Merge pull request kermitt2#317 from csw/bibdata-quoting

Fix BibDataSetContextExtractor to quote replacement text Former-commit-id: 31936a3
elifesciences · May 18, 2018 · 3aa29dd · 3aa29dd
2 parents 2451a58 + 732af07
commit 3aa29dd
Show file tree

Hide file tree

Showing 3 changed files with 33 additions and 1 deletion.
diff --git a/grobid-core/src/main/java/org/grobid/core/utilities/BibDataSetContextExtractor.java b/grobid-core/src/main/java/org/grobid/core/utilities/BibDataSetContextExtractor.java
@@ -83,7 +83,7 @@ private static String extractContextSentence(String cont) {
         Matcher m = REF_PATTERN.matcher(cont);
         if (m.find()) {
             String g = m.group(1);
-            return m.replaceAll(g);
+            return m.replaceAll(Matcher.quoteReplacement(g));
         } else {
             throw new IllegalStateException("Implementation error: no <ref> found in" + cont);
         }

diff --git a/grobid-core/src/test/java/org/grobid/core/utilities/BibDataSetContextExtractorTest.java b/grobid-core/src/test/java/org/grobid/core/utilities/BibDataSetContextExtractorTest.java
@@ -0,0 +1,20 @@
+package org.grobid.core.utilities;
+
+import org.apache.commons.io.IOUtils;
+
+import org.junit.Test;
+
+import java.io.InputStream;
+import java.nio.charset.StandardCharsets;
+
+public class BibDataSetContextExtractorTest {
+
+    @Test
+    public void testRefEscapes() throws Exception {
+        InputStream is = this.getClass().getResourceAsStream("/test/tei-escape.xml");
+        String tei = IOUtils.toString(is, StandardCharsets.UTF_8);
+        is.close();
+        BibDataSetContextExtractor.getCitationReferences(tei);
+    }
+
+}
diff --git a/grobid-core/src/test/resources/test/tei-escape.xml b/grobid-core/src/test/resources/test/tei-escape.xml
@@ -0,0 +1,12 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<TEI xmlns="http://www.tei-c.org/ns/1.0"
+xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+xsi:schemaLocation="http://www.tei-c.org/ns/1.0 /home/lopez/grobid/grobid-home/schemas/xsd/Grobid.xsd"
+ xmlns:xlink="http://www.w3.org/1999/xlink">
+	<text xml:lang="en">
+		<body>
+                    <div xmlns="http://www.tei-c.org/ns/1.0"><p>Lorem ipsum <ref type="bibr" target="#b0">$9,</ref><ref type="bibr" target="#b1">2]</ref>.
+                    </p></div>
+                </body>
+        </text>
+</TEI>