ConversionSupport.pm 52.7 KB
Newer Older
1
=head1 LICENSE
2

3
  Copyright (c) 1999-2011 The European Bioinformatics Institute and
4
  Genome Research Limited.  All rights reserved.
5

6 7
  This software is distributed under a modified Apache license.
  For license details, please see
8

9
    http://www.ensembl.org/info/about/code_licence.html
10

11
=head1 CONTACT
12

13
  Please email comments or questions to the public Ensembl
14
  developers list at <dev@ensembl.org>.
15

16 17
  Questions may also be sent to the Ensembl help desk at
  <helpdesk@ensembl.org>.
18

19
=cut
20

21
=head1 NAME
22

23 24
Bio::EnsEMBL::Utils::ConversionSupport - Utility module for Vega release and
schema conversion scripts
25

26
=head1 SYNOPSIS
27

28 29
  my $serverroot = '/path/to/ensembl';
  my $support = new Bio::EnsEMBL::Utils::ConversionSupport($serverroot);
30

31 32
  # parse common options
  $support->parse_common_options;
33

34 35
  # parse extra options for your script
  $support->parse_extra_options( 'string_opt=s', 'numeric_opt=n' );
36

37 38
  # ask user if he wants to run script with these parameters
  $support->confirm_params;
39

40
  # see individual method documentation for more stuff
41

42 43 44 45 46 47 48 49
=head1 DESCRIPTION

This module is a collection of common methods and provides helper
functions for the Vega release and schema conversion scripts. Amongst
others, it reads options from a config file, parses commandline options
and does logging.

=head1 METHODS
50 51 52

=cut

53 54
package Bio::EnsEMBL::Utils::ConversionSupport;

55 56 57 58 59 60 61 62 63
use strict;
use warnings;
no warnings 'uninitialized';

use Getopt::Long;
use Text::Wrap;
use Bio::EnsEMBL::Utils::Exception qw(throw warning);
use FindBin qw($Bin $Script);
use POSIX qw(strftime);
64
use Cwd qw(abs_path);
Patrick Meidl's avatar
Patrick Meidl committed
65
use DBI;
Web Admin's avatar
Web Admin committed
66
use Data::Dumper;
67 68 69 70 71 72 73 74 75 76 77 78 79 80

=head2 new

  Arg[1]      : String $serverroot - root directory of your ensembl sandbox
  Example     : my $support = new Bio::EnsEMBL::Utils::ConversionSupport(
                                        '/path/to/ensembl');
  Description : constructor
  Return type : Bio::EnsEMBL::Utils::ConversionSupport object
  Exceptions  : thrown if no serverroot is provided
  Caller      : general

=cut

sub new {
Steve Trevanion's avatar
Steve Trevanion committed
81 82 83 84 85 86 87 88 89
  my $class = shift;
  (my $serverroot = shift) or throw("You must supply a serverroot.");
  my $self = {
    '_serverroot'   => $serverroot,
    '_param'        => { interactive => 1 },
    '_warnings'     => 0,
  };
  bless ($self, $class);
  return $self;
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
}

=head2 parse_common_options

  Example     : $support->parse_common_options;
  Description : This method reads options from a configuration file and parses
                some commandline options that are common to all scripts (like
                db connection settings, help, dry-run). Commandline options
                will override config file settings. 

                All options will be accessible via $self->param('name').
  Return type : true on success 
  Exceptions  : thrown if configuration file can't be opened
  Caller      : general

=cut

sub parse_common_options {
Steve Trevanion's avatar
Steve Trevanion committed
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
  my $self = shift;

  # read commandline options
  my %h;
  Getopt::Long::Configure("pass_through");
  &GetOptions( \%h,
	       'dbname|db_name=s',
	       'host|dbhost|db_host=s',
	       'port|dbport|db_port=n',
	       'user|dbuser|db_user=s',
	       'pass|dbpass|db_pass=s',
	       'conffile|conf=s',
	       'logfile|log=s',
	       'logpath=s',
	       'logappend|log_append=s',
	       'verbose|v=s',
	       'interactive|i=s',
	       'dry_run|dry|n=s',
	       'help|h|?',
	     );

  # reads config file
  my $conffile = $h{'conffile'} || $self->serverroot . "/sanger-plugins/vega/conf/ini-files/Conversion.ini";
  $conffile = abs_path($conffile);
  if (-e $conffile) {
    open(CONF, $conffile) or throw( 
      "Unable to open configuration file $conffile for reading: $!");
    my $serverroot = $self->serverroot;
    while (<CONF>) {
      chomp;

      # remove comments
      s/^[#;].*//;
      s/\s+[;].*$//;

      # read options into internal parameter datastructure, removing whitespace
      next unless (/(\w\S*)\s*=\s*(\S*)\s*/);
      my $name = $1;
      my $val = $2;
      if ($val =~ /\$SERVERROOT/) {
	$val =~ s/\$SERVERROOT/$serverroot/g;
	$val = abs_path($val);
      }
      $self->param($name, $val);
152
    }
Steve Trevanion's avatar
Steve Trevanion committed
153 154 155 156 157
    $self->param('conffile', $conffile);
  } elsif ($conffile) {
    warning("Unable to open configuration file $conffile for reading: $!");
  }

158
# override configured parameter with commandline options
Steve Trevanion's avatar
Steve Trevanion committed
159
  map { $self->param($_, $h{$_}) } keys %h;
160 161 162 163 164 165 166 167 168 169 170 171 172 173

  # if logpath & logfile are not se, set them here to /ensemblweb/vega_dev/shared/logs/conversion/DBNAME/SCRIPNAME_NN.log
  if (not (defined($self->param('logpath')))){
    $self->param('logpath', "/ensemblweb/vega_dev/shared/logs/conversion/".$self->param('dbname')."/" );
  }
  if (not (defined($self->param('logfile')))){
    my $log = $Script;
    $log =~ s/.pl//g;
    my $counter;
    for ($counter=1 ; (-e $self->param('logpath')."/".$log."_".sprintf("%03d", $counter).".log"); $counter++){ warn  $self->param('logpath')."/".$log."_".$counter.".log";}

    $self->param('logfile', $log."_".sprintf("%03d", $counter).".log");
  }
  
Steve Trevanion's avatar
Steve Trevanion committed
174
  return(1);
175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
}

=head2 parse_extra_options

  Arg[1-N]    : option descriptors that will be passed on to Getopt::Long
  Example     : $support->parse_extra_options('string_opt=s', 'numeric_opt=n');
  Description : Parse extra commandline options by passing them on to
                Getopt::Long and storing parameters in $self->param('name).
  Return type : true on success
  Exceptions  : none (caugth by $self->error)
  Caller      : general

=cut

sub parse_extra_options {
Steve Trevanion's avatar
Steve Trevanion committed
190 191 192 193 194 195 196 197 198
  my ($self, @params) = @_;
  Getopt::Long::Configure("no_pass_through");
  eval {
    # catch warnings to pass to $self->error
    local $SIG{__WARN__} = sub { die @_; };
    &GetOptions(\%{ $self->{'_param'} }, @params);
  };
  $self->error($@) if $@;
  return(1);
199 200
}

201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
=head2 allowed_params

  Arg[1-N]    : (optional) List of allowed parameters to set
  Example     : my @allowed = $self->allowed_params(qw(param1 param2));
  Description : Getter/setter for allowed parameters. This is used by
                $self->confirm_params() to avoid cluttering of output with
                conffile entries not relevant for a given script. You can use
                $self->get_common_params() as a shortcut to set them.
  Return type : Array - list of allowed parameters
  Exceptions  : none
  Caller      : general

=cut

sub allowed_params {
Steve Trevanion's avatar
Steve Trevanion committed
216 217 218 219 220 221 222 223 224 225 226 227 228
  my $self = shift;

  # setter
  if (@_) {
    @{ $self->{'_allowed_params'} } = @_;
  }

  # getter
  if (ref($self->{'_allowed_params'}) eq 'ARRAY') {
    return @{ $self->{'_allowed_params'} };
  } else {
    return ();
  }
229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
}

=head2 get_common_params

  Example     : my @allowed_params = $self->get_common_params, 'extra_param';
  Description : Returns a list of commonly used parameters in the conversion
                scripts. Shortcut for setting allowed parameters with
                $self->allowed_params().
  Return type : Array - list of common parameters
  Exceptions  : none
  Caller      : general

=cut

sub get_common_params {
Steve Trevanion's avatar
Steve Trevanion committed
244 245 246 247 248 249 250 251 252 253 254 255 256 257
  return qw(
	    conffile
	    dbname
	    host
	    port
	    user
	    pass
	    logpath
	    logfile
	    logappend
	    verbose
	    interactive
	    dry_run
	  );
258 259
}

Steve Trevanion's avatar
Steve Trevanion committed
260
=head2 get_loutre_params
Web Admin's avatar
Web Admin committed
261

Steve Trevanion's avatar
Steve Trevanion committed
262 263 264 265
  Arg         : (optional) return a list to parse or not
  Example     : $support->parse_extra_options($support->get_loutre_params('parse'))
  Description : Returns a list of commonly used loutre db parameters - parse option is
                simply used to distinguish between reporting and parsing parameters
Web Admin's avatar
Web Admin committed
266 267 268 269 270 271 272
  Return type : Array - list of common parameters
  Exceptions  : none
  Caller      : general

=cut

sub get_loutre_params {
Steve Trevanion's avatar
Steve Trevanion committed
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291
  my ($self,$p) = @_;
  if ($p) {
    return qw(
	      loutrehost=s
	      loutreport=s
	      loutreuser=s
	      loutrepass=s
	      loutredbname=s
	    );
  }
  else {
    return qw(
	      loutrehost
	      loutreport
	      loutreuser
	      loutrepass
	      loutredbname
	    );
  }
Web Admin's avatar
Web Admin committed
292 293
}

Steve Trevanion's avatar
Steve Trevanion committed
294 295 296 297 298 299 300 301 302 303 304 305
=head2 remove_vega_params

  Example     : $support->remove_vega_params;
  Description : Removes Vega db conection parameters. Usefull to avoid clutter in log files when
                working exclusively with loutre
  Return type : none
  Exceptions  : none
  Caller      : general

=cut

sub remove_vega_params {
Steve Trevanion's avatar
Steve Trevanion committed
306 307 308 309
  my $self = shift;
  foreach my $param (qw(dbname host port user pass)) {
    $self->{'_param'}{$param} = undef;
  }
Steve Trevanion's avatar
Steve Trevanion committed
310
}
Web Admin's avatar
Web Admin committed
311

312 313 314 315 316 317 318 319 320 321 322 323 324
=head2 confirm_params

  Example     : $support->confirm_params;
  Description : Prints a table of parameters that were collected from config
                file and commandline and asks user to confirm if he wants
                to proceed.
  Return type : true on success
  Exceptions  : none
  Caller      : general

=cut

sub confirm_params {
Steve Trevanion's avatar
Steve Trevanion committed
325 326 327 328 329 330
  my $self = shift;

  # print parameter table
  print "Running script with these parameters:\n\n";
  print $self->list_all_params;

Steve Trevanion's avatar
Steve Trevanion committed
331
  if ($self->param('host') eq 'ensdb-1-10') {
Steve Trevanion's avatar
Steve Trevanion committed
332
    # ask user if he wants to proceed
Steve Trevanion's avatar
Steve Trevanion committed
333
    exit unless $self->user_proceed("**************\n\n You're working on ensdb-1-10! Is that correct and you want to continue ?\n\n**************");
Steve Trevanion's avatar
Steve Trevanion committed
334 335 336 337 338 339
  }
  else {
    # ask user if he wants to proceed
    exit unless $self->user_proceed("Continue?");
  }
  return(1);
340 341 342 343 344 345 346 347 348 349 350 351 352
}

=head2 list_all_params

  Example     : print LOG $support->list_all_params;
  Description : prints a table of the parameters used in the script
  Return type : String - the table to print
  Exceptions  : none
  Caller      : general

=cut

sub list_all_params {
Steve Trevanion's avatar
Steve Trevanion committed
353 354 355 356 357 358
  my $self = shift;
  my $txt = sprintf "    %-21s%-40s\n", qw(PARAMETER VALUE);
  $txt .= "    " . "-"x71 . "\n";
  $Text::Wrap::colums = 72;
  my @params = $self->allowed_params;
  foreach my $key (@params) {
359 360 361 362 363 364
    my @vals = $self->param($key);
    if (@vals) {
      $txt .= Text::Wrap::wrap( sprintf('   %-21s', $key),
				' 'x24,
				join(", ", @vals)
			      ) . "\n";
365
    }
366 367 368
  }
  $txt .= "\n";
  return $txt;
369 370
}

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
=head2 create_commandline_options

  Arg[1]      : Hashref $settings - hashref describing what to do
                Allowed keys:
                    allowed_params => 0|1   # use all allowed parameters
                    exclude => []           # listref of parameters to exclude
                    replace => {param => newval} # replace value of param with
                                                 # newval
  Example     : $support->create_commandline_options({
                    allowed_params => 1,
                    exclude => ['verbose'],
                    replace => { 'dbname' => 'homo_sapiens_vega_33_35e' }
                });
  Description : Creates a commandline options string that can be passed to any
                other script using ConversionSupport.
  Return type : String - commandline options string
  Exceptions  : none
  Caller      : general

=cut

sub create_commandline_options {
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409
  my ($self, $settings) = @_;
  my %param_hash;

  # get all allowed parameters
  if ($settings->{'allowed_params'}) {
    # exclude params explicitly stated
    my %exclude = map { $_ => 1 } @{ $settings->{'exclude'} || [] };
    foreach my $param ($self->allowed_params) {
      unless ($exclude{$param}) {
	my ($first, @rest) = $self->param($param);
	next unless (defined($first));
	
	if (@rest) {
	  $first = join(",", $first, @rest);
	}
	$param_hash{$param} = $first;
      }
410
    }
411
  }
412

413 414 415 416
  # replace values
  foreach my $key (keys %{ $settings->{'replace'} || {} }) {
    $param_hash{$key} = $settings->{'replace'}->{$key};
  }
417

418 419 420 421 422 423
  # create the commandline options string
  my $options_string;
  foreach my $param (keys %param_hash) {
    $options_string .= sprintf("--%s %s ", $param, $param_hash{$param});
  }
  return $options_string;
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438
}

=head2 check_required_params

  Arg[1-N]    : List @params - parameters to check
  Example     : $self->check_required_params(qw(dbname host port));
  Description : Checks $self->param to make sure the requested parameters
                have been set. Dies if parameters are missing.
  Return type : true on success
  Exceptions  : none
  Caller      : general

=cut

sub check_required_params {
439 440 441 442 443 444 445 446 447
  my ($self, @params) = @_;
  my @missing = ();
  foreach my $param (@params) {
    push @missing, $param unless $self->param($param);
  }
  if (@missing) {
    throw("Missing parameters: @missing.\nYou must specify them on the commandline or in your conffile.\n");
  }
  return(1);
448 449
}

450 451 452 453
=head2 user_proceed

  Arg[1]      : (optional) String $text - notification text to present to user
  Example     : # run a code snipped conditionally
454
                if ($support->user_proceed("Run the next code snipped?")) {
455 456
                    # run some code
                }
457

458 459
                # exit if requested by user
                exit unless ($support->user_proceed("Want to continue?"));
460
  Description : If running interactively, the user is asked if he wants to
461 462 463 464
                perform a script action. If he doesn't, this section is skipped
                and the script proceeds with the code. When running
                non-interactively, the section is run by default.
  Return type : TRUE to proceed, FALSE to skip.
465 466 467 468 469
  Exceptions  : none
  Caller      : general

=cut

470
sub user_proceed {
471 472 473 474 475 476 477 478 479 480
  my ($self, $text) = @_;

  if ($self->param('interactive')) {
    print "$text\n" if $text;
    print "[y/N] ";
    my $input = lc(<>);
    chomp $input;
    unless ($input eq 'y') {
      print "Skipping.\n";
      return(0);
481
    }
482
  }
483

484
  return(1);
485 486
}

487 488 489 490 491 492 493
=head2 user_confirm

  Description : DEPRECATED - please use user_proceed() instead

=cut

sub user_confirm {
494 495
  my $self = shift;
  exit unless $self->user_proceed("Continue?");
496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514
}

=head2 read_user_input

  Arg[1]      : (optional) String $text - notification text to present to user
  Example     : my $ret = $support->read_user_input("Choose a number [1/2/3]");
                if ($ret == 1) {
                    # do something
                } elsif ($ret == 2) {
                    # do something else
                }
  Description : If running interactively, the user is asked for input.
  Return type : String - user's input
  Exceptions  : none
  Caller      : general

=cut

sub read_user_input {
515
  my ($self, $text) = @_;
516

517 518 519 520 521 522
  if ($self->param('interactive')) {
    print "$text\n" if $text;
    my $input = <>;
    chomp $input;
    return $input;
  }
523 524
}

525 526 527 528 529 530 531 532 533 534 535 536 537 538 539
=head2 comma_to_list

  Arg[1-N]    : list of parameter names to parse
  Example     : $support->comma_to_list('chromosomes');
  Description : Transparently converts comma-separated lists into arrays (to
                allow different styles of commandline options, see perldoc
                Getopt::Long for details). Parameters are converted in place
                (accessible through $self->param('name')).
  Return type : true on success
  Exceptions  : none
  Caller      : general

=cut

sub comma_to_list {
540 541 542 543 544 545
  my $self = shift;
  foreach my $param (@_) {
    $self->param($param,
		 split (/,/, join (',', $self->param($param))));
  }
  return(1);
546 547
}

548
=head2 list_or_file
549 550

  Arg[1]      : Name of parameter to parse
551
  Example     : $support->list_or_file('gene_stable_id');
552 553 554 555 556 557 558 559 560
  Description : Determines whether a parameter holds a list or it is a filename
                to read the list entries from.
  Return type : true on success
  Exceptions  : thrown if list file can't be opened
  Caller      : general

=cut

sub list_or_file {
561 562 563 564 565 566 567 568 569 570 571 572
  my ($self, $param) = @_;
  my @vals = $self->param($param);
  return unless (@vals);

  my $firstval = $vals[0];
  if (scalar(@vals) == 1 && -e $firstval) {
    # we didn't get a list of values, but a file to read values from
    @vals = ();
    open(IN, $firstval) or throw("Cannot open $firstval for reading: $!");
    while(<IN>){
      chomp;
      push(@vals, $_);
573
    }
574 575 576 577 578
    close(IN);
    $self->param($param, @vals);
  }
  $self->comma_to_list($param);
  return(1);
579 580 581 582 583 584 585 586
}

=head2 param

  Arg[1]      : Parameter name
  Arg[2-N]    : (optional) List of values to set
  Example     : my $dbname = $support->param('dbname');
                $support->param('port', 3306);
587
                $support->param('chromosomes', 1, 6, 'X');
588 589 590 591 592 593 594 595 596 597
  Description : Getter/setter for parameters. Accepts single-value params and
                list params.
  Return type : Scalar value for single-value parameters, array of values for
                list parameters
  Exceptions  : thrown if no parameter name is supplied
  Caller      : general

=cut

sub param {
598 599
  my $self = shift;
  my $name = shift or throw("You must supply a parameter name");
600

601 602 603 604 605
  # setter
  if (@_) {
    if (scalar(@_) == 1) {
      # single value
      $self->{'_param'}->{$name} = shift;
606
    } else {
607 608 609
      # list of values
      undef $self->{'_param'}->{$name};
      @{ $self->{'_param'}->{$name} } = @_;
610
    }
611 612 613 614 615 616 617 618 619 620 621 622
  }

  # getter
  if (ref($self->{'_param'}->{$name}) eq 'ARRAY') {
    # list parameter
    return @{ $self->{'_param'}->{$name} };
  } elsif (defined($self->{'_param'}->{$name})) {
    # single-value parameter
    return $self->{'_param'}->{$name};
  } else {
    return ();
  }
623 624 625 626 627 628 629 630 631 632 633 634 635 636 637
}

=head2 error

  Arg[1]      : (optional) String - error message
  Example     : $support->error("An error occurred: $@");
                exit(0) if $support->error;
  Description : Getter/setter for error messages
  Return type : String - error message
  Exceptions  : none
  Caller      : general

=cut

sub error {
638 639 640
  my $self = shift;
  $self->{'_error'} = shift if (@_);
  return $self->{'_error'};
641 642 643 644 645 646 647 648 649 650 651 652 653 654
}

=head2 warnings

  Example     : print LOG "There were ".$support->warnings." warnings.\n";
  Description : Returns the number of warnings encountered while running the
                script (the warning counter is increased by $self->log_warning).
  Return type : Int - number of warnings
  Exceptions  : none
  Caller      : general

=cut

sub warnings {
655 656
  my $self = shift;
  return $self->{'_warnings'};
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
}

=head2 serverroot

  Arg[1]      : (optional) String - root directory of your ensembl sandbox
  Example     : my $serverroot = $support->serverroot;
  Description : Getter/setter for the root directory of your ensembl sandbox.
                This is set when ConversionSupport object is created, so
                usually only used as a getter.
  Return type : String - the server root directory
  Exceptions  : none
  Caller      : general

=cut

sub serverroot {
673 674 675
  my $self = shift;
  $self->{'_serverroot'} = shift if (@_);
  return $self->{'_serverroot'};
676 677 678 679 680 681
}

=head2 get_database

  Arg[1]      : String $database - the type of database to connect to
                (eg core, otter)
682 683
  Arg[2]      : (optional) String $prefix - the prefix used for retrieving the
                connection settings from the configuration
684 685 686 687 688 689 690 691 692
  Example     : my $db = $support->get_database('core');
  Description : Connects to the database specified.
  Return type : DBAdaptor of the appropriate type
  Exceptions  : thrown if asking for unknown database
  Caller      : general

=cut

sub get_database {
693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
  my $self = shift;
  my $database = shift or throw("You must provide a database");
  my $prefix = shift || '';
  $self->check_required_params(
    "${prefix}host",
    "${prefix}port",
    "${prefix}user",
    "${prefix}dbname",
  );
  my %adaptors = (
    core    => 'Bio::EnsEMBL::DBSQL::DBAdaptor',
    ensembl => 'Bio::EnsEMBL::DBSQL::DBAdaptor',
    evega   => 'Bio::EnsEMBL::DBSQL::DBAdaptor',
    otter   => 'Bio::Otter::DBSQL::DBAdaptor',
    vega    => 'Bio::Otter::DBSQL::DBAdaptor',
    compara => 'Bio::EnsEMBL::DBSQL::DBAdaptor',
    loutre  => 'Bio::Vega::DBSQL::DBAdaptor',
  );
  throw("Unknown database: $database") unless $adaptors{$database};

  $self->dynamic_use($adaptors{$database});
  my $dba = $adaptors{$database}->new(
    -host   => $self->param("${prefix}host"),
    -port   => $self->param("${prefix}port"),
    -user   => $self->param("${prefix}user"),
    -pass   => $self->param("${prefix}pass") || '',
    -dbname => $self->param("${prefix}dbname"),
    -group  => $database,
  );
  #can use this approach to get dna from another db
#  my $dna_db = $adaptors{$database}->new(
#    -host => 'otterlive',
#    -port => '3301',
#    -user => $self->param("${prefix}user"),
#    -pass => $self->param("${prefix}pass"),
#    -dbname => 'loutre_human',
#  );
#  $dba->dnadb($dna_db);

  # otherwise explicitely set the dnadb to itself - by default the Registry assumes
  # a group 'core' for this now
  $dba->dnadb($dba);

  $self->{'_dba'}->{$database} = $dba;
  $self->{'_dba'}->{'default'} = $dba unless $self->{'_dba'}->{'default'};
  return $self->{'_dba'}->{$database};
739 740
}

Patrick Meidl's avatar
Patrick Meidl committed
741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759

=head2 get_dbconnection

  Arg[1]      : (optional) String $prefix - the prefix used for retrieving the
                connection settings from the configuration
  Example     : my $dbh = $self->get_dbconnection;
  Description : Connects to the database server specified. You don't have to
                specify a database name (this is useful for running commands
                like $dbh->do('show databases')).
  Return type : DBI database handle
  Exceptions  : thrown if connection fails
  Caller      : general
  Status      : At Risk

=cut

sub get_dbconnection {
  my $self = shift;
  my $prefix = shift;
Web Admin's avatar
Web Admin committed
760
 
Patrick Meidl's avatar
Patrick Meidl committed
761 762 763 764 765 766 767 768 769 770 771 772 773 774
  $self->check_required_params(
      "${prefix}host",
      "${prefix}port",
      "${prefix}user",
  );

  my $dsn = "DBI:" . ($self->param('driver')||'mysql') .
            ":host=" . $self->param("${prefix}host") .
            ";port=" . $self->param("${prefix}port");

  if ($self->param("${prefix}dbname")) {
    $dsn .= ";dbname=".$self->param("${prefix}dbname");
  }

Steve Trevanion's avatar
Steve Trevanion committed
775 776
#  warn $dsn;

Patrick Meidl's avatar
Patrick Meidl committed
777 778 779 780 781 782 783 784 785
  my $dbh;
  eval{
    $dbh = DBI->connect($dsn, $self->param("${prefix}user"),
      $self->param("${prefix}pass"), {'RaiseError' => 1, 'PrintError' => 0});
  };

  if (!$dbh || $@ || !$dbh->ping) {
    $self->log_error("Could not connect to db server as user ".
      $self->param("${prefix}user") .
Steve Trevanion's avatar
Steve Trevanion committed
786
      " using [$dsn] as a locator:\n" . $DBI::errstr . $@);
Patrick Meidl's avatar
Patrick Meidl committed
787 788 789 790 791 792 793
  }

  $self->{'_dbh'} = $dbh;
  return $self->{'_dbh'};

}

794

795
=head2 dba
796

797 798 799 800 801
  Arg[1]      : (optional) String $database - type of db apaptor to retrieve
  Example     : my $dba = $support->dba;
  Description : Getter for database adaptor. Returns default (i.e. created
                first) db adaptor if no argument is provided.
  Return type : Bio::EnsEMBL::DBSQL::DBAdaptor or Bio::Otter::DBSQL::DBAdaptor
802 803 804 805 806
  Exceptions  : none
  Caller      : general

=cut

807
sub dba {
808 809
  my ($self, $database) = shift;
  return $self->{'_dba'}->{$database} || $self->{'_dba'}->{'default'};
810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825
}

=head2 dynamic_use

  Arg [1]    : String $classname - The name of the class to require/import
  Example    : $self->dynamic_use('Bio::EnsEMBL::DBSQL::DBAdaptor');
  Description: Requires and imports the methods for the classname provided,
               checks the symbol table so that it doesnot re-require modules
               that have already been required.
  Returntype : true on success
  Exceptions : Warns to standard error if module fails to compile
  Caller     : internal

=cut

sub dynamic_use {
826 827 828 829 830 831 832 833 834 835 836 837
  my ($self, $classname) = @_;
  my ($parent_namespace, $module) = $classname =~/^(.*::)(.*)$/ ? ($1,$2) : ('::', $classname);

  no strict 'refs';
  # return if module has already been imported
  return 1 if $parent_namespace->{$module.'::'} && %{ $parent_namespace->{$module.'::'}||{} };

  eval "require $classname";
  throw("Failed to require $classname: $@") if ($@);
  $classname->import();

  return 1;
838 839 840 841
}

=head2 get_chrlength

842 843
  Arg[1]      : (optional) Bio::EnsEMBL::DBSQL::DBAdaptor $dba
  Arg[2]      : (optional) String $version - coord_system version
Web Admin's avatar
Web Admin committed
844 845
  Arg[3]      : (optional) String $type - type of region eg chromsome (defaults to 'toplevel')
  Arg[4]      : (optional) Boolean - return non reference slies as well (required for haplotypes eq 6-COX)
846 847 848 849 850 851 852 853 854 855 856
  Example     : my $chr_length = $support->get_chrlength($dba);
  Description : Get all chromosomes and their length from the database. Return
                chr_name/length for the chromosomes the user requested (or all
                chromosomes by default)
  Return type : Hashref - chromosome_name => length
  Exceptions  : thrown if not passing a Bio::EnsEMBL::DBSQL::DBAdaptor
  Caller      : general

=cut

sub get_chrlength {
857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883
  my ($self, $dba, $version,$type,$include_non_reference) = @_;
  $dba ||= $self->dba;
  $type ||= 'toplevel';
  throw("get_chrlength should be passed a Bio::EnsEMBL::DBSQL::DBAdaptor\n")
    unless ($dba->isa('Bio::EnsEMBL::DBSQL::DBAdaptor'));

  my $sa = $dba->get_SliceAdaptor;

  my @chromosomes = map { $_->seq_region_name } 
    @{ $sa->fetch_all($type, $version,$include_non_reference) };
  my %chr = map { $_ => $sa->fetch_by_region($type, $_, undef, undef, undef, $version)->length } @chromosomes;

  my @wanted = $self->param('chromosomes');
  if (@wanted) {
    # check if user supplied invalid chromosome names
    foreach my $chr (@wanted) {
      my $found = 0;
      foreach my $chr_from_db (keys %chr) {
	if ($chr_from_db eq $chr) {
	  $found = 1;
	  last;
	}
      }
      unless ($found) {
	warning("Didn't find chromosome $chr in database " .
		  $self->param('dbname'));
      }
884 885
    }

886 887 888 889 890 891 892 893 894 895 896 897 898
    # filter to requested chromosomes only
  HASH:
    foreach my $chr_from_db (keys %chr) {
      foreach my $chr (@wanted) {
	if ($chr_from_db eq $chr) {
	  next HASH;
	}
      }
      delete($chr{$chr_from_db});
    }
  }

  return \%chr;
899 900
}

901 902 903 904 905 906 907 908 909 910 911 912 913 914
=head2 get_ensembl_chr_mapping

  Arg[1]      : (optional) Bio::EnsEMBL::DBSQL::DBAdaptor $dba
  Arg[2]      : (optional) String $version - coord_system version
  Example     : my $ensembl_mapping = $support->get_ensembl_chr_mapping($dba);
  Description : Gets a mapping between Vega chromosome names and their
                equivalent Ensembl chromosomes.
  Return type : Hashref - Vega name => Ensembl name
  Exceptions  : thrown if not passing a Bio::EnsEMBL::DBSQL::DBAdaptor
  Caller      : general

=cut

sub get_ensembl_chr_mapping {
Steve Trevanion's avatar
Steve Trevanion committed
915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
  my ($self, $dba, $version) = @_;
  $dba ||= $self->dba;
  throw("get_ensembl_chr_mapping should be passed a Bio::EnsEMBL::DBSQL::DBAdaptor\n") unless ($dba->isa('Bio::EnsEMBL::DBSQL::DBAdaptor'));

  my $sa = $dba->get_SliceAdaptor;
  my @chromosomes = map { $_->seq_region_name } 
    @{ $sa->fetch_all('chromosome', $version) };

  my %chrs;
  foreach my $chr (@chromosomes) {
    my $sr = $sa->fetch_by_region('chromosome', $chr, undef, undef, undef, $version);
    my ($ensembl_name_attr) = @{ $sr->get_all_Attributes('ensembl_name') };
    if ($ensembl_name_attr) {
      $chrs{$chr} = $ensembl_name_attr->value;
    } else {
      $chrs{$chr} = $chr;
931
    }
Steve Trevanion's avatar
Steve Trevanion committed
932 933
  }
  return \%chrs;
934 935
}

936 937 938 939 940 941 942 943 944 945 946 947
=head2 get_taxonomy_id

  Arg[1]      : Bio::EnsEMBL::DBSQL::DBAdaptor $dba
  Example     : my $sid = $support->get_taxonony_id($dba);
  Description : Retrieves the taxononmy ID from the meta table
  Return type : Int - the taxonomy ID
  Exceptions  : thrown if no taxonomy ID is found in the database
  Caller      : general

=cut

sub get_taxonomy_id {
948 949 950 951 952 953 954 955 956
  my ($self, $dba) = @_;
  $dba ||= $self->dba;
  my $sql = 'SELECT meta_value FROM meta WHERE meta_key = "species.taxonomy_id"';
  my $sth = $dba->dbc->db_handle->prepare($sql);
  $sth->execute;
  my ($tid) = $sth->fetchrow_array;
  $sth->finish;
  $self->throw("Could not determine taxonomy_id from database.") unless $tid;
  return $tid;
957 958 959 960 961 962 963 964 965
}

=head2 get_species_scientific_name

  Arg[1]      : Bio::EnsEMBL::DBSQL::DBAdaptor $dba
  Example     : my $species = $support->get_species_scientific_name($dba);
  Description : Retrieves the species scientific name (Genus species) from the
                meta table
  Return type : String - species scientific name
966
  Exceptions  : thrown if species name can not be determined from db
967 968 969 970 971
  Caller      : general

=cut

sub get_species_scientific_name {
972 973 974 975 976 977 978 979 980 981 982 983 984 985 986
  my ($self, $dba) = @_;
  $dba ||= $self->dba;
  my $sql_tmp = "SELECT meta_value FROM meta WHERE meta_key = \'species.classification\' ORDER BY meta_id";
  my $sql = $dba->dbc->add_limit_clause($sql_tmp,2);
  my $sth = $dba->dbc->db_handle->prepare($sql);
  $sth->execute;
  my @sp;
  while (my @row = $sth->fetchrow_array) {
    push @sp, $row[0];
  }
  $sth->finish;
  my $species = join(" ", reverse @sp);
  $self->throw("Could not determine species scientific name from database.")
    unless $species;
  return $species;
987 988
}

989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002
=head2 species

  Arg[1]      : (optional) String $species - species name to set
  Example     : my $species = $support->species;
                my $url = "http://vega.sanger.ac.uk/$species/";
  Description : Getter/setter for species name (Genus_species). If not set, it's
                determined from database's meta table
  Return type : String - species name
  Exceptions  : none
  Caller      : general

=cut

sub species {
1003 1004 1005 1006 1007 1008 1009 1010
  my $self = shift;
  $self->{'_species'} = shift if (@_);
  # get species name from database if not set
  unless ($self->{'_species'}) {
    $self->{'_species'} = join('_',
			       split(/ /, $self->get_species_scientific_name));
  }
  return $self->{'_species'};
1011 1012
}

1013 1014
=head2 sort_chromosomes

1015
  Arg[1]      : (optional) Hashref $chr_hashref - Hashref with chr_name as keys
1016 1017 1018
  Example     : my $chr = { '6-COX' => 1, '1' => 1, 'X' => 1 };
                my @sorted = $support->sort_chromosomes($chr);
  Description : Sorts chromosomes in an intuitive way (numerically, then
1019 1020
                alphabetically). If no chromosome hashref is passed, it's
                retrieve by calling $self->get_chrlength()
1021 1022 1023 1024 1025 1026 1027
  Return type : List - sorted chromosome names
  Exceptions  : thrown if no hashref is provided
  Caller      : general

=cut

sub sort_chromosomes {
1028 1029 1030 1031 1032
  my ($self, $chr_hashref) = @_;
  $chr_hashref = $self->get_chrlength unless ($chr_hashref);
  throw("You have to pass a hashref of your chromosomes")
    unless ($chr_hashref and ref($chr_hashref) eq 'HASH');
  return (sort _by_chr_num keys %$chr_hashref);
1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046
}

=head2 _by_chr_num

  Example     : my @sorted = sort _by_chr_num qw(X, 6-COX, 14, 7);
  Description : Subroutine to use in sort for sorting chromosomes. Sorts
                numerically, then alphabetically
  Return type : values to be used by sort
  Exceptions  : none
  Caller      : internal ($self->sort_chromosomes)

=cut

sub _by_chr_num {
1047 1048 1049 1050 1051 1052 1053
  my @awords = split /-/, $a;
  my @bwords = split /-/, $b;

  my $anum = $awords[0];
  my $bnum = $bwords[0];

  if ($anum !~ /^[0-9]*$/) {
1054
    if ($bnum !~ /^[0-9]*$/) {
1055 1056 1057
      return $anum cmp $bnum;
    } else {
      return 1;
1058
    }
1059 1060 1061 1062
  }
  if ($bnum !~ /^[0-9]*$/) {
    return -1;
  }
1063

1064 1065 1066 1067 1068 1069 1070
  if ($anum <=> $bnum) {
    return $anum <=> $bnum;
  } else {
    if ($#awords == 0) {
      return -1;
    } elsif ($#bwords == 0) {
      return 1;
1071
    } else {
1072
      return $awords[1] cmp $bwords[1];
1073
    }
1074
  }
1075 1076
}

1077 1078 1079 1080
=head2 split_chromosomes_by_size

  Arg[1]      : (optional) Int $cutoff - the cutoff in bp between small and
                large chromosomes
1081 1082
  Arg[2]      : (optional) Boolean to include duplicate regions, ie PAR or not
                (default is no)
1083
  Arg[3]      : (optional) Coordsystem version to retrieve
1084

1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106
  Example     : my $chr_slices = $support->split_chromosomes_by_size;
                foreach my $block_size (keys %{ $chr_slices }) {
                    print "Chromosomes with blocksize $block_size: ";
                    print join(", ", map { $_->seq_region_name }
                                        @{ $chr_slices->{$block_size} });
                }
  Description : Determines block sizes for storing DensityFeatures on
                chromosomes, and return slices for each chromosome. The block
                size is determined so that you have 150 bins for the smallest
                chromosome over 5 Mb in length. For chromosomes smaller than 5
                Mb, an additional smaller block size is used to yield 150 bins
                for the overall smallest chromosome. This will result in
                reasonable resolution for small chromosomes and high
                performance for big ones.
  Return type : Hashref (key: block size; value: Arrayref of chromosome
                Bio::EnsEMBL::Slices)
  Exceptions  : none
  Caller      : density scripts

=cut

sub split_chromosomes_by_size {
1107 1108 1109
  my $self   = shift;
  my $cutoff = shift || 5000000;
  my $dup    = shift || 0;
1110
  my $cs_version = shift;
1111 1112 1113 1114 1115
  my $slice_adaptor = $self->dba->get_SliceAdaptor;
  my $top_slices;
  if ($self->param('chromosomes')) {
    foreach my $chr ($self->param('chromosomes')) {
      push @{ $top_slices }, $slice_adaptor->fetch_by_region('chromosome', $chr);
1116
    }
1117
  } else {
1118
    $top_slices = $slice_adaptor->fetch_all('chromosome',$cs_version,0,$dup);
1119
  }
1120

1121 1122 1123 1124 1125 1126 1127 1128 1129
  my ($big_chr, $small_chr, $min_big_chr, $min_small_chr);
  foreach my $slice (@{ $top_slices }) {
    next if ($slice->length eq 10000); #hack for chrY pseudoslice
    if ($slice->length < $cutoff) {
      if (! $min_small_chr or ($min_small_chr > $slice->length)) {
	$min_small_chr = $slice->length;
      }
      # push small chromosomes onto $small_chr
      push @{ $small_chr }, $slice;
1130
    }
Steve Trevanion's avatar
tidier  
Steve Trevanion committed
1131
    elsif (! $min_big_chr or ($min_big_chr > $slice->length) ){
1132 1133 1134 1135 1136 1137
      $min_big_chr = $slice->length;
    }
    # push _all_ chromosomes onto $big_chr
    push @{ $big_chr }, $slice;
  }
  my $chr_slices;
Steve Trevanion's avatar
tidier  
Steve Trevanion committed
1138
  $chr_slices->{int($min_big_chr/150)}   = $big_chr if $min_big_chr;
1139 1140
  $chr_slices->{int($min_small_chr/150)} = $small_chr if $min_small_chr;
  return $chr_slices;
1141 1142
}

1143 1144 1145 1146
=head2 log

  Arg[1]      : String $txt - the text to log
  Arg[2]      : Int $indent - indentation level for log message
1147
  Example     : my $log = $support->log_filehandle;
1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
                $support->log('Log foo.\n', 1);
  Description : Logs a message to the filehandle initialised by calling
                $self->log_filehandle(). You can supply an indentation level
                to get nice hierarchical log messages.
  Return type : true on success
  Exceptions  : thrown when no filehandle can be obtained
  Caller      : general

=cut

sub log {
1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169
  my ($self, $txt, $indent) = @_;
  $indent ||= 0;

  # strip off leading linebreaks so that indenting doesn't break
  $txt =~ s/^(\n*)//;

  $txt = $1."    "x$indent . $txt;
  my $fh = $self->{'_log_filehandle'};
  throw("Unable to obtain log filehandle") unless $fh;
  print $fh "$txt";
  return(1);
1170 1171 1172 1173 1174 1175
}

=head2 log_warning

  Arg[1]      : String $txt - the warning text to log
  Arg[2]      : Int $indent - indentation level for log message
Steve Trevanion's avatar
Steve Trevanion committed
1176
  Arg[3]      : Bool - add a line break before warning if true
1177
  Example     : my $log = $support->log_filehandle;
1178 1179 1180 1181 1182 1183 1184 1185 1186
                $support->log_warning('Log foo.\n', 1);
  Description : Logs a message via $self->log and increases the warning counter.
  Return type : true on success
  Exceptions  : none
  Caller      : general

=cut

sub log_warning {
1187 1188 1189 1190 1191 1192
  my ($self, $txt, $indent, $break) = @_;
  $txt = "WARNING: " . $txt;
  $txt = "\n$txt" if ($break);
  $self->log($txt, $indent);
  $self->{'_warnings'}++;
  return(1);
1193 1194
}

1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208
=head2 log_error

  Arg[1]      : String $txt - the error text to log
  Arg[2]      : Int $indent - indentation level for log message
  Example     : my $log = $support->log_filehandle;
                $support->log_error('Log foo.\n', 1);
  Description : Logs a message via $self->log and exits the script.
  Return type : none
  Exceptions  : none
  Caller      : general

=cut

sub log_error {
1209 1210 1211 1212 1213
  my ($self, $txt, $indent) = @_;
  $txt = "ERROR: ".$txt;
  $self->log($txt, $indent);
  $self->log("Exiting.\n");
  exit;
1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229
}

=head2 log_verbose

  Arg[1]      : String $txt - the warning text to log
  Arg[2]      : Int $indent - indentation level for log message
  Example     : my $log = $support->log_filehandle;
                $support->log_verbose('Log this verbose message.\n', 1);
  Description : Logs a message via $self->log if --verbose option was used
  Return type : TRUE on success, FALSE if not verbose
  Exceptions  : none
  Caller      : general

=cut

sub log_verbose {
1230 1231 1232 1233
  my ($self, $txt, $indent) = @_;
  return(0) unless $self->param('verbose');
  $self->log($txt, $indent);
  return(1);
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250
}

=head2 log_stamped

  Arg[1]      : String $txt - the warning text to log
  Arg[2]      : Int $indent - indentation level for log message
  Example     : my $log = $support->log_filehandle;
                $support->log_stamped('Log this stamped message.\n', 1);
  Description : Appends timestamp and memory usage to a message and logs it via
                $self->log
  Return type : TRUE on success
  Exceptions  : none
  Caller      : general

=cut

sub log_stamped {
1251 1252 1253 1254 1255 1256
  my ($self, $txt, $indent) = @_;
  # append timestamp and memory usage to log text
  $txt =~ s/(\n*)$//;
  $txt .= " ".$self->date_and_mem.$1;
  $self->log($txt, $indent);
  return(1);
1257 1258
}

1259 1260
=head2 log_filehandle

1261 1262
  Arg[1]      : (optional) String $mode - file access mode
  Example     : my $log = $support->log_filehandle;
1263 1264 1265 1266 1267 1268
                # print to the filehandle
                print $log 'Lets start logging...\n';
                # log via the wrapper $self->log()
                $support->log('Another log message.\n');
  Description : Returns a filehandle for logging (STDERR by default, logfile if
                set from config or commandline). You can use the filehandle
1269 1270 1271 1272
                directly to print to, or use the smart wrapper $self->log().
                Logging mode (truncate or append) can be set by passing the
                mode as an argument to log_filehandle(), or with the
                --logappend commandline option (default: truncate)
1273 1274 1275 1276 1277 1278 1279
  Return type : Filehandle - the filehandle to log to
  Exceptions  : thrown if logfile can't be opened
  Caller      : general

=cut

sub log_filehandle {
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290
  my ($self, $mode) = @_;
  $mode ||= '>';
  $mode = '>>' if ($self->param('logappend'));
  my $fh = \*STDERR;
  if (my $logfile = $self->param('logfile')) {
    if (my $logpath = $self->param('logpath')) {
      unless (-e $logpath) {
	system("mkdir $logpath") == 0 or
	  $self->log_error("Can't create log dir $logpath: $!\n");
      }
      $logfile = "$logpath/$logfile";
1291
    }
1292 1293 1294 1295 1296
    open($fh, "$mode", $logfile) or throw(
      "Unable to open $logfile for writing: $!");
  }
  $self->{'_log_filehandle'} = $fh;
  return $self->{'_log_filehandle'};
1297 1298
}

1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314
=head2 filehandle

  Arg[1]      : String $mode - file access mode
  Arg[2]      : String $file - input or output file
  Example     : my $fh = $support->filehandle('>>', '/path/to/file');
                # print to the filehandle
                print $fh 'Your text goes here...\n';
  Description : Returns a filehandle (*STDOUT for writing, *STDIN for reading
                by default) to print to or read from.
  Return type : Filehandle - the filehandle
  Exceptions  : thrown if file can't be opened
  Caller      : general

=cut

sub filehandle {
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326
  my ($self, $mode, $file) = @_;
  $mode ||= ">";
  my $fh;
  if ($file) {
    open($fh, "$mode", $file) or throw(
      "Unable to open $file for writing: $!");
  } elsif ($mode =~ />/) {
    $fh = \*STDOUT;
  } elsif ($mode =~ /</) {
    $fh = \*STDIN;
  }
  return $fh;
1327 1328
}

1329 1330
=head2 init_log

1331 1332 1333 1334 1335 1336
  Example     : $support->init_log;
  Description : Opens a filehandle to the logfile and prints some header
                information to this file. This includes script name, date, user
                running the script and parameters the script will be running
                with.
  Return type : Filehandle - the log filehandle
1337 1338 1339 1340 1341 1342
  Exceptions  : none
  Caller      : general

=cut

sub init_log {