ConcurrentModificationException in java.util.HashMap

[anonymous]

Hi,

I have recently reported other tickets, and this is what I was doing: #11463 and #11461

This time I just launched JOSM (no internet connection, was just pre-loading it), and did not work.

Cheers,

Repository Root: http://josm.openstreetmap.de/svn
Build-Date: 2015-05-08 01:31:23
Last Changed Author: stoecker
Revision: 8339
Repository UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Relative URL: ^/trunk
URL: http://josm.openstreetmap.de/svn/trunk
Last Changed Date: 2015-05-07 17:30:43 +0200 (Thu, 07 May 2015)
Last Changed Rev: 8339

Identification: JOSM/1.5 (8339 en) Linux Ubuntu 14.04.2 LTS
Memory Usage: 326 MB / 857 MB (262 MB allocated, but free)
Java version: 1.7.0_79, Oracle Corporation, OpenJDK 64-Bit Server VM
Java package: openjdk-7-jre:amd64-7u79-2.5.5-0ubuntu0.14.04.2
VM arguments: [-Djosm.restart=true, -Djava.net.useSystemProxies=true]

Plugins:
- DirectUpload (30892)
- ImportImagePlugin (31114)
- PicLayer (31114)
- editgpx (30892)
- geotools (31068)
- jts (31002)
- log4j (30892)
- opendata (31116)
- utilsplugin2 (31120)

Last errors/warnings:
- E: Broken tagging preset "surface-Surface" - number of items in 'display_values' must be the same as in 'values'
- E: Broken tagging preset "surface-Surface" - number of items in 'display_values' must be the same as in 'values'
- E: java.util.ConcurrentModificationException
- E: Connection to proxy 'wpad.icrc.visitor:8080' for URI 'https://josm.openstreetmap.de/wiki/TestedVersion?format=txt' failed. Exception was: java.net.UnknownHostException: wpad.icrc.visitor
- W: Unable to detect latest version of JOSM: Unknown proxy type : HTTP

java.util.ConcurrentModificationException
	at java.util.HashMap$HashIterator.nextEntry(HashMap.java:922)
	at java.util.HashMap$EntryIterator.next(HashMap.java:962)
	at java.util.HashMap$EntryIterator.next(HashMap.java:960)
	at org.openstreetmap.josm.gui.preferences.ToolbarPreferences.loadActions(ToolbarPreferences.java:952)
	at org.openstreetmap.josm.gui.preferences.ToolbarPreferences.getDefinedActions(ToolbarPreferences.java:980)
	at org.openstreetmap.josm.gui.preferences.ToolbarPreferences.refreshToolbarControl(ToolbarPreferences.java:1034)
	at org.openstreetmap.josm.gui.MainApplication.main(MainApplication.java:454)

[wiktorn]

@team:

this should be easily fixed (I guess) using ConcurrentHashMap, just as Sonar issue suggests:
​http://josm.openstreetmap.de/sonar/issues/search#issues=eb53df2c-c107-469e-8936-9a95f7ad67b0

Shall we (more less) blindly replace all HashMaps with ConcurrentHashMaps?

Based on:
​http://java.dzone.com/articles/java-7-hashmap-vs
​https://ria101.wordpress.com/2011/12/12/concurrenthashmap-avoid-a-common-misuse/

I would go with the recommendation from the latter and do:
new ConcurrentHashMap<>(8, 0.9f, 1); instead of new HashMap<>() as the performance of ConcurrentHashMap is not significantly worse than unsafe HashMap.

Unless we have some better clues about the usage of the map.

[wiktorn]

The second one looks like it was based on Java 6, so this might be not an issue right now.

[Don-vip]

It is still a problem in Java 7:

• ​http://docs.oracle.com/javase/7/docs/api/java/util/concurrent/ConcurrentHashMap.html#ConcurrentHashMap()
• Maybe not in Java 8: ​https://docs.oracle.com/javase/8/docs/api/java/util/concurrent/ConcurrentHashMap.html#ConcurrentHashMap--

[wiktorn]

There is an comment on the second blog post:

Just noting that ConcurrentHashMap was revised in Java 7 to use lazy segment initialization, so the memory waste that used to occur with the default constructor is now mitigated. An empty map now contains a single segment.
However, the lazy initialization is triggered by puts to missing segments and NOT by contention. ConcurrentHashMapV8 which may replace ConcurrentHashMap in Java 8, does not use a fixed lock striping scheme at all, instead each bucket serves as a “stripe” using intrinsic synchronization.

[Don-vip]

OK then the default new ConcurrentHashMap<>() sounds good.

[wiktorn]

In 8422/josm:

Change HashMap to ConcurrentHashMap, to avoid ConcurrentModificationException. Closes #11466

[Don-vip]

There are two problems in the Jenkins build:

• one new Findbugs warning
• one test regression

it seems linked to the handling of empty/null values, it seems it was ok to have empty values before.

[wiktorn]

In 8423/josm:

Fix unit tests and findbugs warning. Closes: #11466

[anonymous]

Referenced in build JOSM 8423