Fix Java 11 deprecation warnings #6145
Conversation
|
#6119 has these changes (and more), but this PR has a set of simple changes that should be easy to merge. |
There was a problem hiding this comment.
@tomwhite I think we can avoid most of the suppressions because they're mostly Class.newInstance which has a replacement in java 8. I thought some of the new Integer calls might be deliberate, but none of them look meaningful on further inspection. Most of them can be just removed entirely though since the compiler will box them if necessary.
| @@ -298,6 +298,7 @@ private static boolean printStackTraceOnUserExceptions() { | |||
| /** | |||
| * Returns the command line program specified, or prints the usage and exits with exit code 1 * | |||
| */ | |||
| @SuppressWarnings("deprecation") | |||
There was a problem hiding this comment.
I think most of the deprecation sites can be fixed by replace class.newInstance() with class.getDeclaredConstructor().newInstance() which seems to exist in 8. It means adding additional exceptions to the list of exceptions caught or thrown but I think that's preferable to keeping the deprecated method and suppressing it.
There was a problem hiding this comment.
In intellij you can use find -> search structurally to find all these sites easily.
| {"-27", new Integer(-27)}, | ||
| {"0", new Integer(0)}, | ||
| {"-27014", new Integer(-27014)}, | ||
| {"27", Integer.valueOf(27)}, |
There was a problem hiding this comment.
I was thinking this might have been a deliberate choice, but looking at it closely I don't think so. This seems fine.
| } | ||
| else if ( annotationsToCheckMaf.get(i).equals(MafOutputRendererConstants.FieldName_End_Position) ) { | ||
| mafFieldValues = maf.getRecords().stream().map(AnnotatedInterval::getEnd).map(x -> new Integer(x)).map(Object::toString).collect(Collectors.toList()); | ||
| mafFieldValues = maf.getRecords().stream().map(AnnotatedInterval::getEnd).map(x -> Integer.valueOf(x)).map(Object::toString).collect(Collectors.toList()); |
There was a problem hiding this comment.
I think these are unnecessary boxing and could just be removed.
| score.addTotalReads(new Integer(tok[7])); | ||
| score.addUnambiguousReads(new Integer(tok[8])); | ||
| score.setReferenceLength(new Long(tok[9])); | ||
| score.addSelfScore(Double.valueOf(tok[5])); |
There was a problem hiding this comment.
this should really be Double.parseDouble instead but it's just a test so it doesn't really matter
| @@ -104,7 +104,7 @@ public void testRandomMaskedKmers() { | |||
| //Generate random indices to mask | |||
| List<Byte> maskIndices = new ArrayList<>(K); | |||
| for (int j = 0; j < K; j++) { | |||
| maskIndices.add(new Byte((byte) j)); | |||
| maskIndices.add(Byte.valueOf((byte) j)); | |||
| @@ -132,7 +132,7 @@ void resizeTest() { | |||
| Assert.assertTrue(hopscotchSet.size() == hashSet.size()); | |||
| LongIterator itrL = hopscotchSet.iterator(); | |||
| while (itrL.hasNext()) { | |||
| Assert.assertTrue(hashSet.contains(new Long(itrL.next()))); | |||
| Assert.assertTrue(hashSet.contains(Long.valueOf(itrL.next()))); | |||
tomwhite
force-pushed
the
tw_fix_java11_deprecation_warnings
branch
from
1f5db31 to
3e30784
Compare
tomwhite
force-pushed
the
tw_fix_java11_deprecation_warnings
branch
from
3e30784 to
2e74072
Compare
Pre-req to testing on Java 11