Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Insight connectors dev 4 4 #1321

Merged
merged 56 commits into from Sep 11, 2013
Merged

Conversation

jburel
Copy link
Member

@jburel jburel commented Jun 28, 2013

This PR combines commit from gh-1278 and gh-1304. Not all commits are applicable to dev_4_4.
This should fix connection issue noticed on develop and at times on dev_4_4.
To test:

  • import data.
  • Browse dataset.
  • View images.
  • etc.

This might require to use the client for few days before the PR can be merged.


--rebased-from #1304
--rebased-from #1278
etc

@pwalczysko
Copy link
Member

Bug: cannot create thumbs in importer.
Workflow:

  • login as user-6
  • ImportAs user-5, group read-write-1 (still logged in as user-6), dv folder on test_images_good
  • the import is successful, the images are okay and their thumbs are okay in Data Manager but
  • the thumbnails cannot be created error is in the importer window (see screenshot below)
  • if you import the same images into the same group, but this time as yourself (=user-6 as user-6), the thumbs in importer are okay (see second screenshot)
    screen shot 2013-07-03 at 10 04 34
    screen shot 2013-07-03 at 10 04 44

@pwalczysko
Copy link
Member

Plates thumbs loading for big plates is still slow (much slower than small images). The 380 well plate loads in 4-5 minutes.
screen shot 2013-07-03 at 10 23 06

This did not repeat though - on other occassion, the loading was fast.

@pwalczysko
Copy link
Member

Connection Destroyed after:

  • change channel name of a plate
  • in the middle of the process, go Switch User and confirm the dialog that you want to disconnect
  • connection destroyed popup appears and is not clickabel

Similar workflow already reported for develop, but I think the problem is sorted there now (see http://trac.openmicroscopy.org.uk/ome/ticket/11032 - the change is that channel name of plate was manipulated instead of big image viewing)

screen shot 2013-07-03 at 10 34 42

@jburel
Copy link
Member Author

jburel commented Jul 3, 2013

@pwalczysko: The last 4 commits should fix the thumbnail after import issue. They do not need to be ported to develop, they are related to the import sequence in dev_4_4

@jburel
Copy link
Member Author

jburel commented Jul 3, 2013

@pwalczysko: Few commits were not backported

@pwalczysko
Copy link
Member

The two problems reported above (connection destroyed and cannot create thumb) are fixed. The problem with slow reading of the plate thumbs cannot be repeated. Works fine.

@joshmoore
Copy link
Member

I'm slightly concerned about rushing this branch. Sorry, I just haven't had a chance to build this locally (and won't before leaving).

@jburel
Copy link
Member Author

jburel commented Jul 3, 2013

@joshmoore, @pwalczysko: I think the PR should remain opened for a while. No need to be merged now (cf. description) since we should not have any development happening

@chris-allan
Copy link
Member

Excluded until we can get the required code from #1345 in. Then this branch can have dev_4_4 merged into it and conflicts resolved.

joshmoore and others added 20 commits July 17, 2013 13:05
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
Connector objects were being created without needed
in some instances. Since the instantiation of an
omero.clients object has high overhead, it's worth
preventing this.

Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
Since all service access calls getConnector(), all logic for
handling the checking of alive sessions and network has been
refactored into the one location. This also prevents multiple
calls.

Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/util/NetworkChecker.java
Most likely, the synchronized methods on OMEROGateway
were hiding this, but a race-condition during startup
allowed dsFactory.getCredentials() to return null.
 * Remove Connector.reconnect() since it leaves services dangling
 * Remove all get*Service() methods from OMEROGateway
 * Refactor all get*Service() methods in Connector
 * Use concurrent storage for services in Connector
 * Close all stateful services as quickly as possible.

Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/agents/fsimporter/ImportResultLoader.java
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OMEROGateway.java
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OmeroImageService.java
	components/insight/SRC/org/openmicroscopy/shoola/env/data/OmeroImageServiceImpl.java
jburel added 16 commits July 17, 2013 13:09
Some thumbnails could take time to load. Display the
progress bar so the user is notified of loading progress.

Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/agents/dataBrowser/ThumbnailLoader.java
Conflicts:
	components/insight/SRC/org/openmicroscopy/shoola/env/data/views/calls/ThumbnailLoader.java
The maximum size of a plane is 2000*2000
see 5b86032
The problem only occurs when import data for another user.
Method was invoked twice.
Resetting the channels name could take some time.
if the user uses the switch user, service no longer valid
could be used.
@jburel
Copy link
Member Author

jburel commented Jul 17, 2013

Rebased the branch and removed the exclude label

@mtbc
Copy link
Member

mtbc commented Jul 21, 2013

Do we want to mark this "on hold" so it hangs around awhile quietly getting accidentally tested?

@jburel
Copy link
Member Author

jburel commented Aug 1, 2013

This PR is in the daily build so indirectly tested.

@joshmoore
Copy link
Member

@jburel: is this a target for 4.4.9? If so, we should look to move it out of ON_HOLD.

@jburel
Copy link
Member Author

jburel commented Sep 4, 2013

@joshmoore: 4.4.9 target.This is a rebase, marked as onhold so it could test w/o specific review. Removing label.

@pwalczysko
Copy link
Member

This is fine on Mac. Nevertheless, I wanted to give it a go on Windows (see #1453 (comment)). As the windows artifacts are not being built atm, I would wait till they do and give it a go.

@pwalczysko
Copy link
Member

Tested on Win 7 Java 1.6. No problems even when challenged (large plate & change settings & switch user).
Ready to merge.

@joshmoore
Copy link
Member

Thanks, @pwalczysko. (And, I guess, @jburel too 😄)

joshmoore added a commit that referenced this pull request Sep 11, 2013
@joshmoore joshmoore merged commit 5db22f6 into ome:dev_4_4 Sep 11, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants