TIKA-3091 prevent npe in PDFParserConfig by initializing

three parameters with default values. # Conflicts: # tika-parsers/src/main/java/org/apache/tika/parser/pdf/PDFParserConfig.java
apache · Apr 14, 2020 · f51ae0a · f51ae0a
1 parent 86cdbb1
commit f51ae0a
Show file tree

Hide file tree

Showing 3 changed files with 47 additions and 2 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
@@ -523,6 +523,17 @@ 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
@@ -114,10 +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.
+ //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.
@@ -238,6 +244,10 @@ 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);
 
@@ -287,6 +297,9 @@ public void configure(PDF2XHTML pdf2XHTML) {
  if (getSpacingTolerance() != null) {
  pdf2XHTML.setSpacingTolerance(getSpacingTolerance());
  }
+ if (getDropThreshold() != null) {
+ pdf2XHTML.setDropThreshold(dropThreshold);
+ }
  pdf2XHTML.setSuppressDuplicateOverlappingText(getSuppressDuplicateOverlappingText());
  }
 
@@ -513,6 +526,14 @@ public void setSpacingTolerance(Float spacingTolerance) {
  this.spacingTolerance = spacingTolerance;
  }
 
+ public Float getDropThreshold() {
+ return dropThreshold;
+ }
+
+ public void setDropThreshold(float dropThreshold) {
+ this.dropThreshold = dropThreshold;
+ }
+
  public AccessChecker getAccessChecker() {
  return accessChecker;
  }
@@ -824,6 +845,7 @@ public boolean equals(Object o) {
  if (getCatchIntermediateIOExceptions() != config.getCatchIntermediateIOExceptions()) return false;
  if (!getAverageCharTolerance().equals(config.getAverageCharTolerance())) return false;
  if (!getSpacingTolerance().equals(config.getSpacingTolerance())) return false;
+ if (!getDropThreshold().equals(config.getDropThreshold())) return false;
  if (!getOcrStrategy().equals(config.getOcrStrategy())) return false;
  if (getOcrImageType() != config.getOcrImageType()) return false;
  if (!getOcrImageFormatName().equals(config.getOcrImageFormatName())) return false;
@@ -844,6 +866,7 @@ public int hashCode() {
  result = 31 * result + (getExtractUniqueInlineImagesOnly() ? 1 : 0);
  result = 31 * result + getAverageCharTolerance().hashCode();
  result = 31 * result + getSpacingTolerance().hashCode();
+ result = 31 * result + getDropThreshold().hashCode();
  result = 31 * result + (getIfXFAExtractOnlyXFA() ? 1 : 0);
  result = 31 * result + ocrStrategy.hashCode();
  result = 31 * result + getOcrDPI();
@@ -869,6 +892,7 @@ public String toString() {
  ", extractUniqueInlineImagesOnly=" + extractUniqueInlineImagesOnly +
  ", averageCharTolerance=" + averageCharTolerance +
  ", spacingTolerance=" + spacingTolerance +
+ ", dropThreshold=" + dropThreshold +
  ", ifXFAExtractOnlyXFA=" + ifXFAExtractOnlyXFA +
  ", ocrStrategy=" + ocrStrategy +
  ", ocrDPI=" + ocrDPI +

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
@@ -1528,6 +1528,16 @@ 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 {