View file File name : showtable Content :#!/usr/bin/env perl # showtable - read data and show it # # Copyright (C) 1996-2013 Alan K. Stebbens <aks@stebbbens.org> # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. # # # showtable [options] [file] # # This script acts as a filter, reading data in columns and displaying # the data in a nicely formatted output. The output can be either # a list style, simple table, boxed table, or HTML-formatted table. # # The input styles accepted are: tab-delimited, comma-delimited, # list-style, simple-style tables, and boxed tables. # # The default is tab-delimited data # # Use -d<char> to set the field separation character. Eg: -d, # # Other options: -table, -html, -box, -list # # If no file is given, STDIN is used by default. # # $Id$ # $Source$ # # Author: Alan K. Stebbens <aks@stebbens.org> # ($DIR,$PROG) = $0 =~ m=^(.*/)?([^/]+)$=; chop($DIR = `pwd`) unless defined($DIR); $DIR =~ s=/$==; $VERSION = '4.5'; unshift(@INC,'.'); unshift(@INC,'blib/lib') if -d 'blib/lib'; # for testing use Data::ShowTable; sub ShowData; sub ReadTableInput; sub ReadListInput; sub NextRow; sub max; sub err { my $fmt = shift; $fmt .= "\n" unless $fmt =~ /\n$/; printf STDERR $fmt,@_; } sub usage { err @_ if $#_ >= 0; err <<EOF; usage: $PROG [options] [file] Read the file, or STDIN by default, and produce a nicely formatted listing of columnar data. By default, columns of data should be separated by tabs, and is assumed to have no titles. Lines of all spaces and any of the characters "-+_=" will be ignored as separator lines, unless -nodashes is given. Options affecting input: -br(eak)=STR set the inter-column break character to "STR". The default is "\t" (a tab). -dSTR set the delimiter string (same as "-break=STR"). -nod(ashes) Do not look for lines of separators, such as dashes, equal signs, or underlines (if they occur, they will be part of the "data". -t(itles)[=NN] Treat the first NN rows of data as column titles; multiple words in the column titles may wrap vertically. If NN is omitted, it defaults to 1. If no -titles is given at all, NN defaults to 0. -st(rip) Strip blanks from around the column values (implied with $InputType eq 'box'). -nos(trip) Don't strip blanks around the column values. -i(nput)=TYPE Set the input type; currently TYPE can be "list", "box", "table", or "simple". Default is "box". The options controlling the output are: -s(imple) use a simple table format (without wrapped columns) -t(able) use a table format (handles wrapped column values) -l(ist) use a list style format -b(ox) use a "boxed" style table -ht(ml) use HTML-formating -noh(eaders) do not output any headers on the tables; -titles=0 implies this option. -ti(tles)=name1,name2,...,nameN Define the output column titles explicitly; "name1" cannot begin with a number. Implies "-titles=0". -w(idth)=NN Define the maximum width of the table; if this not given, the maximum is the sum of all given column widths. -cw(idths)=w1,w2,...,wN Define individual column widths; the last width applies to any remaining columns. Each column may be given as an integer, representing a character count, or a fraction or percentage, representing percentages. -fN1,N2... select fields to display by index; N-M is a range of fields. -fields=Name1, Name2, ... select fields to display by names. -h(elp) Display this string. HTML-only options (their usage implies -html): -noe(scape) Do not perform HTML escape sequences; this will allow data with embedded HTML to be displayed properly. Implies the -html option. -attr(ibutes)='attr1 attr2 ...' Define the table attributes, which get inserted into the <TABLE> token. -t(itle)f(ormats)=fmt1;fmt2;...;fmtN Define the HTML formats for the titles. Implies -html. -d(ata)f(ormats)=fmt1;fmt2;...;fmtN Define the HTML formats for the data values in each column. Implies -html. -url(s)=col1=url1,col2=url2,... Define the URL mappings for given column names. Multiple -url options may be given. EOF exit; } $break_str = "\t"; # tab for default break str $Titles = 0; # assume no input titles $#Titles = -1; # assume no output titles @Fields = (); # no explicit field selections %URLs = (); # no URLs initially $InputType = ''; $Strip_Spaces = ''; $NoStrip_Spaces = ''; $Show_Mode = 'Box'; # by default $Width = 0; # no width limit by default $nofile = 1; while ($_ = shift(@ARGV)) { if (/^-d(\W+)?$/) { # -dSTR $break_str = $1 || shift || usage "$PROG: Missing argument to -d\n"; next; } if (/^-f(\d.*)/) { # -fN1,N2,... $value = $1; # convert possible range spec $value =~ s/\b(\d+)-(\d+)\b/join(',',$1..$2)/e; @Fields = split(/[:, ]/,$value); next; } usage if !index('-help',$_); if (!index('-strip',$_)) { # -strip $Strip_Spaces = 1; $NoStrip_Spaces = ''; } elsif (!index('-nostrip',$_)) { # -nostrip $NoStrip_Spaces = 1; $Strip_Spaces = ''; } elsif (!index('-table',$_) || !index('-simple',$_)) { # -table, -simple $Show_Mode = 'Table'; } elsif (!index('-list', $_)) { # -list $Show_Mode = 'List'; } elsif (!index('-box',$_)) { # -box $Show_Mode = 'Box'; $Strip_Spaces = 1 unless $NoStrip_Spaces; } elsif (!index('-html',$_)) { # -html $Show_Mode = 'HTML'; } elsif (!index('-noescape',$_)) { # -noescape $No_Escape = 1; $Show_Mode = 'HTML'; } elsif (!index('-noheaders',$_)) { # -noheaders $No_Header = 1; } elsif (!index('-nodashes',$_)) { # -nodashes $No_Dashes = 1; } elsif (/^(-\w+)(?:[:=](.*))?/) { # -OPTION=VALUE ($_, $value) = ($1, $2); if (!index('-titles',$_)) { # -titles=NN, -titles=NAME,.. if ($value =~ /^\d/ || $value eq '') { # -titles=NN? $Titles = $value ne '' ? $value : 1; } else { # -titles=name,name2,...,nameN @Titles = split(/[:, ]/,$value); $Titles = 0; # no titles in the datastream } } elsif (!index('-input',$_)) { # -input=TYPE? $value = lc($value); if (!index('box',$value)) { # -input=box? $InputType = 'box'; $Titles = 1 unless $Titles >= 1; } elsif (!index('list',$value)) {# -input=list? $InputType = 'list'; } elsif (!index('table', $value)) { # -input=table? $InputType = 'table'; $Titles = 1 unless $Titles >= 1; } elsif (!index('simple', $value)) {# -input=simple? $InputType = 'simple'; } else { usage "$PROG: bad -input TYPE"; } } elsif (!index('-fields',$_)) { # -fields=Name1, Name2, ... # convert range spec $value =~ s/\b(\d+)-(\d+)\b/join(' ',$1..$2)/e; @Fields = split(/[:, ]+/,$value); } elsif (!index('-break',$_)) { # -break=STR $break_str = $value ne '' ? $value : "\t"; if ($break_str =~ /\\/) { # any escape sequences? $break_str =~ s/"/\\"/g; # yes, escape any quotes $break_str = eval('"'.$break_str.'"'); # evaluate a string } } elsif (!index('-width',$_)) { # -width=NUM $value =~ /^\d+$/ or usage "$PROG: width must be a number."; $Width = 0 + $value; } elsif (!index('-cwidths',$_)) { # -cwidths=N1,N2,...,NM @Cwidths = split(/[,: ]/,$value); if (grep(!/^(\.?\d+|\d+%)?$/,@Cwidths)) { usage "$PROG: bad format for -cwidths."; } } elsif (!index('-titleformats',$_) || !index('-tformats',$_)) { my @tformats = split(/[;|]/,$value); # get the title formats foreach (@tformats) { next if /^<?($HTML_Elements)/i; # HTML element ok? usage "$PROG: bad HTML format element: '%s'", $1; } push(@Tformats,@tformats); # allow multiple options $Show_Mode = 'HTML'; } elsif (!index('-dataformats',$_) || !index('-dformats',$_)) { my @dformats = split(/[;|]/,$value); # get the title formats foreach (@dformats) { next if /^<?(?:$HTML_Elements)/i; # HTML element ok? usage "$PROG: bad HTML format element '%s'", $_; } push(@Dformats,@dformats); # allow multiple options $Show_Mode = 'HTML'; } elsif (!index('-urls',$_)) { # parse COL1=URL1,COL2=URL2,... # or: COL1:URL1,COL2:URL2,... my @urldefs = split(/[,|]/, $value); # get the different URLs my $urldef; foreach $urldef (@urldefs) { my ($col, $url) = split(/[=:]/,$urldef,2); if ($col ne '' && $url ne '') { # both defined? $URLs{$col} = $url; # save the mapping } elsif ($col ne '' || $url ne '') { usage "$PROG: bad URL format element: '%s'", $urldef; } } $Show_Mode = 'HTML'; # make sure } elsif (!index('-attributes',$_)) { $table_attributes .= ' '.$value; # build up the list $Show_Mode = 'HTML'; } else { usage "$PROG: Unknown option: $_"; } } elsif (/^-/) { usage "$PROG: Unknown option: $_"; } elsif (-f) { # does the file exist? open(STDIN,"<$_") || die "$PROG: Can't open $_ for input: $!\n"; ShowData; close(STDIN); $nofile = 0; } else { err "File $_ does not exist."; exit 1; } } ShowData if $nofile; exit; # $type = guess_type $value sub guess_type { local $_ = shift; &PlainText if $Show_Mode eq 'HTML'; # remove HTML effects if -html # so they aren't part of the typing s/^\s+//; s/\s+$//; # trim leading & trailing blanks /^$/ && return 'null'; # null type /^([01tf]|yes|no|on|off)$/i && return 'bool'; # 0, 1, yes, no, on, off, t, f m=^\d{2,4}[-/.]\d{1,2}[-/.]\d{1,2}$= && return 'date'; # mm/dd/yy, yy/mm/dd, yyyy.mm.dd, yyyy-mm-dd m=^\d{1,2}[-/.]\d{1,2}[-/.]\d{2,4}$= && return 'date'; # dd.mm.yyyy, d.m.yy, dd-mm-yy /^\w{3,9} \d{1,2}, \d{4}/ && return 'date'; # mmmm dd, yyyy /^\d{1,2}[- ]\w{3}[- ]\d{2,4}$/ && return 'date'; # dd mmm yyyy /^\d\d:\d\d(:\d\d)?$/ && return 'time'; # hh:mm:ss /^[-+]?\d+$/ && return 'int'; # +-nnnn /^[-+]?[\d.]+(E[+-]?\d+)?$/ && return 'real'; # +-nnn.nnn(E+-nn) /^[-+]?[\d.]\%$/ && return 'pct'; # nn%, nn.n%, -nn.nnn% /^-?\(?\$[ \d,.]+\)?(?: *CR)?$/ && return 'money'; # $ nn,nnn.nn or ($ nnn,nnn.nn) /^-?\(?[\d]+,[\d,]+(\.(\d\d)?)?\)?$/&& return 'money'; # nnn,nnn.nn or ( nnn,nnn.nn) /^['"]|["']$/ && return 'string'; # "xxxx" /^\w+$/ && return 'symbol'; /\n/ && return 'text'; return 'char'; } # Do type conversion sub new_type { my @types = @_; my $type1 = $types[0]; my $type2 = $types[1]; # trivial conversions $type1 eq '' && return $type2; $type2 eq '' && return $type1; $type1 eq $type2 && return $type1; # handle null types $type1 eq 'null' && return $type2; $type2 eq 'null' && return $type1; # These types supercede others grep(/text/i, @types) && return 'text'; grep(/string/i, @types) && return 'string'; grep(/char/i, @types) && return 'char'; # Now do finer-grain conversions grep(/money/i, @types) && return 'money'; grep(/pct/i, @types) && return 'pct'; grep(/real/i, @types) && return 'real'; grep(/int/i, @types) && return 'int'; grep(/symbol/i, @types) && return 'symbol'; grep(/bool/i, @types) && return 'bool'; $type1; # huh? } sub ShowData { @Data = (); # reset titles if they are coming from the data stream #@Titles = () if $Titles > 0; @Types = (); @Widths = (); my ($titles, $data, $maxcols); if ($InputType eq 'list') { ($titles, $data, $maxcols) = ReadListInput; } else { ($titles, $data, $maxcols) = ReadTableInput; } my @titles = @$titles; @Data = @$data; # Make the fields selection if (@Fields) { # build a field index list my ($f,$fx); my %fields = (); my @fields = (); # an index list @fields{@titles} = (0..$#titles); # generate indexes foreach $f (@Fields) { if ($f =~ /^\d+$/) { # a numeric index? $f--; # convert to origin zero push(@fields,$f) if $f <= $maxcols && $f >= 0; next; } # Lookup a field name and use its index if (($fx = $fields{$f}) ne '') { push(@fields,$fx); } } # now, given the field index array, splice each row array according # to the field indexes my $r; for ($r = 0; $r <= $#Data; $r++) { # foreach row of data.. $row = $Data[$r]; # get the row @$row = @$row[@fields]; # slice out the desired fields } $maxcols = $#fields; # set the other vars @titles = @titles[@fields]; # slice out the titles } # Now, map external titles onto the titles array, allowing # user-given titles to override the built-in titles (if any). if ($No_Header) { # if no headers, no titles @titles = (); } elsif ($#Titles >= 0) { # map @Titles onto @titles my ($c,$t); for ($c = 0; $c <= $maxcols; $c++) { if (($t = $Titles[$c]) ne '') { # user-given title $titles[$c] = $t; # overrides built-in title } elsif ($titles[$c] eq '') { # otherwise, use default title $titles[$c] = sprintf("Field_%d", $c + 1); } } } # Scan the column widths, where none are given, and if a total width # was given, calculate a fraction. my $cols_left = length($Width) ? $Width : # maximum columns left $Show_Mode ne 'HTML' ? $ENV{'COLUMNS'} : 0; if ($cols_left > 0) { # cols == 0 means no limit # We must reduce the available columns by the formatting requirements # of the display format $cols_left -= $maxcols * 3 + 2 if $Show_Mode eq 'Box'; $cols_left -= $maxcols * 2 - 1 if $Show_Mode eq 'Table'; my ($c, $w, $cols); for ($c = 0; $c <= $maxcols; $c++) { # for each column.. $w = $c <= $#Cwidths ? $Cwidths[$c] : $Cwidths[$#Cwidths]; if ($w ne '') { # is there a value? # If the width is a fraction or percentage... if ($w =~ /[.%]/ || ($w > 0 && $w < 1.0)) { $w = $1 / 100.0 if $w =~ /^(\d+)%/; # convert NN% to fraction $w = int($cols_left * $w); # convert fraction to width } $w = 1 unless $w > 0; # guarantee 1 col min $cols_left -= $w; # keep an accounting $cols_left = 0 if $cols_left < 0;# keep sane $Widths[$c] = $w; # set the column width } } } my ($c, $r, $val, $row, $type, $newtype); # some vars # Now let's analyze the data for ($c = 0; $c <= $maxcols; $c++) { # Scan each column $type = ''; # assume no default for ($r = 0; $r <= $#Data; $r++) { # scan each row $row = $Data[$r]; # get the row data if (defined($row->[$c])) { $val = $row->[$c]; # get the column data } else { $row->[$c] = ''; } $newtype = guess_type $val; $type = new_type $type, $newtype; } $Types[$c] = $type; } local $theRow; my $theRowSub = sub { &ShowRow( $_[0], \$theRow, \@Data ); }; ShowTable { show_mode => $Show_Mode, max_width => $Width > 0 ? $Width : '', titles => \@titles, types => \@Types, widths => \@Widths, row_sub => $theRowSub, no_escape => $No_Escape, table_attrs => $table_attributes, tformats => \@Tformats, dformats => \@Dformats, url_keys => \%URLs, }; } # ($titles, $data, $maxcols) = ReadTableInput # # $titles is a ref to an array of titles # $data is a ref to an array of rows (arrays of values) # $maxcols is the maximum columns index in any of the rows. # The global $Titles determines how many titles rows there are. # Unless $NoDashes is set, this can be determined imperically # if there are titles and a separator row, or explicitly with # -titles=NN. sub ReadTableInput { my @data; my @titles; my $data; if ($Titles > 0) { @titles = @{&NextRow}; # if there are multiple lines of titles, combine them my ($limit, $r, $c); for ($r = 1; $r < $Titles; $r++) { $data = NextRow; # get the next row of data $limit = max $#titles, $#$data; for ($c = 0; $c <= $limit; $c++) { $titles[$c] .= ' '.$data->[$c]; # combine titles } } foreach (@titles) { # clean up the names s/^\s+//; # trim leading & s/\s+$//; # trailing blanks s/\s+/ /g; # & squeeze middle blanks } } else { $#titles = -1; # no titles at all } my $maxcols = 0; while ($data = NextRow) { push(@data, $data); # collect the data $maxcols = $#$data if $#$data > $maxcols; } (\@titles, \@data, $maxcols); } # Read in next row of data into $_; return \@data. # # If $InputType is 'box', input with "| foo | bar | ... |" is # boxed input, and the bars are removed. Also, if "|<" or ">|" # occur, then the next row of data is appended to the current # row. # # Unboxed input is split by tabs (or the $break_str) sub NextRow { my @row; my ($continue, $append); while (1) { $_ = <STDIN>; last unless $_; chomp; unless ($InputType) { $InputType = 'box' if /^\s*([\+:|*])[-+=:*|_]*$1\s*$/; } next if !$No_Dashes && /^[-_=+ \t]*$/; my @data; # If boxed input, remove boxing characters and split with the # box bar character if (/^\s*([|:]).*$1\s*$/ && $InputType eq 'box') { # The data is boxed--remove boxing data my $bar = $1; my $continue = />\Q$bar\E/; # is there a continuation? my $append |= /\Q$bar\E</; # should this be appended? s/^\s*\Q$bar\E<?\s*//; # remove leading bar (if any) s/\s*>?\Q$bar\E\s*$//; # remove trailing bar (if any) if ($NoStrip_Spaces) { # nostrip? @data = split(/>?\Q$bar\E<?/); } else { @data = split(/\s*>?\Q$bar\E<?\s*/); } } else { # not in box mode if ($break_str eq " \t" or $break_str eq "\t ") { @data = split(' '); } elsif ($Strip_Spaces) { @data = split(/\s*\Q$break_str\E\s*/); } else { @data = split(/\Q$break_str\E/); } if ($InputType eq 'table') { # check for table wraps $append |= grep(s/^<//,@data); # see if any appends $continue = grep(s/>$//,@data); # see if any continuations } } if ($append) { my $maxcol = max($#data, $#row); for ($i = 0; $i < $maxcol; $i++) { $row[$i] .= $data[$i]; } $append = ''; # append done (for now) } else { @row = @data; # initialize the row data } # quit the loop unless we have a continuation of data last unless $continue; $append = $continue; $continue = ''; } $#row >= 0 ? \@row : undef; } # ( $titles, $data, $maxcol) = ReadListInput; # # $titles is a ref to an array of titles. # $data is a ref to an array of rows (arrays of columns) of data. # $maxcol is the maximum number of columns in any one row of the data. # The order of the columns in the resulting array of rows is defined # by the names in the @Titles array, or the order of occurance of the # columns in the lists of column names. sub ReadListInput { my @data; my @titles; my ($key, $lastkey, $val); local $_ = 'go'; my $maxcol = 0; my $titlesref; while (length) { my %row; # Read another "row" of list data while (1) { defined($_ = <STDIN>) or last; last if /^\s*$/; # stop on the row boundary chomp; ($key, $val) = /^\s*([\w_.:\/-]+)?\s*:[ \t]?(.*)/; if ($key ne '') { $row{$lastkey = $key} = $val; push(@titles,$key) unless grep(/^\Q$key\E/,@titles); } else { $val = ' '.$val unless $val =~ /^ / || $row{$lastkey} =~ / $/; $row{$lastkey} .= $val; } } my @row; $titlesref = $#Titles >= 0 ? \@Titles : \@titles; foreach $column (@$titlesref) { if (exists $row{$column}) { push(@row,$row{$column}); } else { push(@row,undef); # same as SQL NULL } } push(@data,\@row); # save the row $maxcol = $#row if $#row > $maxcol; } $maxcol = $#titles if $#titles > $maxcol; $titlesref = $#Titles >= 0 ? \@Titles : \@titles; ($titlesref, \@data, $maxcol); } # $max = max @values; # Return the maximum of a list of values sub max { my $max = shift; foreach (@_) { $max = $_ if $_ > $max; } $max; } __END__ =head1 NAME B<showtable> - Show data in nicely formatted columns =head1 USAGE B<showtable> [-I<options>] [I<file>] =head1 DESCRIPTION B<Showtable> reads an input data stream and displays it in a nicely formatted listing, with exact formatting depending upon the options. The input stream, I<file> or C<STDIN> by default should consist of data separated by tabs or the defined I<separator> character (see B<-d>). The actual output formatting is peformed by the B<ShowTable> module. =head1 OPTIONS There are two general sets of options: those which help determine the format of the input, and those which determine the format of the output. =head2 B<Options affecting input> =over 10 =item B<-break=>I<str> Set the inter-column break string to "I<str>". The default is a tab ("C<\t>"). If B<-strip> is also given, blanks surrounding the break string will also be ignored. =item B<-d>I<str> This is the same as C<-break=>I<str>. =item B<-nod(ashes)> Do not ignore lines of separators, such as dashes, equal signs, or underlines. If B<-nodashes> is given, and these lines do occur in the stream, they will be treated as normal data. =item B<-ti(tles)[=>I<NN>B<]> Treat the first I<NN> rows of data as column titles; multiple words in the column titles may wrap vertically. If I<NN> is omitted, it defaults to 1. No B<-titles> option is the same as B<-titles=0>. =item B<-in(put)=>I<type> Set the input type as I<type>, which can be one of: I<box>, I<list>, I<table>, or I<simple>. A I<simple>-type table is the same as a I<table>-type, but no wrapping characters are recognized. =item B<-s(trip)> Strip blanks from around the column values. =item B<-nos(trip)> Do not strip blanks from the input. Useful if there is formatted or aligned data within a boxed table. =back =head2 B<Options affecting output> =over 10 =item B<-t(able)> Use a I<table> format for output, with wrapping of column values longer than the given or determined column widths. See L<Data::ShowTable> for more details. =item B<-si(mple)> Use a simple table format, without any wrapping of column values. See L<Data::ShowTable> for more details. =item B<-l(ist)> Use a list style format. See L<Data::ShowTable> for more details. =item B<-b(ox)> Use a "boxed" style table. See L<Data::ShowTable> for more details. =item B<-ht(ml)> Use HTML-formating. See L<Data::ShowTable> for more details. =item B<-ti(tles)=>I<name1>B<,>I<name2>B<,...,>I<nameN> Define the column names explicitly. This is useful for naming columns of data from C<STDIN>, when B<showtable> is being used as a filter. The first column name, I<name1>, cannot begin with a digit. This option allows any column titles obtained from the input to be overridden. =item B<-noh(eaders)> Do not output any headers on the tables; B<-titles=0> implies this option. =item B<-f>I<n1>[,I<n2>, ..., I<nN>] Select fields numbered I<n1>, I<n2>, etc., to display. Each I<nN> is a field index, or a range of indexes in the form: C<N>-C<M> The default is to show all the fields in each row. Fields are numbered from 1. An example: to show the first, and three through five fields of the C</etc/passwd> file: showtable -d: -f1,2-5 /etc/passwd =item B<-fields>=I<fname1>[,I<fname2>, ..., I<fnameN>] Select the named fields to display. The field names must be available, either through the data stream, or by using the B<-titles> option. The field names given must match the existing field names I<exactly>. Using the file C</etc/passwd> for another example: to show the same first two fields, by name: showtable -d: -titles=Login,UID -fields=Login,UID /etc/passwd =item B<-w(idth)>=I<num> Set the maximum table width. This value is applied to the variable L<Data::Showtable::Max_Table_Width>. When the total width of all columns to be displayed exceeds this value, all column widths are scaled uniformly. If B<-width> is not given, then for all output but B<-html>, the default value is either "C<COLUMNS>", if defined, or 80, if not. Whith B<-html> mode, there is no default value for B<-width>; in other words, there is no limit to the width. =item B<-cw(idths)>=I<w1>[,I<w2>,...,I<wN>] Set individual column widths to the specified values. Empty column widths imply no maximum width. If the B<-width> option is also given, then the B<-cwidth> column widths can also be given as fractions or percentages. Example: To set the maximum width of the third column to 20 characters: -cw=,,20 =back =head2 B<HTML-only options> (the usage of which implies B<-html>) =over 10 =item B<-noe(scape)> Do not perform HTML escape sequences on the data; this allows embedded HTML text in the data to be displayed properly with the B<-html> option. =item B<-attributes>='I<attr1> I<attr2> ...' Declare the table attributes, which are inserted into the C<TABLE> token. For example, the option: -attributes='BORDER=0 CELLSPACING=2 CELLPADDING=4' would cause the following HTML: <TABLE BORDER=0 CELLSPACING=2 CELLPADDING=4> The default table attributes are: <TABLE BORDER=1 CELLSPACING=1 CELLPADDING=1> =item B<-t(itle)_f(ormats)>=I<fmt1>;I<fmt2>;...;I<fmtN> Set the HTML formats for the column titles. The B<-title_formats> (or just B<-tf>) can be given multiple times, for each column, or formats for multiple columns can be given on the same option separated by semi-colons "C<;>". Each I<fmtN> can itself be multiple HTML items, separated by commas. Each HTML element can be given either as an HTML token (eg: "C<\<BOLD\>>"), or as a plain name (eg: "C<BOLD>"). For example, here is a title format specification for three columns, where the first column title should be bold italic, the second italic, and the third italic in a smaller font: -tf='BOLD,I;I;<FONT SIZE=-2>,I' =item B<-d(ata)_f(formats)>=I<fmt1>;I<fmt2>;...;I<fmtN> The same as B<-title_formats> but applies to the column data. =item B<-url(s)=I<col1>=I<url1>,I<col2>=I<url2>,...> Define a mapping from column names, or indexes, to URLs to be inserted as <A HREF's> around the values for the named columns. Each I<colN> is a column name or index, and each I<urlN> is a string representing the URL to be inserted for the given column. The URL text may contain these substitution strings: B<%K> - will be substituted with the current column name (or I<key>). B<%V> - will be substituted with the current column value. Multiple B<-url> options may be given, if desired, rather than creating one long argument for a single B<-url>. For example: showtable -d: -f1,6 -titles=Login,Homedir \ -url='Login=mailto:%V' \ -url='HomeDir=file:%V' \ /etc/passwd =back =head2 B<Other options> =over 10 =item B<-help> Display some help to the user and quit. =back =head2 B<Boxed Input> If the input type is I<box>, then vertical and horizontal box characters are removed from the input stream, and blanks surrounding the vertical box characters are removed. The vertical box characters (column separaters) are "C<|>" or "C<:>". The The horizontal box characters are "C<+>" and "C<->". Morever, data wrapped within a column is recognized and parsed as one column value, by recognizing the presence of a I<wrapping prefix> or I<wrapping suffix> character. Currently, the wrapping prefix character is "<", and the wrapping suffix character is ">". An example of data wrapped within a column is given here. The table below has just two I<logical> rows of data; with both rows having data wrapped into multiple I<physical> rows. +---------+---------+---------+ | Col 1 | Col 2 | Col 3 | +---------+---------+---------+ | This is>| Another>| Row 1,3>| |< a cont>|< value. |<is also>| |<inued >| |<long. | |<value. | | | |This is >| Item2-2 | Item2-3 | +---------+---------+---------+ =head2 B<List Format> When using the B<-list> or B<-input=list> options, either, or both, the input and output may be in a "list" format, which is implemented using the following syntax: r1c1_name: r1c1_value r1c2_name: r1c2_value ... r1cN_name: r1cN_value r2c1_name: r2c1_value r2c2_name: r2c2_value : r2c2_value_continued ... r2cN_name: r2cN_value rMc1_name: rMc1_value rMc2_name: rMc2_value ... rMcN_name: rMcN_value Each I<row> of data consists of one or more I<columns>, and ends with a blank line. Each I<column> consists of a I<column name>, followed by a colon ":", followed by an optional, single space or tab, followed by the I<column value>, on the same line. Continuation lines of the previous column value consist of one or more space or tab characters, a colon ":", one optional, single space or tab, followed by the continuation value. In the example above, The second column value of the second row was continued. =head2 B<HTML Input with HTML Output> When using B<-html> on data already containing HTML-formatted text, the B<-noescape> option should be used. By default, all input text is assumed I<not> to be HTML-formatted, and is escaped allowing embedded "<", ">" characters, if any, to be displayed correctly. =head1 DEPENDENCIES =over 10 =item B<Data::ShowTable> module Performs the actual output formatting. =back =head1 AUTHOR Alan K. Stebbens I<aks@stebbens.org> =head1 BUGS =over 5 =item * Currently, the box formatting characters are not configurable: '+' for the corners; '-' and '|' for the tops and sides, respectively. In an ideal world, these things would be configurable. =item * The continuation prefix and suffix characters, '<' and '>', respectively, are also not configurable: =item * When reading I<table> input, any data ending with ">" will be considered to be continued by the next row of data. To avoid this, use B<-input=simple>. =item * When selecting noncontiguous fields (ie: B<-f1,4>>) without field names, the default field names will be consecutively numbered from 1, which is counter-intuitive to the original selection. To avoid this, name the fields using the B<-title=...> option. =back =cut