Skip to content
Snippets Groups Projects
ConversionSupport.pm 55 KiB
Newer Older
=head1 LICENSE

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

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

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

=head1 CONTACT

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

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

=cut

=head1 NAME

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

=head1 SYNOPSIS

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

  # parse common options
  $support->parse_common_options;

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

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

  # see individual method documentation for more stuff

=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

=cut

package Bio::EnsEMBL::Utils::ConversionSupport;

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);
use Cwd qw(abs_path);
use DBI;
use Data::Dumper;

my $species_c = 1; #counter to be used for each database connection made

=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 {
  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;
}

=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 {
  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',
               'nolog|nolog=s',
	       'logpath=s',
               'log_base_path=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);
    }
    $self->param('conffile', $conffile);
  }
  elsif ($conffile) {
    warning("Unable to open configuration file $conffile for reading: $!");
  }

# override configured parameter with commandline options
  map { $self->param($_, $h{$_}) } keys %h;

  # if logpath & logfile are not set, set them here to /ensemblweb/vega_dev/shared/logs/conversion/DBNAME/SCRIPNAME_NN.log

  if (defined($self->param('log_base_path')))  {
    my $dbname = $self->param('dbname');
    $dbname =~ s/^vega_//;
    if (not (defined($self->param('logpath')))){
      $self->param('logpath', $self->param('log_base_path')."/".$dbname."/" );
    }
    if (  (not defined $self->param('logfile') ) && (not defined $self->param('nolog') )  ){
      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");
    }
  }
  
  return(1);
}

=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 {
201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726
  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);
}

=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 {
  my $self = shift;

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

  # getter
  if (ref($self->{'_allowed_params'}) eq 'ARRAY') {
    return @{ $self->{'_allowed_params'} };
  } else {
    return ();
  }
}

=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 {
  return qw(
	    conffile
	    dbname
	    host
	    port
	    user
	    pass
            nolog
	    logpath
            log_base_path
	    logfile
	    logappend
	    verbose
	    interactive
	    dry_run
	  );
}

=head2 get_loutre_params

  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
  Return type : Array - list of common parameters
  Exceptions  : none
  Caller      : general

=cut

sub get_loutre_params {
  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
	    );
  }
}

=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 {
  my $self = shift;
  foreach my $param (qw(dbname host port user pass)) {
    $self->{'_param'}{$param} = undef;
  }
}

=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 {
  my $self = shift;

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

  if ($self->param('host') eq 'ensdb-1-10') {
    # ask user if he wants to proceed
    exit unless $self->user_proceed("**************\n\n You're working on ensdb-1-10! Is that correct and you want to continue ?\n\n**************");
  }
  else {
    # ask user if he wants to proceed
    exit unless $self->user_proceed("Continue?");
  }
  return(1);
}

=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 {
  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) {
    my @vals = $self->param($key);
    if (@vals) {
      $txt .= Text::Wrap::wrap( sprintf('   %-21s', $key),
				' 'x24,
				join(", ", @vals)
			      ) . "\n";
    }
  }
  $txt .= "\n";
  return $txt;
}

=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 {
  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;
      }
    }
  }

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

  # 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;
}

=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 {
  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);
}

=head2 user_proceed

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

                # exit if requested by user
                exit unless ($support->user_proceed("Want to continue?"));
  Description : If running interactively, the user is asked if he wants to
                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.
  Exceptions  : none
  Caller      : general

=cut

sub user_proceed {
  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);
    }
  }

  return(1);
}

=head2 user_confirm

  Description : DEPRECATED - please use user_proceed() instead

=cut

sub user_confirm {
  my $self = shift;
  exit unless $self->user_proceed("Continue?");
}

=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 {
  my ($self, $text) = @_;

  if ($self->param('interactive')) {
    print "$text\n" if $text;
    my $input = <>;
    chomp $input;
    return $input;
  }
}

=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 {
  my $self = shift;
  foreach my $param (@_) {
    $self->param($param,
		 split (/,/, join (',', $self->param($param))));
  }
  return(1);
}

=head2 list_or_file

  Arg[1]      : Name of parameter to parse
  Example     : $support->list_or_file('gene_stable_id');
  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 {
  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, $_);
    }
    close(IN);
    $self->param($param, @vals);
  }
  $self->comma_to_list($param);
  return(1);
}

=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);
                $support->param('chromosomes', 1, 6, 'X');
  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 {
  my $self = shift;
  my $name = shift or throw("You must supply a parameter name");

  # setter
  if (@_) {
    if (scalar(@_) == 1) {
      # single value
      $self->{'_param'}->{$name} = shift;
    } else {
      # list of values
      undef $self->{'_param'}->{$name};
      @{ $self->{'_param'}->{$name} } = @_;
    }
  }

  # get
Loading full blame...