#!/usr/local/bin/perl use warnings; use strict; our $VERSION = '1.90'; # Check http://betterthangrep.com/ for updates # These are all our globals. use App::Ack (); MAIN: { if ( $App::Ack::VERSION ne $main::VERSION ) { App::Ack::die( "Program/library version mismatch\n\t$0 is $main::VERSION\n\t$INC{'App/Ack.pm'} is $App::Ack::VERSION" ); } # Do preliminary arg checking; my $env_is_usable = 1; for ( @ARGV ) { last if ( $_ eq '--' ); # Priorities! Get the --thpppt checking out of the way. /^--th[pt]+t+$/ && App::Ack::_thpppt($_); # See if we want to ignore the environment. (Don't tell Al Gore.) if ( $_ eq '--noenv' ) { my @keys = ( 'ACKRC', grep { /^ACK_/ } keys %ENV ); delete @ENV{@keys}; $env_is_usable = 0; } } unshift( @ARGV, App::Ack::read_ackrc() ) if $env_is_usable; App::Ack::load_colors(); if ( exists $ENV{ACK_SWITCHES} ) { App::Ack::warn( 'ACK_SWITCHES is no longer supported. Use ACK_OPTIONS.' ); } if ( !@ARGV ) { App::Ack::show_help(); exit 1; } main(); } sub main { my $opt = App::Ack::get_command_line_options(); $| = 1 if $opt->{flush}; # Unbuffer the output if flush mode if ( App::Ack::input_from_pipe() ) { # We're going into filter mode for ( qw( f g l ) ) { $opt->{$_} and App::Ack::die( "Can't use -$_ when acting as a filter." ); } $opt->{show_filename} = 0; $opt->{regex} = App::Ack::build_regex( defined $opt->{regex} ? $opt->{regex} : shift @ARGV, $opt ); if ( my $nargs = @ARGV ) { my $s = $nargs == 1 ? '' : 's'; App::Ack::warn( "Ignoring $nargs argument$s on the command-line while acting as a filter." ); } my $res = App::Ack::Resource::Basic->new( '-' ); App::Ack::search_resource( $res, $opt ); $res->close(); exit 0; } my $file_matching = $opt->{f} || $opt->{lines}; if ( !$file_matching ) { @ARGV or App::Ack::die( 'No regular expression found.' ); $opt->{regex} = App::Ack::build_regex( defined $opt->{regex} ? $opt->{regex} : shift @ARGV, $opt ); } # check that all regexes do compile fine App::Ack::check_regex( $_ ) for ( $opt->{regex}, $opt->{G} ); my $what = App::Ack::get_starting_points( \@ARGV, $opt ); my $iter = App::Ack::get_iterator( $what, $opt ); App::Ack::filetype_setup(); my $nmatches = 0; App::Ack::set_up_pager( $opt->{pager} ) if defined $opt->{pager}; if ( $opt->{f} ) { App::Ack::print_files( $iter, $opt ); } elsif ( $opt->{l} || $opt->{count} ) { $nmatches = App::Ack::print_files_with_matches( $iter, $opt ); } else { $nmatches = App::Ack::print_matches( $iter, $opt ); } close $App::Ack::fh; exit ($nmatches ? 0 : 1); } =head1 NAME ack - grep-like text finder =head1 SYNOPSIS ack [options] PATTERN [FILE...] ack -f [options] [DIRECTORY...] =head1 DESCRIPTION Ack is designed as a replacement for 99% of the uses of F. Ack searches the named input FILEs (or standard input if no files are named, or the file name - is given) for lines containing a match to the given PATTERN. By default, ack prints the matching lines. Ack can also list files that would be searched, without actually searching them, to let you take advantage of ack's file-type filtering capabilities. =head1 FILE SELECTION I is intelligent about the files it searches. It knows about certain file types, based on both the extension on the file and, in some cases, the contents of the file. These selections can be made with the B<--type> option. With no file selections, I only searches files of types that it recognizes. If you have a file called F, and I doesn't know what a .wango file is, I won't search it. The B<-a> option tells I to select all files, regardless of type. Some files will never be selected by I, even with B<-a>, including: =over 4 =item * Backup files: Files ending with F<~>, or F<#*#> =item * Coredumps: Files matching F =back However, I always searches the files given on the command line, no matter what type. Furthermore, by specifying the B<-u> option all files will be searched. =head1 DIRECTORY SELECTION I descends through the directory tree of the starting directories specified. However, it will ignore the shadow directories used by many version control systems, and the build directories used by the Perl MakeMaker system. You may add or remove a directory from this list with the B<--[no]ignore-dir> option. The option may be repeated to add/remove multiple directories from the ignore list. For a complete list of directories that do not get searched, run F. =head1 WHEN TO USE GREP I trumps I as an everyday tool 99% of the time, but don't throw I away, because there are times you'll still need it. E.g., searching through huge files looking for regexes that can be expressed with I syntax should be quicker with I. If your script or parent program uses I C<--quiet> or C<--silent> or needs exit 2 on IO error, use I. =head1 OPTIONS =over 4 =item B<-a>, B<--all> Operate on all files, regardless of type (but still skip directories like F, F, etc.) =item B<-A I>, B<--after-context=I> Print I lines of trailing context after matching lines. =item B<-B I>, B<--before-context=I> Print I lines of leading context before matching lines. =item B<-C [I]>, B<--context[=I]> Print I lines (default 2) of context around matching lines. =item B<-c>, B<--count> Suppress normal output; instead print a count of matching lines for each input file. If B<-l> is in effect, it will only show the number of lines for each file that has lines matching. Without B<-l>, some line counts may be zeroes. =item B<--color>, B<--nocolor> B<--color> highlights the matching text. B<--nocolor> supresses the color. This is on by default unless the output is redirected. On Windows, this option is off by default unless the L module is installed or the C environment variable is used. =item B<--color-filename=I> Sets the color to be used for filenames. =item B<--color-match=I> Sets the color to be used for matches. =item B<--column> Show the column number of the first match. This is helpful for editors that can place your cursor at a given position. =item B<--env>, B<--noenv> B<--noenv> disables all environment processing. No F<.ackrc> is read and all environment variables are ignored. By default, F considers F<.ackrc> and settings in the environment. =item B<--flush> B<--flush> flushes output immediately. This is off by default unless ack is running interactively (when output goes to a pipe or file). =item B<-f> Only print the files that would be searched, without actually doing any searching. PATTERN must not be specified, or it will be taken as a path to search. =item B<--follow>, B<--nofollow> Follow or don't follow symlinks, other than whatever starting files or directories were specified on the command line. This is off by default. =item B<-G I> Only paths matching I are included in the search. The entire path and filename are matched against I, and I is a Perl regular expression, not a shell glob. The options B<-i>, B<-w>, B<-v>, and B<-Q> do not apply to this I. =item B<-g I> Print files where the relative path + filename matches I. This option is a convenience shortcut for B<-f> B<-G I>. The options B<-i>, B<-w>, B<-v>, and B<-Q> do not apply to this I. =item B<--group>, B<--nogroup> B<--group> groups matches by file name with. This is the default when used interactively. B<--nogroup> prints one result per line, like grep. This is the default when output is redirected. =item B<-H>, B<--with-filename> Print the filename for each match. =item B<-h>, B<--no-filename> Suppress the prefixing of filenames on output when multiple files are searched. =item B<--help> Print a short help statement. =item B<-i>, B<--ignore-case> Ignore case in the search strings. This applies only to the PATTERN, not to the regexes given for the B<-g> and B<-G> options. =item B<--[no]ignore-dir=DIRNAME> Ignore directory (as CVS, .svn, etc are ignored). May be used multiple times to ignore multiple directories. For example, mason users may wish to include B<--ignore-dir=data>. The B<--noignore-dir> option allows users to search directories which would normally be ignored (perhaps to research the contents of F<.svn/props> directories). =item B<--line=I> Only print line I of each file. Multiple lines can be given with multiple B<--line> options or as a comma separated list (B<--line=3,5,7>). B<--line=4-7> also works. The lines are always output in ascending order, no matter the order given on the command line. =item B<-l>, B<--files-with-matches> Only print the filenames of matching files, instead of the matching text. =item B<-L>, B<--files-without-matches> Only print the filenames of files that do I match. This is equivalent to specifying B<-l> and B<-v>. =item B<--match I> Specify the I explicitly. This is helpful if you don't want to put the regex as your first argument, e.g. when executing multiple searches over the same set of files. # search for foo and bar in given files ack file1 t/file* --match foo ack file1 t/file* --match bar =item B<-m=I>, B<--max-count=I> Stop reading a file after I matches. =item B<--man> Print this manual page. =item B<-n> No descending into subdirectories. =item B<-o> Show only the part of each line matching PATTERN (turns off text highlighting) =item B<--output=I> Output the evaluation of I for each line (turns off text highlighting) =item B<--pager=I> Direct ack's output through I. This can also be specified via the C and C environment variables. Using --pager does not suppress grouping and coloring like piping output on the command-line does. =item B<--passthru> Prints all lines, whether or not they match the expression. Highlighting will still work, though, so it can be used to highlight matches while still seeing the entire file, as in: # Watch a log file, and highlight a certain IP address $ tail -f ~/access.log | ack --passthru 123.45.67.89 =item B<--print0> Only works in conjunction with -f, -g, -l or -c (filename output). The filenames are output separated with a null byte instead of the usual newline. This is helpful when dealing with filenames that contain whitespace, e.g. # remove all files of type html ack -f --html --print0 | xargs -0 rm -f =item B<-Q>, B<--literal> Quote all metacharacters in PATTERN, it is treated as a literal. This applies only to the PATTERN, not to the regexes given for the B<-g> and B<-G> options. =item B<--smart-case>, B<--no-smart-case> Ignore case in the search strings if PATTERN contains no uppercase characters. This is similar to C in vim. This option is off by default. B<-i> always overrides this option. This applies only to the PATTERN, not to the regexes given for the B<-g> and B<-G> options. =item B<--sort-files> Sorts the found files lexically. Use this if you want your file listings to be deterministic between runs of I. =item B<--thpppt> Display the all-important Bill The Cat logo. Note that the exact spelling of B<--thpppppt> is not important. It's checked against a regular expression. =item B<--type=TYPE>, B<--type=noTYPE> Specify the types of files to include or exclude from a search. TYPE is a filetype, like I or I. B<--type=perl> can also be specified as B<--perl>, and B<--type=noperl> can be done as B<--noperl>. If a file is of both type "foo" and "bar", specifying --foo and --nobar will exclude the file, because an exclusion takes precedence over an inclusion. Type specifications can be repeated and are ORed together. See I for a list of valid types. =item B<--type-add I=I<.EXTENSION>[,I<.EXT2>[,...]]> Files with the given EXTENSION(s) are recognized as being of (the existing) type TYPE. See also L. =item B<--type-set I=I<.EXTENSION>[,I<.EXT2>[,...]]> Files with the given EXTENSION(s) are recognized as being of type TYPE. This replaces an existing definition for type TYPE. See also L. =item B<-u>, B<--unrestricted> All files and directories (including blib/, core.*, ...) are searched, nothing is skipped. When both B<-u> and B<--ignore-dir> are used, the B<--ignore-dir> option has no effect. =item B<-v>, B<--invert-match> Invert match: select non-matching lines This applies only to the PATTERN, not to the regexes given for the B<-g> and B<-G> options. =item B<--version> Display version and copyright information. =item B<-w>, B<--word-regexp> Force PATTERN to match only whole words. The PATTERN is wrapped with C<\b> metacharacters. This applies only to the PATTERN, not to the regexes given for the B<-g> and B<-G> options. =item B<-1> Stops after reporting first match of any kind. This is different from B<--max-count=1> or B<-m1>, where only one match per file is shown. Also, B<-1> works with B<-f> and B<-g>, where B<-m> does not. =back =head1 THE .ackrc FILE The F<.ackrc> file contains command-line options that are prepended to the command line before processing. Multiple options may live on multiple lines. Lines beginning with a # are ignored. A F<.ackrc> might look like this: # Always sort the files --sort-files # Always color, even if piping to a another program --color # Use "less -r" as my pager --pager=less -r Note that arguments with spaces in them do not need to be quoted, as they are not interpreted by the shell. Basically, each I in the F<.ackrc> file is interpreted as one element of C<@ARGV>. F looks in your home directory for the F<.ackrc>. You can specify another location with the F variable, below. If B<--noenv> is specified on the command line, the F<.ackrc> file is ignored. =head1 Defining your own types ack allows you to define your own types in addition to the predefined types. This is done with command line options that are best put into an F<.ackrc> file - then you do not have to define your types over and over again. In the following examples the options will always be shown on one command line so that they can be easily copy & pasted. I searches for foo in all perl files. I tells you, that perl files are files ending in .pl, .pm, .pod or .t. So what if you would like to include .xs files as well when searching for --perl files? I does this for you. B<--type-add> appends additional extensions to an existing type. If you want to define a new type, or completely redefine an existing type, then use B<--type-set>. I defines the type I to include files with the extensions .e or .eiffel. So to search for all eiffel files containing the word Bertrand use I. As usual, you can also write B<--type=eiffel> instead of B<--eiffel>. Negation also works, so B<--noeiffel> excludes all eiffel files from a search. Redefining also works: I and I<.xs> files no longer belong to the type I. When defining your own types in the F<.ackrc> file you have to use the following: --type-set=eiffel=.e,.eiffel or writing on separate lines --type-set eiffel=.e,.eiffel The following does B work in the F<.ackrc> file: --type-set eiffel=.e,.eiffel In order to see all currently defined types, use I<--help types>, e.g. I Restrictions: =over 4 =item The types 'skipped', 'make', 'binary' and 'text' are considered "builtin" and cannot be altered. =item The shebang line recognition of the types 'perl', 'ruby', 'php', 'python', 'shell' and 'xml' cannot be redefined by I<--type-set>, it is always active. However, the shebang line is only examined for files where the extension is not recognised. Therefore it is possible to say I and only find your shiny new I<.perl> files (and all files with unrecognized extension and perl on the shebang line). =back =head1 ENVIRONMENT VARIABLES For commonly-used ack options, environment variables can make life much easier. These variables are ignored if B<--noenv> is specified on the command line. =over 4 =item ACKRC Specifies the location of the F<.ackrc> file. If this file doesn't exist, F looks in the default location. =item ACK_OPTIONS This variable specifies default options to be placed in front of any explicit options on the command line. =item ACK_COLOR_FILENAME Specifies the color of the filename when it's printed in B<--group> mode. By default, it's "bold green". The recognized attributes are clear, reset, dark, bold, underline, underscore, blink, reverse, concealed black, red, green, yellow, blue, magenta, on_black, on_red, on_green, on_yellow, on_blue, on_magenta, on_cyan, and on_white. Case is not significant. Underline and underscore are equivalent, as are clear and reset. The color alone sets the foreground color, and on_color sets the background color. This option can also be set with B<--color-filename>. =item ACK_COLOR_MATCH Specifies the color of the matching text when printed in B<--color> mode. By default, it's "black on_yellow". This option can also be set with B<--color-match>. See B for the color specifications. =item ACK_PAGER Specifies a pager program, such as C, C or C, to which ack will send its output. Using C does not suppress grouping and coloring like piping output on the command-line does, except that on Windows ack will assume that C does not support color. C overrides C if both are specified. =item ACK_PAGER_COLOR Specifies a pager program that understands ANSI color sequences. Using C does not suppress grouping and coloring like piping output on the command-line does. If you are not on Windows, you never need to use C. =back =head1 ACK & OTHER TOOLS =head2 Vim integration F integrates easily with the Vim text editor. Set this in your F<.vimrc> to use F instead of F: set grepprg=ack\ -a That examples uses C<-a> to search through all files, but you may use other default flags. Now you can search with F and easily step through the results in Vim: :grep Dumper perllib =head2 Emacs integration Phil Jackson put together an F extension that "provides a simple compilation mode ... has the ability to guess what files you want to search for based on the major-mode." L =head2 TextMate integration Pedro Melo is a TextMate user who writes "I spend my day mostly inside TextMate, and the built-in find-in-project sucks with large projects. So I hacked a TextMate command that was using find + grep to use ack. The result is the Search in Project with ack, and you can find it here: L" =head2 Shell and Return Code For greater compatibility with I, I in normal use returns shell return or exit code of 0 only if something is found and 1 if no match is found. (Shell exit code 1 is C<$?=256> in perl with C or backticks.) The I code 2 for errors is not used. 0 is returned if C<-f> or C<-g> are specified, irrespective of number of files found. =cut =head1 DEBUGGING ACK PROBLEMS If ack gives you output you're not expecting, start with a few simple steps. =head2 Use B<--noenv> Your environment variables and F<.ackrc> may be doing things you're not expecting, or forgotten you specified. Use B<--noenv> to ignore your environment and F<.ackrc>. =head2 Use B<-f> to see what files you're scanning The reason I created B<-f> in the first place was as a debugging tool. If ack is not finding matches you think it should find, run F to see what files are being checked. =head1 TIPS =head2 Use the F<.ackrc> file. The F<.ackrc> is the place to put all your options you use most of the time but don't want to remember. Put all your --type-add and --type-set definitions in it. If you like --smart-case, set it there, too. I also set --sort-files there. =head2 Use F<-f> for working with big codesets Ack does more than search files. C will create a list of all the Perl files in a tree, ideal for sending into F. For example: # Change all "this" to "that" in all Perl files in a tree. ack -f --perl | xargs perl -p -i -e's/this/that/g' or if you prefer: perl -p -i -e's/this/thatg/' $(ack -f --perl) =head2 Use F<-Q> when in doubt about metacharacters If you're searching for something with a regular expression metacharacter, most often a period in a filename or IP address, add the -Q to avoid false positives without all the backslashing. See the following example for more... =head2 Use ack to watch log files Here's one I used the other day to find trouble spots for a website visitor. The user had a problem loading F, so I took the access log and scanned it with ack twice. ack -Q aa.bb.cc.dd /path/to/access.log | ack -Q -B5 troublesome.gif The first ack finds only the lines in the Apache log for the given IP. The second finds the match on my troublesome GIF, and shows the previous five lines from the log in each case. =head2 Share your knowledge Join the ack-users mailing list. Send me your tips and I may add them here. =head1 FAQ =head2 Wouldn't it be great if F did search & replace? No, ack will always be read-only. Perl has a perfectly good way to do search & replace in files, using the C<-i>, C<-p> and C<-n> switches. You can certainly use ack to select your files to update. For example, to change all "foo" to "bar" in all PHP files, you can do this form the Unix shell: $ perl -i -p -e's/foo/bar/g' $(ack -f --php) =head1 AUTHOR Andy Lester, C<< >> =head1 BUGS Please report any bugs or feature requests to the issues list at Google Code: L =head1 ENHANCEMENTS All enhancement requests MUST first be posted to the ack-users mailing list at L. I will not consider a request without it first getting seen by other ack users. There is a list of enhancements I want to make to F in the ack issues list at Google Code: L Patches are always welcome, but patches with tests get the most attention. =head1 SUPPORT Support for and information about F can be found at: =over 4 =item * The ack homepage L =item * The ack issues list at Google Code L =item * AnnoCPAN: Annotated CPAN documentation L =item * CPAN Ratings L =item * Search CPAN L =item * Subversion repository L =back =head1 ACKNOWLEDGEMENTS How appropriate to have Inowledgements! Thanks to everyone who has contributed to ack in any way, including Ryan Niebur, Kent Fredric, Mike Morearty, Ingmar Vanhassel, Eric Van Dewoestine, Sitaram Chamarty, Adam James, Richard Carlsson, Pedro Melo, AJ Schuster, Phil Jackson, Michael Schwern, Jan Dubois, Christopher J. Madsen, Matthew Wickline, David Dyck, Jason Porritt, Jjgod Jiang, Thomas Klausner, Uri Guttman, Peter Lewis, Kevin Riggle, Ori Avtalion, Torsten Blix, Nigel Metheringham, GEbor SzabE, Tod Hagan, Michael Hendricks, Evar ArnfjErE Bjarmason, Piers Cawley, Stephen Steneker, Elias Lutfallah, Mark Leighton Fisher, Matt Diephouse, Christian Jaeger, Bill Sully, Bill Ricker, David Golden, Nilson Santos F. Jr, Elliot Shank, Merijn Broeren, Uwe Voelker, Rick Scott, Ask BjErn Hansen, Jerry Gay, Will Coleda, Mike O'Regan, Slaven ReziE<0x107>, Mark Stosberg, David Alan Pisoni, Adriano Ferreira, James Keenan, Leland Johnson, Ricardo Signes and Pete Krawczyk. =head1 COPYRIGHT & LICENSE Copyright 2005-2009 Andy Lester. This program is free software; you can redistribute it and/or modify it under the terms of either: =over 4 =item * the GNU General Public License as published by the Free Software Foundation; either version 1, or (at your option) any later version, or =item * the Artistic License version 2.0. =back =cut