Extend to allow for only 1 corpus to be processed.

clarin-eric · Sep 12, 2023 · b4d08c5 · b4d08c5
1 parent 266d290
commit b4d08c5
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 2 deletions.
diff --git a/Corpora/Sources-TEI/Makefile b/Corpora/Sources-TEI/Makefile
@@ -1,4 +1,8 @@
 ###### Preparing the received corpora
+CORPUS = HU
+one:
+	../../Scripts/parlamint-factorize-corpora.pl ../Taxonomies 'Orig/ParlaMint-${CORPUS}.TEI*'
+	../../Scripts/add-metadata.pl ../Orientations ../Ministers 'Orig/ParlaMint-${CORPUS}.TEI*' '.'
 
 nohup:
 	nohup time make prep > preparation.log &

diff --git a/Scripts/parlamint-factorize-corpora.pl b/Scripts/parlamint-factorize-corpora.pl
@@ -59,7 +59,9 @@
 
 binmode(STDERR, 'utf8');
 
-foreach $corpDir (sort glob "$inDir/ParlaMint-*.TEI*") {
+if ($inDir =~ /ParlaMint-[A-Z-]+\.TEI/) {$corpDirs = $inDir}
+else {$corpDirs = "$inDir/ParlaMint-*.TEI*"}
+foreach $corpDir (sort glob($corpDirs)) {
     my $param = '';
     ($country, $anaSuffix) = $corpDir =~ /ParlaMint-([A-Z-]+)\.TEI(\..+)?/ or die;
     $anaSuffix = '' unless $anaSuffix;
@@ -76,7 +78,8 @@
     # Factorise
     if ($anaSuffix) {
 	#For .ana we will also need the .TEI root file
-        my $teiRoot = "$inDir/ParlaMint-$country.TEI/ParlaMint-$country.xml";
+        my $teiRoot = "$corpDir/ParlaMint-$country.xml";
+	$teiRoot =~ s|\Q$anaSuffix\E||;
         if (-e $teiRoot){$param = " teiRoot=$teiRoot "}
 	else {print STDERR "WARN: $teiRoot not found\n"}
     }