From 745e70e0d79d5b4bed3da1efa2851a86283f29cd Mon Sep 17 00:00:00 2001 From: Giambattista Bloisi Date: Mon, 10 Jul 2023 15:45:49 +0200 Subject: [PATCH] When generating similarities put as 'from' component the one with smaller lexicographic id --- .../src/main/java/eu/dnetlib/pace/util/BlockProcessor.java | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/dhp-pace-core/src/main/java/eu/dnetlib/pace/util/BlockProcessor.java b/dhp-pace-core/src/main/java/eu/dnetlib/pace/util/BlockProcessor.java index 30606bb9b..ef173e270 100644 --- a/dhp-pace-core/src/main/java/eu/dnetlib/pace/util/BlockProcessor.java +++ b/dhp-pace-core/src/main/java/eu/dnetlib/pace/util/BlockProcessor.java @@ -121,7 +121,11 @@ public class BlockProcessor { private void emitOutput(final boolean result, final String idPivot, final String idCurr, final Reporter context) { if (result) { - writeSimilarity(context, idPivot, idCurr); + if (idPivot.compareTo(idCurr) <= 0) { + writeSimilarity(context, idPivot, idCurr); + } else { + writeSimilarity(context, idCurr, idPivot); + } context.incrementCounter(dedupConf.getWf().getEntityType(), "dedupSimilarity (x2)", 1); } else { context.incrementCounter(dedupConf.getWf().getEntityType(), "d < " + dedupConf.getWf().getThreshold(), 1);