Make `t.timestamps` with precision by default. by kamipo · Pull Request #34970 · rails/rails

kamipo · 2019-01-18T01:18:50Z

I'm not sure it is what we want make it in Rails 6.0.
But I've just implemented to ease to discuss that on this PR.

rafaelfranca · 2019-01-18T13:35:08Z

Would this not cause warnings of unused arguments?

Currently Ruby doesn't raise any warnings of unused method arguments for now.

https://travis-ci.org/rails/rails/jobs/481187001

kaspth · 2019-01-18T20:54:29Z

Why do we need to do this?

Just keep db/schema.rb as before.

kaspth · 2019-01-18T20:55:50Z

Why do we need these similar overrides?

It is to inject V5_2::TableDefinition to the table definition instance.

https://github.com/rails/rails/pull/34970/files#diff-2a8be25f82da6b3935cc6a41300a1b01R19`

If the change_table is removed, the test_timestamps_doesnt_set_precision_on_change_table is failed.

https://github.com/rails/rails/pull/34970/files#diff-ec048630132cce280a9445022318906eR140

kaspth · 2019-01-18T20:57:45Z

Should this be supports_datetime_with_precision??

supports_datetime_with_precision? can't be used in the ActiveRecord::TestCase scope.
subsecond_precision_supported? is defined in the top level as a helper.

https://github.com/rails/rails/blob/5fcf8e98808c59bc1496a4048e133a7b32512765/activerecord/test/cases/helper.rb#L43-L45

kamipo · 2019-01-26T14:00:18Z

-    t.datetime "created_at", null: false
-    t.datetime "updated_at", null: false
+    t.datetime "created_at", precision: 6, null: false
+    t.datetime "updated_at", precision: 6, null: false


I've changed to apply new default the same with Action Mailbox.

Currently, if `:datetime` type has a precision, type casting always does round off subseconds regardless of whether necessary or not, it is a bit slow. Since rails#34970, `t.timestamps` with `precision: 6` by default, so `pluck(:created_at)` in newly created app will always be affected by the round off. This avoids the round off if possible, it makes `pluck(:created_at)` about 25% faster. https://gist.github.com/kamipo/e029539f2632aee9f5e711fe66fc8842 Before (0a87d7c with postgresql adapter): ``` Warming up -------------------------------------- users.pluck(:id) 344.000 i/100ms users.pluck(:name) 336.000 i/100ms users.pluck(:created_at) 206.000 i/100ms Calculating ------------------------------------- users.pluck(:id) 3.620k (± 8.5%) i/s - 18.232k in 5.077316s users.pluck(:name) 3.579k (± 9.4%) i/s - 17.808k in 5.020230s users.pluck(:created_at) 2.069k (± 8.0%) i/s - 10.300k in 5.019284s ``` Before (0a87d7c with mysql2 adapter): ``` Warming up -------------------------------------- users.pluck(:id) 245.000 i/100ms users.pluck(:name) 240.000 i/100ms users.pluck(:created_at) 180.000 i/100ms Calculating ------------------------------------- users.pluck(:id) 2.548k (± 9.4%) i/s - 12.740k in 5.066574s users.pluck(:name) 2.513k (± 8.0%) i/s - 12.480k in 5.011260s users.pluck(:created_at) 1.771k (±11.2%) i/s - 8.820k in 5.084473s ``` After (this change with postgresql adapter): ``` Warming up -------------------------------------- users.pluck(:id) 348.000 i/100ms users.pluck(:name) 357.000 i/100ms users.pluck(:created_at) 254.000 i/100ms Calculating ------------------------------------- users.pluck(:id) 3.628k (± 8.2%) i/s - 18.096k in 5.024748s users.pluck(:name) 3.624k (±12.4%) i/s - 17.850k in 5.020959s users.pluck(:created_at) 2.567k (± 7.0%) i/s - 12.954k in 5.081153s ``` After (this change with mysql2 adapter): ``` Warming up -------------------------------------- users.pluck(:id) 268.000 i/100ms users.pluck(:name) 265.000 i/100ms users.pluck(:created_at) 207.000 i/100ms Calculating ------------------------------------- users.pluck(:id) 2.586k (±10.9%) i/s - 12.864k in 5.050546s users.pluck(:name) 2.543k (±10.2%) i/s - 12.720k in 5.067726s users.pluck(:created_at) 2.263k (±14.5%) i/s - 10.971k in 5.004039s ```

Make `t.timestamps` with precision by default rails/rails#34970

andrewhaines · 2020-02-07T17:27:28Z

@kamipo Do you think there's any benefit to documenting that this is incompatible w/ MySQL 5.5? I ran across that issue the other day (the shared ClearDB environments for Heroku don't allow higher versions than 5.5) and it took me a while to figure out what was happening. If so, I'm happy to contribute it.

kamipo · 2020-02-08T06:51:08Z

What is the incompatible you mean?

andrewhaines · 2020-02-08T11:43:45Z

datetime doesn't support precision in MySQL 5.5.

jcoleman · 2023-02-15T21:06:15Z

@kamipo I'm still interested in what the reasoning for this was. Following the links to the discussion it appears this came out of a discussion about MySQL, but this change affected not just MySQL but also databases like Postgres that don't need this change. And I can't find any discussion at all explaining why the change is desirable or was made in the first place.

jcoleman · 2023-02-15T21:07:39Z

In fact the comment actually says this should be for MySQL only and be omitted on PostgreSQL, but that's not what this PR ending up doing.

zzak · 2023-02-16T02:06:29Z

@jcoleman Unfortunately, you might not be able to get a response on a 3 or 4 year old PR. If there is a bug that you've identified, I recommend opening a new issue, or submitting a PR which applies your desired behavior. 🙏

jcoleman · 2023-02-16T02:35:56Z

@zzak Unfortunately getting responses on even new PRs in the Rails project is...not common.

Another issue was opened the same year as this PR asking for an explanation of the change, and no one answered the question there either.

The `timestamps` tests for #34970 and a939506 are mostly overlapped, so consolidate these tests as "test timestamps with implicit defalut".

The `timestamps` tests for rails#34970 and a939506 are mostly overlapped, so consolidate these tests as "test timestamps with implicit defalut".

Seven Migration::Compatibility version classes carried an identical private compatible_table_definition override whose only job was to prepend the class's own TableDefinition module. Compose the modules in a single walk in Migration::Current#compatible_table_definition instead: it prepends the TableDefinition module of every Compatibility version class among the migration's ancestors, oldest version first, which produces the same method lookup order as the per-class overrides. The walk filters the ancestors against the new memoized `Compatibility.version_classes`. A version class now only declares its TableDefinition module. The seven overrides accumulated one version class at a time: * 214dc7d - V5_0 and V4_2 * 57015cd (rails#34970) - V5_2 * e8437a6 (rails#38717) - V6_0 * 867d27d (rails#41084) - V6_1 * da472b6 (rails#43904) - V7_0 * 489e08d (rails#55086) - V8_0 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Seven Migration::Compatibility version classes carried an identical compatible_table_definition override just to prepend their own TableDefinition module. Replace the overrides with a single ancestor walk in Migration::Current, preserving the method lookup order; a version class now only declares its module. The seven overrides accumulated one version class at a time: * 214dc7d - V5_0 and V4_2 * 57015cd (rails#34970) - V5_2 * e8437a6 (rails#38717) - V6_0 * 867d27d (rails#41084) - V6_1 * da472b6 (rails#43904) - V7_0 * 489e08d (rails#55086) - V8_0 Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

Seven Migration::Compatibility version classes carried an identical compatible_table_definition override just to prepend their own TableDefinition module. Replace the overrides with a single ancestor walk in Migration::Current, preserving the method lookup order; a version class now only declares its module. The seven overrides accumulated one version class at a time: * 214dc7d - V5_0 and V4_2 * 57015cd (rails#34970) - V5_2 * e8437a6 (rails#38717) - V6_0 * 867d27d (rails#41084) - V6_1 * da472b6 (rails#43904) - V7_0 * 489e08d (rails#55086) - V8_0 Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

rails-bot Bot added actionmailbox actiontext activerecord labels Jan 18, 2019

kamipo mentioned this pull request Jan 18, 2019

Allow using Action Mailbox on MySQL 5.5 #34956

Merged

kamipo force-pushed the timestamps_with_precision_by_default branch 2 times, most recently from fd2746b to 5fcf8e9 Compare January 18, 2019 03:39

rafaelfranca reviewed Jan 18, 2019

View reviewed changes

rafaelfranca approved these changes Jan 18, 2019

View reviewed changes

kaspth reviewed Jan 18, 2019

View reviewed changes

kaspth approved these changes Jan 24, 2019

View reviewed changes

kamipo force-pushed the timestamps_with_precision_by_default branch from 5fcf8e9 to ec47e62 Compare January 26, 2019 13:13

Make t.timestamps with precision by default

57015cd

kamipo force-pushed the timestamps_with_precision_by_default branch from ec47e62 to 57015cd Compare January 26, 2019 13:51

kamipo commented Jan 26, 2019

View reviewed changes

kamipo merged commit 5a8f0c7 into rails:master Jan 26, 2019

kamipo deleted the timestamps_with_precision_by_default branch January 26, 2019 14:37

kamipo mentioned this pull request May 28, 2019

Don't round off subseconds unless necessary #36352

Merged

suketa added a commit to suketa/rails_sandbox that referenced this pull request Jul 26, 2019

try065

35da934

Make `t.timestamps` with precision by default rails/rails#34970

suketa added a commit to suketa/rails_sandbox that referenced this pull request Jul 26, 2019

try065

971f3dc

Make `t.timestamps` with precision by default rails/rails#34970

6temes mentioned this pull request Aug 24, 2019

Rails 6 migrations: t.timestamps adds precision: 6 #37031

Closed

pezholio mentioned this pull request Dec 18, 2019

Import users from DfE Signin API DFE-Digital/claim-additional-payments-for-teaching#716

Merged

roman-melnyk mentioned this pull request Mar 9, 2020

update rails dependency to include 6.0 braintree/pg_ha_migrations#33

Merged

isc mentioned this pull request Apr 1, 2020

Recognize timestamps with precision as timestamps jeremyevans/sequel#1684

Closed

yskkin mentioned this pull request Jun 4, 2020

Docs: add_column(.. limit: ) re: datetime #31864

Closed

kamipo mentioned this pull request May 3, 2021

Active Record + PostgreSQL: native support for timestamp with time zone #41084

Merged

ttanimichi mentioned this pull request Sep 9, 2021

ActiveRecord where clause ignores the data of the last one sec of the range #43190

Closed

sambostock mentioned this pull request Sep 22, 2022

Fix erroneous nil default precision on virtual datetime columns #46110

Merged

7 of 9 tasks

kamipo added a commit that referenced this pull request Sep 10, 2024

Consolidate timestamps tests for #34970 and a939506

143e58c

The `timestamps` tests for #34970 and a939506 are mostly overlapped, so consolidate these tests as "test timestamps with implicit defalut".

yahonda mentioned this pull request Sep 22, 2026

Compose compatibility TableDefinition modules in Migration::Current #58847

Merged

2 of 4 tasks

Conversation

kamipo commented Jan 18, 2019

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

andrewhaines commented Feb 7, 2020

Uh oh!

kamipo commented Feb 8, 2020

Uh oh!

andrewhaines commented Feb 8, 2020

Uh oh!

jcoleman commented Feb 15, 2023

Uh oh!

jcoleman commented Feb 15, 2023

Uh oh!

zzak commented Feb 16, 2023

Uh oh!

jcoleman commented Feb 16, 2023

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants