Bug: Acquisition Rating etc

[wmoore]

Kenny and I have seen some strange behaviour when editing ratings on Plate-Acquisitions.

Seems like there's some very inconsistent updating of the right-hand panel.

E.g Rate several Acquisitions differently, then select one after the other. Each time I select a different one I get a different rating - not consistently.

Then I got a strange error:
NB: using webstart Insight:

java.lang.Exception: Abnormal termination due to an uncaught exception.
java.lang.NoSuchMethodError: net.n3.nanoxml.IXMLElement.getChildrenNamed(Ljava/lang/String;)Ljava/util/ArrayList;
	at org.openmicroscopy.shoola.agents.util.editorpreview.PreviewModel.parseXmlDescription(PreviewModel.java:95)
	at org.openmicroscopy.shoola.agents.util.editorpreview.PreviewModel.<init>(PreviewModel.java:143)
	at org.openmicroscopy.shoola.agents.util.editorpreview.PreviewPanel.setDescriptionXml(PreviewPanel.java:358)
	at org.openmicroscopy.shoola.agents.util.editorpreview.PreviewPanel.<init>(PreviewPanel.java:341)
	at org.openmicroscopy.shoola.agents.metadata.editor.DocComponent.formatTootTip(DocComponent.java:389)
	at org.openmicroscopy.shoola.agents.metadata.editor.DocComponent.initComponents(DocComponent.java:591)
	at org.openmicroscopy.shoola.agents.metadata.editor.DocComponent.<init>(DocComponent.java:743)
	at org.openmicroscopy.shoola.agents.metadata.editor.DocComponent.<init>(DocComponent.java:755)
...

[wmoore]

Seems each Plate-Acquisition is linked to multiple ratings:

omero=# select * from PlateAcquisitionAnnotationLink order by id desc;
 id  | permissions | version | child | creation_id | external_id | group_id | owner_id | update_id | parent 
-----+-------------+---------+-------+-------------+-------------+----------+----------+-----------+--------
 116 |        -120 |         |  5018 |      142345 |             |        3 |        3 |    142345 |      6
 115 |        -120 |         |  5017 |      142343 |             |        3 |        3 |    142343 |      5
 114 |        -120 |         |  5016 |      142341 |             |        3 |        3 |    142341 |      5
 113 |        -120 |         |  5015 |      142339 |             |        3 |        3 |    142339 |      5
 112 |        -120 |         |  5014 |      142337 |             |        3 |        3 |    142337 |      5
 111 |        -120 |         |  5013 |      142335 |             |        3 |        3 |    142335 |      4
 110 |        -120 |         |  5012 |      142333 |             |        3 |        3 |    142333 |      4
 109 |        -120 |         |  5011 |      142331 |             |        3 |        3 |    142331 |      6
 108 |        -120 |         |  5010 |      142329 |             |        3 |        3 |    142329 |      5
 107 |        -120 |         |  5009 |      142327 |             |        3 |        3 |    142327 |      5
 106 |        -120 |         |  5008 |      142325 |             |        3 |        3 |    142325 |      4
 105 |        -120 |         |  5007 |      142323 |             |        3 |        3 |    142323 |      4
 104 |        -120 |         |  5006 |      142321 |             |        3 |        3 |    142321 |      4
 103 |        -120 |         |  5005 |      142319 |             |        3 |        3 |    142319 |      4
 102 |        -120 |         |  5004 |      142317 |             |        3 |        3 |    142317 |      4
 101 |        -120 |         |  5003 |      142315 |             |        3 |        3 |    142315 |      4

omero=# select id, longvalue from annotation order by id desc;
  id  | longvalue 
------+-----------
 5018 |         5
 5017 |         3
 5016 |         3
 5015 |         3
 5014 |         3
 5013 |         5
 5012 |         5
 5011 |         2
 5010 |         1
 5009 |         4
 5008 |         3
 5007 |         3
 5006 |         3
 5005 |         3
 5004 |         2
 5003 |         1

[wmoore]

Web shows all ratings on PlateAcquisition?

[jburel]

it does not really make sense to show all the ratings, but as discussed that will probably be removed/replaced

[jburel]

For the exception, we need to decide what to do with editor or at list with the library used. That's not the first we see such error b/c of nanoxml

[jburel]

Moved from sprint 2012-11-20 (2)

[jburel]

see #9926

Using web start Insight on howe I get a crash if I try to select the following image:

user: user-3 
group: read-annotate-1 
dataset: avi (id=61)
image: Level003-DB-089.avi

This was initially put down to bad data but checking the same image with the webclient I can select the image and view it with no problems.

Exception attached in comment.

[jburel]

The problem is due to the fact that the jars ordering described in the ivy file is not the order used when using insight/webstart.
Basically it is important to have jhotdraw in the class path before nanoxml (library used by the editor) since jhotdraw uses a different version of nanoxml.

Will: how long will it take to get rid of nanoxml since the editor is no longer under development.
We also need to check what we can do for the webstart build. (might be quicker)

[jmoore]

See grep output in #9926 for one possible solution

[jburel]

Problem has now been fixed in insight but the rating no longer shows up in web.
Testing against howe/user1
Screenshots attached to the ticket.
Code ​https://github.com/jburel/openmicroscopy/tree/rating-9881
Will I am passing the ticket to you. If you need to make some changes it probably makes more sense to use my branch. You can open PR against it when ready.

[wmoore]

Web fixes are in ​https://github.com/openmicroscopy/openmicroscopy/pull/491

[jburel <j.burel@…>]

(In [1abe219b0e3c8e8b833ceeff0d6ff5b8f8d330ba/ome.git] on branch develop) Save one rating per object for plate acquisition (see #9881)