Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

MXParser fixes #185

Merged
merged 4 commits into from May 17, 2022
Merged

MXParser fixes #185

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
89b8ef5
Fix position on MXParser (#121)
joehni Jan 3, 2021
2c6fe7e
Fix start position of PI and comment for all reported errors (#124)
joehni Jan 4, 2021
11c749d
Fix endless loop with invalid PI containing XML (#122)
joehni Jan 3, 2021
0d97c8b
Fix endless loop caused by aborted subsequent PI or comment (#124)
joehni Jan 4, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
12 changes: 8 additions & 4 deletions src/main/java/org/codehaus/plexus/util/xml/pull/MXParser.java
View file
Open in desktop
Expand Up @@ -464,7 +464,7 @@ private void reset()
// System.out.println("reset() called");
location = null;
lineNumber = 1;
columnNumber = 0;
columnNumber = 1;
seenRoot = false;
reachedEnd = false;
eventType = START_DOCUMENT;
Expand Down Expand Up @@ -2988,7 +2988,7 @@ private void parseComment()
posStart = pos;

final int curLine = lineNumber;
final int curColumn = columnNumber;
final int curColumn = columnNumber - 4;
try
{
final boolean normalizeIgnorableWS = tokenize && !roundtripSupported;
Expand Down Expand Up @@ -3109,7 +3109,7 @@ private boolean parsePI()
if ( tokenize )
posStart = pos;
final int curLine = lineNumber;
final int curColumn = columnNumber;
final int curColumn = columnNumber - 2;
int piTargetStart = pos;
int piTargetEnd = -1;
final boolean normalizeIgnorableWS = tokenize && !roundtripSupported;
Expand Down Expand Up @@ -3158,6 +3158,10 @@ else if ( !seenInnerTag )
throw new XmlPullParserException( "processing instruction started on line " + curLine
+ " and column " + curColumn + " was not closed", this, null );
}
else
{
seenInnerTag = false;
}
}
else if ( ch == '<' )
{
Expand Down Expand Up @@ -3960,7 +3964,7 @@ private char more()
fillBuf();
// this return value should be ignored as it is used in epilog parsing ...
if ( reachedEnd )
return (char) -1;
throw new EOFException( "no more data available" + getPositionDescription() );
}
final char ch = buf[pos++];
// line/columnNumber
Expand Down
128 changes: 123 additions & 5 deletions src/test/java/org/codehaus/plexus/util/xml/pull/MXParserTest.java
View file
Open in desktop
Expand Up @@ -353,6 +353,35 @@ public void testValidCharacterReferenceDecimal()
*
* @throws java.lang.Exception if any.
*/
@Test
public void testParserPosition()
throws Exception
{
String input = "<?xml version=\"1.0\" encoding=\"UTF-8\"?><!-- A --> \n <!-- B --><test>\tnnn</test>\n<!-- C\nC -->";

MXParser parser = new MXParser();
parser.setInput( new StringReader( input ) );

assertEquals( XmlPullParser.PROCESSING_INSTRUCTION, parser.nextToken() );
assertPosition( 1, 39, parser );
assertEquals( XmlPullParser.COMMENT, parser.nextToken() );
assertPosition( 1, 49, parser );
assertEquals( XmlPullParser.IGNORABLE_WHITESPACE, parser.nextToken() );
assertPosition( 2, 3, parser ); // end when next token starts
assertEquals( XmlPullParser.COMMENT, parser.nextToken() );
assertPosition( 2, 12, parser );
assertEquals( XmlPullParser.START_TAG, parser.nextToken() );
assertPosition( 2, 18, parser );
assertEquals( XmlPullParser.TEXT, parser.nextToken() );
assertPosition( 2, 23, parser ); // end when next token starts
assertEquals( XmlPullParser.END_TAG, parser.nextToken() );
assertPosition( 2, 29, parser );
assertEquals( XmlPullParser.IGNORABLE_WHITESPACE, parser.nextToken() );
assertPosition( 3, 2, parser ); // end when next token starts
assertEquals( XmlPullParser.COMMENT, parser.nextToken() );
assertPosition( 4, 6, parser );
}

@Test
public void testProcessingInstruction()
throws Exception
Expand Down Expand Up @@ -403,6 +432,37 @@ public void testProcessingInstructionsContainingXml()
*
* @throws java.lang.Exception if any.
*/
@Test
public void testMalformedProcessingInstructionsContainingXmlNoClosingQuestionMark()
throws Exception
{
StringBuffer sb = new StringBuffer();
sb.append( "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n" );
sb.append( "<project />\n" );
sb.append( "<?pi\n" );
sb.append( " <tag>\n" );
sb.append( " </tag>>\n" );

MXParser parser = new MXParser();
parser.setInput( new StringReader( sb.toString() ) );

try
{
assertEquals( XmlPullParser.PROCESSING_INSTRUCTION, parser.nextToken() );
assertEquals( XmlPullParser.IGNORABLE_WHITESPACE, parser.nextToken() );
assertEquals( XmlPullParser.START_TAG, parser.nextToken() );
assertEquals( XmlPullParser.END_TAG, parser.nextToken() );
assertEquals( XmlPullParser.IGNORABLE_WHITESPACE, parser.nextToken() );
assertEquals( XmlPullParser.PROCESSING_INSTRUCTION, parser.nextToken() );

fail( "Should fail since it has invalid PI" );
}
catch ( XmlPullParserException ex )
{
assertTrue( ex.getMessage().contains( "processing instruction started on line 3 and column 1 was not closed" ) );
}
}

@Test
public void testSubsequentProcessingInstructionShort()
throws Exception
Expand Down Expand Up @@ -624,7 +684,7 @@ public void testMalformedProcessingInstructionNoClosingQuestionMark()
}
catch ( XmlPullParserException ex )
{
assertTrue( ex.getMessage().contains( "processing instruction started on line 1 and column 2 was not closed" ) );
assertTrue( ex.getMessage().contains( "processing instruction started on line 1 and column 1 was not closed" ) );
}
}

Expand Down Expand Up @@ -657,7 +717,7 @@ public void testSubsequentMalformedProcessingInstructionNoClosingQuestionMark()
}
catch ( XmlPullParserException ex )
{
assertTrue( ex.getMessage().contains( "processing instruction started on line 1 and column 13 was not closed" ) );
assertTrue( ex.getMessage().contains( "processing instruction started on line 1 and column 12 was not closed" ) );
}
}

Expand All @@ -666,6 +726,58 @@ public void testSubsequentMalformedProcessingInstructionNoClosingQuestionMark()
*
* @throws java.lang.Exception if any.
*/
@Test
public void testSubsequentAbortedProcessingInstruction()
throws Exception
{
MXParser parser = new MXParser();
StringBuilder sb = new StringBuilder();
sb.append( "<project />" );
sb.append( "<?aborted" );

parser.setInput( new StringReader( sb.toString() ) );

try
{
assertEquals( XmlPullParser.START_TAG, parser.next() );
assertEquals( XmlPullParser.END_TAG, parser.next() );
assertEquals( XmlPullParser.PROCESSING_INSTRUCTION, parser.next() );

fail( "Should fail since it has aborted PI" );
}
catch ( XmlPullParserException ex )
{
assertTrue( ex.getMessage().contains( "@1:21" ) );
assertTrue( ex.getMessage().contains( "processing instruction started on line 1 and column 12 was not closed" ) );
}
}

@Test
public void testSubsequentAbortedComment()
throws Exception
{
MXParser parser = new MXParser();
StringBuilder sb = new StringBuilder();
sb.append( "<project />" );
sb.append( "<!-- aborted" );

parser.setInput( new StringReader( sb.toString() ) );

try
{
assertEquals( XmlPullParser.START_TAG, parser.next() );
assertEquals( XmlPullParser.END_TAG, parser.next() );
assertEquals( XmlPullParser.PROCESSING_INSTRUCTION, parser.next() );

fail( "Should fail since it has aborted comment" );
}
catch ( XmlPullParserException ex )
{
assertTrue( ex.getMessage().contains( "@1:24" ) );
assertTrue( ex.getMessage().contains( "comment started on line 1 and column 12 was not closed" ) );
}
}

@Test
public void testMalformedXMLRootElement()
throws Exception
Expand Down Expand Up @@ -900,6 +1012,12 @@ public void testEncodingISO_8859_1_setInputStream()
}
}

private static void assertPosition( int row, int col, MXParser parser )
{
assertEquals( "Current line", row, parser.getLineNumber() );
assertEquals( "Current column", col, parser.getColumnNumber() );
}

/**
* Issue 163: https://github.com/codehaus-plexus/plexus-utils/issues/163
*
Expand Down Expand Up @@ -958,7 +1076,7 @@ public void testCustomEntityNotFoundInText()
}
catch ( XmlPullParserException e )
{
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_TAG seen <root>&otherentity;... @1:19)" ) );
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_TAG seen <root>&otherentity;... @1:20)" ) );
assertEquals( XmlPullParser.START_TAG, parser.getEventType() ); // not an ENTITY_REF
assertEquals( "otherentity", parser.getText() );
}
Expand Down Expand Up @@ -1025,7 +1143,7 @@ public void testCustomEntityNotFoundInAttr()
}
catch ( XmlPullParserException e )
{
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_DOCUMENT seen <root name=\"&otherentity;... @1:25)" ) );
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_DOCUMENT seen <root name=\"&otherentity;... @1:26)" ) );
assertEquals( XmlPullParser.START_DOCUMENT, parser.getEventType() ); // not an ENTITY_REF
assertNull( parser.getText() );
}
Expand Down Expand Up @@ -1060,7 +1178,7 @@ public void testCustomEntityNotFoundInAttrTokenize() throws Exception
}
catch ( XmlPullParserException e )
{
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_DOCUMENT seen <root name=\"&otherentity;... @1:25)" ) );
assertTrue( e.getMessage().contains( "could not resolve entity named 'otherentity' (position: START_DOCUMENT seen <root name=\"&otherentity;... @1:26)" ) );
assertEquals( XmlPullParser.START_DOCUMENT, parser.getEventType() ); // not an ENTITY_REF
assertNull( parser.getText() );
}
Expand Down