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.

Sign up for GitHub

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

Jump to bottom

User specific schema configurations #201

Open
wants to merge 253 commits into
base: feature/schema-tools
Choose a base branch
from
Open

User specific schema configurations #201

wants to merge 253 commits into from

Conversation

raheelriax
Copy link

Added Customized schema configurations for both
Transaction and version tables.
You can now pass schema name as an option to make_versioned method or define it explicitly in versioned attribute of parent model.
If no configuration is passed it will use default configurations

kvesteri added 30 commits March 26, 2014 12:56
@kvesteri
Make tests compatible with older versions of sqlite
8314e78
@kvesteri
Skip mod column creation for assoc tables
b29f090
@kvesteri
Update changes
b0b3bd3
@kvesteri
Fix get_bind schematics, fixes kvesteri#48
95ca466
@kvesteri
Update SA dependency
637d2a2
@kvesteri
Add remove_versioning func, fixes kvesteri#45
896d600
@kvesteri
Update changes
91b7d45
@kvesteri
Fix example
4ba3eaa
@kvesteri
Update docs
3b16e59
@kvesteri
Make versions order by deterministic
b10eb83
@kvesteri
Fix test
f3ac0f3
@kvesteri
Various fixes for inheritance reflection
29a67c3
@kvesteri
Added support for order_by and column_prefix mapper args
3fd0e0a
@kvesteri
Add tests for concrete inheritance
95da74d
@kvesteri
Add tests for mapper args
bcb88ed
@kvesteri
Fix tx id condition
992cdfb
@kvesteri
Refactor fetcher
63f22c4
@kvesteri
Refactor fetcher
291e4c2
@kvesteri
Remove futile import
8c9adee
@kvesteri
Refactor fetcher and unit of work classes
a9e4ba8
@kvesteri
Bump version
03886ba
@kvesteri
Fixed many to many support for engine bind
9fa2208
@kvesteri
Update SA-Utils dependency
e833a37
@kvesteri
Fix primary key alias handling
ddf216d
@kvesteri
Add skipped test case
403451c
@kvesteri
Fix many-to-many issue with custom tx column
1949dd0
@kvesteri
Bump version
6088e2b
@kvesteri
Update requirements
bd42aa7
@kvesteri
Added order_by arg ignore
b9e0097
@kvesteri
Fix code block
649442c
Fernando Cezar and others added 16 commits May 16, 2018 16:47
track cloned connections
acef01b 
Hotfix/track cloned connections (kvesteri#2)

Hotfix/track cloned connections (kvesteri#3)

change iterations

use only ConnectionFairy as indexes

create new dict entry for clones

create entry using old connection

create entry using session

revert to event listening strategy

remove cloned connections on rollback

treat None case when cleaning connections

Hotfix/track cloned connections (kvesteri#2)

Hotfix/track cloned connections (kvesteri#3)

change iterations

use only ConnectionFairy as indexes

create new dict entry for clones

create entry using old connection

create entry using session

revert to event listening strategy

remove cloned connections on rollback

treat None case when cleaning connections
@kvesteri
Merge pull request kvesteri#167 from netcriptus/hotfix/track-cloned-c…
8658a89 
…onnections

track cloned connections
@kvesteri
Bump version
d4ed900
@fuhrysteve
Reraise original exception instead of a new one
79b9876 
Raising a new KeyError means you lose the original stacktrace, which
is going to be more useful since it'll contain more information about
the relevant context for this error.
@kvesteri
Merge pull request kvesteri#189 from fuhrysteve/reraise_original_exce…
5a3fbcf 
…ption

Reraise original exception instead of a new one
@vault
Test for leaked connections on external rollback
8496cdd
@vault
Check connection.closed before connection.connection
3a5b967
@vault
Invalidate units_of_work on Engine rollback event
4ded9b3
@kvesteri
Merge pull request kvesteri#196 from vault/fix-connections-leaking
345be03 
Fix ResourceClosedErrors from connections leaking when using an external transaction
@kvesteri
Bump version
cedad54
@raheelriax
Added Custom schema_name options for version tables, adds more flexib…
88d1b25 
…ility in handling table name
@raheelriax
Added Transaction table options
6ecf476
@raheelriax
Merge branch 'develop' into hotfix/version-table-fix
4dce399
@raheelriax
Fixed version_table_schema_issue
a0d8417
@raheelriax
Merge branch 'hotfix/version-table-fix' into develop
5b73670
@raheelriax
Added Universal options for version tables schema
1a06240
@chris-mouthwatch
Copy link

@raheelriax is there any estimate on when this PR could be merged?

@chris-mouthwatch
Copy link

pinging @kvesteri on this.

@marksteward
Copy link
Collaborator

This looks like it needs some serious rebasing.

@chris-mouthwatch
Copy link

@marksteward @kvesteri do you have any suggestions on how to generate transaction_id in a custom way rather than using seq?
We are using schema in postgresql and I think:

  • for long-term, this PR will resolve the issue.
  • for short-term, maybe using UUID or something like that could be used instead of transaction_id_seq.

What do you think?

@marksteward
Copy link
Collaborator

What aspect of this PR will fix that? I'm happy to look at a smaller PR.

I don't think UUID makes sense as the primary key for Transaction, you'd still need some way of sequencing them that's stable and unambiguously ordered (i.e. not date).

Why can't you use a sequence currently? I think we might be able to remove the explicit Sequence in recent versions of SQLAlchemy if that helps.

@chris-mouthwatch
Copy link

@marksteward
So the problem is that we have applied versioning to the tables in communication schema in postgres.
The problem is that it seems like it's trying to use public.transaction_id_seq instead of communication.transaction_id_seq and that's why we are seeing this error.
image
image

@chris-mouthwatch
Copy link

So if we can append schema name i.e. communication.transaction_id_seq then it would resolve the issue.
What do you think, @marksteward ?

@chris-mouthwatch
Copy link

Actually, adding this in the back resolved the issue.
image

@marksteward
Copy link
Collaborator

I think we might just need to reference Base.metadata in the Sequence declaration. I'll give that a go this evening.

@chris-mouthwatch
Copy link

Great, thanks.

@jun-828
Copy link

jun-828 commented Aug 10, 2022

Thanks, guys. @marksteward I think, one of these two will work:

  • Reference the schema information in Base.metadata as you mentioned.
  • If first option will take too much effort, then we can just remove nextVal(transaction_seq_id) when we run INSERT INTO transactions as it is already autoincrement int type.

In anyway, I can't wait to see this fix coming. Thanks for all your supports btw.

@chris-mouthwatch
Copy link

@marksteward any chance that you could take a look on this?

@chris-mouthwatch
Copy link

@marksteward did you have a chance to look into this one?
You can just give me a hint and I can try

@marksteward
Copy link
Collaborator

I've not forgotten about this but ran out of time today.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.