File::Listing::Ftpcopy - parse directory listing using ftpparse from ftpcopy
version 0.09
# traditional interface
use File::Listing::Ftpcopy qw(parse_dir);
$ENV{LANG} = "C"; # dates in non-English locales not supported
for (parse_dir(`ls -l`)) {
($name, $type, $size, $mtime, $mode) = @$_;
next if $type ne 'f'; # plain file
#...
}
# directory listing can also be read from a file
open(LISTING, "zcat ls-lR.gz|");
$dir = parse_dir(\*LISTING, '+0000');
# ftpparse interface
use 5.010;
use Parse::Listing::Ftpcopy qw( ftpparse SIZE_UNKNOWN );
foreach my $line (`ls -l`)
{
chomp $line;
my $h = ftpparse($line);
next unless defined $h;
say "name : ", $h->{name}
say "size : ", $h->{size} if $h->{sizetype} != SIZE_UNKNOWN;
}
This module provides functions for parsing the output of directory listings of the sort generated by an FTP server. It is intended to provide a mostly drop in replacement for the parse_dir function from File::Listing (although see CAVEATS below) that uses the ftpparse function from ftpcopy (see URL in the SEE ALSO section below) instead of the Perl implementation provided by File::Listing. ftpparse is written in C, and so may or may not be faster, although probably unnoticeable unless you are parsing a recursive directory listing of a large system, and if you have to do that maybe you should rethink your approach anyway.
Where this module may come in handy over File::Listing is that it understands the output from a different subset of systems. For my purposes, ftpparse understands VMS listings, on the other hand, File::Listing understands Apache listings, neither module understands both. If you know ahead of time which system you are going to be dealing with you can use either this module or File::Listing, or if you do not know ahead of time, you can try each and use the results from which ever one actually works (or works best).
This module supports the following file listings:
This module also provides a direct interface to the ftpparse function as well.
my $dir = parse_dir( $listing ); my $dir = parse_dir( $listing, $time_zone ); my $dir = parse_dir( $listing, $time_zone, $type ); my $dir = parse_dir( $listing, $time_zone, $type, $error);
The first argument ($listing) is the directory listing to parse. It can be a scalar, a reference to an array of directory lines or a glob representing a filehandle to read the directory listing from.
The second argument ($time_zone) is used when parsing the time stamps in the listing. If the value is undefined, then the local time zone is assumed.
The third argument ($type) is ignored, but included here for compatibility with File::Listing.
The fourth argument ($error) specifies how unparseable lines should be treated. Values can be 'ignore', 'warn' or a code reference. 'warn' means that the perl warn function will be called. If a code reference is passed, then this routine will be called and the return value from it will be incorporated in the listing. The default is 'ignore'.
For each file found in the listing it returns an array ref
foreach my $fileinfo (parse_dir($listing))
{
($name, $type, $size, $mtime, $mode) = @$fileinfo;
# ...
}
The first element ($name) is the name of the file.
The second element ($size) is the size of the file.
The third element ($mtime) is the modification time of the file.
The forth element ($mode) is supposed to be the permission bits of the file, but ftpparse ignores the permission information so this is always undef.
Any field which could not be determined by the algorithm will be undef.
my $hash = ftpparse( $line );
Parse a single line from an FTP listing. Returns a hash ref of information about the file found in that line, or undef if no file information was found about the file.
Here is the information found in the hash ref:
The name of the file
The size of the file
What format the size refers to, one of
The size could not be determined (size is set to 0)
The size assumes a binary transfer (TYPE I)
The size assumes an ASCII transfer (TYPE A)
This is currently unused, but could theoretically be used in the future.
0 if the file is definitely not a directory. 1 otherwise.
0 if the file is definitely not a regular file or symlink, which can be retrieved. 1 otherwise.
If the file is a symlink, then this contains the target name (or at least part of the target name) of the symlink.
The detected format of the listing, one of:
The modification time as the number of non-leap seconds since the epoch.
Information about the mtime field, one of:
modification time is undetermined
time is correct for the current time zone
Time zone and seconds are unknown
Time zone and time of day are unknown
time zone is unknown
Because ftpparse is written in C and the bindings to it are in XS, so a C compiler is required.
Because ftpparse does not parse out permission information, the mode field is always undef.
ftpparse uses a different algorithm, and uses a different interface under the covers, it recognizes a different subset of system listings, and may interpret them differently so this module is not, and does not pretend to be 100% compatible with File::Listing.
Internally ftpparse assumes GMT if it can't determine the time zone from the listing, and doesn't provide an interface for specifying another time zone if you do happen to know what the remote server's time zone is. File::Listing assumes the listing is for the local time zone unless you specify one through the calling interface. In order to get the expected behavior for parse_dir, this module jumps through some extra hoops to support the File::Listing interface. To avoid these hoops use the ftpparse interface instead.
The ftpparse function from ftpcopy is based on ftpparse by Daniel J. Bernsteins. Bernsteins' version is incompatible with GPL, and possibly other open source licenses. The ftpparse function from ftpcopy was written by Uwe Ohse and is mostly public domain, but there was one dependent C source file which was licensed under GPL 2, so I am licensing this whole distribution under GPL 2.
C code: Uwe Ohse
XS and Perl code: Graham Ollis <plicease@cpan.org>
Copyright 2002 by Uwe Ohse
Copyright 2013 by Graham Ollis
This is free software, licensed under the GNU General Public License, Version 2, June 1991
Some source files marked as public domain are in the public domain.