| 1 |
#!perl |
| 2 |
|
| 3 |
use strict; |
| 4 |
use warnings; |
| 5 |
use version; |
| 6 |
use File::Basename; |
| 7 |
use Getopt::Long; |
| 8 |
use List::Util 'max'; |
| 9 |
use Pod::Usage; |
| 10 |
use Readonly; |
| 11 |
use Text::Autoformat 'autoformat'; |
| 12 |
use Text::RecordParser; |
| 13 |
use Text::TabularDisplay; |
| 14 |
|
| 15 |
Readonly our $VERSION => 1.15; |
| 16 |
Readonly my $DASH => q{-}; |
| 17 |
Readonly my $EMPTY_STR => q{}; |
| 18 |
Readonly my $TAB => qq{\t}; |
| 19 |
Readonly my $NEWLINE => qq{\n}; |
| 20 |
Readonly my $WIDTH => 78; |
| 21 |
|
| 22 |
my $comment_start = ''; |
| 23 |
my $fs = $TAB; |
| 24 |
my $rs = $NEWLINE; |
| 25 |
my $headers = ''; |
| 26 |
my $no_headers = 0; |
| 27 |
my $show_vertically = 0; |
| 28 |
my $no_pager = 0; |
| 29 |
my $strip_quotes = 0; |
| 30 |
|
| 31 |
my ( $show_fields, $list, $limit, @where, $help, $man_page, $show_version ); |
| 32 |
GetOptions( |
| 33 |
'c|comment:s' => \$comment_start, |
| 34 |
'fs:s' => \$fs, |
| 35 |
'rs:s' => \$rs, |
| 36 |
'f|fields:s' => \$show_fields, |
| 37 |
'l|list' => \$list, |
| 38 |
'i|limit:i' => \$limit, |
| 39 |
'w|where:s' => \@where, |
| 40 |
'v|vertical' => \$show_vertically, |
| 41 |
'strip-quotes' => \$strip_quotes, |
| 42 |
'n|no-headers' => \$no_headers, |
| 43 |
'h|headers:s' => \$headers, |
| 44 |
'no-pager' => \$no_pager, |
| 45 |
'help' => \$help, |
| 46 |
'man' => \$man_page, |
| 47 |
'version' => \$show_version, |
| 48 |
) or pod2usage; |
| 49 |
|
| 50 |
if ( $help || $man_page ) { |
| 51 |
pod2usage({ |
| 52 |
-exitval => 0, |
| 53 |
-verbose => $man_page ? 2 : 1 |
| 54 |
}); |
| 55 |
}; |
| 56 |
|
| 57 |
if ( $show_version ) { |
| 58 |
my $prog = basename( $0 ); |
| 59 |
print "$prog $VERSION\n"; |
| 60 |
exit 0; |
| 61 |
} |
| 62 |
|
| 63 |
my $file = shift or pod2usage('No file'); |
| 64 |
my $p = Text::RecordParser->new( |
| 65 |
field_separator => $fs, |
| 66 |
record_separator => $rs, |
| 67 |
comment => $comment_start ? qr/^$comment_start/ : undef, |
| 68 |
); |
| 69 |
|
| 70 |
if ( $strip_quotes ) { |
| 71 |
$p->field_filter( sub { s/^["']//; s/["']$//; $_ } ); |
| 72 |
} |
| 73 |
|
| 74 |
if ( $file eq $DASH ) { |
| 75 |
$p->fh( \*STDIN ); |
| 76 |
} |
| 77 |
else { |
| 78 |
$p->filename( $file ); |
| 79 |
} |
| 80 |
|
| 81 |
my @fields; |
| 82 |
if ( @fields = split( /\s*,\s*/, $headers ) ) { |
| 83 |
$p->bind_fields( @fields ); |
| 84 |
} |
| 85 |
elsif ( !$no_headers ) { |
| 86 |
$p->bind_header; |
| 87 |
@fields = $p->field_list; |
| 88 |
} |
| 89 |
|
| 90 |
if ( $list ) { |
| 91 |
pod2usage(q[Can't list fields with --no-headers]) if $no_headers; |
| 92 |
my $tab = Text::TabularDisplay->new('Field No.', 'Field'); |
| 93 |
my $i = 1; |
| 94 |
$tab->add( $i++, $_ ) for @fields; |
| 95 |
print $tab->render, $NEWLINE; |
| 96 |
exit 0; |
| 97 |
} |
| 98 |
|
| 99 |
my %where; |
| 100 |
for ( @where ) { |
| 101 |
if ( /([\w\d]+)\s*(==|eq|>=?|<=?|=~)\s*(.*)/ ) { |
| 102 |
my $field = $1; |
| 103 |
my $op = $2; |
| 104 |
my $value = $3; |
| 105 |
|
| 106 |
unless ( $no_headers ) { |
| 107 |
my %available = map { $_, 1 } @fields; |
| 108 |
next unless $available{ $field }; |
| 109 |
} |
| 110 |
|
| 111 |
$field-- if $field =~ /^\d+$/; |
| 112 |
$where{ $field } = [ $op, $value ]; |
| 113 |
} |
| 114 |
} |
| 115 |
|
| 116 |
if ( $show_fields ) { |
| 117 |
my @show = map { $_ =~ m/^\d+$/ && @fields |
| 118 |
? $_-1 < scalar @fields ? $_ : () : $_ } |
| 119 |
map { $_ =~ m/(\d+)-(\d+)/ ? ( $1..$2 ) : $_ } |
| 120 |
split /,/, $show_fields; |
| 121 |
my @numbers = grep { /^(\d+)$/ } @show; |
| 122 |
|
| 123 |
if ( $no_headers ) { |
| 124 |
@fields = @show; |
| 125 |
} |
| 126 |
else { |
| 127 |
if ( scalar @show == scalar @numbers ) { # all numbers |
| 128 |
@numbers = map { $_ - 1 } @numbers; |
| 129 |
@fields = @fields[ @numbers ]; |
| 130 |
} |
| 131 |
elsif ( @show ) { |
| 132 |
my %available = map { $_, 1 } @fields; |
| 133 |
my @temp = @fields; |
| 134 |
@fields = map { $available{ $_ } ? $_ : () } @show; |
| 135 |
} |
| 136 |
|
| 137 |
if ( !@fields ) { |
| 138 |
die "No fields match in list '$show_fields'\n"; |
| 139 |
} |
| 140 |
} |
| 141 |
} |
| 142 |
|
| 143 |
my $fh; |
| 144 |
my $pager = $ENV{'PAGER'}; |
| 145 |
if ( !$no_pager && $pager ) { |
| 146 |
open $fh, "| $pager"; |
| 147 |
} |
| 148 |
else { |
| 149 |
$fh = \*STDOUT; |
| 150 |
} |
| 151 |
|
| 152 |
my $tab = Text::TabularDisplay->new( @fields ); |
| 153 |
|
| 154 |
my $max_col_length = 0; |
| 155 |
my $num_records = 0; |
| 156 |
my $separator = "************ Record %s ************\n"; |
| 157 |
if ( $no_headers ) { |
| 158 |
my @field_names; |
| 159 |
RECORD: |
| 160 |
while ( my @data = $p->fetchrow_array ) { |
| 161 |
if ( !@fields ) { |
| 162 |
@fields = ( 0..$#data ); |
| 163 |
@field_names = map { 'Field' . ($_+1) } @fields; |
| 164 |
$max_col_length = max( map { length $_ } @field_names ); |
| 165 |
$tab->columns( @field_names ); |
| 166 |
} |
| 167 |
|
| 168 |
for my $field ( keys %where ) { |
| 169 |
my ( $op, $value ) = @{ $where{ $field } }; |
| 170 |
my $cmd = "'$data[ ($field - 1) ]' $op $value"; |
| 171 |
next RECORD unless eval $cmd; |
| 172 |
} |
| 173 |
|
| 174 |
$num_records++; |
| 175 |
if ( $show_vertically ) { |
| 176 |
printf $fh $separator, $num_records; |
| 177 |
for my $i ( @fields ) { |
| 178 |
printf $fh "%${max_col_length}s: %s\n", |
| 179 |
$field_names[ $i ], |
| 180 |
defined $data[ $i ] ? $data[ $i ] : q{}; |
| 181 |
} |
| 182 |
} |
| 183 |
else { |
| 184 |
$tab->add( map { $data[ $_ ] } @fields ); |
| 185 |
} |
| 186 |
|
| 187 |
last if $limit && $num_records >= $limit; |
| 188 |
} |
| 189 |
} |
| 190 |
else { |
| 191 |
$max_col_length = max map { $_ ? length $_ : 0 } $p->field_list; |
| 192 |
|
| 193 |
RECORD: |
| 194 |
while ( my $data = $p->fetchrow_hashref ) { |
| 195 |
for my $field ( keys %where ) { |
| 196 |
my ( $op, $value ) = @{ $where{ $field } }; |
| 197 |
my $cmd = "'$data->{ $field }' $op $value"; |
| 198 |
next RECORD unless eval $cmd; |
| 199 |
} |
| 200 |
|
| 201 |
$num_records++; |
| 202 |
if ( $show_vertically ) { |
| 203 |
printf $fh $separator, $num_records; |
| 204 |
for my $field ( @fields ) { |
| 205 |
next unless $field; |
| 206 |
my $v = defined $data->{ $field } ? $data->{ $field } : q{}; |
| 207 |
if ( length $v > $WIDTH ) { |
| 208 |
( $v = autoformat( $v, { left => $max_col_length + 3 } ) ) |
| 209 |
=~ s/^\s+|\s+$//g; |
| 210 |
} |
| 211 |
|
| 212 |
printf $fh "%${max_col_length}s: %s\n", $field, $v; |
| 213 |
} |
| 214 |
} |
| 215 |
else { |
| 216 |
$tab->add( map { $data->{ $_ } } @fields ); |
| 217 |
} |
| 218 |
last if $limit && $num_records >= $limit; |
| 219 |
} |
| 220 |
} |
| 221 |
|
| 222 |
if ( !$show_vertically ) { |
| 223 |
print $fh $tab->render; |
| 224 |
} |
| 225 |
|
| 226 |
print $fh $num_records |
| 227 |
? sprintf( |
| 228 |
"\n%s record%s returned\n", $num_records, $num_records > 1 |
| 229 |
? 's' : $EMPTY_STR |
| 230 |
) |
| 231 |
: "\nNo records returned\n"; |
| 232 |
|
| 233 |
close $fh; |
| 234 |
|
| 235 |
__END__ |
| 236 |
|
| 237 |
# ------------------------------------------------------------------- |
| 238 |
|
| 239 |
# $Id: tablify,v 1.14 2006/03/07 17:20:00 kclark Exp $ |
| 240 |
|
| 241 |
=pod |
| 242 |
|
| 243 |
=head1 NAME |
| 244 |
|
| 245 |
tablify - turn a delimited text file into a text table |
| 246 |
|
| 247 |
=head1 SYNOPSIS |
| 248 |
|
| 249 |
tablify [options] file |
| 250 |
|
| 251 |
Options: |
| 252 |
|
| 253 |
-h|--help Show help |
| 254 |
-c|--comment Define the beginning of a (single-line) comment |
| 255 |
-n|--no-headers Assume first line is data, not headers |
| 256 |
--no-pager Do not use $ENV{'PAGER'} even if defined |
| 257 |
--strip-quotes Strip " or ' around fields |
| 258 |
-l|--list List the fields in the file (for use with -f) |
| 259 |
-f|--fields=f1[,f2] Show only fields in comma-separated list; |
| 260 |
when used in conjunction with "no-headers" |
| 261 |
the list should be field numbers (starting at 1); |
| 262 |
otherwise, should be field names |
| 263 |
-w|where=f<cmp>v Apply the "cmp" Perl operator to restrict output |
| 264 |
where field "f" matches the value "v"; acceptable |
| 265 |
operators include ==, eq, >, >=, <=, and =~ |
| 266 |
-v|--vertical Show records vertically |
| 267 |
-i|--limit=n Limit to given number of records |
| 268 |
--fs=x Use "x" as the field separator |
| 269 |
(default is tab "\t") |
| 270 |
--rs=x Use "x" as the record separator |
| 271 |
(default is newline "\n") |
| 272 |
--as-html Create an HTML table instead of plain text |
| 273 |
--headers Comma-separated list of names matching |
| 274 |
the number of columns |
| 275 |
|
| 276 |
=head1 DESCRIPTION |
| 277 |
|
| 278 |
This script is essentially a quick way to parse a delimited text file |
| 279 |
and view it as a nice ASCII table. By selecting only certain B<fields>, |
| 280 |
employing a B<where> clause to only select records where a field matches |
| 281 |
some value, and using the B<limit> to only see some of the output, you |
| 282 |
almost have a mini-database front-end for a simple text file. |
| 283 |
|
| 284 |
=head1 EXAMPLES |
| 285 |
|
| 286 |
Given a data file like this: |
| 287 |
|
| 288 |
name,rank,serial_no,is_living,age |
| 289 |
George,General,190293,0,64 |
| 290 |
Dwight,General,908348,0,75 |
| 291 |
Attila,Hun,,0,56 |
| 292 |
Tojo,Emporor,,0,87 |
| 293 |
Tommy,General,998110,1,54 |
| 294 |
|
| 295 |
To find the fields you can reference, use the B<list> option: |
| 296 |
|
| 297 |
$ tablify --fs ',' -l people.dat |
| 298 |
+-----------+-----------+ |
| 299 |
| Field No. | Field | |
| 300 |
+-----------+-----------+ |
| 301 |
| 1 | name | |
| 302 |
| 2 | rank | |
| 303 |
| 3 | serial_no | |
| 304 |
| 4 | is_living | |
| 305 |
| 5 | age | |
| 306 |
+-----------+-----------+ |
| 307 |
|
| 308 |
To extract just the name and serial numbers, use the B<fields> option: |
| 309 |
|
| 310 |
$ tablify --fs ',' -f name,serial_no people.dat |
| 311 |
+--------+-----------+ |
| 312 |
| name | serial_no | |
| 313 |
+--------+-----------+ |
| 314 |
| George | 190293 | |
| 315 |
| Dwight | 908348 | |
| 316 |
| Attila | | |
| 317 |
| Tojo | | |
| 318 |
| Tommy | 998110 | |
| 319 |
+--------+-----------+ |
| 320 |
5 records returned |
| 321 |
|
| 322 |
To extract the first through third fields and the fifth field (where |
| 323 |
field numbers start at "1" -- tip: use the B<list> option to quickly |
| 324 |
determine field numbers), use this syntax for B<fields>: |
| 325 |
|
| 326 |
$ tablify --fs ',' -f 1-3,5 people.dat |
| 327 |
+--------+---------+-----------+------+ |
| 328 |
| name | rank | serial_no | age | |
| 329 |
+--------+---------+-----------+------+ |
| 330 |
| George | General | 190293 | 64 | |
| 331 |
| Dwight | General | 908348 | 75 | |
| 332 |
| Attila | Hun | | 56 | |
| 333 |
| Tojo | Emporor | | 87 | |
| 334 |
| Tommy | General | 998110 | 54 | |
| 335 |
+--------+---------+-----------+------+ |
| 336 |
5 records returned |
| 337 |
|
| 338 |
To select only the ones with six serial numbers, use a B<where> |
| 339 |
clause: |
| 340 |
|
| 341 |
$ tablify --fs ',' -w 'serial_no=~/^\d{6}$/' people.dat |
| 342 |
+--------+---------+-----------+-----------+------+ |
| 343 |
| name | rank | serial_no | is_living | age | |
| 344 |
+--------+---------+-----------+-----------+------+ |
| 345 |
| George | General | 190293 | 0 | 64 | |
| 346 |
| Dwight | General | 908348 | 0 | 75 | |
| 347 |
| Tommy | General | 998110 | 1 | 54 | |
| 348 |
+--------+---------+-----------+-----------+------+ |
| 349 |
3 records returned |
| 350 |
|
| 351 |
To find Dwight's record, you would do this: |
| 352 |
|
| 353 |
$ tablify --fs ',' -w 'name eq "Dwight"' people.dat |
| 354 |
+--------+---------+-----------+-----------+------+ |
| 355 |
| name | rank | serial_no | is_living | age | |
| 356 |
+--------+---------+-----------+-----------+------+ |
| 357 |
| Dwight | General | 908348 | 0 | 75 | |
| 358 |
+--------+---------+-----------+-----------+------+ |
| 359 |
1 record returned |
| 360 |
|
| 361 |
To find the name of all the people with a serial number who are living: |
| 362 |
|
| 363 |
$ tablify --fs ',' -f name -w 'is_living==1' -w 'serial_no>0' people.dat |
| 364 |
+-------+ |
| 365 |
| name | |
| 366 |
+-------+ |
| 367 |
| Tommy | |
| 368 |
+-------+ |
| 369 |
1 record returned |
| 370 |
|
| 371 |
To filter outside of program and simply format the results, use "-" as |
| 372 |
the last argument to force reading of STDIN (and probably assume no |
| 373 |
headers): |
| 374 |
|
| 375 |
$ grep General people.dat | tablify --fs ',' -f 1-3 --no-headers - |
| 376 |
+---------+--------+--------+ |
| 377 |
| Field1 | Field2 | Field3 | |
| 378 |
+---------+--------+--------+ |
| 379 |
| General | 190293 | 0 | |
| 380 |
| General | 908348 | 0 | |
| 381 |
| General | 998110 | 1 | |
| 382 |
+---------+--------+--------+ |
| 383 |
3 records returned |
| 384 |
|
| 385 |
When dealing with data lacking field names, you can specify "no-headers" |
| 386 |
and then refer to fields by number (starting at one), e.g.: |
| 387 |
|
| 388 |
$ tail -5 people.dat | tablify --fs ',' --no-headers -w '3 eq "General"' - |
| 389 |
+--------+---------+--------+--------+--------+ |
| 390 |
| Field1 | Field2 | Field3 | Field4 | Field5 | |
| 391 |
+--------+---------+--------+--------+--------+ |
| 392 |
| George | General | 190293 | 0 | 64 | |
| 393 |
| Dwight | General | 908348 | 0 | 75 | |
| 394 |
| Tommy | General | 998110 | 1 | 54 | |
| 395 |
+--------+---------+--------+--------+--------+ |
| 396 |
3 records returned |
| 397 |
|
| 398 |
If your file has many fields which are hard to see across the screen, |
| 399 |
consider using the vertical display with "-v" or "--vertical", e.g.: |
| 400 |
|
| 401 |
$ tablify --fs ',' -v --limit 1 people.dat |
| 402 |
************ Record 1 ************ |
| 403 |
name: George |
| 404 |
rank: General |
| 405 |
serial_no: 190293 |
| 406 |
is_living: 0 |
| 407 |
age : 64 |
| 408 |
|
| 409 |
1 record returned |
| 410 |
|
| 411 |
=head1 SEE ALSO |
| 412 |
|
| 413 |
=over 4 |
| 414 |
|
| 415 |
=item * Text::RecordParser |
| 416 |
|
| 417 |
=item * Text::TabularDisplay |
| 418 |
|
| 419 |
=item * DBD::CSV |
| 420 |
|
| 421 |
Although I don't DBD::CSV this module, the idea was much the inspiration |
| 422 |
for this. I just didn't want to have to install DBI and DBD::CSV to |
| 423 |
get this kind of functionality. I think my interface is simpler. |
| 424 |
|
| 425 |
=back |
| 426 |
|
| 427 |
=head1 AUTHOR |
| 428 |
|
| 429 |
Ken Youens-Clark E<lt>[email protected]<gt>. |
| 430 |
|
| 431 |
=head1 LICENSE AND COPYRIGHT |
| 432 |
|
| 433 |
Copyright (C) 2006-10 Ken Youens-Clark. All rights reserved. |
| 434 |
|
| 435 |
This program is free software; you can redistribute it and/or modify |
| 436 |
it under the terms of the GNU General Public License as published by |
| 437 |
the Free Software Foundation; version 2. |
| 438 |
|
| 439 |
This program is distributed in the hope that it will be useful, but |
| 440 |
WITHOUT ANY WARRANTY; without even the implied warranty of |
| 441 |
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 442 |
General Public License for more details. |
| 443 |
|
| 444 |
=cut |