[
https://issues.apache.org/jira/browse/PDFBOX-4472?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16782351#comment-16782351
]
Tilman Hausherr commented on PDFBOX-4472:
-----------------------------------------
I don't have a good idea for a test either, but from the javadoc, it makes
sense:
{quote}
Like most collection classes, this class is not synchronized. A synchronized
WeakHashMap may be constructed using the Collections.synchronizedMap method.
{quote}
The warning that "It is imperative that the user manually synchronize on the
returned map when iterating over any of its collection views" doesn't apply in
our code, because we don't iterate.
> Thread stuck in SoftReferenceCache.get
> --------------------------------------
>
> Key: PDFBOX-4472
> URL: https://issues.apache.org/jira/browse/PDFBOX-4472
> Project: PDFBox
> Issue Type: Bug
> Components: JBIG2
> Affects Versions: 3.0.2 JBIG2
> Reporter: Arnaud Jeansen
> Priority: Major
> Attachments:
> 0001-PDFBOX-4472-Enforce-access-to-the-weak-cache-through.patch
>
>
> We use pdfbox-jbig2 3.02 in a multi-threaded environment, and the
> SoftReferenceCache implementation is a problem because of its usage of an
> unsynchronized WeakHashMap.
> Today, we found two threads blocking each other with the same stack trace:
> {code:java}
> pool-95-thread-1 - priority:5 - threadId:0x00007f882c109800 - nativeId:0x3d81
> - nativeId (decimal):15745 - state:RUNNABLE
> stackTrace:
> java.lang.Thread.State: RUNNABLE
> at java.util.WeakHashMap.get(WeakHashMap.java:403)
> at
> org.apache.pdfbox.jbig2.util.cache.SoftReferenceCache.get(SoftReferenceCache.java:40)
> at org.apache.pdfbox.jbig2.JBIG2ImageReader.read(JBIG2ImageReader.java:243)
> at javax.imageio.ImageReader.read(ImageReader.java:939)
> at sun.reflect.GeneratedMethodAccessor3824.invoke(Unknown Source)
> at
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.lang.reflect.Method.invoke(Method.java:498)
> at
> org.icepdf.core.pobjects.graphics.images.JBig2Decoder.decodeJbig2(JBig2Decoder.java:161)
> at
> org.icepdf.core.pobjects.graphics.images.JBig2Decoder.decode(JBig2Decoder.java:80)
> at
> org.icepdf.core.pobjects.graphics.images.ImageStream.getImage(ImageStream.java:94)
> at
> org.icepdf.core.pobjects.graphics.images.references.ImageStreamReference.call(ImageStreamReference.java:74)
> at
> org.icepdf.core.pobjects.graphics.images.references.ImageStreamReference.call(ImageStreamReference.java:40)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748)
> Locked ownable synchronizers:
> - <0x0000000600019a48> (a java.util.concurrent.ThreadPoolExecutor$Worker)
> {code}
> I couldn't manage to write a unit test to reproduce the problem
> unfortunately. But the patch is straightforward and attached to this ticket.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
