add -external-bin-dir for giza++/mgiza directory. Ignore -bin-dir and -script-root-dir

2 files changed, 28 insertions, 25 deletions

diff --git a/scripts/ems/experiment.perl b/scripts/ems/experiment.perl
index cb4518dde..b0e71c969 100755
--- a/scripts/ems/experiment.perl
+++ b/scripts/ems/experiment.perl
@@ -1994,6 +1994,7 @@ sub get_training_setting {
     my ($step) = @_;
     my $dir = &check_and_get("GENERAL:working-dir");
     my $training_script = &check_and_get("TRAINING:script");
+    my $external_bin_dir = &check_backoff_and_get("TRAINING:external-bin-dir");
     my $scripts = &check_backoff_and_get("TUNING:moses-script-dir");
     my $reordering = &get("TRAINING:lexicalized-reordering");
     my $input_extension = &check_backoff_and_get("TRAINING:input-extension");
@@ -2016,7 +2017,7 @@ sub get_training_setting {
     $cmd .= "-dont-zip ";
     $cmd .= "-first-step $step " if $step>1;
     $cmd .= "-last-step $step "  if $step<9;
-    $cmd .= "-scripts-root-dir $scripts ";
+    $cmd .= "-external-bin-dir $external_bin_dir " if defined($external_bin_dir);
     $cmd .= "-f $input_extension -e $output_extension ";
     $cmd .= "-alignment $alignment ";
     $cmd .= "-max-phrase-length $phrase_length " if $phrase_length;
diff --git a/scripts/training/train-model.perl.missing_bin_dir b/scripts/training/train-model.perl.missing_bin_dir
index 48911a891..b0cb26275 100755
--- a/scripts/training/train-model.perl.missing_bin_dir
+++ b/scripts/training/train-model.perl.missing_bin_dir
@@ -17,9 +17,9 @@ if ($SCRIPTS_ROOTDIR eq '') {
     $SCRIPTS_ROOTDIR = dirname(__FILE__);
 }
 $SCRIPTS_ROOTDIR =~ s/\/training$//;
-$SCRIPTS_ROOTDIR = $ENV{"SCRIPTS_ROOTDIR"} if defined($ENV{"SCRIPTS_ROOTDIR"});
+#$SCRIPTS_ROOTDIR = $ENV{"SCRIPTS_ROOTDIR"} if defined($ENV{"SCRIPTS_ROOTDIR"});
 
-my($_ROOT_DIR, $_CORPUS_DIR, $_GIZA_E2F, $_GIZA_F2E, $_MODEL_DIR, $_TEMP_DIR, $_SORT_BUFFER_SIZE, $_SORT_BATCH_SIZE,  $_SORT_COMPRESS, $_SORT_PARALLEL, $_CORPUS,
+my($_EXTERNAL_BINDIR, $_ROOT_DIR, $_CORPUS_DIR, $_GIZA_E2F, $_GIZA_F2E, $_MODEL_DIR, $_TEMP_DIR, $_SORT_BUFFER_SIZE, $_SORT_BATCH_SIZE,  $_SORT_COMPRESS, $_SORT_PARALLEL, $_CORPUS,
    $_CORPUS_COMPRESSION, $_FIRST_STEP, $_LAST_STEP, $_F, $_E, $_MAX_PHRASE_LENGTH,
    $_LEXICAL_FILE, $_NO_LEXICAL_WEIGHTING, $_VERBOSE, $_ALIGNMENT,
    $_ALIGNMENT_FILE, $_ALIGNMENT_STEM, @_LM, $_EXTRACT_FILE, $_GIZA_OPTION, $_HELP, $_PARTS,
@@ -36,17 +36,14 @@ my($_ROOT_DIR, $_CORPUS_DIR, $_GIZA_E2F, $_GIZA_F2E, $_MODEL_DIR, $_TEMP_DIR, $_
    $_MEMSCORE, $_FINAL_ALIGNMENT_MODEL,
    $_CONTINUE,$_MAX_LEXICAL_REORDERING,$_DO_STEPS,
    $_ADDITIONAL_INI,
-   $_DICTIONARY, $_EPPEX);
+   $_DICTIONARY, $_EPPEX, $IGNORE);
 my $_CORES = 1;
 
 my $debug = 0; # debug this script, do not delete any files in debug mode
 
-# the following line is set installation time by 'make release'.  BEWARE!
-my $BINDIR="/Users/hieuhoang/workspace/bin/training-tools/";
-
 $_HELP = 1
     unless &GetOptions('root-dir=s' => \$_ROOT_DIR,
-		       'bin-dir=s' => \$BINDIR, # allow to override default bindir path
+		       'external-bin-dir=s' => \$_EXTERNAL_BINDIR,
 		       'corpus-dir=s' => \$_CORPUS_DIR,
 		       'corpus=s' => \$_CORPUS,
 		       'f=s' => \$_F,
@@ -93,7 +90,8 @@ $_HELP = 1
 		       'generation-factors=s' => \$_GENERATION_FACTORS,
 		       'decoding-steps=s' => \$_DECODING_STEPS,
 		       'decoding-graph-backoff=s' => \$_DECODING_GRAPH_BACKOFF,
-		       'scripts-root-dir=s' => \$SCRIPTS_ROOTDIR,
+    			 'bin-dir=s' => \$IGNORE,
+		       'scripts-root-dir=s' => \$IGNORE,
 		       'factor-delimiter=s' => \$_FACTOR_DELIMITER,
 		       'phrase-translation-table=s' => \@_PHRASE_TABLE,
 		       'generation-corpus=s' => \$_GENERATION_CORPUS,
@@ -144,6 +142,10 @@ For more, please check manual or contact koehn\@inf.ed.ac.uk\n";
   exit(1);
 }
 
+if (defined($IGNORE)) {
+  print STDERR "WARNING: Do not specify -bin-dir or -scripts-root-dir anymore. These variable are ignored and will be deleted soon";
+}
+
 $_HIERARCHICAL = 1 if $_SOURCE_SYNTAX || $_TARGET_SYNTAX;
 $_XML = 1 if $_SOURCE_SYNTAX || $_TARGET_SYNTAX;
 my $___FACTOR_DELIMITER = $_FACTOR_DELIMITER;
@@ -187,24 +189,24 @@ foreach my $step (@step_conf) {
 
 
 # supporting binaries from other packages
-my $MGIZA_MERGE_ALIGN = "$BINDIR/merge_alignment.py";
+my $MGIZA_MERGE_ALIGN = "$_EXTERNAL_BINDIR/merge_alignment.py";
 my $GIZA;
 my $SNT2COOC;
 
 if(!defined $_MGIZA ){
-	$GIZA = "$BINDIR/GIZA++";
-	if (-x "$BINDIR/snt2cooc.out") {
-  	$SNT2COOC = "$BINDIR/snt2cooc.out";
-	} elsif (-x "$BINDIR/snt2cooc") { # Since "snt2cooc.out" and "snt2cooc" work the same   
-		$SNT2COOC = "$BINDIR/snt2cooc";
+	$GIZA = "$_EXTERNAL_BINDIR/GIZA++";
+	if (-x "$_EXTERNAL_BINDIR/snt2cooc.out") {
+  	$SNT2COOC = "$_EXTERNAL_BINDIR/snt2cooc.out";
+	} elsif (-x "$_EXTERNAL_BINDIR/snt2cooc") { # Since "snt2cooc.out" and "snt2cooc" work the same   
+		$SNT2COOC = "$_EXTERNAL_BINDIR/snt2cooc";
 	}
 	print STDERR "Using single-thread GIZA\n";
 } else {
-  $GIZA = "$BINDIR/mgiza";
-	if (-x "$BINDIR/snt2cooc") {
-  	$SNT2COOC = "$BINDIR/snt2cooc";
-  } elsif (-x "$BINDIR/snt2cooc.out") { # Important for users that use MGIZA and copy only the "mgiza" file to $BINDIR 
-    $SNT2COOC = "$BINDIR/snt2cooc.out";
+  $GIZA = "$_EXTERNAL_BINDIR/mgiza";
+	if (-x "$_EXTERNAL_BINDIR/snt2cooc") {
+  	$SNT2COOC = "$_EXTERNAL_BINDIR/snt2cooc";
+  } elsif (-x "$_EXTERNAL_BINDIR/snt2cooc.out") { # Important for users that use MGIZA and copy only the "mgiza" file to $_EXTERNAL_BINDIR
+    $SNT2COOC = "$_EXTERNAL_BINDIR/snt2cooc.out";
   }
 	print STDERR "Using multi-thread GIZA\n";	
   if (!defined($_MGIZA_CPUS)) {
@@ -214,9 +216,9 @@ if(!defined $_MGIZA ){
 }
 
 # override
-$SNT2COOC = "$BINDIR/$_SNT2COOC" if defined($_SNT2COOC);
+$SNT2COOC = "$_EXTERNAL_BINDIR/$_SNT2COOC" if defined($_SNT2COOC);
 
-my $MKCLS = "$BINDIR/mkcls";
+my $MKCLS = "$_EXTERNAL_BINDIR/mkcls";
 
 # parallel extract
 my $SPLIT_EXEC = `gsplit --help 2>/dev/null`; 
@@ -278,8 +280,8 @@ my $BZCAT = "bzcat";
 # do a sanity check to make sure we can find the necessary binaries since
 # these are not installed by default
 # not needed if we start after step 2
-die("ERROR: Cannot find mkcls, GIZA++/mgiza, & snt2cooc.out/snt2cooc in $BINDIR.\nDid you install this script using 'make release'?") unless ((!$STEPS[2]) ||
-                                       (-x $GIZA && defined($SNT2COOC) && -x $MKCLS));
+die("ERROR: Cannot find mkcls, GIZA++/mgiza, & snt2cooc.out/snt2cooc in $_EXTERNAL_BINDIR.\nYou MUST specify the parameter -external-bin-dir") unless ((!$STEPS[2]) ||
+                                       (defined($_EXTERNAL_BINDIR) && -x $GIZA && defined($SNT2COOC) && -x $MKCLS));
 
 # set varibles to defaults or from options
 my $___ROOT_DIR = ".";
@@ -1099,7 +1101,7 @@ sub run_single_snt2cooc {
     my($dir,$e,$f,$vcb_e,$vcb_f,$train) = @_;
     print STDERR "(2.1a) running snt2cooc $f-$e @ ".`date`."\n";
     safesystem("mkdir -p $dir") or die("ERROR");
-    if ($SNT2COOC eq "$BINDIR/snt2cooc.out") {
+    if ($SNT2COOC eq "$_EXTERNAL_BINDIR/snt2cooc.out") {
     print "$SNT2COOC $vcb_e $vcb_f $train > $dir/$f-$e.cooc\n";
     safesystem("$SNT2COOC $vcb_e $vcb_f $train > $dir/$f-$e.cooc") or die("ERROR");
     } else {