    printf "file %s updated at %s\n", $file,
        scalar localtime($write_secs);

Posix strftime function

Function strftime is essentially POSIX re-implementation of functionality of Unix utility date. See POSIX::strftime :POSIX -

POSIX is part of standard Perl library.

For example

use POSIX qw(strftime);
print strftime '%Y-%m-%d-%H-%M-%S', localtime;

From manual

Convert date and time information to string. Returns the string.


        strftime(fmt, sec, min, hour, mday, mon, year,
                 wday = -1, yday = -1, isdst = -1)

The month (mon), weekday (wday), and yearday (yday) begin at zero, i.e., January is 0, not 1; Sunday is 0, not 1; January 1st is 0, not 1. The year (year) is given in years since 1900, i.e., the year 1995 is 95; the year 2001 is 101. Consult your system's strftime() manpage for details about these and the other arguments.

If you want your code to be portable, your format (fmt) argument should use only the conversion specifiers defined by the ANSI C standard (C89, to play safe). These are aAbBcdHIjmMpSUwWxXyYZ%. But even then, the results of some of the conversion specifiers are non-portable.

For example, the specifiers aAbBcpZ change according to the locale settings of the user, and both how to set locales (the locale names) and what output to expect are non-standard. The specifier c changes according to the timezone settings of the user and the timezone computation rules of the operating system. The Z specifier is notoriously unportable since the names of timezones are non-standard. Sticking to the numeric specifiers is the safest route.

The given arguments are made consistent as though by calling mktime() before calling your system's strftime() function, except that the isdst value is not affected.

The string for Tuesday, December 12, 1995.

        $str = POSIX::strftime( "%A, %B %d, %Y",
                                 0, 0, 0, 12, 11, 95, 2 );
        print "$str\n";

Another more complex example:

use POSIX; 
my $file = "/etc/passwd"; 
my $date = POSIX::strftime(  
                 ( stat $file )[9] 

Here's a reasonably straightforward method using POSIX (which is a core module) and localtime to avoid twiddling with the details of what stat returns (see Edg's answer for why that's a pain):

use POSIX qw/strftime/; 
my @records; 
opendir my $dh, '.' or die "Can't open current directory for reading: $!"; 
while (my $item = readdir $dh) { 
    next unless -f $item && $item !~ m/^\./; 
    push @records, [$item, (stat $item)[9]]; 
for my $record (@records) { 
    print "File: $record->[0]\n"; 
    print "Mtime: " . strftime("%d%m%Y", localtime $record->[1]), "\n"; 

Top Visited
Past week
Past month


Old News ;-)

Recommended Links


Google matched content

Softpanorama Recommended

Top articles


Top articles