Hi, pgmodeler version 0.9.2 has been released and it provides support for PostgreSQL 11 and add many bug fixes and improvements. See: https://github.com/pgmodeler/pgmodeler/blob/v0.9.2/CHANGELOG.md Best regards, Nico.
Source RPM: (none) => pgmodeler-0.9.1-3.mga7.src.rpmAssignee: bugsquad => nicolas.salguero
Suggested advisory: ======================== The updated packages provide support for PostgreSQL 11 and add many bug fixes and improvements. References: https://github.com/pgmodeler/pgmodeler/blob/v0.9.2/CHANGELOG.md https://pgmodeler.io/blog/2019/12/26/pgmodeler-092-stable-is-here-for-my-pride-and-joy ======================== Updated packages in core/updates_testing: ======================== pgmodeler-0.9.2-1.mga7 pgmodeler-devel-0.9.2-1.mga7 from SRPMS: pgmodeler-0.9.2-1.mga7.src.rpm
Assignee: nicolas.salguero => qa-bugsStatus: NEW => ASSIGNED
MGA7-64 Plasma on Lenovo B50 No installation issues As I tested the mediawiki update a few days ago, using postgres, I launched pgmodeler, connected to the local instance of postgres, its mediawiki database and imported the whole data structure of it. Nice!!! I then saved the whole thing, using the "File - Save as" menu, closed the file, used dolphin to locate the dbm file, clicked and pgmodeler opened with the whole schema, but complained the model haddn't been saved. I looked around the documentation, but found nothing on it. I like the tool a lot, but not being able to properly save it, is a nogo for me. But it just seems a stupid question .....
CC: (none) => herman.viaene
Ah yes indeed, I also saw that in the test I made now. But if you open pgModeler, click "Open Model" and choose your model, do you see the error popup? In my test, it does not appear and the model is correctly loaded so I think the problem is not with the "save" function but with opening the model outside the software.
@ Nicolas Tried to open the model from within the application, and indeed the problem does nit replicate in this way. So I agree with you the problem is not with the "save" function. But the problem is even worse: if I give at the CLI, making sure I am in the folder where the model file resides. $ pgmodeler mediawiki.dbm then I get an error box : "Could not load the database model file mediawiki.dbm. Check the error stack to see details. You can try to fix it in order to make it loadable again." I can then either clickon "Fix model" or "Cancel" If I choose "Fix model" then the process aborts with Loading input file: /usr/bin/mediawiki.dbm Fixed model file: /usr/bin/mediawiki_fixed.dbm Extracting objects' XML... [0] src/pgmodelercli.cpp (716) void PgModelerCli::extractObjectXML() [FileDirectoryNotAccessed] Could not access the file or directory `/usr/bin/mediawiki.dbm'! Make sure that it exists or if the user has access permissions on it! ** pgmodeler-cli aborted due to critical error(s). ** So it seems the program looks for the file to be opened in the wrong place. This is a serious bug IMnotsoHO, but in the veiw that this version should support Postgres 11, I think we should let it go. If you do not agree, please remove the OK marker.
Whiteboard: (none) => MGA7-64-OK
Upstream released a new tarball for version 0.9.2 with additional fixes so pgmodeler-0.9.2-2.mga7 will be available soon.
Whiteboard: MGA7-64-OK => feedback
Suggested advisory: ======================== The updated packages provide support for PostgreSQL 11 and add many bug fixes and improvements. References: https://github.com/pgmodeler/pgmodeler/blob/v0.9.2/CHANGELOG.md https://pgmodeler.io/blog/2019/12/26/pgmodeler-092-stable-is-here-for-my-pride-and-joy ======================== Updated packages in core/updates_testing: ======================== pgmodeler-0.9.2-2.mga7 pgmodeler-devel-0.9.2-2.mga7 from SRPMS: pgmodeler-0.9.2-2.mga7.src.rpm
Whiteboard: feedback => (none)
Installed new version and got exactly same resuts (and errors) as per Comment and above Leaving to Nicolas whether to OK this
I tried with pgmodeler-0.9.1-3.mga7 and got the same error with CLI. Regarding the popup error that appears when a dbm is opened outside pgmodeler, I found this upstream bug report: https://github.com/pgmodeler/pgmodeler/issues/1370. Maybe we can wait a bit so that, if I find an upstream commit solving that issue, I will add it.
Assignee: qa-bugs => nicolas.salguero
I added a patch that makes pgmodeler waiting 1 second before loading the model when someone opens a model outside the software. I also added an upstream patch for issue 1366 (when import function code replace some " with "e). Suggested advisory: ======================== The updated packages provide support for PostgreSQL 11 and add many bug fixes and improvements. References: https://github.com/pgmodeler/pgmodeler/blob/v0.9.2/CHANGELOG.md https://pgmodeler.io/blog/2019/12/26/pgmodeler-092-stable-is-here-for-my-pride-and-joy https://github.com/pgmodeler/pgmodeler/issues/1366 https://github.com/pgmodeler/pgmodeler/issues/1370 ======================== Updated packages in core/updates_testing: ======================== pgmodeler-0.9.2-3.mga7 pgmodeler-devel-0.9.2-3.mga7 from SRPMS: pgmodeler-0.9.2-3.mga7.src.rpm
Assignee: nicolas.salguero => qa-bugs
Improvement, but not completely Click on dbm file in Dolphin: model opens correctly, no fuss anymore. At CLI: $ pgmodeler kf5.kio.core: "Kon de map tags:/ niet binnengaan." (could nt enter folder tags:/) qt.qpa.xcb: QXcbConnection: XCB error: 3 (BadWindow), sequence: 1432, resource id: 33581707, major code: 40 (TranslateCoords), minor code: 0 but pgmodeler opens and from its dialogue, the dbm opens up OK. but ]$ pgmodeler mediawiki.dbm qt.qpa.xcb: QXcbConnection: XCB error: 3 (BadWindow), sequence: 956, resource id: 33581979, major code: 40 (TranslateCoords), minor code: 0 comes up wth the sameas previously "Could not load the database model file mediawiki.dbm. Check the error stack to see details. You can try to fix it in order to make it loadable again." I can then either clickon "Fix model" or "Cancel" the "Fix" aborts as before, and the pgmodeler opens empty, but there I can open the model with the Open-dailogue
At CLI, it is simpler to specify an absolute path. PgModeler always starts in "/usr/lib(64)/pgmodeler" so, if you want to use a relative path, the path has to begin there.
I can see that, but it is not a "usual way" of doing things at the CLI IMHO. If you find this satisfactory, I will nott object your OK.
I agree with you that this behaviour is not usual but it seems to be the behaviour of pgModeler since the beginning so I think it is time to OK.
Thank you for your efforts, Gentlemen. Since you seem to be in agreement, I'll go ahead and validate. Advisory in Comment 9.
Keywords: (none) => validated_updateCC: (none) => andrewsfarm, sysadmin-bugs
CC: (none) => tmbKeywords: (none) => advisory
An update for this issue has been pushed to the Mageia Updates repository. https://advisories.mageia.org/MGAA-2020-0023.html
Status: ASSIGNED => RESOLVEDResolution: (none) => FIXED