diff --git a/src/main/java/org/json/XML.java b/src/main/java/org/json/XML.java index a94c3fc..9d42bb3 100644 --- a/src/main/java/org/json/XML.java +++ b/src/main/java/org/json/XML.java @@ -431,6 +431,9 @@ public class XML { && jsonObject.opt(config.getcDataTagName()) != null) { context.accumulate(tagName, jsonObject.opt(config.getcDataTagName())); } else { + if (!config.shouldTrimWhiteSpace()) { + removeEmpty(jsonObject, config); + } context.accumulate(tagName, jsonObject); } } @@ -445,6 +448,48 @@ public class XML { } } } + /** + * This method removes any JSON entry which has the key set by XMLParserConfiguration.cDataTagName + * and contains whitespace as this is caused by whitespace between tags. See test XMLTest.testNestedWithWhitespaceTrimmingDisabled. + * @param jsonObject JSONObject which may require deletion + * @param config The XMLParserConfiguration which includes the cDataTagName + */ + private static void removeEmpty(final JSONObject jsonObject, final XMLParserConfiguration config) { + if (jsonObject.has(config.getcDataTagName())) { + final Object s = jsonObject.get(config.getcDataTagName()); + if (s instanceof String) { + if (isStringAllWhiteSpace(s.toString())) { + jsonObject.remove(config.getcDataTagName()); + } + } + else if (s instanceof JSONArray) { + final JSONArray sArray = (JSONArray) s; + for (int k = sArray.length()-1; k >= 0; k--){ + final Object eachString = sArray.get(k); + if (eachString instanceof String) { + String s1 = (String) eachString; + if (isStringAllWhiteSpace(s1)) { + sArray.remove(k); + } + } + } + if (sArray.isEmpty()) { + jsonObject.remove(config.getcDataTagName()); + } + } + } + } + + private static boolean isStringAllWhiteSpace(final String s) { + for (int k = 0; k forceList; + + /** + * Flag to indicate whether white space should be trimmed when parsing XML. + * The default behaviour is to trim white space. When this is set to false, inputting XML + * with tags that are the same as the value of cDataTagName is unsupported. It is recommended to set cDataTagName + * to a distinct value in this case. + */ + private boolean shouldTrimWhiteSpace; + /** * Default parser configuration. Does not keep strings (tries to implicitly convert - * values), and the CDATA Tag Name is "content". + * values), and the CDATA Tag Name is "content". Trims whitespace. */ public XMLParserConfiguration () { super(); @@ -71,6 +80,7 @@ public class XMLParserConfiguration extends ParserConfiguration { this.convertNilAttributeToNull = false; this.xsiTypeMap = Collections.emptyMap(); this.forceList = Collections.emptySet(); + this.shouldTrimWhiteSpace = true; } /** @@ -172,7 +182,7 @@ public class XMLParserConfiguration extends ParserConfiguration { // item, a new map instance should be created and if possible each value in the // map should be cloned as well. If the values of the map are known to also // be immutable, then a shallow clone of the map is acceptable. - return new XMLParserConfiguration( + final XMLParserConfiguration config = new XMLParserConfiguration( this.keepStrings, this.cDataTagName, this.convertNilAttributeToNull, @@ -181,6 +191,8 @@ public class XMLParserConfiguration extends ParserConfiguration { this.maxNestingDepth, this.closeEmptyTag ); + config.shouldTrimWhiteSpace = this.shouldTrimWhiteSpace; + return config; } /** @@ -327,7 +339,23 @@ public class XMLParserConfiguration extends ParserConfiguration { return clonedConfiguration; } + /** + * Sets whether whitespace should be trimmed inside of tags. *NOTE* Do not use this if + * you expect your XML tags to have names that are the same as cDataTagName as this is unsupported. + * cDataTagName should be set to a distinct value in these cases. + * @param shouldTrimWhiteSpace boolean to set trimming on or off. Off is default. + * @return same instance of configuration with empty tag config updated + */ + public XMLParserConfiguration withShouldTrimWhitespace(boolean shouldTrimWhiteSpace){ + XMLParserConfiguration clonedConfiguration = this.clone(); + clonedConfiguration.shouldTrimWhiteSpace = shouldTrimWhiteSpace; + return clonedConfiguration; + } + public boolean isCloseEmptyTag() { return this.closeEmptyTag; } + public boolean shouldTrimWhiteSpace() { + return this.shouldTrimWhiteSpace; + } } diff --git a/src/main/java/org/json/XMLTokener.java b/src/main/java/org/json/XMLTokener.java index 957498c..bc18b31 100644 --- a/src/main/java/org/json/XMLTokener.java +++ b/src/main/java/org/json/XMLTokener.java @@ -20,6 +20,8 @@ public class XMLTokener extends JSONTokener { */ public static final java.util.HashMap entity; + private XMLParserConfiguration configuration = XMLParserConfiguration.ORIGINAL; + static { entity = new java.util.HashMap(8); entity.put("amp", XML.AMP); @@ -45,6 +47,16 @@ public class XMLTokener extends JSONTokener { super(s); } + /** + * Construct an XMLTokener from a Reader and an XMLParserConfiguration. + * @param r A source reader. + * @param configuration the configuration that can be used to set certain flags + */ + public XMLTokener(Reader r, XMLParserConfiguration configuration) { + super(r); + this.configuration = configuration; + } + /** * Get the text in the CDATA block. * @return The string up to the ]]>. @@ -83,7 +95,7 @@ public class XMLTokener extends JSONTokener { StringBuilder sb; do { c = next(); - } while (Character.isWhitespace(c)); + } while (Character.isWhitespace(c) && configuration.shouldTrimWhiteSpace()); if (c == 0) { return null; } @@ -97,7 +109,9 @@ public class XMLTokener extends JSONTokener { } if (c == '<') { back(); - return sb.toString().trim(); + if (configuration.shouldTrimWhiteSpace()) { + return sb.toString().trim(); + } else return sb.toString(); } if (c == '&') { sb.append(nextEntity(c)); diff --git a/src/test/java/org/json/junit/XMLConfigurationTest.java b/src/test/java/org/json/junit/XMLConfigurationTest.java index ffdc20c..ba8418c 100755 --- a/src/test/java/org/json/junit/XMLConfigurationTest.java +++ b/src/test/java/org/json/junit/XMLConfigurationTest.java @@ -1181,4 +1181,4 @@ public class XMLConfigurationTest { assertTrue("Error: " +e.getMessage(), false); } } -} \ No newline at end of file +} diff --git a/src/test/java/org/json/junit/XMLTest.java b/src/test/java/org/json/junit/XMLTest.java index 823a065..9ae1ee2 100644 --- a/src/test/java/org/json/junit/XMLTest.java +++ b/src/test/java/org/json/junit/XMLTest.java @@ -1323,6 +1323,80 @@ public class XMLTest { "parameter of the XMLParserConfiguration used"); } } + @Test + public void testWithWhitespaceTrimmingDisabled() { + String originalXml = " Test Whitespace String \t "; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration().withShouldTrimWhitespace(false)); + String expectedJsonString = "{\"testXml\":\" Test Whitespace String \t \"}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + @Test + public void testNestedWithWhitespaceTrimmingDisabled() { + String originalXml = + "\n"+ + "\n"+ + "
\n"+ + " Sherlock Holmes \n"+ + "
\n"+ + "
"; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration().withShouldTrimWhitespace(false)); + String expectedJsonString = "{\"addresses\":{\"address\":{\"name\":\" Sherlock Holmes \"}}}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + @Test + public void shouldTrimWhitespaceDoesNotSupportTagsEqualingCDataTagName() { + // When using withShouldTrimWhitespace = true, input containing tags with same name as cDataTagName is unsupported and should not be used in conjunction + String originalXml = + "\n"+ + "\n"+ + "
\n"+ + " Sherlock Holmes \n"+ + "
\n"+ + "
"; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration().withShouldTrimWhitespace(false).withcDataTagName("content")); + String expectedJsonString = "{\"addresses\":{\"address\":[[\"\\n \",\" Sherlock Holmes \",\"\\n \"]]}}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + @Test + public void shouldTrimWhitespaceEnabledDropsTagsEqualingCDataTagNameButValueRemains() { + String originalXml = + "\n"+ + "\n"+ + "
\n"+ + " Sherlock Holmes \n"+ + "
\n"+ + "
"; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration().withShouldTrimWhitespace(true).withcDataTagName("content")); + String expectedJsonString = "{\"addresses\":{\"address\":\"Sherlock Holmes\"}}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + @Test + public void testWithWhitespaceTrimmingEnabled() { + String originalXml = " Test Whitespace String \t "; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration().withShouldTrimWhitespace(true)); + String expectedJsonString = "{\"testXml\":\"Test Whitespace String\"}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + @Test + public void testWithWhitespaceTrimmingEnabledByDefault() { + String originalXml = " Test Whitespace String \t "; + + JSONObject actualJson = XML.toJSONObject(originalXml, new XMLParserConfiguration()); + String expectedJsonString = "{\"testXml\":\"Test Whitespace String\"}"; + JSONObject expectedJson = new JSONObject(expectedJsonString); + Util.compareActualVsExpectedJsonObjects(actualJson,expectedJson); + } + }