Improper matching performed in RelationshipPairingVisitor's PairOneToManys #313
Conversation
|
Thanks, I'll review it shortly |
|
Just checking in. Is it possible to know what time frame this bug fix (assuming approved) is likely to merged? Thank you. |
|
Sorry for such long delay, I'll merge it in this weekend. About new NuGet 2015-05-29 11:29 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
I apologise for being a bother (I'm British I have to apologise). I realise maintenance of fluent-nhibernate is likely a side task/project and other tasks are likely taking priority. But I urge you to quickly take some time to look at this issue. In our deployment this mismatch is the cause of several one to many relationships being mistakenly made cascade save-update. Thank you for your time. |
|
It's me who should apologize - I screwed up again. Fix will be merged & 2015-06-10 11:58 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
| @@ -72,7 +72,8 @@ namespace :source do | |||
| info.product_name = 'FluentNHibernate' | |||
| info.description = commit_hash[0..(commit_hash.length - 3)] | |||
| info.copyright = "Copyright 2008-#{Time.new.year} James Gregory and contributors (Paul Batum, Hudson Akridge et al). All rights reserved." | |||
| info.namespaces = ['System.Security'] | |||
| info.namespaces = ['System.Security','System.Runtime.CompilerServices'] | |||
| info.custom_attributes :InternalsVisibleTo => "FluentNHibernate.Testing" | |||
There was a problem hiding this comment.
Why InternalsVisibleTo?
There was a problem hiding this comment.
Oh, I see - because you used PropertyMember
|
Any updates on this one @Ignition ? |
|
I didn't have email notifications turned on, only just seen this. Is there something particular you needed me to resolve? The abstract class was just copying the style from other tests that exist in the solution, I can change that if you like. Plus I have noticed some misnamed variables that may need changing. I don't think I can do anything about If you want me to do any changes to the supporting test then let me know and I'll get that sorted ASAP (I've enabled notifications now). |
|
Oh, I see. 2015-07-10 19:49 GMT+04:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
I've committed those changes. |
|
Ok, thanks. I'll merge this and try to release new versions this week. 2015-07-13 17:31 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
Hi, hope all is well. Just a friendly checkup, hope you haven't forgot about me and my pull request :) |
|
Yep, I merged it in separate release branches on my fork, NuGets will be 2015-08-17 19:00 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
Another friendly nudge :) |
|
Sorry, I screwed up again. 2015-09-23 14:02 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
Bugging you |
|
Will finally have some time tonight. 2015-10-01 11:46 GMT+03:00 Gareth Andrew Lloyd notifications@github.com:
Yours faithfully, |
|
Just pushed the packages (both .NET 3.5 and 4.0 versions), release 2015-10-05 12:56 GMT+03:00 Gleb Chermennov thebitterend77@gmail.com:
Yours faithfully, |
An entity (A) with a one to many collection pointing to an entity (B) that has the corresponding many to one reference back (to A) may in some cases be pared incorrectly.
If another entity (C) has a many to one reference back to the original entity (A), but (A) has no collection for (C) then when attempting to pair (A->B) we can actually get (A->C).
This happens when (C)s property to (A) has a name which alphabetically before (B)s property to (A).
I have provided a fix and a test, the test may need modification as it requires
PropertyMemberand hence the addedInternalsVisibleToin the RakeFile.If the fix is approved, I am in need of the fix to be cherry-picked back into 1.4, thank you.