Bug: Server class should become as a part of the common location

[atarkowska]

webadmin.custom_models should moved to webgateway, as it is needed by any app that wants to interact with the servers
14:56:29 Carlos Neves: (though settings.py is all we need, to be honest, but still). Its not webadmin specific at all
Ola: Does it make sense to move it to 'common lib' like decorators, see ticket #6342

[cneves]

I don't think common is the correct place for it because the Server class is used for accessing the configured server connections, thus being as close to omero.gateway as one can be but being web specific. Thus webgateway == omero.gateway web wrapper, this would be the logical place to put this, imo.

I am assuming my understanding of what the common app is correct, a base app holding all media needed to comply with the omeroweb framework. Having that, putting code that is used at the blitz connection handling layer there sounds to me like we're polluting it.

[jmoore]

Moved from sprint 2012-01-31 (7)

[jburel]

Moved from sprint 2012-02-14 (8)

[jburel]

Moved from sprint 2012-02-28 (9)

[atarkowska]

Chris, could you please decide what I should do with it? I can handle it when I get back

[jburel]

Moved from sprint 2012-03-13 (10)

[jburel]

Chris is off, moving to next sprint

[jburel]

Moved from sprint 2012-04-24 (13)

[jburel]

Moved from sprint 2012-05-08 (14)

[jburel]

Moved from sprint 2012-05-22 (15)

[jburel]

Moved from sprint 2012-06-05 (16)

[jburel]

Moved from sprint 2012-06-19 (17)

[jburel]

Moved from sprint 2012-07-03 (18)

[jmoore]

Per Chris, moving to 4.4.1

[cxallan]

In progress branch is available here:

• ​https://github.com/chris-allan/openmicroscopy/commits/7780_server

[wmoore]

Seems like this is nearly done Chris? ​https://github.com/chris-allan/openmicroscopy/commit/33af9d6367027d39b64469e69d018751fc43b81c

Keep it in 4.4.1, yes?

[jmoore]

Re-assigning to Chris for clarification. I'm tempted to just push.

[cneves]

Cherry-picked on top of current develop and fixed conflicts, also added some unittest fixes to cope with the changes to the library but webclient doesn't really have any proper unittesting and webadmin has a few failing tests that are probably failing for a long time now and are unrelated to these changes.

Current work being done on ​https://github.com/cneves/openmicroscopy/tree/tickets/7780_server and next step I'll incorporate this into JCB code where we have much better code coverage.

[cneves]

The branch above has been tested on the commercial side and so far it doesn't seem to be breaking anything.

Chris:

• you have the time remaining on this ticket to 0.2; is there something else you want to get moved around or was that just for testing / cleanup
• The logger.info('Haha') thing on connector.py, along with other logging attempts I did there made me realize logging isn't outputting anything on this module. Do you know why?

[cxallan]

Replying to cneves-x:

The branch above has been tested on the commercial side and so far it doesn't seem to be breaking anything.

Chris:

• you have the time remaining on this ticket to 0.2; is there something else you want to get moved around or was that just for testing / cleanup

Pretty much just the testing and branch cleanup. I would go ahead and completely clean this up getting rid of the ugly merge commits and just cherry-pick the two commits and rewrite the history with your changes.

• The logger.info('Haha') thing on connector.py, along with other logging attempts I did there made me realize logging isn't outputting anything on this module. Do you know why?

I don't know why and it was one of the things I wanted to investigate. The reason I put that in there was that I was having trouble debugging things and there was no logging coming from the module at all. That should be stripped out and cleaned up.

I also endevoured for there to be no code changes required for this largely other than the migration itself. Are the find() changes things that have just been broken on the open source side?

[cneves]

Replying to cxallan:

Chris:

• you have the time remaining on this ticket to 0.2; is there something else you want to get moved around or was that just for testing / cleanup

Pretty much just the testing and branch cleanup. I would go ahead and completely clean this up getting rid of the ugly merge commits and just cherry-pick the two commits and rewrite the history with your changes.

Doing that now.

• The logger.info('Haha') thing on connector.py, along with other logging attempts I did there made me realize logging isn't outputting anything on this module. Do you know why?

I don't know why and it was one of the things I wanted to investigate. The reason I put that in there was that I was having trouble debugging things and there was no logging coming from the module at all. That should be stripped out and cleaned up.

I'm investigating this, but since time is a little pressed I'll probably move this to a new ticket

I also endevoured for there to be no code changes required for this largely other than the migration itself. Are the find() changes things that have just been broken on the open source side?

This change is much older than your branch but unittesting for webclient and webadmin being all but ignored lately the issue was never noticed. In the end it is unrelated to the ticket but needed for it to be correctly merged into develop, so I'm keeping the fixes unless you'd rather I do two tickets and two dependent PRs.

[cneves]

​https://github.com/openmicroscopy/openmicroscopy/pull/362 and ​https://trac.openmicroscopy.org.uk/ome/ticket/9608

[jmoore]

Carlos, once you've opened a PR for review, go ahead and close the ticket.

[Carlos Neves <carlos@…>]

(In [1b811e48f97cb83039eb88e8ba69db9bccea6dc9/ome.git] on branch develop) Move Server to omeroweb.connector. (See #7780)

[Carlos Neves <carlos@…>]

(In [3260278b4633cf931777df141a095dfbf9df569b/ome.git] on branch develop) Aggregating scattered methods into omeroweb's Connector class (see #7780)

Taken from ​https://github.com/chris-allan/openmicroscopy/commit/033df2f930b7e27274f935d449dde0e6a5879817