diff options
| author | Jeroen Vermeulen <jtv@precisiontranslationtools.com> | 2015-05-17 16:04:04 +0300 |
|---|---|---|
| committer | Jeroen Vermeulen <jtv@precisiontranslationtools.com> | 2015-05-17 16:04:04 +0300 |
| commit | a25193cc5da6a271c4676e843db80ee19221822a (patch) | |
| tree | db764696f2e663016a9ab87a95846501c585b4aa /scripts/ems | |
| parent | 108da163745699e724f581c3d33b1e8a5e26a9e9 (diff) | |
Fix a lot of lint, mostly trailing whitespace.
This is lint reported by the new lint-checking functionality in beautify.py.
(We can change to a different lint checker if we have a better one, but it
would probably still flag these same problems.)
Lint checking can help a lot, but only if we get the lint under control.
Diffstat (limited to 'scripts/ems')
48 files changed, 634 insertions, 634 deletions
diff --git a/scripts/ems/experiment.perl b/scripts/ems/experiment.perl index 5d68e409c..62b039124 100755 --- a/scripts/ems/experiment.perl +++ b/scripts/ems/experiment.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl # Experiment Management System # Documentation at http://www.statmt.org/moses/?n=FactoredTraining.EMS @@ -30,7 +30,7 @@ my ($CONFIG_FILE, $DELETE_CRASHED, $DELETE_VERSION ); - + my $SLEEP = 2; my $META = "$RealBin/experiment.meta"; @@ -63,7 +63,7 @@ die("experiment.perl -config config-file [-exec] [-no-graph]") 'no-graph' => \$NO_GRAPH); if (! -e "steps") { `mkdir -p steps`; } -die("error: could not find config file") +die("error: could not find config file") unless ($CONFIG_FILE && -e $CONFIG_FILE) || ($CONTINUE && -e &steps_file("config.$CONTINUE",$CONTINUE)) || ($DELETE_CRASHED && -e &steps_file("config.$DELETE_CRASHED",$DELETE_CRASHED)) || @@ -159,7 +159,7 @@ exit(); # graph that depicts steps of the experiment, with depedencies sub init_agenda_graph() { - my $dir = &check_and_get("GENERAL:working-dir"); + my $dir = &check_and_get("GENERAL:working-dir"); my $graph_file = &steps_file("graph.$VERSION",$VERSION); open(PS,">".$graph_file.".ps") or die "Cannot open: $!"; @@ -208,7 +208,7 @@ sub detect_if_cluster { $CLUSTER = 1; print "running on a cluster\n" if $CLUSTER; } - } + } } sub detect_if_multicore { @@ -230,7 +230,7 @@ sub read_meta { my ($module,$step); while(<META>) { s/\#.*$//; # strip comments - next if /^\s*$/; + next if /^\s*$/; while (/\\\s*$/) { $_ .= <META>; s/\s*\\\s*[\n\r]*\s+/ /; @@ -369,7 +369,7 @@ sub read_config { else { print STDERR "BUGGY CONFIG LINE ($line_count): $_"; $error++; - } + } } } die("$error ERROR".(($error>1)?"s":"")." IN CONFIG FILE") if $error; @@ -406,9 +406,9 @@ sub read_config { s/\$\{$pattern\}/$o/ if $escaped; s/\$$pattern/$o/ unless $escaped; print "$_\n" if $VERBOSE; - if (/\$/) { + if (/\$/) { print "more resolving needed\n" if $VERBOSE; - $resolve = 1; + $resolve = 1; } } } @@ -419,12 +419,12 @@ sub read_config { # check if specified files exist $error = 0; foreach my $parameter (keys %CONFIG) { - foreach (@{$CONFIG{$parameter}}) { + foreach (@{$CONFIG{$parameter}}) { next if $parameter =~ /temp-dir/; next if (!/^\// || -e); # ok if not file, or exists - my $file = $_; + my $file = $_; $file =~ s/ .+$//; # remove switches - my $gz = $file; $gz =~ s/\.gz$//; + my $gz = $file; $gz =~ s/\.gz$//; next if -e $gz; # ok if non gzipped exists next if `find $file* -maxdepth 0 -follow`; # ok if stem print STDERR "$parameter: file $_ does not exist!\n"; @@ -515,15 +515,15 @@ sub find_steps_for_module { print "\t\tproduces $out\n" if $VERBOSE; next unless defined($NEEDED{$out}) || (defined($FINAL_STEP) && $FINAL_STEP eq $step); print "\t\tneeded\n" if $VERBOSE; - - # if output of a step is specified, you do not have + + # if output of a step is specified, you do not have # to execute that step if(defined($CONFIG{$out})) { $GIVEN{$out} = $step; next; } print "\t\toutput not specified in config\n" if $VERBOSE; - + # not needed, if optional and not specified if (defined($STEP_IGNORE{$defined_step})) { my $next = 0; @@ -560,16 +560,16 @@ sub find_steps_for_module { # OK, add step to the list - push @DO_STEP,$step; + push @DO_STEP,$step; $STEP_LOOKUP{$step} = $#DO_STEP; print "\tdo-step: $step\n" if $VERBOSE; - - # mark as pass step (where no action is taken), if step is + + # mark as pass step (where no action is taken), if step is # optional and nothing needs to be do done if (defined($STEP_PASS{$defined_step})) { my $flag = 1; foreach my $pass (@{$STEP_PASS{$defined_step}}) { - $flag = 0 + $flag = 0 if &backoff_and_get(&extend_local_name($module,$set,$pass)); } $PASS{$#DO_STEP}++ if $flag; @@ -578,12 +578,12 @@ sub find_steps_for_module { if (defined($STEP_PASS_IF{$defined_step})) { my $flag = 0; foreach my $pass (@{$STEP_PASS_IF{$defined_step}}) { - $flag = 1 + $flag = 1 if &backoff_and_get(&extend_local_name($module,$set,$pass)); } $PASS{$#DO_STEP}++ if $flag; } - + # special case for passing: steps that only affect factor 0 if (defined($ONLY_FACTOR_0{$defined_step})) { my $FACTOR = &backoff_and_get_array("LM:$set:factors"); @@ -625,7 +625,7 @@ sub find_steps_for_module { print "\n\t\tcross-directed to $in\n" if $VERBOSE; } elsif(defined($CONFIG{$in})) { - print "\n\t\t... but that is specified\n" if $VERBOSE; + print "\n\t\t... but that is specified\n" if $VERBOSE; } else { push @{$NEEDED{$in}}, $#DO_STEP; @@ -639,11 +639,11 @@ sub find_steps_for_module { sub check_producability { my ($module,$set,$output) = @_; - + # find $output requested as input by step in $module/$set my @OUT = &construct_input($module,$set,$output); - - # if multiple outputs (due to multiple sets merged into one), + + # if multiple outputs (due to multiple sets merged into one), # only one needs to exist foreach my $out (@OUT) { print "producable? $out\n" if $VERBOSE; @@ -687,12 +687,12 @@ sub construct_input { # potentially multiple input files my @IN; - + # input from same module if ($in !~ /([^:]+):(\S+)/) { push @IN, &construct_name($module,$set,$in); } - + # input from previous model, multiple elsif ($MODULE_TYPE{$1} eq "multiple") { my @SETS = &get_sets($1); @@ -711,7 +711,7 @@ sub construct_input { else { push @IN,$in; } - + return @IN; } @@ -750,7 +750,7 @@ sub delete_version { # check which versions are already deleted my %ALREADY_DELETED; - my $dir = &check_and_get("GENERAL:working-dir"); + my $dir = &check_and_get("GENERAL:working-dir"); open(VERSION,"ls $dir/steps/*/deleted.* 2>/dev/null|"); while(<VERSION>) { /deleted\.(\d+)/; @@ -787,7 +787,7 @@ sub delete_version { chomp($step_file); my $step = &get_step_from_step_file($step_file); next if $USED_BY_OTHERS{$step}; - &delete_step($step,$DELETE_VERSION); + &delete_step($step,$DELETE_VERSION); } # orphan killing: delete steps in deleted versions, if they were only preserved because this version needed them @@ -808,12 +808,12 @@ sub get_step_from_step_file { $step =~ s/_/:/g; return $step; } - + sub delete_step { my ($step_name,$version) = @_; my ($module,$set,$step) = &deconstruct_name($step_name); - my $step_file = &versionize(&step_file2($module,$set,$step),$version); + my $step_file = &versionize(&step_file2($module,$set,$step),$version); print "delete step $step_file\n"; `rm $step_file $step_file.*` if $EXECUTE; @@ -839,7 +839,7 @@ sub delete_output { if (-e $file) { print "\tdelete file $file\n"; `rm $file` if $EXECUTE; - } + } # delete files that have additional extension $file =~ /^(.+)\/([^\/]+)$/; my ($dir,$f) = ($1,$2); @@ -861,7 +861,7 @@ sub delete_output { # RE-USE # look for completed step jobs from previous experiments sub find_re_use { - my $dir = &check_and_get("GENERAL:working-dir"); + my $dir = &check_and_get("GENERAL:working-dir"); return unless -e "$dir/steps"; for(my $i=0;$i<=$#DO_STEP;$i++) { @@ -953,13 +953,13 @@ sub find_re_use { $change = 1; print "\tpassed step $DO_STEP[$_] used in re-use run $run -> fail\n" if $VERBOSE; } - } + } } # re-use step has to exist for this run if (! defined($RE_USE[$parent]{$reuse_run})) { print "\tno previous step -> fail\n" if $VERBOSE; delete($RE_USE[$i]{$run}); - $change = 1; + $change = 1; } } } @@ -967,7 +967,7 @@ sub find_re_use { } } - # summarize and convert hashes into integers for to be re-used + # summarize and convert hashes into integers for to be re-used print "\nSTEP SUMMARY:\n"; open(RE_USE,">".&steps_file("re-use.$VERSION",$VERSION)) or die "Cannot open: $!"; for(my $i=$#DO_STEP;$i>=0;$i--) { @@ -998,7 +998,7 @@ sub find_dependencies { } for(my $i=0;$i<=$#DO_STEP;$i++) { my $step = $DO_STEP[$i]; - $step =~ /^(.+:)[^:]+$/; + $step =~ /^(.+:)[^:]+$/; my $module_set = $1; foreach my $needed_by (@{$NEEDED{$module_set.$STEP_OUT{&defined_step($step)}}}) { print "$needed_by needed by $i\n" if $VERBOSE; @@ -1020,9 +1020,9 @@ sub draw_agenda_graph { print DOT " ranksep=0;\n"; for(my $i=0;$i<=$#DO_STEP;$i++) { my $step = $DO_STEP[$i]; - $step =~ /^(.+):[^:]+$/; + $step =~ /^(.+):[^:]+$/; my $module_set = $1; - push @{$M{$module_set}},$i; + push @{$M{$module_set}},$i; } my $i = 0; my (@G,%GIVEN_NUMBER); @@ -1049,7 +1049,7 @@ sub draw_agenda_graph { } else { my $step = $DO_STEP[$i]; - $step =~ s/^.+:([^:]+)$/$1/; + $step =~ s/^.+:([^:]+)$/$1/; $step .= " (".$RE_USE[$i].")" if $RE_USE[$i]; my $color = "green"; @@ -1058,7 +1058,7 @@ sub draw_agenda_graph { $color = "lightblue" if $RE_USE[$i] && $RE_USE[$i] != $VERSION; $color = "red" if defined($CRASHED{$i}); $color = "lightyellow" if defined($PASS{$i}); - + print DOT " $i [label=\"$step\",shape=box,fontsize=10,height=0,style=filled,fillcolor=\"$color\"];\n"; } } @@ -1087,7 +1087,7 @@ sub draw_agenda_graph { sub define_step { my ($step) = @_; - my $dir = &check_and_get("GENERAL:working-dir"); + my $dir = &check_and_get("GENERAL:working-dir"); `mkdir -p $dir` if ! -e $dir; my @STEP; if ($step eq "all") { @@ -1104,14 +1104,14 @@ sub define_step { next if &define_template($i); if ($DO_STEP[$i] =~ /^CORPUS:(.+):factorize$/) { &define_corpus_factorize($i); - } + } elsif ($DO_STEP[$i] eq 'SPLITTER:train') { &define_splitter_train($i); - } + } elsif ($DO_STEP[$i] =~ /^LM:(.+):factorize$/) { &define_lm_factorize($i,$1); } - elsif ($DO_STEP[$i] =~ /^LM:(.+):randomize$/ || + elsif ($DO_STEP[$i] =~ /^LM:(.+):randomize$/ || $DO_STEP[$i] eq 'INTERPOLATED-LM:randomize') { &define_lm_randomize($i,$1); } @@ -1178,7 +1178,7 @@ sub define_step { } elsif ($DO_STEP[$i] eq 'TUNING:factorize-input') { &define_tuningevaluation_factorize($i); - } + } elsif ($DO_STEP[$i] eq 'TUNING:factorize-input-devtest') { &define_tuningevaluation_factorize($i); } @@ -1193,7 +1193,7 @@ sub define_step { } elsif ($DO_STEP[$i] =~ /^EVALUATION:(.+):factorize-input$/) { &define_tuningevaluation_factorize($i); - } + } elsif ($DO_STEP[$i] =~ /^EVALUATION:(.+):filter$/) { &define_tuningevaluation_filter($1,$i); } @@ -1225,7 +1225,7 @@ sub define_step { } } -# LOOP that executes the steps +# LOOP that executes the steps # including checks, if needed to be executed, waiting for completion, and error detection sub execute_steps { @@ -1266,7 +1266,7 @@ sub execute_steps { print "number of steps doable or running: ".(scalar keys %DO)." at ".`date`; foreach my $step (keys %DO) { print "\t".($DO{$step}==2?"running: ":"doable: ").$DO_STEP[$step]."\n"; } return unless scalar keys %DO; - + # execute new step my $done = 0; foreach my $i (keys %DO) { @@ -1284,7 +1284,7 @@ sub execute_steps { # cluster job submission if ($CLUSTER && (!&is_qsub_script($i) || (&backoff_and_get($DO_STEP[$i].":jobs") && (&backoff_and_get($DO_STEP[$i].":jobs")==1)))) { $DO{$i}++; - my $qsub_args = &get_qsub_args($DO_STEP[$i]); + my $qsub_args = &get_qsub_args($DO_STEP[$i]); print "\texecuting $step via qsub $qsub_args ($active active)\n"; my $qsub_command="qsub $qsub_args -S /bin/bash -e $step.STDERR -o $step.STDOUT $step"; print "\t$qsub_command\n" if $VERBOSE; @@ -1306,8 +1306,8 @@ sub execute_steps { } # update state - &draw_agenda_graph() unless $done; - + &draw_agenda_graph() unless $done; + # sleep until one more step is done while(! $done) { sleep($SLEEP); @@ -1328,7 +1328,7 @@ sub execute_steps { } } `touch $running_file`; - } + } } } @@ -1356,17 +1356,17 @@ sub get_qsub_args { # instead of submited as jobs. here we check for that. sub is_qsub_script { my ($i) = @_; - return (defined($QSUB_STEP{$i}) || + return (defined($QSUB_STEP{$i}) || defined($QSUB_SCRIPT{&defined_step($DO_STEP[$i])})); } -# write the info file that is consulted to check if +# write the info file that is consulted to check if # a steps has to be redone, even if it was run before sub write_info { my ($i) = @_; my $step = $DO_STEP[$i]; my $module_set = $step; $module_set =~ s/:[^:]+$//; - + open(INFO,">".&versionize(&step_file($i)).".INFO") or die "Cannot open: $!"; my %VALUE = &get_parameters_relevant_for_re_use($i); foreach my $parameter (keys %VALUE) { @@ -1426,19 +1426,19 @@ sub check_info { if (defined($ONLY_EXISTENCE_MATTERS{"$module:$step"}{$parameter})) { print "existence ok\n" if $VERBOSE; } - elsif (&match_info_strings($VALUE{$parameter},$INFO{$parameter})) { - print "ok\n" if $VERBOSE; + elsif (&match_info_strings($VALUE{$parameter},$INFO{$parameter})) { + print "ok\n" if $VERBOSE; } - else { + else { print "mismatch\n" if $VERBOSE; - return 0; + return 0; } } print "\tall parameters match\n" if $VERBOSE; return 1; } -sub match_info_strings { +sub match_info_strings { my ($current,$old) = @_; $current =~ s/ $//; $old =~ s/ $//; @@ -1479,7 +1479,7 @@ sub get_parameters_relevant_for_re_use { my ($out,@INPUT) = &get_output_and_input($i); my $actually_used = "USED"; foreach my $in_file (@INPUT) { - $actually_used .= " ".$in_file; + $actually_used .= " ".$in_file; } $VALUE{"INPUT"} = $actually_used; @@ -1591,7 +1591,7 @@ sub step_file2 { my ($module,$set,$step) = @_; my $dir = &check_and_get("GENERAL:working-dir"); `mkdir -p $dir/steps` if ! -e "$dir/steps"; - my $file = "$dir/steps/$module" . ($set ? ("_".$set) : "") . "_$step"; + my $file = "$dir/steps/$module" . ($set ? ("_".$set) : "") . "_$step"; return $file; } @@ -1609,7 +1609,7 @@ sub defined_step_id { sub defined_step { my ($step) = @_; - my $defined_step = $step; + my $defined_step = $step; $defined_step =~ s/:.+:/:/; return $defined_step; } @@ -1659,7 +1659,7 @@ sub define_corpus_factorize { my ($output,$input) = &get_output_and_input($step_id); my $input_extension = &check_backoff_and_get("TRAINING:input-extension"); my $output_extension = &check_backoff_and_get("TRAINING:output-extension"); - + my $dir = &check_and_get("GENERAL:working-dir"); my $temp_dir = &check_and_get("INPUT-FACTOR:temp-dir") . ".$VERSION"; my $cmd = "mkdir -p $temp_dir\n" @@ -1673,7 +1673,7 @@ sub define_corpus_factorize { "$output.$output_extension", &check_backoff_and_get_array("TRAINING:output-factors"), $step_id); - + &create_step($step_id,$cmd); } @@ -1689,7 +1689,7 @@ sub define_tuningevaluation_factorize { . &factorize_one_language("INPUT-FACTOR",$input,$output, &check_backoff_and_get_array("TRAINING:input-factors"), $step_id); - + &create_step($step_id,$cmd); } @@ -1700,12 +1700,12 @@ sub define_lm_factorize { my ($output,$input) = &get_output_and_input($step_id); print "LM:$set:factors\n" if $VERBOSE; my $factor = &check_backoff_and_get_array("LM:$set:factors"); - + my $dir = &check_and_get("GENERAL:working-dir"); my $temp_dir = &check_and_get("INPUT-FACTOR:temp-dir") . ".$VERSION"; my $cmd = "mkdir -p $temp_dir\n" . &factorize_one_language("OUTPUT-FACTOR",$input,$output,$factor,$step_id); - + &create_step($step_id,$cmd); } @@ -1715,12 +1715,12 @@ sub define_interpolated_lm_factorize_tuning { my ($output,$input) = &get_output_and_input($step_id); my $factor = &check_backoff_and_get_array("TRAINING:output-factors"); - + my $dir = &check_and_get("GENERAL:working-dir"); my $temp_dir = &check_and_get("INPUT-FACTOR:temp-dir") . ".$VERSION"; my $cmd = "mkdir -p $temp_dir\n" . &factorize_one_language("OUTPUT-FACTOR",$input,$output,$factor,$step_id); - + &create_step($step_id,$cmd); } @@ -1732,7 +1732,7 @@ sub define_splitter_train { my $output_splitter = &get("GENERAL:output-splitter"); my $input_extension = &check_backoff_and_get("SPLITTER:input-extension"); my $output_extension = &check_backoff_and_get("SPLITTER:output-extension"); - + my $cmd = ""; if ($input_splitter) { $cmd .= "$input_splitter -train -model $output.$input_extension -corpus $input.$input_extension\n"; @@ -1747,7 +1747,7 @@ sub define_splitter_train { sub define_lm_train_randomized { my ($step_id,$set) = @_; my $training = &check_backoff_and_get("LM:$set:rlm-training"); - my $order = &check_backoff_and_get("LM:$set:order"); + my $order = &check_backoff_and_get("LM:$set:order"); my ($output,$input) = &get_output_and_input($step_id); $output =~ /^(.+)\/([^\/]+)$/; @@ -1765,7 +1765,7 @@ sub define_lm_randomize { my ($module,$set,$stepname) = &deconstruct_name($DO_STEP[$step_id]); my $randomizer = &check_backoff_and_get("$module:$set:lm-randomizer"); - my $order = &check_backoff_and_get("$module:$set:order"); + my $order = &check_backoff_and_get("$module:$set:order"); my ($output,$input) = &get_output_and_input($step_id); $output =~ /^(.+)\/([^\/]+)$/; @@ -1782,7 +1782,7 @@ sub factorize_one_language { my $temp_dir = &check_and_get("INPUT-FACTOR:temp-dir") . ".$VERSION"; my $parallelizer = &get("GENERAL:generic-parallelizer"); my ($module,$set,$stepname) = &deconstruct_name($DO_STEP[$step_id]); - + my ($cmd,$list) = (""); foreach my $factor (@{$FACTOR}) { if ($factor eq "word") { @@ -1791,7 +1791,7 @@ sub factorize_one_language { else { my $script = &check_and_get("$type:$factor:factor-script"); my $out = "$outfile.$factor"; - if ($parallelizer && defined($PARALLELIZE{&defined_step($DO_STEP[$step_id])}) + if ($parallelizer && defined($PARALLELIZE{&defined_step($DO_STEP[$step_id])}) && ( (&get("$module:jobs") && $CLUSTER) || (&get("$module:cores") && $MULTICORE))) { my $subdir = $module; @@ -1803,7 +1803,7 @@ sub factorize_one_language { $qflags="--queue-flags \"$qsub_args\"" if ($CLUSTER && $qsub_args); $cmd .= "$parallelizer $qflags -in $infile -out $out -cmd '$script %s %s $temp_dir/$subdir' -jobs ".&get("$module:jobs")." -tmpdir $temp_dir/$subdir\n"; $QSUB_STEP{$step_id}++; - } + } elsif ($MULTICORE) { $cmd .= "$parallelizer -in $infile -out $out -cmd '$script %s %s $temp_dir/$subdir' -cores ".&get("$module:cores")." -tmpdir $temp_dir/$subdir\n"; } @@ -1825,9 +1825,9 @@ sub define_tuning_tune { my $use_mira = &backoff_and_get("TUNING:use-mira", 0); my $word_alignment = &backoff_and_get("TRAINING:include-word-alignment-in-rules"); my $tmp_dir = &get_tmp_file("TUNING","","tune"); - - # the last 3 variables are only used for mira tuning - my ($tuned_config,$config,$input,$reference,$config_devtest,$input_devtest,$reference_devtest, $filtered_config) = &get_output_and_input($step_id); + + # the last 3 variables are only used for mira tuning + my ($tuned_config,$config,$input,$reference,$config_devtest,$input_devtest,$reference_devtest, $filtered_config) = &get_output_and_input($step_id); $config = $filtered_config if $filtered_config; @@ -1853,13 +1853,13 @@ sub define_tuning_tune { my $mira_config = "$tmp_dir/mira-config.$VERSION."; my $mira_config_log = $mira_config."log"; $mira_config .= "cfg"; - + write_mira_config($mira_config,$tmp_dir,$config,$input,$reference,$config_devtest,$input_devtest,$reference_devtest); #$cmd = "$tuning_script -config $mira_config -exec >& $mira_config_log"; # we want error messages in top-level log file $cmd = "$tuning_script -config $mira_config -exec "; - # write script to select the best set of weights after training for the specified number of epochs --> + # write script to select the best set of weights after training for the specified number of epochs --> # cp to tuning/tmp.?/moses.ini my $script_filename = "$tmp_dir/selectBestWeights."; my $script_filename_log = $script_filename."log"; @@ -1881,7 +1881,7 @@ sub define_tuning_tune { my $decoder_settings = &backoff_and_get("TUNING:decoder-settings"); $decoder_settings = "" unless $decoder_settings; $decoder_settings .= " -v 0 " unless $CLUSTER && $jobs && $jobs>1; - + my $tuning_settings = &backoff_and_get("TUNING:tuning-settings"); $tuning_settings = "" unless $tuning_settings; @@ -1890,7 +1890,7 @@ sub define_tuning_tune { $cmd .= " --continue" if $tune_continue; $cmd .= " --skip-decoder" if $skip_decoder; $cmd .= " --inputtype $tune_inputtype" if defined($tune_inputtype); - + my $qsub_args = &get_qsub_args($DO_STEP[$step_id]); $cmd .= " --queue-flags=\"$qsub_args\"" if ($CLUSTER && $qsub_args); $cmd .= " --jobs $jobs" if $CLUSTER && $jobs && $jobs>1; @@ -1898,14 +1898,14 @@ sub define_tuning_tune { $tuning_dir =~ s/\/[^\/]+$//; $cmd .= "\nmkdir -p $tuning_dir"; } - + $cmd .= "\ncp $tmp_dir/moses.ini $tuned_config"; &create_step($step_id,$cmd); } sub write_mira_config { - my ($config_filename,$expt_dir,$tune_filtered_ini,$input,$reference,$devtest_filtered_ini,$input_devtest,$reference_devtest) = @_; + my ($config_filename,$expt_dir,$tune_filtered_ini,$input,$reference,$devtest_filtered_ini,$input_devtest,$reference_devtest) = @_; my $moses_src_dir = &check_and_get("GENERAL:moses-src-dir"); my $mira_src_dir = &backoff_and_get("GENERAL:mira-src-dir"); my $tuning_decoder_settings = &check_and_get("TUNING:decoder-settings"); @@ -1916,7 +1916,7 @@ sub write_mira_config { my $use_jackknife = &backoff_and_get("TUNING:use-jackknife"); # are we tuning a meta feature? - my $tune_meta_feature = &backoff_and_get("TUNING:tune-meta-feature"); + my $tune_meta_feature = &backoff_and_get("TUNING:tune-meta-feature"); my $tune_filtered_ini_start; if (!$use_jackknife) { @@ -1927,13 +1927,13 @@ sub write_mira_config { # apply start weights to filtered ini file, and pass the new ini to mira print "DEBUG: $RealBin/support/substitute-weights.perl $start_weights $tune_filtered_ini $tune_filtered_ini_start \n"; system("$RealBin/support/substitute-weights.perl $start_weights $tune_filtered_ini $tune_filtered_ini_start"); - } + } } # do we want to continue an interrupted experiment? my $continue_expt = &backoff_and_get("TUNING:continue-expt"); my $continue_epoch = &backoff_and_get("TUNING:continue-epoch"); - my $continue_weights = &backoff_and_get("TUNING:continue-weights"); + my $continue_weights = &backoff_and_get("TUNING:continue-weights"); # mira config file open(CFG, ">$config_filename"); @@ -1956,7 +1956,7 @@ sub write_mira_config { print CFG "tune-meta-feature=1 \n" if ($tune_meta_feature); print CFG "jackknife=1 \n" if ($use_jackknife); print CFG "wait-for-bleu=1 \n\n"; - #print CFG "decoder-settings=".$tuning_decoder_settings."\n\n"; + #print CFG "decoder-settings=".$tuning_decoder_settings."\n\n"; print CFG "[train] \n"; print CFG "trainer=\${moses-home}/bin/mira \n"; if ($use_jackknife) { @@ -1972,7 +1972,7 @@ sub write_mira_config { } else { print CFG $input.".only$i, " if $i<9; - print CFG $input.".only$i" if $i==9; + print CFG $input.".only$i" if $i==9; } } print CFG "\n"; @@ -1999,14 +1999,14 @@ sub write_mira_config { print CFG "moses-ini-file=".$tune_filtered_ini."\n"; } } - print CFG "decoder-settings=".$tuning_decoder_settings." -text-type \"dev\"\n"; - print CFG "hours=48 \n"; + print CFG "decoder-settings=".$tuning_decoder_settings." -text-type \"dev\"\n"; + print CFG "hours=48 \n"; if ($parallel_settings) { foreach my $setting (split(" ", $parallel_settings)) { print CFG $setting."\n"; } } - print CFG "extra-args=".$tuning_settings."\n\n"; + print CFG "extra-args=".$tuning_settings."\n\n"; print CFG "[devtest] \n"; if (&get("TRAINING:hierarchical-rule-set")) { print CFG "moses=\${moses-home}/bin/moses_chart \n"; @@ -2019,7 +2019,7 @@ sub write_mira_config { print CFG "input-file=".$input_devtest."\n"; print CFG "reference-file=".$reference_devtest."\n"; print CFG "moses-ini-file=".$devtest_filtered_ini."\n"; - print CFG "decoder-settings=".$tuning_decoder_settings." -text-type \"devtest\"\n"; + print CFG "decoder-settings=".$tuning_decoder_settings." -text-type \"devtest\"\n"; print CFG "hours=12 \nextra-args= \nskip-dev=1 \nskip-devtest=0 \nskip-submit=0 \n"; close(CFG); } @@ -2052,7 +2052,7 @@ sub write_selectBestMiraWeights { print SCR "} \n\n"; print SCR "print STDERR \"Best weights according to BLEU on devtest set: \$best_weights \\n\"; \n"; print SCR "system(\"cp \$best_weights $weight_out_file\"); \n\n"; - + close(SCR); system("chmod u+x $script_filename"); } @@ -2118,7 +2118,7 @@ sub define_training_symmetrize_giza { my $method = &check_and_get("TRAINING:alignment-symmetrization-method"); my $cmd = &get_training_setting(3); my $alignment_stem = &versionize(&long_file_name("aligned","model","")); - + $cmd .= "-giza-e2f $giza -giza-f2e $giza_inv "; $cmd .= "-alignment-file $aligned "; $cmd .= "-alignment-stem $alignment_stem "; @@ -2129,17 +2129,17 @@ sub define_training_symmetrize_giza { sub define_training_build_suffix_array { my ($step_id) = @_; - + my $scripts = &check_and_get("GENERAL:moses-script-dir"); - + my ($model, $aligned,$corpus) = &get_output_and_input($step_id); my $sa_exec_dir = &check_and_get("TRAINING:suffix-array"); my $input_extension = &check_backoff_and_get("TRAINING:input-extension"); my $output_extension = &check_backoff_and_get("TRAINING:output-extension"); my $method = &check_and_get("TRAINING:alignment-symmetrization-method"); - + my $glue_grammar_file = &versionize(&long_file_name("glue-grammar","model","")); - + my $cmd = "$scripts/training/wrappers/adam-suffix-array/suffix-array-create.sh $sa_exec_dir $corpus.$input_extension $corpus.$output_extension $aligned.$method $model $glue_grammar_file"; &create_step($step_id,$cmd); @@ -2222,7 +2222,7 @@ sub define_training_extract_phrases { $cmd .= "-alignment-stem $alignment_stem "; $cmd .= "-extract-file $extract "; $cmd .= "-corpus $corpus "; - + if (&get("TRAINING:hierarchical-rule-set")) { my $no_glue_grammar = &get("TRAINING:no-glue-grammar"); if (!defined($no_glue_grammar) || $no_glue_grammar eq "false") { @@ -2327,7 +2327,7 @@ sub define_training_build_ttable { $cmd .= "-ghkm-parts-of-speech-file $parts_of_speech_labels_file "; } } - + &create_step($step_id,$cmd); } @@ -2346,7 +2346,7 @@ sub define_domain_feature_score_option { else { return "-score-options '--Domain$method $domains' "; } -} +} sub define_training_build_reordering { my ($step_id) = @_; @@ -2406,8 +2406,8 @@ sub define_training_sigtest_filter { chop($filtered_table); } $raw_table =~ s/\s*\-\S+\s*//; # remove switch - $filtered_table =~ s/\s*\-\S+\s*//; - + $filtered_table =~ s/\s*\-\S+\s*//; + my $cmd = "zcat $raw_table.gz | $moses_src_dir/contrib/sigtest-filter/filter-pt -e $suffix_array.$output_extension -f $suffix_array.$input_extension $sigtest_filter $hierarchical_flag | gzip - > $filtered_table.gz\n"; &create_step($step_id,$cmd); } @@ -2461,7 +2461,7 @@ sub get_config_tables { # additional settings for hierarchical models my $extract_version = $VERSION; if (&get("TRAINING:hierarchical-rule-set")) { - $extract_version = $RE_USE[$STEP_LOOKUP{"TRAINING:extract-phrases"}] + $extract_version = $RE_USE[$STEP_LOOKUP{"TRAINING:extract-phrases"}] if defined($STEP_LOOKUP{"TRAINING:extract-phrases"}); my $no_glue_grammar = &get("TRAINING:no-glue-grammar"); if (!defined($no_glue_grammar) || $no_glue_grammar eq "false") { @@ -2506,10 +2506,10 @@ sub define_training_create_config { if($transliteration_pt){ $cmd .= "-transliteration-phrase-table $transliteration_pt "; - } + } if ($osm) { - my $osm_settings = &get("TRAINING:operation-sequence-model-settings"); + my $osm_settings = &get("TRAINING:operation-sequence-model-settings"); if ($osm_settings =~ /-factor *(\S+)/){ $cmd .= "-osm-model $osm/ -osm-setting $1 "; } @@ -2547,7 +2547,7 @@ sub define_training_create_config { $type = 5 if (&get("INTERPOLATED-LM:rlm") || &backoff_and_get("INTERPOLATED-LM:lm-randomizer")); - # manually set type + # manually set type $type = &get("INTERPOLATED-LM:type") if &get("INTERPOLATED-LM:type"); # go through each interpolated language model @@ -2588,7 +2588,7 @@ sub define_training_create_config { &backoff_and_get("LM:$set:rlm-training") || &backoff_and_get("LM:$set:lm-randomizer")); - # manually set type + # manually set type $type = &backoff_and_get("LM:$set:type") if (&backoff_and_get("LM:$set:type")); # binarized by INTERPOLATED-LM @@ -2596,7 +2596,7 @@ sub define_training_create_config { $lm_file =~ s/\.lm/\.binlm/; $type = 1; $type = &get("INTERPOLATED-LM:type") if &get("INTERPOLATED-LM:type"); - } + } # which factor is the model trained on? my $factor = 0; @@ -2631,7 +2631,7 @@ sub define_interpolated_lm_interpolate { } } - # go through language models by factor and order + # go through language models by factor and order my ($icount,$ILM_SETS) = &get_interpolated_lm_sets(); foreach my $factor (keys %{$ILM_SETS}) { foreach my $order (keys %{$$ILM_SETS{$factor}}) { @@ -2643,8 +2643,8 @@ sub define_interpolated_lm_interpolate { foreach my $id_set (@{$$ILM_SETS{$factor}{$order}}) { my ($id,$set) = split(/ /,$id_set,2); $lm_list .= $LM[$id].","; - if (defined($weights)) { - die("ERROR: no interpolation weight set for $factor:$order:$set (factor:order:set)") + if (defined($weights)) { + die("ERROR: no interpolation weight set for $factor:$order:$set (factor:order:set)") unless defined($WEIGHT{"$factor:$order:$set"}); $weight_list .= $WEIGHT{"$factor:$order:$set"}.","; } @@ -2708,7 +2708,7 @@ sub define_interpolated_lm_process { my $tool = &check_backoff_and_get("INTERPOLATED-LM:lm-${stepname}r"); my $FACTOR = &backoff_and_get_array("TRAINING:output-factors"); - # go through language models by factor and order + # go through language models by factor and order my ($icount,$ILM_SETS) = &get_interpolated_lm_sets(); my $cmd = ""; foreach my $factor (keys %{$ILM_SETS}) { @@ -2729,7 +2729,7 @@ sub define_interpolated_lm_process { $name = "$interpolated_lm$suffix"; $name_processed = "$processed_lm$suffix"; } - $cmd .= "$tool $name $name_processed\n"; + $cmd .= "$tool $name $name_processed\n"; } } @@ -2768,7 +2768,7 @@ sub get_interpolated_lm_sets { my $icount=0; foreach my $set (@LM_SETS) { my $order = &check_backoff_and_get("LM:$set:order"); - + my $factor = 0; if (&backoff_and_get("TRAINING:output-factors") && &backoff_and_get("LM:$set:factors")) { @@ -2873,7 +2873,7 @@ sub get_table_name_settings { push @NAME,"$default.$f"; # push @NAME,"$dir/model/$table.$VERSION.$f"; } - + # get specified names, if any if (&get("TRAINING:$table")) { my @SPECIFIED_NAME = @{$CONFIG{"TRAINING:$table"}}; @@ -2890,7 +2890,7 @@ sub get_table_name_settings { $cmd .= "-$table $name "; } return $cmd; -} +} sub get_factor_id { my ($type) = @_; @@ -2908,7 +2908,7 @@ sub encode_factor_definition { my $encoded; foreach my $mapping (split(/,\s*/,$definition)) { my ($in,$out) = split(/\s*->\s*/,$mapping); - $encoded .= + $encoded .= &encode_factor_list($IN,$in)."-". &encode_factor_list($OUT,$out)."+"; } @@ -2941,22 +2941,22 @@ sub define_tuningevaluation_filter { $binarizer = &backoff_and_get("EVALUATION:$set:ttable-binarizer") unless $tuning_flag; $binarizer = &backoff_and_get("TUNING:ttable-binarizer") if $tuning_flag; my $report_precision_by_coverage = !$tuning_flag && &backoff_and_get("EVALUATION:$set:report-precision-by-coverage"); - - # occasionally, lattices and conf nets need to be able - # to filter phrase tables, we can provide sentences/ngrams + + # occasionally, lattices and conf nets need to be able + # to filter phrase tables, we can provide sentences/ngrams # in a separate file my $input_filter; $input_filter = &get("EVALUATION:$set:input-filter") unless $tuning_flag; $input_filter = &get("TUNING:input-filter") if $tuning_flag; #print "filter: $input_filter \n"; $input_filter = $input unless $input_filter; - + my $settings = &backoff_and_get("EVALUATION:$set:filter-settings") unless $tuning_flag; $settings = &backoff_and_get("TUNING:filter-settings") if $tuning_flag; $settings = "" unless $settings; $binarizer .= " -no-alignment-info" if defined ($binarizer) && !$hierarchical && defined $word_alignment && $word_alignment eq "no"; - + $settings .= " -Binarizer \"$binarizer\"" if $binarizer; $settings .= " --Hierarchical" if $hierarchical; @@ -2992,13 +2992,13 @@ sub define_tuningevaluation_filter { $config = $tuning_flag ? "$dir/tuning/moses.table.ini.$VERSION" : "$dir/evaluation/$set.moses.table.ini.$VERSION"; $cmd = "touch $config\n"; $delete_config = 1; - + $cmd .= &get_config_tables($config,$reordering_table,$phrase_translation_table,undef,$domains); if (&get("TRAINING:in-decoding-transliteration")) { $cmd .= "-transliteration-phrase-table $dir/model/transliteration-phrase-table.$VERSION "; - } + } $cmd .= "-lm 0:3:$config:8\n"; # dummy kenlm 3-gram model on factor 0 @@ -3009,12 +3009,12 @@ sub define_tuningevaluation_filter { if ($sa_exec_dir) { # suffix array $cmd .= "$scripts/training/wrappers/adam-suffix-array/suffix-array-extract.sh $sa_exec_dir $phrase_translation_table $input_filter $filter_dir $sa_extractors \n"; - + my $escaped_filter_dir = $filter_dir; $escaped_filter_dir =~ s/\//\\\\\//g; $cmd .= "cat $config | sed s/10\\ 0\\ 0\\ 7.*/10\\ 0\\ 0\\ 7\\ $escaped_filter_dir/g > $filter_dir/moses.ini \n"; - # kind of a hack -- the correct thing would be to make the generation of the config file ($filter_dir/moses.ini) - # set the PhraseDictionaryALSuffixArray's path to the filtered directory rather than to the suffix array itself + # kind of a hack -- the correct thing would be to make the generation of the config file ($filter_dir/moses.ini) + # set the PhraseDictionaryALSuffixArray's path to the filtered directory rather than to the suffix array itself $cmd .= "sed -i 's%path=$phrase_translation_table%path=$filter_dir%' $filter_dir/moses.ini\n"; } else { @@ -3022,7 +3022,7 @@ sub define_tuningevaluation_filter { $cmd .= "$scripts/training/filter-model-given-input.pl"; $cmd .= " $filter_dir $config $input_filter $settings\n"; } - + # clean-up $cmd .= "rm $config" if $delete_config; @@ -3033,7 +3033,7 @@ sub define_evaluation_decode { my ($set,$step_id) = @_; my $scripts = &check_and_get("GENERAL:moses-script-dir"); my $dir = &check_and_get("GENERAL:working-dir"); - + my ($system_output, $config,$input,$filtered_config) = &get_output_and_input($step_id); $config = $filtered_config if $filtered_config; @@ -3051,12 +3051,12 @@ sub define_evaluation_decode { my $hierarchical = &get("TRAINING:hierarchical-rule-set"); my $word_alignment = &backoff_and_get("TRAINING:include-word-alignment-in-rules"); my $post_decoding_transliteration = &get("TRAINING:post-decoding-transliteration"); - - # If Transliteration Module is to be used as post-decoding step ... + + # If Transliteration Module is to be used as post-decoding step ... if (defined($post_decoding_transliteration) && $post_decoding_transliteration eq "yes"){ $settings .= " -output-unknowns $system_output.oov"; } - + # specify additional output for analysis if (defined($report_precision_by_coverage) && $report_precision_by_coverage eq "yes") { @@ -3095,7 +3095,7 @@ sub define_evaluation_decode { $cmd .= "mkdir -p $dir/evaluation/tmp.$set.$VERSION\n"; $cmd .= "cd $dir/evaluation/tmp.$set.$VERSION\n"; if (defined $moses_parallel) { - $cmd .= $moses_parallel; + $cmd .= $moses_parallel; } else { $cmd .= "$scripts/generic/moses-parallel.pl"; } @@ -3105,7 +3105,7 @@ sub define_evaluation_decode { $cmd .= " -config $config"; $cmd .= " -input-file $input"; $cmd .= " --jobs $jobs"; - $cmd .= " -decoder-parameters \"$settings\" > $system_output"; + $cmd .= " -decoder-parameters \"$settings\" > $system_output"; $cmd .= " -n-best-file $system_output.best$nbest_size -n-best-size $nbest" if $nbest; } else { @@ -3252,12 +3252,12 @@ sub define_reporting_report { my $scripts = &check_and_get("GENERAL:moses-script-dir"); my $cmd = "$scripts/ems/support/report-experiment-scores.perl"; - + # get scores that were produced foreach my $parent (@{$DEPENDENCY[$step_id]}) { - my ($parent_module,$parent_set,$parent_step) + my ($parent_module,$parent_set,$parent_step) = &deconstruct_name($DO_STEP[$parent]); - + my $file = &get_default_file($parent_module,$parent_set,$parent_step); $cmd .= " set=$parent_set,type=$parent_step,file=$file"; } @@ -3282,7 +3282,7 @@ sub get_output_and_input { my $output = &get_default_file(&deconstruct_name($step)); my @INPUT; - if (defined($USES_INPUT{$step_id})) { + if (defined($USES_INPUT{$step_id})) { for(my $i=0; $i<scalar @{$USES_INPUT{$step_id}}; $i++) { # get name of input file needed my $in_file = $USES_INPUT{$step_id}[$i]; @@ -3293,9 +3293,9 @@ sub get_output_and_input { my $prev_step = ""; # print "\tlooking up in_file $in_file\n"; foreach my $parent (@{$DEPENDENCY[$step_id]}) { - my ($parent_module,$parent_set,$parent_step) + my ($parent_module,$parent_set,$parent_step) = &deconstruct_name($DO_STEP[$parent]); - my $parent_file + my $parent_file = &construct_name($parent_module,$parent_set, $STEP_OUT{&defined_step($DO_STEP[$parent])}); if ($in_file eq $parent_file) { @@ -3367,12 +3367,12 @@ sub define_template { if ($single_cmd =~ /^ln /) { $new_cmd .= $single_cmd."\n"; } - elsif ($single_cmd =~ /^.+$/) { + elsif ($single_cmd =~ /^.+$/) { # find IN and OUT files $single_cmd =~ /(EMS_IN_EMS\S*)/ || die("ERROR: could not find EMS_IN_EMS in $single_cmd"); my $in = $1; - $single_cmd =~ /(EMS_OUT_EMS\S*)/ + $single_cmd =~ /(EMS_OUT_EMS\S*)/ || die("ERROR: could not find OUT in $single_cmd"); my $out = $1; # replace IN and OUT with %s @@ -3388,13 +3388,13 @@ sub define_template { my $qsub_args = &get_qsub_args($DO_STEP[$step_id]); $qflags="--queue-flags \"$qsub_args\"" if ($CLUSTER && $qsub_args); $new_cmd .= "$parallelizer $qflags -in $in -out $out -cmd '$single_cmd' -jobs ".&get("$module:jobs")." -tmpdir $dir/$tmp_dir\n"; - } + } if ($MULTICORE) { $new_cmd .= "$parallelizer -in $in -out $out -cmd '$single_cmd' -cores ".&get("$module:cores")." -tmpdir $dir/$tmp_dir\n"; } } } - + $cmd = $new_cmd; $QSUB_STEP{$step_id}++; } @@ -3468,12 +3468,12 @@ sub create_step { my $file = &versionize(&step_file2($module,$set,$step)); my $dir = &check_and_get("GENERAL:working-dir"); my $subdir = $module; - $subdir =~ tr/A-Z/a-z/; + $subdir =~ tr/A-Z/a-z/; $subdir = "evaluation" if $subdir eq "reporting"; $subdir = "lm" if $subdir eq "interpolated-lm"; open(STEP,">$file") or die "Cannot open: $!"; print STEP "#!/bin/bash\n\n"; - print STEP "PATH=\"".$ENV{"PATH"}."\"\n"; + print STEP "PATH=\"".$ENV{"PATH"}."\"\n"; print STEP "cd $dir\n"; print STEP "echo 'starting at '`date`' on '`hostname`\n"; print STEP "mkdir -p $dir/$subdir\n\n"; @@ -3481,7 +3481,7 @@ sub create_step { print STEP "echo 'finished at '`date`\n"; print STEP "touch $file.DONE\n"; close(STEP); -} +} sub get { return &check_and_get($_[0],"allow_undef"); @@ -3541,7 +3541,7 @@ sub check_backoff_and_get_array { sub get_specified_or_default_file { my ($specified_module,$specified_set,$specified_parameter, $default_module, $default_set, $default_step) = @_; - my $specified = + my $specified = &construct_name($specified_module,$specified_set,$specified_parameter); if (defined($CONFIG{$specified})) { print "\t\texpanding $CONFIG{$specified}[0]\n" if $VERBOSE; @@ -3624,7 +3624,7 @@ sub long_file_name { $file = "$dir/$file"; } - my $module_working_dir_parameter = + my $module_working_dir_parameter = $module . ($set ne "" ? ":$set" : "") . ":working-dir"; if (defined($CONFIG{$module_working_dir_parameter})) { @@ -3634,7 +3634,7 @@ sub long_file_name { } sub compute_version_number { - my $dir = &check_and_get("GENERAL:working-dir"); + my $dir = &check_and_get("GENERAL:working-dir"); $VERSION = 1; return unless -e $dir; open(LS,"find $dir/steps -maxdepth 1 -follow |"); diff --git a/scripts/ems/fix-info.perl b/scripts/ems/fix-info.perl index 8f83d4ccf..abe58fe83 100755 --- a/scripts/ems/fix-info.perl +++ b/scripts/ems/fix-info.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -8,7 +8,7 @@ $step = "*" unless defined($step); die("fix-info.perl file [step]") unless defined($file); die("file not found") unless -e $file; -die("full path!") unless $file =~ /^\//; +die("full path!") unless $file =~ /^\//; my @filestat = stat($file); my $newtime = $filestat[9]; @@ -21,7 +21,7 @@ while(my $info = <LS>) { if (/$file .*\[/) { $changed++; s/($file) (.*\[)\d+/$1 $2$newtime/g; - } + } } if ($changed) { print "updating $info\n"; diff --git a/scripts/ems/support/analysis.perl b/scripts/ems/support/analysis.perl index cea2657c9..f4d5a55b4 100755 --- a/scripts/ems/support/analysis.perl +++ b/scripts/ems/support/analysis.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -29,7 +29,7 @@ if (!&GetOptions('system=s' => \$system, # raw output from decoder 'search-graph=s' => \$search_graph, # visualization of search graph 'hierarchical' => \$hierarchical) || # hierarchical model? !defined($dir)) { - die("ERROR: syntax: analysis.perl -system FILE -reference FILE -dir DIR [-input FILE] [-input-corpus FILE] [-ttable FILE] [-score-options SETTINGS] [-segmentation FILE] [-output-corpus FILE] [-alignment-file FILE] [-biconcor BIN]"); + die("ERROR: syntax: analysis.perl -system FILE -reference FILE -dir DIR [-input FILE] [-input-corpus FILE] [-ttable FILE] [-score-options SETTINGS] [-segmentation FILE] [-output-corpus FILE] [-alignment-file FILE] [-biconcor BIN]"); } `mkdir -p $dir`; @@ -95,7 +95,7 @@ if (defined($segmentation)) { # coverage analysis my (%INPUT_PHRASE,%CORPUS_COVERED,%TTABLE_COVERED,%TTABLE_ENTROPY); -if (!defined($coverage_dir) && (defined($ttable) || defined($corpus))) { +if (!defined($coverage_dir) && (defined($ttable) || defined($corpus))) { if (!defined($input)) { die("ERROR: when specifying either ttable or input-corpus, please also specify input\n"); } @@ -170,7 +170,7 @@ sub input_phrases { $line = &get_factor_phrase($factor,$line); &extract_n_grams($line,\%INPUT_PHRASE); } - close(INPUT); + close(INPUT); } # reduce a factorized phrase into the factors of interest @@ -279,11 +279,11 @@ sub bleu_annotation { } if (ref($REFERENCE[$i]) eq 'ARRAY') { foreach my $ref (@{$REFERENCE[$i]}) { - print OUT "\t".$ref; + print OUT "\t".$ref; } } else { - print OUT "\t".$REFERENCE[$i] + print OUT "\t".$REFERENCE[$i] } print OUT "\n"; } @@ -301,7 +301,7 @@ sub add_match { my $ref_count = 0; $ref_count = $REF_NGRAM{$length}{$ngram} if defined($REF_NGRAM{$length}{$ngram}); my $match_count = ($sys_count > $ref_count) ? $ref_count : $sys_count; - + $$CORRECT{$length}{$ngram} += $match_count; $$TOTAL{$length}{$ngram} += $sys_count; #print "$length:$ngram $sys_count $ref_count\n"; @@ -345,7 +345,7 @@ sub ttable_coverage { # handling hierarchical $in =~ s/ \[[^ \]]+\]$//; # remove lhs nt next if $in =~ /\[[^ \]]+\]\[[^ \]]+\]/; # only consider flat rules - $in = &get_factor_phrase($factor,$in) if defined($factor) && $factor eq "0"; + $in = &get_factor_phrase($factor,$in) if defined($factor) && $factor eq "0"; $scores = $COLUMN[4] if defined($hierarchical); #scalar @COLUMN == 5; my @IN = split(/ /,$in); $size = scalar @IN; @@ -473,7 +473,7 @@ sub input_annotation { #$ttable_entropy = 0 unless defined($ttable_entropy); $ttable_covered = 0 unless defined($ttable_covered); $corpus_covered = 0 unless defined($corpus_covered); - + if (defined($TTABLE_COVERED{$length}{$phrase})) { printf OUT "%d-%d:%d:%d:%.5f ",$start,$start+$length-1,$corpus_covered,$ttable_covered,$ttable_entropy; } @@ -481,7 +481,7 @@ sub input_annotation { } print OUT "\n"; } - close(INPUT); + close(INPUT); close(OUT); } @@ -532,7 +532,7 @@ sub extract_n_grams { $sentence =~ s/\s+/ /g; $sentence =~ s/^ //; $sentence =~ s/ $//; - + my @WORD = split(/ /,$sentence); for(my $length=1;$length<=$MAX_LENGTH;$length++) { for(my $i=0;$i<=scalar(@WORD)-$length;$i++) { @@ -604,8 +604,8 @@ sub precision_by_coverage { defined($REF_NGRAM{1}{$ngram})) { my $ref_count = $REF_NGRAM{1}{$ngram}; my $sys_count = $SYS_NGRAM{1}{$ngram}; - $PREC_NGRAM{1}{$ngram} = - ($ref_count >= $sys_count) ? 1 : $ref_count/$sys_count; + $PREC_NGRAM{1}{$ngram} = + ($ref_count >= $sys_count) ? 1 : $ref_count/$sys_count; } } close(REPORT); @@ -615,10 +615,10 @@ sub precision_by_coverage { while($line =~ /([^|]+) \|(\d+)\-(\d+)\|\s*(.*)$/) { my ($output,$from,$to) = ($1,$2,$3); $line = $4; - + # bug fix: 1-1 unknown word mappings get alignment point if ($from == $to && # one - scalar(split(/ /,$output)) == 1 && # to one + scalar(split(/ /,$output)) == 1 && # to one !defined($ALIGNED{$from})) { # but not aligned push @{$ALIGNED{$from}},$output_pos; } @@ -631,11 +631,11 @@ sub precision_by_coverage { my ($precision,$deleted,$length) = (0,0,0); - # unaligned? note as deleted + # unaligned? note as deleted if (!defined($ALIGNED{$i})) { $deleted = 1; } - # aligned + # aligned else { foreach my $o (@{$ALIGNED{$i}}) { $precision += $PREC_NGRAM{1}{$OUTPUT[$o]}; @@ -649,12 +649,12 @@ sub precision_by_coverage { $DELETED_BY_WORD{$word} += $deleted; $PREC_BY_WORD{$word} += $precision; $LENGTH_BY_WORD{$word} += $length; - $TOTAL_BY_WORD{$word}++; + $TOTAL_BY_WORD{$word}++; $DELETED_BY_COVERAGE{$coverage} += $deleted; $PREC_BY_COVERAGE{$coverage} += $precision; $LENGTH_BY_COVERAGE{$coverage} += $length; - $TOTAL_BY_COVERAGE{$coverage}++; + $TOTAL_BY_COVERAGE{$coverage}++; if ($precision_by_coverage_factor) { $DELETED_BY_FACTOR{$FACTOR[$i]} += $deleted; @@ -662,9 +662,9 @@ sub precision_by_coverage { $PREC_BY_FACTOR{$FACTOR[$i]} += $precision; $PREC_BY_FACTOR_COVERAGE{$FACTOR[$i]}{$coverage} += $precision; $LENGTH_BY_FACTOR{$FACTOR[$i]} += $length; - $LENGTH_BY_FACTOR_COVERAGE{$FACTOR[$i]}{$coverage} += $length; - $TOTAL_BY_FACTOR{$FACTOR[$i]}++; - $TOTAL_BY_FACTOR_COVERAGE{$FACTOR[$i]}{$coverage}++; + $LENGTH_BY_FACTOR_COVERAGE{$FACTOR[$i]}{$coverage} += $length; + $TOTAL_BY_FACTOR{$FACTOR[$i]}++; + $TOTAL_BY_FACTOR_COVERAGE{$FACTOR[$i]}{$coverage}++; } } } @@ -853,10 +853,10 @@ sub hs_scan_line { # process a single sentence for hierarchical segmentation sub hs_process { my ($sentence,$DERIVATION,$STATS) = @_; - + my $DROP_RULE = shift @{$DERIVATION}; # get rid of S -> S </s> my $max = $$DERIVATION[0]{'end'}; - + # consolidate glue rules into one rule my %GLUE_RULE; $GLUE_RULE{'start'} = 1; @@ -867,10 +867,10 @@ sub hs_process { while(1) { my $RULE = shift @{$DERIVATION}; if (scalar(@{$$RULE{'rule_rhs'}}) == 2 && - ($$RULE{'rule_lhs'} eq "S" && + ($$RULE{'rule_lhs'} eq "S" && $$RULE{'rule_rhs'}[0] eq "S" && $$RULE{'rule_rhs'}[1] eq "X") || - ($$RULE{'rule_lhs'} eq "Q" && + ($$RULE{'rule_lhs'} eq "Q" && $$RULE{'rule_rhs'}[0] eq "Q")) { unshift @{$GLUE_RULE{'spans'}},$$RULE{'spans'}[1]; push @{$GLUE_RULE{'rule_rhs'}}, $$RULE{'rule_rhs'}[1]; @@ -883,17 +883,17 @@ sub hs_process { last; } } - unshift @{$DERIVATION}, \%GLUE_RULE; + unshift @{$DERIVATION}, \%GLUE_RULE; $$STATS{'glue-rule'} += $x; - + # create chart my %CHART; foreach my $RULE (@{$DERIVATION}) { $CHART{$$RULE{'start'}}{$$RULE{'end'}} = $RULE; } - + # compute depth - &hs_compute_depth(1,$max,0,\%CHART); + &hs_compute_depth(1,$max,0,\%CHART); my $max_depth = 0; foreach my $RULE (@{$DERIVATION}) { next unless defined($$RULE{'depth'}); # better: delete offending rule S -> S <s> @@ -901,17 +901,17 @@ sub hs_process { } &hs_recompute_depth(1,$max,\%CHART,$max_depth); $$STATS{'depth'} += $max_depth; - + # build matrix of divs - + my @MATRIX; &hs_create_out_span(1,$max,\%CHART,\@MATRIX); print OUTPUT_TREE &hs_output_matrix($sentence,\@MATRIX,$max_depth); - + my @MATRIX_IN; &hs_create_in_span(1,$max,\%CHART,\@MATRIX_IN); print INPUT_TREE &hs_output_matrix($sentence,\@MATRIX_IN,$max_depth); - + # number rules and get their children my $id = 0; foreach my $RULE (@{$DERIVATION}) { @@ -920,10 +920,10 @@ sub hs_process { $$RULE{'id'} = $id++; } &hs_get_children(1,$max,\%CHART); - + foreach my $RULE (@{$DERIVATION}) { next unless defined($$RULE{'start_div'}); # better: delete offending rule S -> S <s> - + print NODE $sentence." "; print NODE $$RULE{'depth'}." "; print NODE $$RULE{'start_div'}." ".$$RULE{'end_div'}." "; @@ -963,11 +963,11 @@ sub hs_output_matrix { $class = "]"; } elsif ($OPEN[$d]) { - $class = "-"; + $class = "-"; } $out .= $class; } - $out .= "\t"; + $out .= "\t"; $out .= $$SPAN{'lhs'} if defined($$SPAN{'lhs'}); $out .= "\t"; $out .= $$SPAN{'rhs'} if defined($$SPAN{'rhs'}); @@ -984,9 +984,9 @@ sub hs_output_matrix { sub hs_rule_type { my ($RULE) = @_; - + my $type = ""; - + # output side my %NT; my $total_word_count = 0; @@ -998,7 +998,7 @@ sub hs_rule_type { $word_count = 0; my $nt = chr(97+$nt_count++); $NT{$$RULE{'alignment'}{$i}} = $nt; - $type .= $nt; + $type .= $nt; } else { $word_count++; @@ -1006,9 +1006,9 @@ sub hs_rule_type { } } $type .= $word_count if $word_count > 0; - + $type .= ":".$total_word_count.":".$nt_count.":"; - + # input side $word_count = 0; $total_word_count = 0; @@ -1039,7 +1039,7 @@ sub hs_compute_depth { my $RULE = $$CHART{$start}{$end}; $$RULE{'depth'} = $depth; - + for(my $i=0;$i<scalar @{$$RULE{'rule_rhs'}};$i++) { # non-terminals if (defined($$RULE{'alignment'}{$i})) { @@ -1057,7 +1057,7 @@ sub hs_recompute_depth { return 0; } my $RULE = $$CHART{$start}{$end}; - + my $min_sub_depth = $max_depth+1; for(my $i=0;$i<scalar @{$$RULE{'rule_rhs'}};$i++) { # non-terminals @@ -1079,10 +1079,10 @@ sub hs_get_children { return -1; } my $RULE = $$CHART{$start}{$end}; - + my @CHILDREN = (); $$RULE{'children'} = \@CHILDREN; - + for(my $i=0;$i<scalar @{$$RULE{'rule_rhs'}};$i++) { # non-terminals if (defined($$RULE{'alignment'}{$i})) { @@ -1091,7 +1091,7 @@ sub hs_get_children { push @CHILDREN, $child unless $child == -1; } } - return $$RULE{'id'}; + return $$RULE{'id'}; } # create the span annotation for an output sentence @@ -1102,7 +1102,7 @@ sub hs_create_out_span { return; } my $RULE = $$CHART{$start}{$end}; - + my %SPAN; $SPAN{'start'} = $start; $SPAN{'end'} = $end; @@ -1130,7 +1130,7 @@ sub hs_create_out_span { $SPAN{'end'} = $end; $SPAN{'depth'} = $$RULE{'depth'}; push @{$MATRIX},\%SPAN; - $THIS_SPAN = \%SPAN; + $THIS_SPAN = \%SPAN; } $$THIS_SPAN{'rhs'} .= " " if defined($$THIS_SPAN{'rhs'}); $$THIS_SPAN{'rhs'} .= $$RULE{"rule_rhs"}[$i]; @@ -1150,7 +1150,7 @@ sub hs_create_in_span { return; } my $RULE = $$CHART{$start}{$end}; - + my %SPAN; $SPAN{'start'} = $start; $SPAN{'end'} = $end; @@ -1160,7 +1160,7 @@ sub hs_create_in_span { push @{$MATRIX},\%SPAN; $$RULE{'start_div_in'} = $#{$MATRIX}; my $THIS_SPAN = \%SPAN; - + my $terminal = 1; # in input order ... for(my $i=0;$i<scalar(@{$$RULE{'spans'}});$i++) { @@ -1177,7 +1177,7 @@ sub hs_create_in_span { $SPAN{'end'} = $end; $SPAN{'depth'} = $$RULE{'depth'}; push @{$MATRIX},\%SPAN; - $THIS_SPAN = \%SPAN; + $THIS_SPAN = \%SPAN; } $$THIS_SPAN{'rhs'} .= " " if defined($$THIS_SPAN{'rhs'}); $$THIS_SPAN{'rhs'} .= $$SUBSPAN{'word'}; @@ -1204,7 +1204,7 @@ sub process_search_graph { $heuristic_rule_score = $rule_score; # hmmmm.... } else { - die("ERROR: buggy search graph line: $_"); + die("ERROR: buggy search graph line: $_"); } chop($alignment) if $alignment; chop($children) if $children; diff --git a/scripts/ems/support/berkeley-process.sh b/scripts/ems/support/berkeley-process.sh index 42b8ba9c3..e68056c96 100755 --- a/scripts/ems/support/berkeley-process.sh +++ b/scripts/ems/support/berkeley-process.sh @@ -1,6 +1,6 @@ #!/bin/sh -if [ $# -lt 8 ] +if [ $# -lt 8 ] then echo "Usage: $0 <\"java options\"> <berkeleyaligner jar file> <input file stem> <previous berkeley param dir> <output directory> <source lang> <target lang> <alignment name (i.e. 'berk' or 'low-posterior')> <posterior threshold> [aligner options...]" exit 1 @@ -23,7 +23,7 @@ shift shift shift shift -shift +shift JAVA_CMD="/usr/local/share/java/bin/java $JAVA_OPTS -jar $JAR -Data.trainSources $INFILE.list -Main.loadParamsDir $PARAMDIR -exec.execDir $OUTNAME -Main.loadLexicalModelOnly false -Data.englishSuffix $SLANG -Data.foreignSuffix $TLANG -exec.create true -Main.saveParams false -Main.alignTraining true -Main.forwardModels HMM -Main.reverseModels HMM -Main.mode JOINT -Main.iters 0 -Data.testSources -EMWordAligner.posteriorDecodingThreshold $POSTERIOR $@" echo "Running $JAVA_CMD" @@ -37,8 +37,8 @@ gzip $OUTNAME/training.$TLANG-$SLANG.A3 #now shift the output perl -e " -use strict; -while (<STDIN>) { +use strict; +while (<STDIN>) { chomp(); my @pairs = split(\" \"); for (my \$i=0;\$i<scalar(@pairs);\$i++) { diff --git a/scripts/ems/support/berkeley-train.sh b/scripts/ems/support/berkeley-train.sh index 57d2963fc..96f6b648c 100755 --- a/scripts/ems/support/berkeley-train.sh +++ b/scripts/ems/support/berkeley-train.sh @@ -1,6 +1,6 @@ #!/bin/sh -if [ $# -lt 6 ] +if [ $# -lt 6 ] then echo "Usage: $0 <\"java options\"> <berkeleyaligner jar file> <input file stem> <output directory> <source lang> <target lang> [aligner options...]" exit 1 @@ -20,4 +20,4 @@ shift shift echo $INFILE > $INFILE.list -/usr/local/share/java/bin/java $JAVA_OPTS -jar $JAR -Data.trainSources $INFILE.list -exec.execDir $OUTDIR -Data.englishSuffix $SLANG -Data.foreignSuffix $TLANG -exec.create true -Main.SaveParams true -Main.alignTraining false -Data.testSources $@ +/usr/local/share/java/bin/java $JAVA_OPTS -jar $JAR -Data.trainSources $INFILE.list -exec.execDir $OUTDIR -Data.englishSuffix $SLANG -Data.foreignSuffix $TLANG -exec.create true -Main.SaveParams true -Main.alignTraining false -Data.testSources $@ diff --git a/scripts/ems/support/build-domain-file-from-subcorpora.perl b/scripts/ems/support/build-domain-file-from-subcorpora.perl index f166c8927..085fd2629 100755 --- a/scripts/ems/support/build-domain-file-from-subcorpora.perl +++ b/scripts/ems/support/build-domain-file-from-subcorpora.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -7,7 +7,7 @@ use strict; # (helper for domain adatpation) # Creates a file with domain names and end line numbers for different domains -# within the cleaned training corpus. This file is used by various domain +# within the cleaned training corpus. This file is used by various domain # adaptation methods. my ($extension,@SUBCORPORA) = @ARGV; diff --git a/scripts/ems/support/build-sparse-features.perl b/scripts/ems/support/build-sparse-features.perl index 3f4b505d5..79fc1e394 100755 --- a/scripts/ems/support/build-sparse-features.perl +++ b/scripts/ems/support/build-sparse-features.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -18,7 +18,7 @@ foreach my $feature_spec (split(/,\s*/,$specification)) { my @SPEC = split(/\s+/,$feature_spec); my $factor = ($SPEC[0] eq 'word-translation') ? "0-0" : "0"; - $factor = $1 if $feature_spec =~ / factor ([\d\-]+)/; + $factor = $1 if $feature_spec =~ / factor ([\d\-]+)/; $feature_spec =~ s/ factor ([\d\-]+)//; if ($SPEC[0] eq 'target-word-insertion') { @@ -107,7 +107,7 @@ sub create_top_words { open(TOP,">$file"); for(my $i=0; $i<$count && $i<scalar(@SORTED); $i++) { $SORTED[$i] =~ /^\d+ (.+)$/; - print TOP "$1\n"; + print TOP "$1\n"; } close(TOP); diff --git a/scripts/ems/support/consolidate-training-data.perl b/scripts/ems/support/consolidate-training-data.perl index 170ba999c..4ab7f82cf 100755 --- a/scripts/ems/support/consolidate-training-data.perl +++ b/scripts/ems/support/consolidate-training-data.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl # $Id: consolidate-training-data.perl 928 2009-09-02 02:58:01Z philipp $ diff --git a/scripts/ems/support/fast-align-in-parts.perl b/scripts/ems/support/fast-align-in-parts.perl index fa501b454..f777d7e52 100755 --- a/scripts/ems/support/fast-align-in-parts.perl +++ b/scripts/ems/support/fast-align-in-parts.perl @@ -24,7 +24,7 @@ die("ERROR - usage: fast-align-in-parts.perl -bin FAST_ALIGN_BIN -i PARALLEL_COR && $MAX_LINES > 0; die("ERROR - input file does not exist: $IN") unless -e $IN; die("ERROR - fast_align binary does not exist: $BIN") unless -e $BIN; - + chomp(my $line_count = `cat $IN | wc -l`); # not more than maximal number of lines -> just run it regulary diff --git a/scripts/ems/support/generic-multicore-parallelizer.perl b/scripts/ems/support/generic-multicore-parallelizer.perl index e5a12adce..0f7910603 100755 --- a/scripts/ems/support/generic-multicore-parallelizer.perl +++ b/scripts/ems/support/generic-multicore-parallelizer.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -6,7 +6,7 @@ use strict; my $cores = 8; my $serial = 1; my ($infile,$outfile,$cmd,$tmpdir); -my $parent = $$; +my $parent = $$; use Getopt::Long qw(:config pass_through no_ignore_case); GetOptions('cores=i' => \$cores, @@ -27,7 +27,7 @@ die("ERROR: you need to specify a tempdir with -tmpdir") unless $tmpdir; # create split input files my $sentenceN = `cat $infile | wc -l`; -my $splitN = int(($sentenceN+($cores*$serial)-0.5) / ($cores*$serial)); +my $splitN = int(($sentenceN+($cores*$serial)-0.5) / ($cores*$serial)); print STDERR "split -a 3 -l $splitN $infile $tmpdir/in-$parent-\n"; `split -a 4 -l $splitN $infile $tmpdir/in-$parent-`; diff --git a/scripts/ems/support/generic-parallelizer.perl b/scripts/ems/support/generic-parallelizer.perl index fd7fb2552..811a99bde 100755 --- a/scripts/ems/support/generic-parallelizer.perl +++ b/scripts/ems/support/generic-parallelizer.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -25,7 +25,7 @@ $qflags = "" unless $qflags; # create split input files my $sentenceN = `cat $infile | wc -l`; -my $splitN = int(($sentenceN+$jobs-0.5) / $jobs); +my $splitN = int(($sentenceN+$jobs-0.5) / $jobs); `split -a 2 -l $splitN $infile $tmpdir/in-$$-`; # find out the names of the jobs @@ -56,7 +56,7 @@ foreach my $job (@JOB){ # get qsub ID my @QSUB_ID; -foreach my $job (@JOB){ +foreach my $job (@JOB){ `cat $tmpdir/job-$$-$job.log` =~ /Your job (\d+) / or die "ERROR: Can't read log of job $tmpdir/job-$$-$job.log"; push @QSUB_ID,$1; diff --git a/scripts/ems/support/input-from-sgm.perl b/scripts/ems/support/input-from-sgm.perl index 223996676..18000581a 100755 --- a/scripts/ems/support/input-from-sgm.perl +++ b/scripts/ems/support/input-from-sgm.perl @@ -1,9 +1,9 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; -die("ERROR syntax: input-from-sgm.perl < in.sgm > in.txt") +die("ERROR syntax: input-from-sgm.perl < in.sgm > in.txt") unless scalar @ARGV == 0; while(my $line = <STDIN>) { @@ -17,7 +17,7 @@ while(my $line = <STDIN>) { $line !~ /<seg[^>]+>\s*(.*)\s*<\/seg>/i) { my $next_line = <STDIN>; $line .= $next_line; - chop($line); + chop($line); } if ($line =~ /<seg[^>]+>\s*(.*)\s*<\/seg>/i) { my $input = $1; diff --git a/scripts/ems/support/interpolate-lm.perl b/scripts/ems/support/interpolate-lm.perl index a2fe62b22..7d52fd877 100755 --- a/scripts/ems/support/interpolate-lm.perl +++ b/scripts/ems/support/interpolate-lm.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -28,8 +28,8 @@ die("interpolate-lm.perl --tuning set --name out-lm --lm lm0,lm1,lm2,lm3 [--sril # check and set default to unset parameters die("ERROR: please specify output language model name --name") unless defined($NAME); -die("ERROR: please specify tuning set with --tuning") unless defined($TUNING); -die("ERROR: please specify language models with --lm") unless defined($LM); +die("ERROR: please specify tuning set with --tuning") unless defined($TUNING); +die("ERROR: please specify language models with --lm") unless defined($LM); die("ERROR: can't read $TUNING") unless -e $TUNING; die("ERROR: did not find srilm dir") unless -e $SRILM; die("ERROR: cannot run ngram") unless -x $SRILM."/ngram"; @@ -152,7 +152,7 @@ sub interpolate { $mix =~ /best lambda \(([\d\. e-]+)\)/ || die("ERROR: computing lambdas failed: $mix"); @LAMBDA = split(/ /,$1); } - + # create new language model print STDERR "creating new language model...\n"; my $i = 0; @@ -196,7 +196,7 @@ sub saferun3 { print STDERR "Executing: @_\n"; my $wtr = gensym(); my $rdr = gensym(); - my $err = gensym(); + my $err = gensym(); my $pid = open3($wtr, $rdr, $err, @_); close($wtr); my $gotout = ""; diff --git a/scripts/ems/support/lmplz-wrapper.perl b/scripts/ems/support/lmplz-wrapper.perl index f36d2d9e0..df503754f 100755 --- a/scripts/ems/support/lmplz-wrapper.perl +++ b/scripts/ems/support/lmplz-wrapper.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/mml-filter.perl b/scripts/ems/support/mml-filter.perl index c50725aae..51bc4cda5 100755 --- a/scripts/ems/support/mml-filter.perl +++ b/scripts/ems/support/mml-filter.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/mml-score.perl b/scripts/ems/support/mml-score.perl index 449d6a05c..6f7b724ea 100755 --- a/scripts/ems/support/mml-score.perl +++ b/scripts/ems/support/mml-score.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/mml-train.perl b/scripts/ems/support/mml-train.perl index 1f0548082..dcc998711 100755 --- a/scripts/ems/support/mml-train.perl +++ b/scripts/ems/support/mml-train.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/prepare-fast-align.perl b/scripts/ems/support/prepare-fast-align.perl index 54c124af0..80fec36b2 100755 --- a/scripts/ems/support/prepare-fast-align.perl +++ b/scripts/ems/support/prepare-fast-align.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/reference-from-sgm.perl b/scripts/ems/support/reference-from-sgm.perl index 595226bf1..ebb9ae4ae 100755 --- a/scripts/ems/support/reference-from-sgm.perl +++ b/scripts/ems/support/reference-from-sgm.perl @@ -1,9 +1,9 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; -die("ERROR syntax: reference-from-sgm.perl ref src out") +die("ERROR syntax: reference-from-sgm.perl ref src out") unless scalar @ARGV == 3; my ($ref,$src,$txt) = @ARGV; diff --git a/scripts/ems/support/remove-segmentation-markup.perl b/scripts/ems/support/remove-segmentation-markup.perl index d6333f813..a0bd61fff 100755 --- a/scripts/ems/support/remove-segmentation-markup.perl +++ b/scripts/ems/support/remove-segmentation-markup.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; @@ -6,7 +6,7 @@ use strict; $|++; while(<STDIN>) { - s/ \|\d+\-\d+\| / /g; - s/ \|\d+\-\d+\|$//; + s/ \|\d+\-\d+\| / /g; + s/ \|\d+\-\d+\|$//; print $_; } diff --git a/scripts/ems/support/report-experiment-scores.perl b/scripts/ems/support/report-experiment-scores.perl index ef64d4c2d..b649951ce 100755 --- a/scripts/ems/support/report-experiment-scores.perl +++ b/scripts/ems/support/report-experiment-scores.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl # $Id: report-experiment-scores.perl 407 2008-11-10 14:43:31Z philipp $ diff --git a/scripts/ems/support/run-command-on-multiple-refsets.perl b/scripts/ems/support/run-command-on-multiple-refsets.perl index c3db3c4dc..1e914b44b 100755 --- a/scripts/ems/support/run-command-on-multiple-refsets.perl +++ b/scripts/ems/support/run-command-on-multiple-refsets.perl @@ -1,9 +1,9 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; -die("ERROR: syntax: run-command-on-multiple-refsets.perl cmd in out") +die("ERROR: syntax: run-command-on-multiple-refsets.perl cmd in out") unless scalar @ARGV == 3; my ($cmd,$in,$out) = @ARGV; diff --git a/scripts/ems/support/run-wade.perl b/scripts/ems/support/run-wade.perl index 25cda3bb3..175948b98 100755 --- a/scripts/ems/support/run-wade.perl +++ b/scripts/ems/support/run-wade.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/split-sentences.perl b/scripts/ems/support/split-sentences.perl index f1af451b3..02a1e2315 100755 --- a/scripts/ems/support/split-sentences.perl +++ b/scripts/ems/support/split-sentences.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl # Based on Preprocessor written by Philipp Koehn @@ -97,21 +97,21 @@ sub preprocess { $text =~ s/ \n/\n/g; $text =~ s/^ //g; $text =~ s/ $//g; - + #####add sentence breaks as needed##### - + #non-period end of sentence markers (?!) followed by sentence starters. $text =~ s/([?!]) +([\'\"\(\[\¿\¡\p{IsPi}]*[\p{IsUpper}])/$1\n$2/g; - + #multi-dots followed by sentence starters $text =~ s/(\.[\.]+) +([\'\"\(\[\¿\¡\p{IsPi}]*[\p{IsUpper}])/$1\n$2/g; - + # add breaks for sentences that end with some sort of punctuation inside a quote or parenthetical and are followed by a possible sentence starter punctuation and upper case $text =~ s/([?!\.][\ ]*[\'\"\)\]\p{IsPf}]+) +([\'\"\(\[\¿\¡\p{IsPi}]*[\ ]*[\p{IsUpper}])/$1\n$2/g; - + # add breaks for sentences that end with some sort of punctuation are followed by a sentence starter punctuation and upper case $text =~ s/([?!\.]) +([\'\"\(\[\¿\¡\p{IsPi}]+[\ ]*[\p{IsUpper}])/$1\n$2/g; - + # special punctuation cases are covered. Check all remaining periods. my $word; my $i; @@ -125,32 +125,32 @@ sub preprocess { if($prefix && $NONBREAKING_PREFIX{$prefix} && $NONBREAKING_PREFIX{$prefix} == 1 && !$starting_punct) { #not breaking; } elsif ($words[$i] =~ /(\.)[\p{IsUpper}\-]+(\.+)$/) { - #not breaking - upper case acronym + #not breaking - upper case acronym } elsif($words[$i+1] =~ /^([ ]*[\'\"\(\[\¿\¡\p{IsPi}]*[ ]*[\p{IsUpper}0-9])/) { #the next word has a bunch of initial quotes, maybe a space, then either upper case or a number $words[$i] = $words[$i]."\n" unless ($prefix && $NONBREAKING_PREFIX{$prefix} && $NONBREAKING_PREFIX{$prefix} == 2 && !$starting_punct && ($words[$i+1] =~ /^[0-9]+/)); #we always add a return for these unless we have a numeric non-breaker and a number start } - + } $text = $text.$words[$i]." "; } - + #we stopped one token from the end to allow for easy look-ahead. Append it now. $text = $text.$words[$i]; - + # clean up spaces at head and tail of each line as well as any double-spacing $text =~ s/ +/ /g; $text =~ s/\n /\n/g; $text =~ s/ \n/\n/g; $text =~ s/^ //g; $text =~ s/ $//g; - + #add trailing break $text .= "\n" unless $text =~ /\n$/; - + return $text; - + } diff --git a/scripts/ems/support/submit-grid.perl b/scripts/ems/support/submit-grid.perl index 9997241e7..a0967f9a5 100755 --- a/scripts/ems/support/submit-grid.perl +++ b/scripts/ems/support/submit-grid.perl @@ -9,7 +9,7 @@ use File::Basename; my $continue = 0; -my $args = ""; +my $args = ""; my $config; GetOptions("continue=i" => \$continue, diff --git a/scripts/ems/support/substitute-filtered-tables-and-weights.perl b/scripts/ems/support/substitute-filtered-tables-and-weights.perl index 681d251c7..13be52c6b 100755 --- a/scripts/ems/support/substitute-filtered-tables-and-weights.perl +++ b/scripts/ems/support/substitute-filtered-tables-and-weights.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/substitute-filtered-tables.perl b/scripts/ems/support/substitute-filtered-tables.perl index e7d9f55f8..c5ebabded 100755 --- a/scripts/ems/support/substitute-filtered-tables.perl +++ b/scripts/ems/support/substitute-filtered-tables.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; @@ -45,7 +45,7 @@ while(my $line = <STDIN>) { elsif ($feature_section && $line =~ /LexicalReordering/) { print $arr[$ind]."\n"; ++$ind; - } + } else { print "$line\n"; } diff --git a/scripts/ems/support/substitute-weights.perl b/scripts/ems/support/substitute-weights.perl index 42357ed1e..b692f3f85 100755 --- a/scripts/ems/support/substitute-weights.perl +++ b/scripts/ems/support/substitute-weights.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; @@ -18,7 +18,7 @@ while(my $line = <BASEINI>) { } elsif ($line =~ /\[[a-zA-Z0-9\-]*\]/) { $inWeightSection = 0; - } + } if (!$inWeightSection) { print OUT "$line\n" unless $line =~ /dense weights for feature functions/; @@ -48,7 +48,7 @@ while(my $line = <WEIGHTINI>) { elsif ($line =~ /\[[a-zA-Z0-9\-]*\]/) { print OUT "\n" if $inWeightSection; $inWeightSection = 0; - } + } if ($inWeightSection && $line !~ /^\s*$/) { print OUT "$line\n"; diff --git a/scripts/ems/support/symmetrize-fast-align.perl b/scripts/ems/support/symmetrize-fast-align.perl index 90621dea9..9f7fec248 100755 --- a/scripts/ems/support/symmetrize-fast-align.perl +++ b/scripts/ems/support/symmetrize-fast-align.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/thot-lm-wrapper.perl b/scripts/ems/support/thot-lm-wrapper.perl index 222623c5b..59d483e65 100755 --- a/scripts/ems/support/thot-lm-wrapper.perl +++ b/scripts/ems/support/thot-lm-wrapper.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/tree-converter-wrapper.perl b/scripts/ems/support/tree-converter-wrapper.perl index a37654cf1..aae55991a 100755 --- a/scripts/ems/support/tree-converter-wrapper.perl +++ b/scripts/ems/support/tree-converter-wrapper.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/support/wrap-xml.perl b/scripts/ems/support/wrap-xml.perl index 28708a62a..52190309a 100755 --- a/scripts/ems/support/wrap-xml.perl +++ b/scripts/ems/support/wrap-xml.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/web/analysis.php b/scripts/ems/web/analysis.php index 00bb9e15f..57776dd22 100644 --- a/scripts/ems/web/analysis.php +++ b/scripts/ems/web/analysis.php @@ -1,4 +1,4 @@ -<?php +<?php # main page frame, triggers the loading of parts function show_analysis() { @@ -122,7 +122,7 @@ function precision_by_coverage() { $log_info[$log_count]["precision"] += $item[1]; $log_info[$log_count]["delete"] += $item[2]; $log_info[$log_count]["length"] += $item[3]; - $log_info[$log_count]["total"] += $item[4]; + $log_info[$log_count]["total"] += $item[4]; } print "<h4>By log<sub>2</sub>-count in the training corpus</h4>"; precision_by_coverage_graph("byCoverage",$log_info,$total,$img_width,SORT_NUMERIC); @@ -159,7 +159,7 @@ function precision_by_coverage_factored($img_width,$total,$file,$factor_id) { $log_info_factored[$factor][$log_count]["precision"] += $item[2]; $log_info_factored[$factor][$log_count]["delete"] += $item[3]; $log_info_factored[$factor][$log_count]["length"] += $item[4]; - $log_info_factored[$factor][$log_count]["total"] += $item[5]; + $log_info_factored[$factor][$log_count]["total"] += $item[5]; } print "<h4>By factor ".factor_name("input",$factor_id)."</h4>"; precision_by_coverage_graph("byFactor",$info_factored_sum,$total,$img_width,SORT_STRING); @@ -203,7 +203,7 @@ function precision_by_word($type) { if ($byCoverage != -2 && $byCoverage != $log_count) { continue; } - + //# filter for factor $word = $item[5]; if ($byFactor != "false" && $byFactor != $item[6]) { @@ -231,7 +231,7 @@ function precision_by_word($type) { function precision_by_coverage_latex($name,$log_info,$total,$img_width,$sort_type) { $keys = array_keys($log_info); sort($keys,$sort_type); - + $img_width /= 100; print "<div id=\"LatexToggle$name\" onClick=\"document.getElementById('Latex$name').style.display = 'block'; this.style.display = 'none';\" style=\"display:none;\"><font size=-2>(show LaTeX)</font></div>\n"; print "<div id=\"Latex$name\" style=\"display:none;\">\n"; @@ -242,7 +242,7 @@ function precision_by_coverage_latex($name,$log_info,$total,$img_width,$sort_typ $height = 1.8-$line/10*1.8; print "\\draw[thin,lightgray] (0.2,-$height) "; print "node[anchor=east,black] {".$line."0\\%} -- "; - print "($img_width,-$height) ;<br>\n"; + print "($img_width,-$height) ;<br>\n"; } print "% co-ordinates for deletion<br>\n"; for($line=0;$line<=3;$line++) { @@ -251,7 +251,7 @@ function precision_by_coverage_latex($name,$log_info,$total,$img_width,$sort_typ if ($line != 0) { print "node[anchor=east,black] {".$line."0\\%} "; } - print "-- ($img_width,-$height) ;<br>\n"; + print "-- ($img_width,-$height) ;<br>\n"; } print "% boxes<br>\n"; @@ -265,13 +265,13 @@ function precision_by_coverage_latex($name,$log_info,$total,$img_width,$sort_typ $width += $x; $height += $y; - + print "\\filldraw[very thin,gray] ($x,-$y) rectangle($width,-$height) ;<br>"; print "\\draw[very thin,black] ($x,-$y) rectangle($width,-$height);<br>"; if ($width-$x>.1) { print "\\draw (".(($x+$width)/2).",-1.8) node[anchor=north,black] {".$i."};<br>"; } - + $del_ratio = $log_info[$i]["delete"]/$log_info[$i]["total"]; $height = $del_ratio*1.80; @@ -281,10 +281,10 @@ function precision_by_coverage_latex($name,$log_info,$total,$img_width,$sort_typ print "\\filldraw[very thin,lightgray] ($x,-2) rectangle($width,-$height);<br>\n"; print "\\draw[very thin,black] ($x,-2) rectangle($width,-$height);<br>\n"; - $total_so_far += $log_info[$i]["total"]; + $total_so_far += $log_info[$i]["total"]; } print "\\end{tikzpicture}</code>"; - print "</div>"; + print "</div>"; } function precision_by_coverage_graph($name,$log_info,$total,$img_width,$sort_type) { @@ -351,7 +351,7 @@ ctx.font = '9px serif'; print "ctx.fillRect ($x, 200, $width, $height);"; $total_so_far += $log_info[$i]["total"]; - + if ($width>3) { print "ctx.fillStyle = \"rgb(0,0,0)\";"; // print "ctx.rotate(-1.5707);"; @@ -410,7 +410,7 @@ function ngram_summary() { $info["$type-1-correct"], $info["$type-2-correct"], $info["$type-3-correct"], - $info["$type-4-correct"]); + $info["$type-4-correct"]); printf("<tr><td> </td><td>%.1f%s</td><td>%.1f%s</td><td>%.1f%s</td><td>%.1f%s</td></tr>\n", $info["$type-1-correct"]/$info["$type-1-total"]*100,'%', $info["$type-2-correct"]/$info["$type-2-total"]*100,'%', @@ -483,7 +483,7 @@ function ngram_show($type) { $data = file(get_current_analysis_filename("basic","n-gram-$type.$order")); for($i=0;$i<count($data);$i++) { $item = split("\t",$data[$i]); - $line["total"] = $item[0]; + $line["total"] = $item[0]; $line["correct"] = $item[1]; $line["ngram"] = $item[2]; $ngram[] = $line; @@ -496,7 +496,7 @@ function ngram_show($type) { $sort = 'ratio_worst'; $smooth = 1; } - + // sort index for($i=0;$i<count($ngram);$i++) { if ($sort == "abs_worst") { @@ -806,7 +806,7 @@ function segmentation_summary() { if (array_key_exists($in,$count) && array_key_exists($out,$count[$in])) { $c = $count[$in][$out]; - } + } else { $c = 0; } printf("<td align=right nowrap>%d (%.1f%s)</td>",$c,100*$c/$total,"%"); } @@ -885,8 +885,8 @@ function bleu_show() { $count = $_GET['count']; if ($count == 0) { $count = 5; } - $filter = ""; - if (array_key_exists("filter",$_GET)) { + $filter = ""; + if (array_key_exists("filter",$_GET)) { $filter = base64_decode($_GET['filter']); } @@ -924,7 +924,7 @@ function bleu_show() { if ($filter != "") { print "; filter: '$filter'"; } - + sentence_annotation($count,$filter); print "<p align=center><A HREF=\"javascript:show('bleu','" . $_GET['sort'] . "',5+$count,'".base64_encode($filter)."')\">5 more</A> | "; print "<A HREF=\"javascript:show('bleu','" . $_GET['sort'] . "',10+$count,'".base64_encode($filter)."')\">10 more</A> | "; @@ -950,28 +950,28 @@ function sentence_annotation($count,$filter) { $word = explode(" ",$item[0]); $keep = 0; for($j=0;$j<count($word);$j++) { - if ($word[$j] == $filter) { - $keep = 1; + if ($word[$j] == $filter) { + $keep = 1; } } - if (!$keep) { $filtered[$i] = 1; } + if (!$keep) { $filtered[$i] = 1; } } - } + } } - - # load bleu scores + + # load bleu scores $data = file(get_current_analysis_filename("basic","bleu-annotation")); for($i=0;$i<count($data);$i++) { $item = split("\t",$data[$i]); if (! array_key_exists($item[1],$filtered)) { - $line["bleu"] = $item[0]; - $line["id"] = $item[1]; + $line["bleu"] = $item[0]; + $line["id"] = $item[1]; $line["system"] = $item[2]; - $line["reference"] = ""; + $line["reference"] = ""; for($j=3;$j<count($item);$j++) { if ($j>3) { $line["reference"] .= "<br>"; }; $line["reference"] .= $item[$j]; - } + } $bleu[] = $line; } } @@ -987,7 +987,7 @@ function sentence_annotation($count,$filter) { else if ($sort == "worst" || $sort == "75") { $a_idx = $a["bleu"]; $b_idx = $b["bleu"]; - if ($a_idx == $b_idx) { + if ($a_idx == $b_idx) { $a_idx = $b["id"]; $b_idx = $a["id"]; } @@ -995,7 +995,7 @@ function sentence_annotation($count,$filter) { else if ($sort == "best" || $sort == "avg" || $sort == "25") { $a_idx = -$a["bleu"]; $b_idx = -$b["bleu"]; - if ($a_idx == $b_idx) { + if ($a_idx == $b_idx) { $a_idx = $a["id"]; $b_idx = $b["id"]; } @@ -1021,7 +1021,7 @@ function sentence_annotation($count,$filter) { $retained = array(); for($i=$offset;$i<$count+$offset && $i<count($bleu);$i++) { - $line = $bleu[$i]; + $line = $bleu[$i]; $retained[$line["id"]] = 1; } @@ -1056,7 +1056,7 @@ function sentence_annotation($count,$filter) { list($sentence,$brackets,$nt,$words) = split("\t",$data[$i]); if ($sentence != $last_sentence) { $span = 0; } $last_sentence = $sentence; - if (array_key_exists($sentence,$retained)) { + if (array_key_exists($sentence,$retained)) { $segmentation[$sentence][$span]["brackets"] = $brackets; # $segmentation[$sentence][$span]["nt"] = $nt; $segmentation[$sentence][$span]["words"] = rtrim($words); @@ -1083,7 +1083,7 @@ function sentence_annotation($count,$filter) { list($sentence,$brackets,$nt,$words) = split("\t",$data[$i]); if ($sentence != $last_sentence) { $span = 0; } $last_sentence = $sentence; - if (array_key_exists($sentence,$retained)) { + if (array_key_exists($sentence,$retained)) { $segmentation_out[$sentence][$span]["brackets"] = $brackets; $segmentation_out[$sentence][$span]["nt"] = $nt; $segmentation_out[$sentence][$span]["words"] = rtrim($words); @@ -1109,7 +1109,7 @@ function sentence_annotation($count,$filter) { list($sentence,$depth,$start_div,$end_div,$start_div_in,$end_div_in,$children) = split(" ",$data[$i]); if ($sentence != $last_sentence) { $n = 0; } $last_sentence = $sentence; - if (array_key_exists($sentence,$retained)) { + if (array_key_exists($sentence,$retained)) { $node[$sentence][$n]['depth'] = $depth; $node[$sentence][$n]['start_div'] = $start_div; $node[$sentence][$n]['end_div'] = $end_div; @@ -1119,10 +1119,10 @@ function sentence_annotation($count,$filter) { $n++; } } - } + } # display - if ($filter != "") { + if ($filter != "") { print " (".(count($input)-count($filtered))." retaining)"; } print "</font><BR>\n"; @@ -1130,7 +1130,7 @@ function sentence_annotation($count,$filter) { $biconcor = get_biconcor_version($dir,$set,$id); //print "<div id=\"debug\">$sort / $offset</div>"; for($i=$offset;$i<$count+$offset && $i<count($bleu);$i++) { - $line = $bleu[$i]; + $line = $bleu[$i]; $search_graph_dir = get_current_analysis_filename("basic","search-graph"); if (file_exists($search_graph_dir) && file_exists($search_graph_dir."/graph.".$line["id"])) { $state = return_state_for_link(); @@ -1279,7 +1279,7 @@ function input_annotation($sentence,$input,$segmentation,$filter) { print "<tr><td colspan=".($sep_end-$sep_start)."><div style=\"position:relative; z-index:1;\">"; for($j=$sep_start;$j<$sep_end;$j++) { if ($segmentation && array_key_exists($j,$segmentation["input_start"])) { - $id = $segmentation["input_start"][$j]; + $id = $segmentation["input_start"][$j]; print "<span id=\"input-$sentence-$id\" style=\"border-color:#000000; border-style:solid; border-width:1px;\" onmouseover=\"highlight_phrase($sentence,$id);\" onmouseout=\"lowlight_phrase($sentence,$id);\">"; } if (array_key_exists($j,$coverage)) { @@ -1413,7 +1413,7 @@ function annotation_hierarchical($sentence,$segmentation,$segmentation_out,$node print "<span style=\"opacity:0\">|</span>"; } - $span_word = array(); + $span_word = array(); if ($words != "") { $span_word = split(" ",$words); } for($w=0;$w<count($span_word);$w++) { if ($w > 0) { print " "; } diff --git a/scripts/ems/web/analysis_diff.php b/scripts/ems/web/analysis_diff.php index 2f0947e13..214ae1592 100644 --- a/scripts/ems/web/analysis_diff.php +++ b/scripts/ems/web/analysis_diff.php @@ -1,4 +1,4 @@ -<?php +<?php function diff_analysis() { global $task,$user,$setup,$id,$id2,$set; @@ -15,7 +15,7 @@ function diff_analysis() { print "Run $id2 ($c2) vs $id ($c)"; } print "</h4>"; - + ?><script language="javascript" src="javascripts/prototype.js"></script> <script language="javascript" src="javascripts/scriptaculous.js"></script> <script> @@ -96,9 +96,9 @@ function precision_by_coverage_diff() { $log_info[$log_count]["precision"] += $item[1]; $log_info[$log_count]["delete"] += $item[2]; $log_info[$log_count]["length"] += $item[3]; - $log_info[$log_count]["total"] += $item[4]; + $log_info[$log_count]["total"] += $item[4]; } - $log_info_new = $log_info; + $log_info_new = $log_info; // load base data $data = file(get_current_analysis_filename("precision","precision-by-corpus-coverage")); @@ -154,7 +154,7 @@ function precision_by_coverage_diff_factored($img_width,$total,$file,$factor_id) $log_info_factored[$factor][$log_count]["precision"] += $item[2]; $log_info_factored[$factor][$log_count]["delete"] += $item[3]; $log_info_factored[$factor][$log_count]["length"] += $item[4]; - $log_info_factored[$factor][$log_count]["total"] += $item[5]; + $log_info_factored[$factor][$log_count]["total"] += $item[5]; } $info_factored_new = $info_factored; $info_factored_sum_new = $info_factored_sum; @@ -225,7 +225,7 @@ function precision_by_word_diff($type) { if ($byCoverage != -2 && $byCoverage != $log_count) { continue; } - + //# filter for factor $word = $item[5]; if ($byFactor != "false" && $byFactor != $item[6]) { @@ -258,7 +258,7 @@ function precision_by_word_diff($type) { if ($byCoverage != -2 && $byCoverage != $log_count) { continue; } - + //# filter for factor $word = $item[5]; if ($byFactor != "false" && $byFactor != $item[6]) { @@ -319,7 +319,7 @@ ctx.fillRect (0, 0, $size, $size); $surface = $item[5]; $word[$surface] = array(); $word[$surface]["precision"] = $item[0]; # number of precise translations - $word[$surface]["delete"] = $item[1]; # number of deleted + $word[$surface]["delete"] = $item[1]; # number of deleted $word[$surface]["total"] = $item[2]; # number of all translations $word[$surface]["coverage"] = $item[4]; # count in training corpus if ($item[4] == 0) { $log_count = -1; } @@ -369,7 +369,7 @@ ctx.fillRect (0, 0, $size, $size); $matrix[$base][$alt]["coverage2"] = 0; } # ignore mismatches in source words due to tokenization / casing - if (array_key_exists($surface,$word)) { + if (array_key_exists($surface,$word)) { $matrix[$base][$alt]["precision1"] += $word[$surface]["precision"]; $matrix[$base][$alt]["delete1"] += $word[$surface]["delete"]; $matrix[$base][$alt]["total1"] += $word[$surface]["total"]; @@ -413,14 +413,14 @@ ctx.fillRect (0, 0, $size, $size); $prec_imp = (int)(sqrt($prec1-$prec2)); $prec_color = "255,100,100"; } - else { + else { $prec_base = (int)(sqrt($prec2)); $prec_imp = (int)(sqrt($prec2-$prec1)); $prec_color = "100,255,100"; } $prec_base_top = (int)(($total-$prec_base)/2); $prec_imp_top = (int)(($total-$prec_imp)/2); - + $del1 = $matrix[$base][$alt]["delete1"]*$scale; $del2 = $matrix[$base][$alt]["delete2"]*$scale; if ($del1 > $del2) { @@ -428,7 +428,7 @@ ctx.fillRect (0, 0, $size, $size); $del_imp = $del1-$del2; $del_color = "150,100,255"; } - else { + else { $del_base = $del2; $del_imp = $del2-$del1; $del_color = "100,200,200"; @@ -470,7 +470,7 @@ ctx.fillRect (0, ".($total+$del_base_height).", $total, $del_imp_height); function precision_by_coverage_diff_matrix_details() { $alt = $_GET["alt"]; $base = $_GET["base"]; - + $impact_total = 0; $data = file(get_current_analysis_filename("precision","precision-by-input-word")); $word = array(); $class = array(); @@ -483,7 +483,7 @@ function precision_by_coverage_diff_matrix_details() { $surface = $item[5]; $word[$surface] = array(); $word[$surface]["precision"] = $item[0]; # number of precise translations - $word[$surface]["delete"] = $item[1]; # number of deleted + $word[$surface]["delete"] = $item[1]; # number of deleted $word[$surface]["total"] = $item[2]; # number of all translations $word[$surface]["coverage"] = $item[4]; # count in training corpus } @@ -502,7 +502,7 @@ function precision_by_coverage_diff_matrix_details() { $surface = $item[5]; if ($log_count-$base == $alt && array_key_exists($surface,$word)) { $precision = $item[0]; # number of precise translations - $delete = $item[1]; # number of deleted + $delete = $item[1]; # number of deleted $total = $item[3]; # number of all translations + deletions $coverage = $item[4]; # count in training corpus $surface = $item[5]; @@ -527,17 +527,17 @@ function precision_by_coverage_diff_matrix_details() { } } sort($all_out); - foreach($all_out as $out) { $o = explode("\t",$out); print $o[1]; } + foreach($all_out as $out) { $o = explode("\t",$out); print $o[1]; } print "</table>"; } function precision_by_coverage_diff_graph($name,$log_info,$log_info_new,$total,$img_width,$sort_type) { $keys = array_keys($log_info); sort($keys,$sort_type); - + print "<div id=\"Toggle$name\" onClick=\"document.getElementById('Table$name').style.display = 'none'; this.style.display = 'none';\" style=\"display:none;\"><font size=-2>(hide table)</font></div>\n"; print "<div id=\"Table$name\" style=\"display:none;\">\n"; - print "<table border=1><tr><td align=center> </td><td align=center colspan=3>Precision</td><td align=center colspan=2>Precision Impact</td><td align=center colspan=3>Delete</td><td align=center colspan=2>Delete Impact</td><td align=center>Length</td></tr>\n"; + print "<table border=1><tr><td align=center> </td><td align=center colspan=3>Precision</td><td align=center colspan=2>Precision Impact</td><td align=center colspan=3>Delete</td><td align=center colspan=2>Delete Impact</td><td align=center>Length</td></tr>\n"; foreach ($keys as $i) { if (array_key_exists($i,$log_info)) { print "<tr><td align=center>$i</td>"; @@ -595,7 +595,7 @@ ctx.font = '9px serif'; print "ctx.fillRect ($x, 250, $width, $height);"; $total_so_far += $log_info[$i]["total"]; - + if ($width>3) { print "ctx.fillStyle = \"rgb(0,0,0)\";"; // print "ctx.rotate(-1.5707);"; @@ -763,9 +763,9 @@ function bleu_diff_annotation() { $data = file(get_analysis_filename($dir,$set,$idx?$id2:$id,"basic","bleu-annotation")); for($i=0;$i<count($data);$i++) { $item = split("\t",$data[$i]); - $annotation[$item[1]]["bleu$idx"] = $item[0]; - $annotation[$item[1]]["system$idx"] = $item[2]; - $annotation[$item[1]]["reference"] = $item[3]; + $annotation[$item[1]]["bleu$idx"] = $item[0]; + $annotation[$item[1]]["system$idx"] = $item[2]; + $annotation[$item[1]]["reference"] = $item[3]; $annotation[$item[1]]["id"] = $item[1]; } } @@ -825,7 +825,7 @@ function bleu_diff_annotation() { // display for($i=0;$i<$count && $i<count($annotation);$i++) { - $line = $annotation[$i]; + $line = $annotation[$i]; print "<font size=-2>[src]</font> ".$input[$line["id"]]."<br>"; $word_with_score1 = split(" ",$line["system1"]); @@ -840,7 +840,7 @@ function bleu_diff_annotation() { $matched_with_score1 = preg_replace('/D/',"",$matched_with_score); bleu_line_diff( $word_with_score1, $matched1, $matched_with_score1 ); - print "<font size=-2>[".$id."-".$line["id"].":".$line["bleu0"]."]</font> "; + print "<font size=-2>[".$id."-".$line["id"].":".$line["bleu0"]."]</font> "; $matched0 = preg_replace('/I/',"",$matched); $matched_with_score0 = preg_replace('/I/',"",$matched_with_score); bleu_line_diff( $word_with_score0, $matched0, $matched_with_score0 ); @@ -875,14 +875,14 @@ function ngram_diff($type) { // load data $order = $_GET['order']; - + for($idx=0;$idx<2;$idx++) { $data = file(get_analysis_filename($dir,$set,$idx?$id2:$id,"basic","n-gram-$type.$order")); for($i=0;$i<count($data);$i++) { $item = split("\t",$data[$i]); - $ngram_hash[$item[2]]["total$idx"] = $item[0]; + $ngram_hash[$item[2]]["total$idx"] = $item[0]; $ngram_hash[$item[2]]["correct$idx"] = $item[1]; - } + } unset($data); } @@ -893,7 +893,7 @@ function ngram_diff($type) { $sort = 'ratio_worse'; $smooth = 1; } - + error_reporting(E_ERROR); // otherwise undefined counts trigger notices // sort index @@ -914,12 +914,12 @@ function ngram_diff($type) { + (2*$item["correct0"] - $item["total0"]); } else if ($sort == "ratio_worse") { - $item["index"] = + $item["index"] = ($item["correct1"] + $smooth) / ($item["total1"] + $smooth) - ($item["correct0"] + $smooth) / ($item["total0"] + $smooth); } else if ($sort == "ratio_better") { - $item["index"] = + $item["index"] = - ($item["correct1"] + $smooth) / ($item["total1"] + $smooth) + ($item["correct0"] + $smooth) / ($item["total0"] + $smooth); } @@ -1010,7 +1010,7 @@ function ngram_diff($type) { } else { printf("<td align=right>%+d</td><td>(%d)</td><td align=right>%+d</td><td>(%d)</td></tr>", $ok_diff,$ok,$wrong_diff,$wrong); - } + } } print "</table>\n"; } diff --git a/scripts/ems/web/bilingual-concordance.css b/scripts/ems/web/bilingual-concordance.css index 4648a21dd..f9941175e 100644 --- a/scripts/ems/web/bilingual-concordance.css +++ b/scripts/ems/web/bilingual-concordance.css @@ -17,57 +17,57 @@ text-align: center; } -table.biconcor { - table-layout: fixed; - padding: 0px; - margin: 0px; +table.biconcor { + table-layout: fixed; + padding: 0px; + margin: 0px; } -tr.biconcor { - padding: 0px; - margin: 0px; +tr.biconcor { + padding: 0px; + margin: 0px; } -td.biconcor { - white-space: nowrap; - overflow: hidden; - padding: 0px; - margin: 0px; +td.biconcor { + white-space: nowrap; + overflow: hidden; + padding: 0px; + margin: 0px; } -td.pp_source_left { +td.pp_source_left { font-size: 70%; - text-align: right; + text-align: right; } -td.pp_target_left { +td.pp_target_left { font-size: 70%; - text-align: right; + text-align: right; } -td.pp_source { +td.pp_source { font-size: 70%; - font-weight: bold; + font-weight: bold; } -td.pp_target { +td.pp_target { font-size: 70%; - font-weight: bold; + font-weight: bold; } -td.mismatch_target { +td.mismatch_target { font-size: 70%; text-align: center; } -td.pp_source_right { +td.pp_source_right { font-size: 70%; - border-style:solid; - border-width:0px 2px 0px 0px ; - border-color: black; + border-style:solid; + border-width:0px 2px 0px 0px ; + border-color: black; } -td.pp_target_right { +td.pp_target_right { font-size: 70%; } @@ -88,11 +88,11 @@ span.mismatch_misaligned { } span.mismatch_aligned { - font-weight: bold; + font-weight: bold; } td.pp_more { font-size: 70%; color: navy; - text-align: center; + text-align: center; } diff --git a/scripts/ems/web/comment.php b/scripts/ems/web/comment.php index 04628ea4f..dcd51ab00 100644 --- a/scripts/ems/web/comment.php +++ b/scripts/ems/web/comment.php @@ -1,4 +1,4 @@ -<?php +<?php $fp = fopen("comment","a"); fwrite($fp,$_GET{'run'} . ";" . $_GET{'text'} . "\n"); fclose($fp); diff --git a/scripts/ems/web/diff.php b/scripts/ems/web/diff.php index 71e732af1..f440d3240 100644 --- a/scripts/ems/web/diff.php +++ b/scripts/ems/web/diff.php @@ -37,7 +37,7 @@ function compute_diff($base,$change) { foreach ($all_parameters as $parameter) { if (!array_key_exists($parameter,$parameter_base)) { $parameter_base[$parameter] = ""; - } + } if (!array_key_exists($parameter,$parameter_change)) { $parameter_change[$parameter] = ""; } diff --git a/scripts/ems/web/hierarchical-segmentation.css b/scripts/ems/web/hierarchical-segmentation.css index 47f2c2693..e66555070 100644 --- a/scripts/ems/web/hierarchical-segmentation.css +++ b/scripts/ems/web/hierarchical-segmentation.css @@ -10,7 +10,7 @@ div.leaf { border:1px solid black; margin:2px; padding:0px; - font-weight: normal; + font-weight: normal; background-color: white; } div.empty { @@ -34,7 +34,7 @@ div.continued { margin-right: 0px; padding: 0px; padding-top: 0px; - font-weight: normal; + font-weight: normal; background-color: white; } div.opening { @@ -44,7 +44,7 @@ div.opening { margin:2px; margin-right: 0px; padding: 0px; - font-weight: normal; + font-weight: normal; background-color: white; } div.closing { @@ -54,6 +54,6 @@ div.closing { margin:2px; margin-left: 0px; padding: 0px; - font-weight: normal; + font-weight: normal; background-color: white; } diff --git a/scripts/ems/web/hierarchical-segmentation.js b/scripts/ems/web/hierarchical-segmentation.js index b4eb206ce..819ed121e 100644 --- a/scripts/ems/web/hierarchical-segmentation.js +++ b/scripts/ems/web/hierarchical-segmentation.js @@ -1,6 +1,6 @@ -var nodeIn = []; +var nodeIn = []; var nodeOut = []; -var nodeChildren = []; +var nodeChildren = []; var max_depth = []; var span_count_in = []; var span_count_out = []; diff --git a/scripts/ems/web/index.php b/scripts/ems/web/index.php index d216b114a..9c918a96a 100644 --- a/scripts/ems/web/index.php +++ b/scripts/ems/web/index.php @@ -29,7 +29,7 @@ else if (array_key_exists("setup",$_POST) || array_key_exists("setup",$_GET)) { if (array_key_exists("show",$_GET)) { show(); } else if (array_key_exists("diff",$_GET)) { diff(); } - else if (array_key_exists("analysis",$_GET)) { + else if (array_key_exists("analysis",$_GET)) { $action = $_GET["analysis"]; $set = $_GET["set"]; $id = $_GET["id"]; @@ -60,7 +60,7 @@ else if (array_key_exists("setup",$_POST) || array_key_exists("setup",$_GET)) { if ($match[2] == $set) { $id_array[] = $match[1]; } - } + } } if (count($id_array) != 2) { print "ERROR: comp 2!"; diff --git a/scripts/ems/web/javascripts/scriptaculous-js-1.8.3/src/unittest.js b/scripts/ems/web/javascripts/scriptaculous-js-1.8.3/src/unittest.js index 33a0c7157..cd3d19b39 100644 --- a/scripts/ems/web/javascripts/scriptaculous-js-1.8.3/src/unittest.js +++ b/scripts/ems/web/javascripts/scriptaculous-js-1.8.3/src/unittest.js @@ -19,10 +19,10 @@ Event.simulateMouse = function(element, eventName) { metaKey: false }, arguments[2] || {}); var oEvent = document.createEvent("MouseEvents"); - oEvent.initMouseEvent(eventName, true, true, document.defaultView, - options.buttons, options.pointerX, options.pointerY, options.pointerX, options.pointerY, + oEvent.initMouseEvent(eventName, true, true, document.defaultView, + options.buttons, options.pointerX, options.pointerY, options.pointerX, options.pointerY, options.ctrlKey, options.altKey, options.shiftKey, options.metaKey, 0, $(element)); - + if(this.mark) Element.remove(this.mark); this.mark = document.createElement('div'); this.mark.appendChild(document.createTextNode(" ")); @@ -34,10 +34,10 @@ Event.simulateMouse = function(element, eventName) { this.mark.style.height = "5px;"; this.mark.style.borderTop = "1px solid red;"; this.mark.style.borderLeft = "1px solid red;"; - + if(this.step) alert('['+new Date().getTime().toString()+'] '+eventName+'/'+Test.Unit.inspect(options)); - + $(element).dispatchEvent(oEvent); }; @@ -55,7 +55,7 @@ Event.simulateKey = function(element, eventName) { }, arguments[2] || {}); var oEvent = document.createEvent("KeyEvents"); - oEvent.initKeyEvent(eventName, true, true, window, + oEvent.initKeyEvent(eventName, true, true, window, options.ctrlKey, options.altKey, options.shiftKey, options.metaKey, options.keyCode, options.charCode ); $(element).dispatchEvent(oEvent); @@ -123,7 +123,7 @@ Test.Unit.Logger.prototype = { _toHTML: function(txt) { return txt.escapeHTML().replace(/\n/g,"<br/>"); }, - addLinksToResults: function(){ + addLinksToResults: function(){ $$("tr.failed .nameCell").each( function(td){ // todo: limit to children of this.log td.title = "Run only this test"; Event.observe(td, 'click', function(){ window.location.search = "?tests=" + td.innerHTML;}); @@ -162,7 +162,7 @@ Test.Unit.Runner.prototype = { if(/^test/.test(testcase)) { this.tests.push( new Test.Unit.Testcase( - this.options.context ? ' -> ' + this.options.titles[testcase] : testcase, + this.options.context ? ' -> ' + this.options.titles[testcase] : testcase, testcases[testcase], testcases["setup"], testcases["teardown"] )); } @@ -203,7 +203,7 @@ Test.Unit.Runner.prototype = { }, postResults: function() { if (this.options.resultsURL) { - new Ajax.Request(this.options.resultsURL, + new Ajax.Request(this.options.resultsURL, { method: 'get', parameters: 'result=' + this.getResult(), asynchronous: false }); } }, @@ -240,9 +240,9 @@ Test.Unit.Runner.prototype = { errors += this.tests[i].errors; } return ( - (this.options.context ? this.options.context + ': ': '') + - this.tests.length + " tests, " + - assertions + " assertions, " + + (this.options.context ? this.options.context + ': ': '') + + this.tests.length + " tests, " + + assertions + " assertions, " + failures + " failures, " + errors + " errors"); } @@ -258,7 +258,7 @@ Test.Unit.Assertions.prototype = { }, summary: function() { return ( - this.assertions + " assertions, " + + this.assertions + " assertions, " + this.failures + " failures, " + this.errors + " errors" + "\n" + this.messages.join("\n")); @@ -284,55 +284,55 @@ Test.Unit.Assertions.prototype = { }, assert: function(expression) { var message = arguments[1] || 'assert: got "' + Test.Unit.inspect(expression) + '"'; - try { expression ? this.pass() : + try { expression ? this.pass() : this.fail(message); } catch(e) { this.error(e); } }, assertEqual: function(expected, actual) { var message = arguments[2] || "assertEqual"; try { (expected == actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, assertInspect: function(expected, actual) { var message = arguments[2] || "assertInspect"; try { (expected == actual.inspect()) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, assertEnumEqual: function(expected, actual) { var message = arguments[2] || "assertEnumEqual"; - try { $A(expected).length == $A(actual).length && + try { $A(expected).length == $A(actual).length && expected.zip(actual).all(function(pair) { return pair[0] == pair[1] }) ? - this.pass() : this.fail(message + ': expected ' + Test.Unit.inspect(expected) + + this.pass() : this.fail(message + ': expected ' + Test.Unit.inspect(expected) + ', actual ' + Test.Unit.inspect(actual)); } catch(e) { this.error(e); } }, assertNotEqual: function(expected, actual) { var message = arguments[2] || "assertNotEqual"; - try { (expected != actual) ? this.pass() : + try { (expected != actual) ? this.pass() : this.fail(message + ': got "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, - assertIdentical: function(expected, actual) { - var message = arguments[2] || "assertIdentical"; - try { (expected === actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + - '", actual "' + Test.Unit.inspect(actual) + '"'); } - catch(e) { this.error(e); } + assertIdentical: function(expected, actual) { + var message = arguments[2] || "assertIdentical"; + try { (expected === actual) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + '", actual "' + Test.Unit.inspect(actual) + '"'); } + catch(e) { this.error(e); } }, - assertNotIdentical: function(expected, actual) { - var message = arguments[2] || "assertNotIdentical"; - try { !(expected === actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + - '", actual "' + Test.Unit.inspect(actual) + '"'); } - catch(e) { this.error(e); } + assertNotIdentical: function(expected, actual) { + var message = arguments[2] || "assertNotIdentical"; + try { !(expected === actual) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + '", actual "' + Test.Unit.inspect(actual) + '"'); } + catch(e) { this.error(e); } }, assertNull: function(obj) { var message = arguments[1] || 'assertNull'; - try { (obj==null) ? this.pass() : + try { (obj==null) ? this.pass() : this.fail(message + ': got "' + Test.Unit.inspect(obj) + '"'); } catch(e) { this.error(e); } }, @@ -353,38 +353,38 @@ Test.Unit.Assertions.prototype = { }, assertType: function(expected, actual) { var message = arguments[2] || 'assertType'; - try { - (actual.constructor == expected) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + try { + (actual.constructor == expected) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + (actual.constructor) + '"'); } catch(e) { this.error(e); } }, assertNotOfType: function(expected, actual) { var message = arguments[2] || 'assertNotOfType'; - try { - (actual.constructor != expected) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + try { + (actual.constructor != expected) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + (actual.constructor) + '"'); } catch(e) { this.error(e); } }, assertInstanceOf: function(expected, actual) { var message = arguments[2] || 'assertInstanceOf'; - try { - (actual instanceof expected) ? this.pass() : + try { + (actual instanceof expected) ? this.pass() : this.fail(message + ": object was not an instance of the expected type"); } - catch(e) { this.error(e); } + catch(e) { this.error(e); } }, assertNotInstanceOf: function(expected, actual) { var message = arguments[2] || 'assertNotInstanceOf'; - try { - !(actual instanceof expected) ? this.pass() : + try { + !(actual instanceof expected) ? this.pass() : this.fail(message + ": object was an instance of the not expected type"); } - catch(e) { this.error(e); } + catch(e) { this.error(e); } }, assertRespondsTo: function(method, obj) { var message = arguments[2] || 'assertRespondsTo'; try { - (obj[method] && typeof obj[method] == 'function') ? this.pass() : + (obj[method] && typeof obj[method] == 'function') ? this.pass() : this.fail(message + ": object doesn't respond to [" + method + "]"); } catch(e) { this.error(e); } }, @@ -393,7 +393,7 @@ Test.Unit.Assertions.prototype = { try { var m = obj[method]; if(!m) m = obj['is'+method.charAt(0).toUpperCase()+method.slice(1)]; - m() ? this.pass() : + m() ? this.pass() : this.fail(message + ": method returned false"); } catch(e) { this.error(e); } }, @@ -402,17 +402,17 @@ Test.Unit.Assertions.prototype = { try { var m = obj[method]; if(!m) m = obj['is'+method.charAt(0).toUpperCase()+method.slice(1)]; - !m() ? this.pass() : + !m() ? this.pass() : this.fail(message + ": method returned true"); } catch(e) { this.error(e); } }, assertRaise: function(exceptionName, method) { var message = arguments[2] || 'assertRaise'; - try { + try { method(); this.fail(message + ": exception expected but none was raised"); } catch(e) { - ((exceptionName == null) || (e.name==exceptionName)) ? this.pass() : this.error(e); + ((exceptionName == null) || (e.name==exceptionName)) ? this.pass() : this.error(e); } }, assertElementsMatch: function() { @@ -434,7 +434,7 @@ Test.Unit.Assertions.prototype = { var startAt = new Date(); (iterations || 1).times(operation); var timeTaken = ((new Date())-startAt); - this.info((arguments[2] || 'Operation') + ' finished ' + + this.info((arguments[2] || 'Operation') + ' finished ' + iterations + ' iterations in ' + (timeTaken/1000)+'s' ); return timeTaken; }, @@ -444,7 +444,7 @@ Test.Unit.Assertions.prototype = { this.assertNotNull(element); if(element.style && Element.getStyle(element, 'display') == 'none') return false; - + return this._isVisible(element.parentNode); }, assertNotVisible: function(element) { @@ -457,7 +457,7 @@ Test.Unit.Assertions.prototype = { var startAt = new Date(); (iterations || 1).times(operation); var timeTaken = ((new Date())-startAt); - this.info((arguments[2] || 'Operation') + ' finished ' + + this.info((arguments[2] || 'Operation') + ' finished ' + iterations + ' iterations in ' + (timeTaken/1000)+'s' ); return timeTaken; } @@ -468,7 +468,7 @@ Object.extend(Object.extend(Test.Unit.Testcase.prototype, Test.Unit.Assertions.p initialize: function(name, test, setup, teardown) { Test.Unit.Assertions.prototype.initialize.bind(this)(); this.name = name; - + if(typeof test == 'string') { test = test.gsub(/(\.should[^\(]+\()/,'#{0}this,'); test = test.gsub(/(\.should[^\(]+)\(this,\)/,'#{1}(this)'); @@ -478,7 +478,7 @@ Object.extend(Object.extend(Test.Unit.Testcase.prototype, Test.Unit.Assertions.p } else { this.test = test || function() {}; } - + this.setup = setup || function() {}; this.teardown = teardown || function() {}; this.isWaiting = false; @@ -519,23 +519,23 @@ Test.setupBDDExtensionMethods = function(){ shouldNotBeAn: 'assertNotOfType', shouldBeNull: 'assertNull', shouldNotBeNull: 'assertNotNull', - + shouldBe: 'assertReturnsTrue', shouldNotBe: 'assertReturnsFalse', shouldRespondTo: 'assertRespondsTo' }; - var makeAssertion = function(assertion, args, object) { + var makeAssertion = function(assertion, args, object) { this[assertion].apply(this,(args || []).concat([object])); }; - - Test.BDDMethods = {}; - $H(METHODMAP).each(function(pair) { - Test.BDDMethods[pair.key] = function() { - var args = $A(arguments); - var scope = args.shift(); - makeAssertion.apply(scope, [pair.value, args, this]); }; + + Test.BDDMethods = {}; + $H(METHODMAP).each(function(pair) { + Test.BDDMethods[pair.key] = function() { + var args = $A(arguments); + var scope = args.shift(); + makeAssertion.apply(scope, [pair.value, args, this]); }; }); - + [Array.prototype, String.prototype, Number.prototype, Boolean.prototype].each( function(p){ Object.extend(p, Test.BDDMethods) } ); @@ -543,7 +543,7 @@ Test.setupBDDExtensionMethods = function(){ Test.context = function(name, spec, log){ Test.setupBDDExtensionMethods(); - + var compiledSpec = {}; var titles = {}; for(specName in spec) { @@ -557,7 +557,7 @@ Test.context = function(name, spec, log){ var body = spec[specName].toString().split('\n').slice(1); if(/^\{/.test(body[0])) body = body.slice(1); body.pop(); - body = body.map(function(statement){ + body = body.map(function(statement){ return statement.strip() }); compiledSpec[testName] = body.join('\n'); diff --git a/scripts/ems/web/javascripts/sound.js b/scripts/ems/web/javascripts/sound.js index a286eb98e..0a63379fb 100644 --- a/scripts/ems/web/javascripts/sound.js +++ b/scripts/ems/web/javascripts/sound.js @@ -56,4 +56,4 @@ if(Prototype.Browser.Gecko && navigator.userAgent.indexOf("Win") > 0){ Sound.template = new Template('<embed type="audio/x-pn-realaudio-plugin" style="height:0" id="sound_#{track}_#{id}" src="#{url}" loop="false" autostart="true" hidden="true"/>'); else Sound.play = function(){}; -}
\ No newline at end of file +}
\ No newline at end of file diff --git a/scripts/ems/web/javascripts/unittest.js b/scripts/ems/web/javascripts/unittest.js index 33a0c7157..cd3d19b39 100644 --- a/scripts/ems/web/javascripts/unittest.js +++ b/scripts/ems/web/javascripts/unittest.js @@ -19,10 +19,10 @@ Event.simulateMouse = function(element, eventName) { metaKey: false }, arguments[2] || {}); var oEvent = document.createEvent("MouseEvents"); - oEvent.initMouseEvent(eventName, true, true, document.defaultView, - options.buttons, options.pointerX, options.pointerY, options.pointerX, options.pointerY, + oEvent.initMouseEvent(eventName, true, true, document.defaultView, + options.buttons, options.pointerX, options.pointerY, options.pointerX, options.pointerY, options.ctrlKey, options.altKey, options.shiftKey, options.metaKey, 0, $(element)); - + if(this.mark) Element.remove(this.mark); this.mark = document.createElement('div'); this.mark.appendChild(document.createTextNode(" ")); @@ -34,10 +34,10 @@ Event.simulateMouse = function(element, eventName) { this.mark.style.height = "5px;"; this.mark.style.borderTop = "1px solid red;"; this.mark.style.borderLeft = "1px solid red;"; - + if(this.step) alert('['+new Date().getTime().toString()+'] '+eventName+'/'+Test.Unit.inspect(options)); - + $(element).dispatchEvent(oEvent); }; @@ -55,7 +55,7 @@ Event.simulateKey = function(element, eventName) { }, arguments[2] || {}); var oEvent = document.createEvent("KeyEvents"); - oEvent.initKeyEvent(eventName, true, true, window, + oEvent.initKeyEvent(eventName, true, true, window, options.ctrlKey, options.altKey, options.shiftKey, options.metaKey, options.keyCode, options.charCode ); $(element).dispatchEvent(oEvent); @@ -123,7 +123,7 @@ Test.Unit.Logger.prototype = { _toHTML: function(txt) { return txt.escapeHTML().replace(/\n/g,"<br/>"); }, - addLinksToResults: function(){ + addLinksToResults: function(){ $$("tr.failed .nameCell").each( function(td){ // todo: limit to children of this.log td.title = "Run only this test"; Event.observe(td, 'click', function(){ window.location.search = "?tests=" + td.innerHTML;}); @@ -162,7 +162,7 @@ Test.Unit.Runner.prototype = { if(/^test/.test(testcase)) { this.tests.push( new Test.Unit.Testcase( - this.options.context ? ' -> ' + this.options.titles[testcase] : testcase, + this.options.context ? ' -> ' + this.options.titles[testcase] : testcase, testcases[testcase], testcases["setup"], testcases["teardown"] )); } @@ -203,7 +203,7 @@ Test.Unit.Runner.prototype = { }, postResults: function() { if (this.options.resultsURL) { - new Ajax.Request(this.options.resultsURL, + new Ajax.Request(this.options.resultsURL, { method: 'get', parameters: 'result=' + this.getResult(), asynchronous: false }); } }, @@ -240,9 +240,9 @@ Test.Unit.Runner.prototype = { errors += this.tests[i].errors; } return ( - (this.options.context ? this.options.context + ': ': '') + - this.tests.length + " tests, " + - assertions + " assertions, " + + (this.options.context ? this.options.context + ': ': '') + + this.tests.length + " tests, " + + assertions + " assertions, " + failures + " failures, " + errors + " errors"); } @@ -258,7 +258,7 @@ Test.Unit.Assertions.prototype = { }, summary: function() { return ( - this.assertions + " assertions, " + + this.assertions + " assertions, " + this.failures + " failures, " + this.errors + " errors" + "\n" + this.messages.join("\n")); @@ -284,55 +284,55 @@ Test.Unit.Assertions.prototype = { }, assert: function(expression) { var message = arguments[1] || 'assert: got "' + Test.Unit.inspect(expression) + '"'; - try { expression ? this.pass() : + try { expression ? this.pass() : this.fail(message); } catch(e) { this.error(e); } }, assertEqual: function(expected, actual) { var message = arguments[2] || "assertEqual"; try { (expected == actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, assertInspect: function(expected, actual) { var message = arguments[2] || "assertInspect"; try { (expected == actual.inspect()) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, assertEnumEqual: function(expected, actual) { var message = arguments[2] || "assertEnumEqual"; - try { $A(expected).length == $A(actual).length && + try { $A(expected).length == $A(actual).length && expected.zip(actual).all(function(pair) { return pair[0] == pair[1] }) ? - this.pass() : this.fail(message + ': expected ' + Test.Unit.inspect(expected) + + this.pass() : this.fail(message + ': expected ' + Test.Unit.inspect(expected) + ', actual ' + Test.Unit.inspect(actual)); } catch(e) { this.error(e); } }, assertNotEqual: function(expected, actual) { var message = arguments[2] || "assertNotEqual"; - try { (expected != actual) ? this.pass() : + try { (expected != actual) ? this.pass() : this.fail(message + ': got "' + Test.Unit.inspect(actual) + '"'); } catch(e) { this.error(e); } }, - assertIdentical: function(expected, actual) { - var message = arguments[2] || "assertIdentical"; - try { (expected === actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + - '", actual "' + Test.Unit.inspect(actual) + '"'); } - catch(e) { this.error(e); } + assertIdentical: function(expected, actual) { + var message = arguments[2] || "assertIdentical"; + try { (expected === actual) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + '", actual "' + Test.Unit.inspect(actual) + '"'); } + catch(e) { this.error(e); } }, - assertNotIdentical: function(expected, actual) { - var message = arguments[2] || "assertNotIdentical"; - try { !(expected === actual) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + - '", actual "' + Test.Unit.inspect(actual) + '"'); } - catch(e) { this.error(e); } + assertNotIdentical: function(expected, actual) { + var message = arguments[2] || "assertNotIdentical"; + try { !(expected === actual) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + '", actual "' + Test.Unit.inspect(actual) + '"'); } + catch(e) { this.error(e); } }, assertNull: function(obj) { var message = arguments[1] || 'assertNull'; - try { (obj==null) ? this.pass() : + try { (obj==null) ? this.pass() : this.fail(message + ': got "' + Test.Unit.inspect(obj) + '"'); } catch(e) { this.error(e); } }, @@ -353,38 +353,38 @@ Test.Unit.Assertions.prototype = { }, assertType: function(expected, actual) { var message = arguments[2] || 'assertType'; - try { - (actual.constructor == expected) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + try { + (actual.constructor == expected) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + (actual.constructor) + '"'); } catch(e) { this.error(e); } }, assertNotOfType: function(expected, actual) { var message = arguments[2] || 'assertNotOfType'; - try { - (actual.constructor != expected) ? this.pass() : - this.fail(message + ': expected "' + Test.Unit.inspect(expected) + + try { + (actual.constructor != expected) ? this.pass() : + this.fail(message + ': expected "' + Test.Unit.inspect(expected) + '", actual "' + (actual.constructor) + '"'); } catch(e) { this.error(e); } }, assertInstanceOf: function(expected, actual) { var message = arguments[2] || 'assertInstanceOf'; - try { - (actual instanceof expected) ? this.pass() : + try { + (actual instanceof expected) ? this.pass() : this.fail(message + ": object was not an instance of the expected type"); } - catch(e) { this.error(e); } + catch(e) { this.error(e); } }, assertNotInstanceOf: function(expected, actual) { var message = arguments[2] || 'assertNotInstanceOf'; - try { - !(actual instanceof expected) ? this.pass() : + try { + !(actual instanceof expected) ? this.pass() : this.fail(message + ": object was an instance of the not expected type"); } - catch(e) { this.error(e); } + catch(e) { this.error(e); } }, assertRespondsTo: function(method, obj) { var message = arguments[2] || 'assertRespondsTo'; try { - (obj[method] && typeof obj[method] == 'function') ? this.pass() : + (obj[method] && typeof obj[method] == 'function') ? this.pass() : this.fail(message + ": object doesn't respond to [" + method + "]"); } catch(e) { this.error(e); } }, @@ -393,7 +393,7 @@ Test.Unit.Assertions.prototype = { try { var m = obj[method]; if(!m) m = obj['is'+method.charAt(0).toUpperCase()+method.slice(1)]; - m() ? this.pass() : + m() ? this.pass() : this.fail(message + ": method returned false"); } catch(e) { this.error(e); } }, @@ -402,17 +402,17 @@ Test.Unit.Assertions.prototype = { try { var m = obj[method]; if(!m) m = obj['is'+method.charAt(0).toUpperCase()+method.slice(1)]; - !m() ? this.pass() : + !m() ? this.pass() : this.fail(message + ": method returned true"); } catch(e) { this.error(e); } }, assertRaise: function(exceptionName, method) { var message = arguments[2] || 'assertRaise'; - try { + try { method(); this.fail(message + ": exception expected but none was raised"); } catch(e) { - ((exceptionName == null) || (e.name==exceptionName)) ? this.pass() : this.error(e); + ((exceptionName == null) || (e.name==exceptionName)) ? this.pass() : this.error(e); } }, assertElementsMatch: function() { @@ -434,7 +434,7 @@ Test.Unit.Assertions.prototype = { var startAt = new Date(); (iterations || 1).times(operation); var timeTaken = ((new Date())-startAt); - this.info((arguments[2] || 'Operation') + ' finished ' + + this.info((arguments[2] || 'Operation') + ' finished ' + iterations + ' iterations in ' + (timeTaken/1000)+'s' ); return timeTaken; }, @@ -444,7 +444,7 @@ Test.Unit.Assertions.prototype = { this.assertNotNull(element); if(element.style && Element.getStyle(element, 'display') == 'none') return false; - + return this._isVisible(element.parentNode); }, assertNotVisible: function(element) { @@ -457,7 +457,7 @@ Test.Unit.Assertions.prototype = { var startAt = new Date(); (iterations || 1).times(operation); var timeTaken = ((new Date())-startAt); - this.info((arguments[2] || 'Operation') + ' finished ' + + this.info((arguments[2] || 'Operation') + ' finished ' + iterations + ' iterations in ' + (timeTaken/1000)+'s' ); return timeTaken; } @@ -468,7 +468,7 @@ Object.extend(Object.extend(Test.Unit.Testcase.prototype, Test.Unit.Assertions.p initialize: function(name, test, setup, teardown) { Test.Unit.Assertions.prototype.initialize.bind(this)(); this.name = name; - + if(typeof test == 'string') { test = test.gsub(/(\.should[^\(]+\()/,'#{0}this,'); test = test.gsub(/(\.should[^\(]+)\(this,\)/,'#{1}(this)'); @@ -478,7 +478,7 @@ Object.extend(Object.extend(Test.Unit.Testcase.prototype, Test.Unit.Assertions.p } else { this.test = test || function() {}; } - + this.setup = setup || function() {}; this.teardown = teardown || function() {}; this.isWaiting = false; @@ -519,23 +519,23 @@ Test.setupBDDExtensionMethods = function(){ shouldNotBeAn: 'assertNotOfType', shouldBeNull: 'assertNull', shouldNotBeNull: 'assertNotNull', - + shouldBe: 'assertReturnsTrue', shouldNotBe: 'assertReturnsFalse', shouldRespondTo: 'assertRespondsTo' }; - var makeAssertion = function(assertion, args, object) { + var makeAssertion = function(assertion, args, object) { this[assertion].apply(this,(args || []).concat([object])); }; - - Test.BDDMethods = {}; - $H(METHODMAP).each(function(pair) { - Test.BDDMethods[pair.key] = function() { - var args = $A(arguments); - var scope = args.shift(); - makeAssertion.apply(scope, [pair.value, args, this]); }; + + Test.BDDMethods = {}; + $H(METHODMAP).each(function(pair) { + Test.BDDMethods[pair.key] = function() { + var args = $A(arguments); + var scope = args.shift(); + makeAssertion.apply(scope, [pair.value, args, this]); }; }); - + [Array.prototype, String.prototype, Number.prototype, Boolean.prototype].each( function(p){ Object.extend(p, Test.BDDMethods) } ); @@ -543,7 +543,7 @@ Test.setupBDDExtensionMethods = function(){ Test.context = function(name, spec, log){ Test.setupBDDExtensionMethods(); - + var compiledSpec = {}; var titles = {}; for(specName in spec) { @@ -557,7 +557,7 @@ Test.context = function(name, spec, log){ var body = spec[specName].toString().split('\n').slice(1); if(/^\{/.test(body[0])) body = body.slice(1); body.pop(); - body = body.map(function(statement){ + body = body.map(function(statement){ return statement.strip() }); compiledSpec[testName] = body.join('\n'); diff --git a/scripts/ems/web/lib.php b/scripts/ems/web/lib.php index 6d936ddf4..68c58860b 100644 --- a/scripts/ems/web/lib.php +++ b/scripts/ems/web/lib.php @@ -21,7 +21,7 @@ function load_experiment_info() { file_exists($dir."/steps/1")) { $topd = dir($dir."/steps"); while (false !== ($run = $topd->read())) { - if (preg_match('/^([0-9]+)$/',$run,$match) + if (preg_match('/^([0-9]+)$/',$run,$match) && $run>0 && !file_exists("$dir/steps/$run/deleted.$run")) { $d = dir($dir."/steps/$run"); @@ -49,7 +49,7 @@ function load_experiment_info() { } $experiment[$id]->start = $stat[9]; } - + reset($experiment); while (list($id,$info) = each($experiment)) { if (file_exists("$dir/evaluation/report.$id")) { @@ -57,15 +57,15 @@ function load_experiment_info() { foreach ($f as $line_num => $line) { if (preg_match('/^(.+): (.+)/',$line,$match)) { $experiment[$id]->result[$match[1]] = $match[2]; - if (!$evalset || !array_key_exists($match[1],$evalset)) { - $evalset[$match[1]] = 0; + if (!$evalset || !array_key_exists($match[1],$evalset)) { + $evalset[$match[1]] = 0; } $evalset[$match[1]]++; } } } } - + krsort($experiment); uksort($evalset,"evalsetsort"); } @@ -81,7 +81,7 @@ function load_parameter($run) { if (file_exists($dir."/steps/new") || file_exists($dir."/steps/$run")) { $file = file("$dir/steps/$run/parameter.$run"); - } + } else { $file = file("$dir/steps/parameter.$run"); } @@ -121,7 +121,7 @@ function process_file_entry($dir,$entry) { if (file_exists($file.".STDOUT")) { $stat2 = stat($file.".STDOUT"); } if ($stat2[9] > $stat[9]) { $stat = $stat2; } $time = $stat[9]; - + if (!$experiment || !array_key_exists($run,$experiment) || !property_exists($experiment[$run],"last_step_time") || $time > $experiment[$run]->last_step_time) { @@ -142,7 +142,7 @@ function get_analysis_version($dir,$set,$id) { #while(list($type,$i) = each($analysis_version[$id][$set])) { # print "$type=$i "; #} - #print ") FROM CACHE<br>"; + #print ") FROM CACHE<br>"; return $analysis_version[$id][$set]; } $analysis_version[$id][$set]["basic"] = 0; @@ -188,7 +188,7 @@ function get_analysis_version($dir,$set,$id) { file_exists("$dir/model/biconcor.$match[1]")) { $analysis_version[$id][$set]["biconcor"] = $match[1]; } - } + } } # legacy stuff below... @@ -225,7 +225,7 @@ function get_analysis_version($dir,$set,$id) { #while(list($type,$i) = each($analysis_version[$id][$set])) { # print "$type=$i "; #} - #print ") ZZ<br>"; + #print ") ZZ<br>"; return $analysis_version[$id][$set]; } diff --git a/scripts/ems/web/overview.php b/scripts/ems/web/overview.php index ce0434bb8..534c7d8c0 100644 --- a/scripts/ems/web/overview.php +++ b/scripts/ems/web/overview.php @@ -39,7 +39,7 @@ function overview() { $report_info = "$dir/steps/$id/REPORTING_report.$id.INFO"; // does the analysis file exist? if (file_exists($analysis)) { - if (!array_key_exists($set,$has_analysis)) { + if (!array_key_exists($set,$has_analysis)) { $has_analysis[$set] = 0; } $has_analysis[$set]++; @@ -138,7 +138,7 @@ new Ajax.Updater("<?php print "$module_step[0]-$module_step[1]-$id"; ?>", '?setS } } } - else { $score = ""; } + else { $score = ""; } } } print "var best_score = [];\n"; @@ -158,9 +158,9 @@ function getHTTPObject(){ alert("Your browser does not support AJAX."); return null; } -} +} function createCommentBox( runID ) { - document.getElementById("run-" + runID).innerHTML = "<form onsubmit=\"return false;\"><input id=\"comment-" + runID + "\" name=\"comment-" + runID + "\" size=30><br><input type=submit onClick=\"addComment('" + runID + "');\" value=\"Add Comment\"></form>"; + document.getElementById("run-" + runID).innerHTML = "<form onsubmit=\"return false;\"><input id=\"comment-" + runID + "\" name=\"comment-" + runID + "\" size=30><br><input type=submit onClick=\"addComment('" + runID + "');\" value=\"Add Comment\"></form>"; if (currentComment[runID]) { document.getElementById("comment-" + runID).value = currentComment[runID]; } @@ -196,7 +196,7 @@ function highlightBest() { for (run in score) { var column = "score-"+run+"-"+set; if ($(column)) { - if (score[run][set] == best_score[set]) { + if (score[run][set] == best_score[set]) { $(column).setStyle({ backgroundColor: '#a0ffa0'}); } else if (score[run][set]+1 >= best_score[set]) { @@ -219,7 +219,7 @@ function highlightLine( id ) { $(column).setStyle({ backgroundColor: '#ffffff'}); } else { - if (score[run][set] < score[id][set]-1) { + if (score[run][set] < score[id][set]-1) { $(column).setStyle({ backgroundColor: '#ffa0a0'}); } else if (score[run][set] < score[id][set]) { @@ -234,7 +234,7 @@ function highlightLine( id ) { } } } - } + } } function lowlightAll() { for (run in score) { @@ -298,13 +298,13 @@ function output_score($id,$info) { preg_match('/([\d\(\)\.\s]+) (METEOR[\-c]*)/',$each_score[$i],$match)) { if ($i>0) { print "<BR>"; } $opened_a_tag = 0; - if ($set != "avg") { + if ($set != "avg") { if (file_exists("$dir/evaluation/$set.cleaned.$id")) { - print "<a href=\"?$state&show=evaluation/$set.cleaned.$id\">"; + print "<a href=\"?$state&show=evaluation/$set.cleaned.$id\">"; $opened_a_tag = 1; } else if (file_exists("$dir/evaluation/$set.output.$id")) { - print "<a href=\"?$state&show=evaluation/$set.output.$id\">"; + print "<a href=\"?$state&show=evaluation/$set.output.$id\">"; $opened_a_tag = 1; } } @@ -336,7 +336,7 @@ function tune_status($id) { if (! file_exists($dir."/tuning/tmp.".$id)) { return ""; } $d = dir($dir."/tuning/tmp.".$id); while (false !== ($entry = $d->read())) { - if (preg_match('/run(\d+).moses.ini/',$entry,$match) + if (preg_match('/run(\d+).moses.ini/',$entry,$match) && $match[1] > $max_iteration) { $max_iteration = $match[1]; } @@ -383,7 +383,7 @@ function show() { } $fullname = $dir."/steps/".$extra.$_GET["show"]; - if (preg_match("/\//",$_GET["show"])) { + if (preg_match("/\//",$_GET["show"])) { $fullname = $dir."/".$_GET["show"]; } if (preg_match("/graph/",$fullname)) { diff --git a/scripts/ems/web/progress.perl b/scripts/ems/web/progress.perl index fd742e410..fa2ce9e8f 100755 --- a/scripts/ems/web/progress.perl +++ b/scripts/ems/web/progress.perl @@ -1,4 +1,4 @@ -#!/usr/bin/env perl +#!/usr/bin/env perl use warnings; use strict; diff --git a/scripts/ems/web/sgviz.js b/scripts/ems/web/sgviz.js index 03ad4741a..3e63260ee 100644 --- a/scripts/ems/web/sgviz.js +++ b/scripts/ems/web/sgviz.js @@ -54,7 +54,7 @@ function process_hypotheses() { // // INITIALIZATION -// +// function index_hypotheses_by_cell() { // init edge_lists @@ -93,7 +93,7 @@ function find_reachable_hypotheses_recursive( id ) { find_reachable_hypotheses_recursive( children[c] ); } } - + function compute_best_derivation_scores() { for(var from=0; from<length; from++ ) { cell_derivation_score[from] = Array(); @@ -147,7 +147,7 @@ function click_menu( id, force_flag ) { return; } menu_processing = 1; - + if (current_menu_selection == 1) { best_derivation(0); } if (current_menu_selection == 2) { unannotate_cells(); } if (current_menu_selection == 3) { unannotate_cells(); } @@ -155,7 +155,7 @@ function click_menu( id, force_flag ) { if (current_menu_selection == 5) { remove_non_terminal_treemap(0); } if (current_menu_selection == 6 && SORT_OPTION != 3) { remove_hypothesis_overview(); } if (current_menu_selection == 6 && SORT_OPTION == 3) { remove_hypothesis_overview(); remove_non_terminal_treemap(); } - if (current_menu_selection > 0) { + if (current_menu_selection > 0) { highlight_menu_button( current_menu_selection, 0 ); } @@ -196,8 +196,8 @@ function draw_menu_button( id, label ) { var content = document.createTextNode( label ); button_label.appendChild( content ); button_label.setAttribute("onclick","click_menu(" + id + ",0);") - - chart.appendChild( button_label ); + + chart.appendChild( button_label ); } function highlight_menu_button( id, on_off ) { @@ -251,8 +251,8 @@ function draw_option_button( rule_option, id, label ) { button_label.setAttribute("pointer-events", "none"); var content = document.createTextNode( label ); button_label.appendChild( content ); - - chart.appendChild( button_label ); + + chart.appendChild( button_label ); } function draw_sort_button( id, label ) { @@ -294,8 +294,8 @@ function draw_sort_button( id, label ) { button_label.setAttribute("pointer-events", "none"); var content = document.createTextNode( label ); button_label.appendChild( content ); - - chart.appendChild( button_label ); + + chart.appendChild( button_label ); } function click_sort( id ) { @@ -319,17 +319,17 @@ var show_scores = 0; var show_id = 0; var show_derivation = 0; function click_option( id ) { - if (id == 1) { - show_scores = !show_scores; + if (id == 1) { + show_scores = !show_scores; highlight_option_button( 0, 1, show_scores ); } - if (id == 2) { - show_derivation = !show_derivation; + if (id == 2) { + show_derivation = !show_derivation; color_cells(); highlight_option_button( 0, 2, show_derivation ); } - if (id == 3) { - show_id = !show_id; + if (id == 3) { + show_id = !show_id; highlight_option_button( 0, 3, show_id ); } if (current_menu_selection > 0) { @@ -340,15 +340,15 @@ function click_option( id ) { var show_hyp_score = 0; var show_derivation_score = 0; function click_rule_option( id ) { - if (id == 1) { - show_hyp_score = !show_hyp_score; + if (id == 1) { + show_hyp_score = !show_hyp_score; highlight_option_button( 1, 1, show_hyp_score ); } - if (id == 2) { - show_derivation_score = !show_derivation_score; + if (id == 2) { + show_derivation_score = !show_derivation_score; highlight_option_button( 1, 2, show_derivation_score ); } - if (id == 3) { + if (id == 3) { if (ZOOM > 0) { ZOOM = 0; } @@ -377,7 +377,7 @@ function draw_chart() { for (var from=0;from<length;from++) { for(var width=1; width<=length-from; width++) { var to = from + width - 1; - + // logical container var container = document.createElementNS(xmlns,"svg"); container.setAttribute("id", "cell-container-" + from + "-" + to); @@ -385,7 +385,7 @@ function draw_chart() { var transform = document.createElementNS(xmlns,"g"); transform.setAttribute("id", "cell-" + from + "-" + to); container.appendChild( transform ); - + // yellow box for the cell var cell = document.createElementNS(xmlns,"rect"); cell.setAttribute("id", "cellbox-" + from + "-" + to); @@ -403,7 +403,7 @@ function draw_chart() { cell.setAttribute("onclick","click_cell(" + from + "," + to + ");") transform.appendChild( cell ); } - + // box for the input word var input_box = document.createElementNS(xmlns,"rect"); input_box.setAttribute("id", "inputbox-" + from); @@ -415,8 +415,8 @@ function draw_chart() { input_box.setAttribute("height", CELL_HEIGHT/2); //cell.setAttribute("opacity", .75); input_box.setAttribute("fill", INPUT_REGULAR_COLOR); - chart.appendChild( input_box ); - + chart.appendChild( input_box ); + // input word input_word = document.createElementNS(xmlns,"text"); input_word.setAttribute("id", "input-" + from); @@ -426,7 +426,7 @@ function draw_chart() { input_word.setAttribute("text-anchor", "middle"); var content = document.createTextNode( input[from] ); input_word.appendChild( content ); - chart.appendChild( input_word ); + chart.appendChild( input_word ); } assign_chart_coordinates(); } @@ -435,7 +435,7 @@ function assign_chart_coordinates() { for (var from=0;from<length;from++) { for(var width=1; width<=length-from; width++) { var to = from + width - 1; - + var x = from*CELL_WIDTH + (width-1)*CELL_WIDTH/2; var y = (length-width)*CELL_HEIGHT*(1-ZOOM); //alert("(x,y) = (" + length + "," + width + "), width = " + ZOOM + ", height = " + (1-ZOOM)); @@ -505,7 +505,7 @@ var current_from = -1; var current_to; function hover_cell( from, to ) { if (current_from >= 0) { - highlight_input( current_from, current_to, 0) + highlight_input( current_from, current_to, 0) } highlight_input( from, to, 1) current_from = from; @@ -534,7 +534,7 @@ function highlight_input( from, to, on_off ) { } } -// +// // VISUALIZATION OF CHART CELLS // @@ -570,7 +570,7 @@ function annotate_cells_with_rulecount() { function annotate_cells_with_derivation_score() { for (var from=0;from<length;from++) { for(var width=1; width<=length-from; width++) { - var to = from + width - 1; + var to = from + width - 1; var score = cell_derivation_score[from][to]; if (score < -9e9) { score = "dead end"; } annotate_cell( from, to, score, 15 ) @@ -596,12 +596,12 @@ function annotate_cell( from, to, label, font_size ) { cell_label.setAttribute("pointer-events", "none"); cell_label.setAttribute("text-anchor", "middle"); var content = document.createTextNode(line[i]); - cell_label.appendChild( content ); + cell_label.appendChild( content ); cell_label_group.appendChild( cell_label ); } var cell = document.getElementById("cell-" + from + "-" + to); - cell.appendChild( cell_label_group ); + cell.appendChild( cell_label_group ); } function unannotate_cells() { @@ -624,7 +624,7 @@ function unannotate_cell( from, to ) { function non_terminal_treemap( with_hyps ) { for (var from=0;from<length;from++) { for(var width=1; width<=length-from; width++) { - var to = from + width - 1; + var to = from + width - 1; // get nt counts var lhs = new Array(); var lhs_list = new Array(); @@ -639,7 +639,7 @@ function non_terminal_treemap( with_hyps ) { lhs[nt]++; } } - // sort + // sort function sortByCount(a,b) { return lhs[b] - lhs[a]; } @@ -652,7 +652,7 @@ function non_terminal_treemap( with_hyps ) { function remove_non_terminal_treemap() { for (var from=0;from<length;from++) { for(var width=1; width<=length-from; width++) { - var to = from + width - 1; + var to = from + width - 1; var cell = document.getElementById("cell-" + from + "-" + to); var done = false; var j=0; @@ -693,7 +693,7 @@ function treemap_cell( from, to, label, count, total, with_hyps ) { rect.setAttribute("stroke", "black"); rect.setAttribute("stroke-width", "0.5"); rect.setAttribute("onclick","click_menu(" + id + ",0);") - cell.appendChild( rect ); + cell.appendChild( rect ); x += width * count[label[i]] / total; } } @@ -749,7 +749,7 @@ function treemap_squarify( from, to, label, count, total, with_hyps ) { rect.setAttribute("pointer-events", "none"); rect.setAttribute("stroke", "black"); rect.setAttribute("stroke-width", "0.5"); - cell.appendChild( rect ); + cell.appendChild( rect ); // hypotheses if (with_hyps) { var hyp_list = Array(); @@ -760,15 +760,15 @@ function treemap_squarify( from, to, label, count, total, with_hyps ) { hyp_list.push( id ); } } - hypothesis_in_rect( this_width * scale_factor - 2, - this_height * scale_factor - 2, - CELL_MARGIN + (offset_x + cum_x) * scale_factor + 1, - CELL_MARGIN + (offset_y + cum_y) * scale_factor + 1, + hypothesis_in_rect( this_width * scale_factor - 2, + this_height * scale_factor - 2, + CELL_MARGIN + (offset_x + cum_x) * scale_factor + 1, + CELL_MARGIN + (offset_y + cum_y) * scale_factor + 1, cell, hyp_list ); } // label var font_size = Math.min( Math.round(this_width * scale_factor / label[j].length * 1.3), - Math.round(this_height * scale_factor )); + Math.round(this_height * scale_factor )); if (font_size > 20) { font_size = 20; } if (font_size >= 3) { var rect_label = document.createElementNS(xmlns,"text"); @@ -782,7 +782,7 @@ function treemap_squarify( from, to, label, count, total, with_hyps ) { rect_label.setAttribute("pointer-events", "none"); var content = document.createTextNode( label[j] ); rect_label.appendChild( content ); - cell.appendChild( rect_label ); + cell.appendChild( rect_label ); } if (adding_on_left) { cum_y += this_height; } else { cum_x += this_width; } @@ -829,7 +829,7 @@ function best_derivation( on_off ) { best_derivation_recurse( best_id, on_off, -1, -1, 0 ); } -function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos ) { +function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos ) { var from = edge[id][FROM]; var to = edge[id][TO]; @@ -846,10 +846,10 @@ function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos else { unannotate_cell( from, to ); } - + // highlight hyp highlight_hyp( id, on_off ); - + // arrow to parent if (parent_from >= 0) { if (on_off) { @@ -860,7 +860,7 @@ function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos chart.removeChild(arrow); } } - + var child_order = Array(); if (edge[id][ALIGNMENT] != "") { var alignment = edge[id][ALIGNMENT].split(" "); @@ -880,7 +880,7 @@ function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos child_order[target_source[1]] = i; } } - + // recurse var covered = new Array; var children = get_children( id ); @@ -888,7 +888,7 @@ function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos var child = children[c]; for( var i=edge[child][FROM]; i<=edge[child][TO]; i++ ) { covered[i] = 1; - } + } best_derivation_recurse( child, on_off, from, to, children.length == 1 ? 0.5 : child_order[c]/(children.length-1.0) ); } @@ -900,10 +900,10 @@ function best_derivation_recurse( id, on_off, parent_from, parent_to, child_pos } else { var arrow = document.getElementById("arrow-word-" + i); - chart.removeChild(arrow); + chart.removeChild(arrow); } } - } + } } function make_arrow( id, parent_from, parent_to, from, to, word_flag, position ) { @@ -1034,7 +1034,7 @@ function hypothesis_in_rect( width, height, offset_x, offset_y, parent_element, for (var i=0; i<hyp_list.length;i++) { id = hyp_list[i]; - + //alert("adding circle (" + (x + diameter/2) + "," + (y + diameter/2) + ") - " + (diameter/2) ); var hyp = document.createElementNS(xmlns,"circle"); hyp.setAttribute("id", "hyp-" + id); @@ -1045,7 +1045,7 @@ function hypothesis_in_rect( width, height, offset_x, offset_y, parent_element, hyp.setAttribute("onmouseover","hover_hyp(" + id + ");") hyp.setAttribute("onmouseout","unhover_hyp(" + id + ");") parent_element.appendChild( hyp ); - + x += diameter; if (++column >= row_size) { column = 0; @@ -1114,7 +1114,7 @@ function hyp_color( id, on_off ) { if (on_off) { var color = "#ff0000"; if (edge[id][RECOMBINED]>0) { color = "#808080"; } - else if (id in reachable) { color = "#00c000"; } + else if (id in reachable) { color = "#00c000"; } return color; } var color = "#ffc0c0"; @@ -1136,26 +1136,26 @@ function get_rule( id ) { var source_target = alignment[i].split("-"); nt_label.push(output[source_target[1]]); } - + var rule = edge[id][LHS]+"\u2192"; var children = get_children(id); var pos = edge[id][FROM]; for (var i=0; i<children.length; i++) { if (pos != edge[id][FROM]) { rule += " "; } var child = children[i]; - for(;pos<edge[child][FROM];pos++) { - rule += (input[pos].length <= 10) ? input[pos] : input[pos].substr(0,8) + "."; + for(;pos<edge[child][FROM];pos++) { + rule += (input[pos].length <= 10) ? input[pos] : input[pos].substr(0,8) + "."; rule += " "; } rule += nt_label[i]; - rule += (edge[child][FROM] == edge[child][TO]) ? + rule += (edge[child][FROM] == edge[child][TO]) ? "[" + edge[child][FROM] + "]" : "[" + edge[child][FROM] + "-" + edge[child][TO] + "]"; pos = edge[child][TO]+1; } - for(;pos<=edge[id][TO];pos++) { + for(;pos<=edge[id][TO];pos++) { if (pos != edge[id][FROM]) { rule += " "; } - rule += (input[pos].length <= 10) ? input[pos] : input[pos].substr(0,8) + "."; + rule += (input[pos].length <= 10) ? input[pos] : input[pos].substr(0,8) + "."; } return rule; @@ -1176,10 +1176,10 @@ function show_rules( from, to ) { cell.setAttribute("stroke-width", "3"); current_rule_from = from; current_rule_to = to; - + best_hyp_score = -9e9; best_derivation_score = cell_derivation_score[from][to]; - + var rule_hash = Array(); var rule_count = Array(); rule_list = Array(); @@ -1196,16 +1196,16 @@ function show_rules( from, to ) { rule_count[rule_hash[rule]]++; } edge2rule[id] = rule_hash[rule]; - - if (edge[id][HYP_SCORE] > best_hyp_score) { - best_hyp_score = edge[id][HYP_SCORE]; + + if (edge[id][HYP_SCORE] > best_hyp_score) { + best_hyp_score = edge[id][HYP_SCORE]; } } function sortByRuleCount( a, b ) { return rule_count[rule_hash[b]] - rule_count[rule_hash[a]]; } rule_list = rule_list.sort(sortByRuleCount); - + RULE_HEIGHT = 15; RULE_FONT_SIZE = 11; // squeeze if too many rules @@ -1214,7 +1214,7 @@ function show_rules( from, to ) { RULE_HEIGHT = Math.floor( RULE_HEIGHT * factor ); RULE_FONT_SIZE = Math.ceil( RULE_FONT_SIZE * factor ); } - + draw_rule_options(); for(var i=-1; i<rule_list.length; i++) { draw_rule(from, to, i); @@ -1243,7 +1243,7 @@ function unshow_rules() { finished = 0; for(var i=1; !finished; i++) { var old = document.getElementById("rule-option-" + i); - if (old != null) { + if (old != null) { chart.removeChild( old ); var old = document.getElementById("rule-option-label-" + i); chart.removeChild( old ); @@ -1306,12 +1306,12 @@ function click_rule( from, to, rule_id ) { // highlight current rule if (current_rule_id>=0) { var rule_label = document.getElementById("rule-"+current_rule_id); - rule_label.setAttribute("style", "font-size: "+RULE_FONT_SIZE+"; font-family: Verdana, Arial;"); + rule_label.setAttribute("style", "font-size: "+RULE_FONT_SIZE+"; font-family: Verdana, Arial;"); } var rule_label = document.getElementById("rule-"+rule_id); rule_label.setAttribute("style", "font-size: "+RULE_FONT_SIZE+"; font-family: Verdana, Arial; font-weight: bold;"); current_rule_id = rule_id; - + // first get all the data output_list = Array(); var output_hash = Array(); @@ -1332,7 +1332,7 @@ function click_rule( from, to, rule_id ) { // create index for children var children = get_children( id ); for(var j=0;j<children.length;j++) { - // init children indices if needed + // init children indices if needed if (j > children_list.length-1) { children_hash.push([]); children_list.push([]); @@ -1346,7 +1346,7 @@ function click_rule( from, to, rule_id ) { } } } - + // sort function sortBySecond(a,b) { asplit = a.split("|"); @@ -1361,7 +1361,7 @@ function click_rule( from, to, rule_id ) { for(var i=0;i<children.length;i++) { children_list[i].sort(sortHypByScore); } - + // select dimensions of rule cube axis = Array(); axis.push(output_list); @@ -1394,7 +1394,7 @@ function draw_rule_cube(z_pos_string) { if (z_pos_string != "") { z_pos = z_pos_string.split(","); } - + // draw rube cube var old = document.getElementById("rule-cube"); if (old != null) { chart.removeChild( old ); } @@ -1439,7 +1439,7 @@ function draw_rule_cube(z_pos_string) { } else if (max_length+8 > CHART_HEIGHT/9) { RULE_CUBE_HYP_SIZE = 9; - RULE_CUBE_FONT_SIZE = 7; + RULE_CUBE_FONT_SIZE = 7; } else { RULE_CUBE_HYP_SIZE = CHART_HEIGHT/(max_length+8); @@ -1455,7 +1455,7 @@ function draw_rule_cube(z_pos_string) { rule_cube.setAttribute("x", CHART_WIDTH - 30); rule_cube.setAttribute("y", 0); chart.appendChild( rule_cube ); - + // draw y axis var label = get_rule_axis_name(dimension_order[0]); draw_rule_row(-1,label); @@ -1470,7 +1470,7 @@ function draw_rule_cube(z_pos_string) { // draw x axis if (axis.length > 1) { var label = get_rule_axis_name(dimension_order[1]); - draw_rule_column(-1,label); + draw_rule_column(-1,label); for(var x=0; x<axis[dimension_order[1]].length && x<CHART_HEIGHT/9-10; x++) { var label = get_rule_axis_label(dimension_order[1], x); draw_rule_column(x,label); @@ -1478,7 +1478,7 @@ function draw_rule_cube(z_pos_string) { if (axis[dimension_order[1]].length > CHART_HEIGHT/9-10) { draw_rule_column(Math.ceil(CHART_HEIGHT/9-10),"(more, "+axis[dimension_order[1]].length+" total)"); } - } + } // draw hyps for(var y=0; y<axis[dimension_order[0]].length && y<(CHART_HEIGHT-Z_HEIGHT)/9-10; y++) { @@ -1493,8 +1493,8 @@ function draw_rule_cube(z_pos_string) { } } } - - + + // draw z-axes var pos_offset = axis[dimension_order[0]].length+2; for(var z=2;z<dimension_order.length;z++) { @@ -1506,7 +1506,7 @@ function draw_rule_cube(z_pos_string) { } pos_offset += axis[dimension_order[z]].length+2; } - + // report summary statistics var message = output_list.length + " output phrases"; message += "<br>DEBUG: " + axis.length; @@ -1514,7 +1514,7 @@ function draw_rule_cube(z_pos_string) { for(var i=0;i<children_list.length;i++) { message += "<br>" + children_list[i].length + " hyps for NT" + (i+1); } - //draw_rule_message(message); + //draw_rule_message(message); } function find_hyp_by_rule(position, dimension_order) { @@ -1524,14 +1524,14 @@ function find_hyp_by_rule(position, dimension_order) { var match = 1; for(var p=0; p<position.length; p++) { if (dimension_order[p] == 0) { - if (output_list[position[p]] != edge[id][OUTPUT]+"|"+edge[id][HEURISTIC_RULE_SCORE]) { + if (output_list[position[p]] != edge[id][OUTPUT]+"|"+edge[id][HEURISTIC_RULE_SCORE]) { match = 0; } } else { var nt_number = dimension_order[p]-1; - if (children_list[nt_number][position[p]] != children[nt_number]) { - match = 0; + if (children_list[nt_number][position[p]] != children[nt_number]) { + match = 0; } } } @@ -1580,7 +1580,7 @@ function get_full_output( id ) { var source_target = alignment[i].split("-"); nonterminal[source_target[1]] = children[i]; } - + var full_output = ""; for(var i=0;i<output.length;i++) { if (nonterminal[i] === undefined) { @@ -1647,7 +1647,7 @@ function draw_rule_z( z,total_z, z_pos, pos,pos_offset, label ) { else { rule_label.setAttribute("style", "font-size: "+(RULE_CUBE_FONT_SIZE-2)+"; font-family: Verdana, Arial; font-weight: bold;"); } - + var content = document.createTextNode( label ); rule_label.appendChild( content ); var rule_cube = document.getElementById("rule-cube"); @@ -1686,7 +1686,7 @@ function rule_hyp_color( id, on_off ) { } else { derivation_score_color = get_score_from_color(best_derivation_score-edge[id][DERIVATION_SCORE]); - } + } } return "#" + inactive_color + derivation_score_color + hyp_score_color; } diff --git a/scripts/ems/web/sgviz.php b/scripts/ems/web/sgviz.php index bac9289ba..9fccadf60 100644 --- a/scripts/ems/web/sgviz.php +++ b/scripts/ems/web/sgviz.php @@ -39,7 +39,7 @@ new Ajax.Request('?analysis=sgviz_data' method: "post" }); </script></body></html> -<?php +<?php // read graph //$file = get_current_analysis_filename("basic","search-graph")."/graph.$sentence"; //$handle = fopen($file,"r"); @@ -55,7 +55,7 @@ function sgviz_data($sentence) { $file = get_current_analysis_filename("basic","search-graph")."/graph.$sentence"; $handle = fopen($file,"r"); - while (($line = fgets($handle)) !== false) { + while (($line = fgets($handle)) !== false) { $e = explode("\t",addslashes(chop($line))); $edge[$e[0]] = array($e[1],$e[2],$e[3],$e[4],$e[5],$e[6],$e[7],$e[8],$e[9],$e[10]); } |