TIKA-3091 prevent npe in PDFParserConfig by initializing

three parameters with default values.
MrToast · Apr 14, 2020 · d811de9 · d811de9
1 parent e479995
commit d811de9
Show file tree

Hide file tree

Showing 3 changed files with 31 additions and 3 deletions.
diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java b/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParser.java
@@ -503,6 +503,17 @@ void setExtractInlineImages(boolean extractInlineImages) {
         defaultConfig.setExtractInlineImages(extractInlineImages);
     }
 
+    @Field
+    void setAverageCharTolerance(float averageCharTolerance) {
+        defaultConfig.setAverageCharTolerance(averageCharTolerance);
+    }
+
+    @Field
+    void setSpacingTolerance(float spacingTolerance) {
+        defaultConfig.setSpacingTolerance(spacingTolerance);
+    }
+
+
     @Field
     void setCatchIntermediateExceptions(boolean catchIntermediateExceptions) {
         defaultConfig.setCatchIntermediateIOExceptions(catchIntermediateExceptions);

diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java b/tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
@@ -114,13 +114,16 @@ private static OCR_STRATEGY parse(String s) {
     private boolean extractMarkedContent = false;
 
     //The character width-based tolerance value used to estimate where spaces in text should be added
-    private Float averageCharTolerance;
+    //Default taken from PDFBox.
+    private Float averageCharTolerance = 0.5f;
 
     //The space width-based tolerance value used to estimate where spaces in text should be added
-    private Float spacingTolerance;
+    //Default taken from PDFBox.
+    private Float spacingTolerance = 0.3f;
 
     // The multiplication factor for line height to decide when a new paragraph starts.
-    private float dropThreshold;
+    //Default taken from PDFBox.
+    private float dropThreshold = 2.5f;
 
     //If the PDF has an XFA element, process only that and skip extracting
     //content from elsewhere in the document.
@@ -235,6 +238,10 @@ private void init(InputStream is) {
 
         setSetKCMS(getBooleanProp(props.getProperty("setKCMS"), false));
 
+        setAverageCharTolerance(getFloatProp(props.getProperty("averageCharTolerance"), averageCharTolerance));
+        setSpacingTolerance(getFloatProp(props.getProperty("spacingTolerance"), spacingTolerance));
+        setDropThreshold(getFloatProp(props.getProperty("dropThreshold"), dropThreshold));
+
         boolean checkExtractAccessPermission = getBooleanProp(props.getProperty("checkExtractAccessPermission"), false);
         boolean allowExtractionForAccessibility = getBooleanProp(props.getProperty("allowExtractionForAccessibility"), true);
 

diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/pdf/PDFParserTest.java
@@ -1512,6 +1512,16 @@ public void testUnmappedUnicodeStats() throws Exception {
 
     }
 
+    @Test
+    public void testNPEInPDFParserConfig() {
+        //TIKA-3091
+        PDFParserConfig config = new PDFParserConfig();
+        //don't care about values; want to make sure no NPE is thrown
+        String txt = config.toString();
+        config.hashCode();
+        config.equals(new PDFParserConfig());
+    }
+
     @Test //TIKA-3041
     @Ignore("turn back on if we add file from PDFBOX-52")
     public void testPDFBox52() throws Exception {