From 2922957611a4d3de8a0bf36ef918c1520208600b Mon Sep 17 00:00:00 2001 From: Andrew Wen Date: Sat, 9 Sep 2023 17:25:55 -0500 Subject: [PATCH] Don't attempt splits for zero-length lines --- pom.xml | 2 +- .../ohnlp/medtagger/backbone/MedTaggerBackboneTransform.java | 2 +- .../ohnlp/medtagger/backbone/SectionResolutionTransform.java | 3 +++ 3 files changed, 5 insertions(+), 2 deletions(-) diff --git a/pom.xml b/pom.xml index d852e7c..7a0c45e 100644 --- a/pom.xml +++ b/pom.xml @@ -6,7 +6,7 @@ org.ohnlp.medtagger medtagger - 1.0.69 + 1.0.70 The MedTagger biomedical information extraction pipeline diff --git a/src/main/java/org/ohnlp/medtagger/backbone/MedTaggerBackboneTransform.java b/src/main/java/org/ohnlp/medtagger/backbone/MedTaggerBackboneTransform.java index a143582..1c145d1 100644 --- a/src/main/java/org/ohnlp/medtagger/backbone/MedTaggerBackboneTransform.java +++ b/src/main/java/org/ohnlp/medtagger/backbone/MedTaggerBackboneTransform.java @@ -264,7 +264,7 @@ public void init() throws IOException, InvalidXMLException, URISyntaxException, ae.add(AnalysisEngineFactory.createEngineDescription(RuleContextAnnotator.class, "context_ruleset", contextURI.toString())); } } else { - if (mode.equals(RunMode.STANDALONE_DICT_ONLY) || mode.equals(RunMode.STANDALONE_DICT_ONLY)) { + if (mode.equals(RunMode.STANDALONE_DICT_ONLY) || mode.equals(RunMode.GENERAL_CLINICAL)) { if (this.context.equalsIgnoreCase("DEFAULT")) { ae.add(AnalysisEngineFactory.createEngineDescription(RuleContextAnnotator.class)); } else { diff --git a/src/main/java/org/ohnlp/medtagger/backbone/SectionResolutionTransform.java b/src/main/java/org/ohnlp/medtagger/backbone/SectionResolutionTransform.java index b0f64da..539f03d 100644 --- a/src/main/java/org/ohnlp/medtagger/backbone/SectionResolutionTransform.java +++ b/src/main/java/org/ohnlp/medtagger/backbone/SectionResolutionTransform.java @@ -62,6 +62,9 @@ public void process(ProcessContext pc) { // First, split by newlines Arrays.stream(raw.split("\\r?\\n")).forEach(s -> { // Now split by colon and add first part to candidates + if (s.trim().length() == 0) { + return; + } String cand = s.split(":")[0].trim(); if (cand.length() > 0) { // And output