[JBoss JIRA] Created: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
2:29 p.m.
graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Fix For: 0.7
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different subtype
of Location, and only one of these objects has ID properties, those locations should be
equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
9:56 a.m.
New subject: [JBoss JIRA] Updated: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch updated DNA-551:
------------------------------
Priority: Blocker (was: Major)
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Priority: Blocker
Fix For: 0.7
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
7:43 p.m.
New subject: [JBoss JIRA] Updated: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch updated DNA-551:
------------------------------
Attachment: DNA-551-patch1.txt
Attached a (still incomplete) patch that JPAV has been working on. John says one test is
still failing, and I haven't yet reviewed it.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
11:37 a.m.
New subject: [JBoss JIRA] Commented: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch commented on DNA-551:
-----------------------------------
I've reviewed the patch, and there are several problems.
One federation unit test in 'dna-graph' is failing because the patch changes the
behavior of some of the Location.create(...) methods that take an identification Property.
The patch was looking for the case where the supplied ID property has a name of
"dna:uuid" and created a LocationWithPathAndUuid using the path and the first
value of the ID property. So if the ID property contains multiple UUIDs (as is the case
in the federation case, when merging nodes with the same path but different UUIDs from
multiple sources), then the UUID values other than the first were lost.
Another problem with the patch is that the Location.create(Path,UUID) method was checking
for a null UUID and creating a Location properly (with just the Path) but was not
returning it, so the method continued to create a LocationWithPathAndUuid using the path
and null UUID reference. This of course fails on the assertion (if assertions are enabled
when running the tests).
Fixing these (locally) resulted in no unit test failures in 'dna-graph' or
'dna-repository'. However, even with these fixes there is one TCK test failure in
WorkspaceCloneTest.testCloneNodesTwice(), with the error:
javax.jcr.PathNotFoundException: Unable to find
"{}testroot/{}node2/{}node1/{}node3"; lowest existing path is
"/{}testroot/{}node2": Unable to find
"{}testroot/{}node2/{}node1/{}node3"; lowest existing path is
"/{}testroot/{}node2": Unable to find
"{}testroot/{}node2/{}node1/{}node3"; lowest existing path is
"/{}testroot/{}node2"
at org.jboss.dna.jcr.JcrWorkspace.clone(JcrWorkspace.java:401)
at
org.apache.jackrabbit.test.api.WorkspaceCloneTest.testCloneNodesTwice(WorkspaceCloneTest.java:64)
I'll continue to investigate this error.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:01 p.m.
New subject: [JBoss JIRA] Commented: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch commented on DNA-551:
-----------------------------------
The error was actually an unrelated bug in GraphSession (DNA-559). The fix has been
committed, and incorporating that fix to these changes gets us farther. However, there
still is another error in one of the unit tests in 'dna-connector-filesystem'.
The problem appears to be in AbstractConnectorTest, where one use of
Location.isSame(Location) was not changed to use 'equals()'. Fixing that makes
all unit tests and integration tests pass (via a complete integration build).
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
1:03 p.m.
New subject: [JBoss JIRA] Updated: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch updated DNA-551:
------------------------------
Attachment: DNA-551-patch2.txt
Uploaded the patch file "DNA-551-patch2.txt" that contains all fixes to pass all
tests.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt, DNA-551-patch2.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
2:19 p.m.
New subject: [JBoss JIRA] Resolved: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch resolved DNA-551.
-------------------------------
Resolution: Done
Assignee: Johnny Verhaeg
Applied the second version of the patch ('DNA-551-patch2.txt'), after all unit and
integration tests passed. Basically, the Location.equals(...) and Location.isSame(...)
method behaviors were switched, and the various uses corrected. Now, the
Location.equals(...) and Location.hashCode() methods behave in the traditional way, and
can be used as keys in Maps (even when you might have partial Location objects and the Map
might contain actual Location objects, or vice versa).
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Assignee: Johnny Verhaeg
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt, DNA-551-patch2.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:53 p.m.
New subject: [JBoss JIRA] Reopened: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch reopened DNA-551:
-------------------------------
Found a couple of places where Location.isSame(Location) is still being used that should
be Location.equals(Object), including several places in JavaDoc.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Assignee: Johnny Verhaeg
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt, DNA-551-patch2.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:53 p.m.
New subject: [JBoss JIRA] Commented: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch commented on DNA-551:
-----------------------------------
Fixed a couple of calls to 'Location.isSame()' that should have been fixed
earlier.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Assignee: Johnny Verhaeg
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt, DNA-551-patch2.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
3:55 p.m.
New subject: [JBoss JIRA] Resolved: (DNA-551) graph.Location's equals(Object) and hashCode() Methods Are Not Consistent
[
https://jira.jboss.org/jira/browse/DNA-551?page=com.atlassian.jira.plugin...
]
Randall Hauch resolved DNA-551.
-------------------------------
Resolution: Done
Hopefully that is it, so marking as resolved for now.
graph.Location's equals(Object) and hashCode() Methods Are Not
Consistent
-------------------------------------------------------------------------
Key: DNA-551
URL: https://jira.jboss.org/jira/browse/DNA-551
Project: DNA
Issue Type: Bug
Components: Graph
Affects Versions: 0.6
Reporter: Dan Florian
Assignee: Johnny Verhaeg
Priority: Blocker
Fix For: 0.7
Attachments: DNA-551-patch1.txt, DNA-551-patch2.txt
Location.hashCode() and Location.equals(Object) need to be consistent. The current
implementation of the base class and subclasses prevent logically equal locations (same
path or same ID properties) from being used java.util.Map objects.
It should be true that if one subtype of Location has the same path as a different
subtype of Location, and only one of these objects has ID properties, those locations
should be equal and should have the same hashCode.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
5988
days inactive
6005
days old
9 comments
2 participants
participants (2)
-
Dan Florian (JIRA) -
Randall Hauch (JIRA)