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

Issue788 durack1 remove deprecated folders and files #1341

Closed

Conversation

durack1
Copy link
Member

@durack1 durack1 commented May 21, 2015

Fix #788 - remove deprecated folders and files

Starting with files and will continue with tests, docs and esg subdirs..

Resolving the duplication and obsolete files in images and resources will be resolved by @doutriaux1 in #1353

The issue of updating the license info will be dealt with in #1354

@durack1
Copy link
Member Author

durack1 commented May 22, 2015

@aashish24 when I looked at this last night it seemed to indicate a cmake build problem - wanna take a peek here?

@aashish24
Copy link
Contributor

This is weird. @jbeezley, any suggestion?

@durack1
Copy link
Member Author

durack1 commented May 26, 2015

@aashish24 @jbeezley any ideas what's going wrong here?

@jbeezley
Copy link
Contributor

CMake Error at /usr/share/cmake-2.8/Modules/CPack.cmake:395 (message):
  CPack package description file:
  "/home/buildbot/buildbot/uvcdat-garant-linux-release/source/docs/README.txt"
  could not be found.
Call Stack (most recent call first):
  /usr/share/cmake-2.8/Modules/CPack.cmake:399 (cpack_check_file_exists)
  CMakeLists.txt:810 (include)


CMake Error at /usr/share/cmake-2.8/Modules/CPack.cmake:395 (message):
  CPack license resource file:
  "/home/buildbot/buildbot/uvcdat-garant-linux-release/source/docs/Legal.txt"
  could not be found.
Call Stack (most recent call first):
  /usr/share/cmake-2.8/Modules/CPack.cmake:400 (cpack_check_file_exists)
  CMakeLists.txt:810 (include)

It is looking for docs/README.txt and docs/Legal.txt. I don't know enough about CPack to say why or if it is possible override the path to these files, but I would guess if you added them back, it would probably work.

@doutriaux1
Copy link
Contributor

@durack1 you yanked too many files, please try to build before submitting PR. You need to restore (at least) the two files that @jbeezley mentioned. Actually I would create a symbolic link for te README file to the official README.

@durack1
Copy link
Member Author

durack1 commented May 28, 2015

@jbeezley it seems that there are just two VCS/VTK issues only occurring on the test-laptop setup

@jbeezley
Copy link
Contributor

I have seen these resize errors from time to time. I suspect there is a race condition popping up. It can be tricky to determine when a resize event is done rendering. I have no idea at all if this is related, but there was a bug in VTK-Web fixed recently that went to great lengths to handle window resizes on X11.

@durack1
Copy link
Member Author

durack1 commented May 29, 2015

@jbeezley will #1348 that @dlonie proposed pull across these changes into the UV-CDAT/master? It'd be nice to get these tests behaving properly..

@alliepiper
Copy link
Contributor

I see Chris's commit does get pulled in by that branch, but it gets reverted:

commit 30eedf071db7252b956a0364f40af2bb98abe149
Author: Chris Harris <[email protected]>
Date:   Thu Apr 9 10:06:34 2015 -0400

    Revert "Wait for resize notification in SetSize(...)"

    Fix in vtkWeb code instead.

    This reverts commit 39fedcd67301c7975fc44dcd910d7cf1f23dc031.

Not sure what's up with that.

@jbeezley
Copy link
Contributor

Yeah, I think it caused other problems, so he fixed the issue in the vtkWeb python source. I linked the patch just to emphasize that syncing resizes is not easy in general. I doubt it would have helped here anyway.

I'm not sure it is worth trying to fix this until we have cleaned up the vtk pipeline in the vcs code. We could probably add an explicit sleep after the resize but before the screenshot is taken to reduce the chance that it happens.

@durack1
Copy link
Member Author

durack1 commented Jun 12, 2015

@doutriaux1 I think this is ready to go - can you review please?
@chaosphere2112 I added a time.sleep(2) to your testing/vcs/vtk_ui/test_vtk_ui_manager_resize.py file - can you check this is ok before the PR is merged..?

@doutriaux1
Copy link
Contributor

@durack1 why did you add a time.sleep? Was it failing for you?

@doutriaux1
Copy link
Contributor

@chaosphere2112 while we are on the vtk_ui thing. I was wondering if there's anyway to make these test run in "background" mode, they keep poping up and make my windows loose focus, very annoying.
Same for dv3d.

@durack1
Copy link
Member Author

durack1 commented Jun 12, 2015

@doutriaux1 yep, the test appeared to be checking for the resize before it was completed - see the dialogue above for more info.. The sleep statement seems to have resolved the race condition, at least in this PR..

@chaosphere2112
Copy link
Contributor

"Due to #1148, have to render on screen when resizing"

@dlonie, any progress with the bug? If not, then, nope, sorry Charles.

@chaosphere2112
Copy link
Contributor

If the test is passing with the sleep, then that's fine with me. I assume that the size stored in the object is the size that it's shrinking to, not the size that it's literally currently at; I could restructure that test a bit to listen to the notifications of window resizing, perhaps. That might be a bit easier/better.

@durack1
Copy link
Member Author

durack1 commented Jun 15, 2015

@chaosphere2112 it seems like the canvas is inheriting something along the right hand side pixel column, and only for the test-laptop config..

@durack1
Copy link
Member Author

durack1 commented Jun 16, 2015

@doutriaux1 can you check and merge this, it'll solve the vcs test problem which appeared in the previous #1380 build (I also added the sleep(1) from this PR into #1380 to get it passing)

@durack1 durack1 force-pushed the issue788_durack1_removeDeprecatedFoldersAndFiles branch from f8c65c1 to 63452ce Compare June 17, 2015 16:24
@durack1 durack1 force-pushed the issue788_durack1_removeDeprecatedFoldersAndFiles branch from bd237b8 to 4601c8a Compare June 17, 2015 17:55
@durack1
Copy link
Member Author

durack1 commented Jun 17, 2015

@doutriaux1 I've finished up with this, please go ahead and checkout and then build on top of this PR.. As noted in #1353 there is a bit more cleaning up to do..

@aashish24
Copy link
Contributor

👍 for doing this awesome stuff...

@durack1
Copy link
Member Author

durack1 commented Jun 24, 2015

@aashish24 if you can comment on #1353 that would be helpful. I'm not sure if there are standard directory names/conventions that should be considered when renaming/moving/removing things.. In particular I'm thinking about the eventual creation of Mac OS X *.dmg files with their standard Contents, Resources etc subdirs..

@durack1 durack1 added this to the 2.3 milestone Jul 7, 2015
@aashish24
Copy link
Contributor

@doutriaux1 this seems to be a useful branch. I think we should really merge (and if needed update it). what you think?

@doutriaux1
Copy link
Contributor

some of the deleted files need to stay (licences, legal etc..) I need to carefully review it (and update it), can't merge now. We can clause and do it from scratch later if you want.

@durack1
Copy link
Member Author

durack1 commented Sep 2, 2015

@aashish24 @doutriaux1 as noted up the top of the PR:

Resolving the duplication and obsolete files in images and resources will be resolved by @doutriaux1 in #1353

The issue of updating the license info will be dealt with in #1354 (this requires regeneration, as the UC license is not longer active for LLNS)

So I think this PR is ok to merge..

@doutriaux1
Copy link
Contributor

I'm not sure if the esg module is no longer used, I need to ask ESGF people. Too many uncertainties. This really needs to be redone from scratch by me. Closing. Sorry.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Remove/update deprecated files
6 participants