Set `precision` of `sqlalchemy.Float` in `RDBStorage` table definition by toshihikoyanase · Pull Request #3327 · optuna/optuna

toshihikoyanase · 2022-02-22T10:57:28Z

Motivation

Related to #3227.
I found pandas set the precision option of sqlalchemy.Float to 53 in .https://github.com/pandas-dev/pandas/blob/af8ad6d8b0db8ea98dab98ae77a48092ba3832f8/pandas/io/sql.py#L1149-L1186. This change will change the column type of the trial value from float to double when I confirmed the table definition in MySQL 5.8.

I guess the value 53 comes from IEEE 754, but I'm not sure.

Current master

CREATE TABLE `trial_values` (
  `trial_value_id` int(11) NOT NULL AUTO_INCREMENT,
  `trial_id` int(11) NOT NULL,
  `objective` int(11) NOT NULL,
  `value` float NOT NULL,
  PRIMARY KEY (`trial_value_id`),
  UNIQUE KEY `trial_id` (`trial_id`,`objective`),
  CONSTRAINT `trial_values_ibfk_1` FOREIGN KEY (`trial_id`) REFERENCES `trials` (`trial_id`)
) ENGINE=InnoDB AUTO_INCREMENT=2 DEFAULT CHARSET=latin1;

This PR

CREATE TABLE `trial_values` (
  `trial_value_id` int(11) NOT NULL AUTO_INCREMENT,
  `trial_id` int(11) NOT NULL,
  `objective` int(11) NOT NULL,
  `value` double NOT NULL,
  PRIMARY KEY (`trial_value_id`),
  UNIQUE KEY `trial_id` (`trial_id`,`objective`),
  CONSTRAINT `trial_values_ibfk_1` FOREIGN KEY (`trial_id`) REFERENCES `trials` (`trial_id`)
) ENGINE=InnoDB AUTO_INCREMENT=2 DEFAULT CHARSET=latin1;

Description of the changes

Set precision=53 to sqlalchemy.Float
- intermediate_value of trial_intermediate_values table
- param_value of trial_params table
- value of trial_values table
Change nullable of intermediate_value of trial_intermediate_values table
- c.f., Accept nan in trial.report #3348

Data types of the other RDB

SQLite3

REAL. The value is a floating point value, stored as an 8-byte IEEE floating point number.
https://www.sqlite.org/datatype3.html

PostgreSQL

PostgreSQL also supports the SQL-standard notations float and float(p) for specifying inexact numeric types. Here, p specifies the minimum acceptable precision in binary digits. PostgreSQL accepts float(1) to float(24) as selecting the real type, while float(25) to float(53) select double precision. Values of p outside the allowed range draw an error. float with no precision specified is taken to mean double precision.
https://www.postgresql.org/docs/current/datatype-numeric.html

TODO

Add migration script
Check the behavior of the other database (i.e., SQLite3 and PostgreSQL.)
Resolve conflict

github-actions · 2022-03-08T23:05:21Z

This pull request has not seen any recent activity.

toshihikoyanase · 2022-03-10T05:08:35Z

I'm still working on this PR.

codecov-commenter · 2022-03-10T13:49:45Z

Codecov Report

Merging #3327 (f29ef89) into master (c245717) will decrease coverage by 0.01%.
The diff coverage is 76.92%.

@@            Coverage Diff             @@
##           master    #3327      +/-   ##
==========================================
- Coverage   91.48%   91.47%   -0.02%     
==========================================
  Files         158      159       +1     
  Lines       12241    12270      +29     
==========================================
+ Hits        11199    11224      +25     
- Misses       1042     1046       +4

Impacted Files	Coverage Δ
optuna/storages/_rdb/alembic/versions/v3.0.0.b.py	`71.42% <71.42%> (ø)`
optuna/storages/_rdb/alembic/env.py	`75.00% <100.00%> (ø)`
optuna/storages/_rdb/models.py	`98.79% <100.00%> (+<0.01%)`	⬆️
optuna/study/study.py	`96.19% <0.00%> (ø)`
optuna/trial/_trial.py	`100.00% <0.00%> (ø)`
optuna/samplers/_qmc.py	`95.09% <0.00%> (ø)`
optuna/trial/_frozen.py	`97.91% <0.00%> (ø)`
optuna/samplers/_grid.py	`100.00% <0.00%> (ø)`
optuna/samplers/_cmaes.py	`98.97% <0.00%> (ø)`
optuna/importance/_base.py	`96.36% <0.00%> (ø)`
... and 25 more

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

github-actions · 2022-03-28T23:05:21Z

This pull request has not seen any recent activity.

himkt · 2022-04-12T05:14:48Z

[memo]

[at]c-bata suggested to use filter warning in a migration script, which avoids to show deprecation messages to users.
(It had to do in my PR as well. 🙇)

toshihikoyanase · 2022-04-14T07:26:14Z

#3348 requires migration script and it will be conflict with this PR. I'll restart this PR after #3348 is merged or closed.

toshihikoyanase · 2022-04-27T07:58:05Z

optuna/storages/_rdb/alembic/env.py

 from sqlalchemy import pool

-import optuna
+import optuna.storages._rdb.models


This change was required to avoid the following import error.

$ alembic revision --autogenerate --rev-id v3.0.0.b Traceback (most recent call last): File "/Users/yanase/pfn/code/optuna/.venv/bin/alembic", line 8, in <module> sys.exit(main()) File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/config.py", line 588, in main CommandLine(prog=prog).main(argv=argv) File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/config.py", line 582, in main self.run_cmd(cfg, options) File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/config.py", line 559, in run_cmd fn( File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/command.py", line 227, in revision script_directory.run_env() File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/script/base.py", line 563, in run_env util.load_python_file(self.dir, "env.py") File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/util/pyfiles.py", line 92, in load_python_file module = load_module_py(module_id, path) File "/Users/yanase/pfn/code/optuna/.venv/lib/python3.9/site-packages/alembic/util/pyfiles.py", line 108, in load_module_py spec.loader.exec_module(module) # type: ignore File "<frozen importlib._bootstrap_external>", line 850, in exec_module File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed File "alembic/env.py", line 25, in <module> target_metadata = optuna.storages._rdb.models.BaseModel.metadata AttributeError: module 'optuna.storages._rdb' has no attribute 'models'

toshihikoyanase · 2022-04-27T08:00:26Z

Resolved the conflict
Changed nullable of intermediate values column from False to True (c.f., Accept nan in trial.report #3348)
Update import line of alembic/env.py

toshihikoyanase · 2022-04-28T07:53:29Z

I confirmed the change by using @himkt 's optuna-2e. The corresponding branch is this one.

To dump DB data, please use the following commands:

# SQLite3
echo ".dump" | sqlite3 data/sample.db

# MySQL
docker compose run --rm mysql bash
mysqldump -u root -h mysql -B optuna -p

# PostgreSQL
docker compose run --rm postgresql bash
pg_dump -U root -h postgresql optuna > pg.sql

toshihikoyanase · 2022-04-28T07:54:07Z

@c-bata @contramundum53 Could you review this PR, please?

xadrianzetx · 2022-05-07T13:43:21Z

I guess this means we should relax following limits to actually store double precision values.

optuna/optuna/storages/_rdb/storage.py

Lines 59 to 60 in f00abac

    
           _RDB_MAX_FLOAT = np.finfo(np.float32).max 
        
           _RDB_MIN_FLOAT = np.finfo(np.float32).min

Ref #3238 (comment)

c-bata · 2022-05-08T14:07:59Z

@xadrianzetx Yes, you are correct 👍 Actually I'm planning to fix the problem you pointed.
https://github.com/optuna/optuna/compare/master...c-bata:rdb-inf-handling?expand=1

xadrianzetx · 2022-05-08T14:23:03Z

Yeah, that's what I had in mind. My first solution was a bit too hacky.

c-bata

Thank you for your pull request. LGTM with nits!

It might be a bit controversial that this PR also contains a schema migration for #3348. I think it's acceptable since those schema changes are executed in one query like the following.

SQL queries of forward migration

2022-05-10T05:12:41.383794Z	   26 Query	ALTER TABLE trial_intermediate_values MODIFY intermediate_value FLOAT(53) NULL
2022-05-10T05:12:41.456748Z	   26 Query	ALTER TABLE trial_params MODIFY param_value FLOAT(53) NULL
2022-05-10T05:12:41.532337Z	   26 Query	ALTER TABLE trial_values MODIFY value FLOAT(53) NOT NULL
2022-05-10T05:12:41.605596Z	   26 Query	UPDATE alembic_version SET version_num='v3.0.0.b' WHERE alembic_version.version_num = 'v3.0.0.a'
2022-05-10T05:12:41.608029Z	   26 Query	COMMIT

SQL queries of backward migration

2022-05-10T05:11:38.376333Z	   25 Query	ALTER TABLE trial_intermediate_values MODIFY intermediate_value FLOAT NOT NULL
2022-05-10T05:11:38.466837Z	   25 Query	ALTER TABLE trial_params MODIFY param_value FLOAT NULL
2022-05-10T05:11:38.539438Z	   25 Query	ALTER TABLE trial_values MODIFY value FLOAT NOT NULL
2022-05-10T05:11:38.613731Z	   25 Query	UPDATE alembic_version SET version_num='v3.0.0.a' WHERE alembic_version.version_num = 'v3.0.0.b'
2022-05-10T05:11:38.616350Z	   25 Query	COMMIT

Here is a note about how table schemas are changed on SQLite3, MySQL, and PostgreSQL.

SQLite3

In SQLite3, the only data type that stores floating point numbers is REAL. REAL is a double precision floating point number, so the precision does not change before or after this PR.

Before running migration

sqlite> .schema trial_intermediate_values
CREATE TABLE trial_intermediate_values (
	trial_intermediate_value_id INTEGER NOT NULL,
	trial_id INTEGER NOT NULL,
	step INTEGER NOT NULL,
	intermediate_value FLOAT NOT NULL,
	PRIMARY KEY (trial_intermediate_value_id),
	UNIQUE (trial_id, step),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id)
);
sqlite> .schema trial_values
CREATE TABLE trial_values (
	trial_value_id INTEGER NOT NULL,
	trial_id INTEGER NOT NULL,
	objective INTEGER NOT NULL,
	value FLOAT NOT NULL,
	PRIMARY KEY (trial_value_id),
	UNIQUE (trial_id, objective),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id)
);
sqlite> .schema trial_params
CREATE TABLE trial_params (
	param_id INTEGER NOT NULL,
	trial_id INTEGER,
	param_name VARCHAR(512),
	param_value FLOAT,
	distribution_json TEXT,
	PRIMARY KEY (param_id),
	UNIQUE (trial_id, param_name),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id)
);

After running migration

sqlite> .schema trial_intermediate_values
CREATE TABLE IF NOT EXISTS "trial_intermediate_values" (
	trial_intermediate_value_id INTEGER NOT NULL,
	trial_id INTEGER NOT NULL,
	step INTEGER NOT NULL,
	intermediate_value FLOAT,
	PRIMARY KEY (trial_intermediate_value_id),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id),
	UNIQUE (trial_id, step)
);
sqlite> .schema trial_values
CREATE TABLE IF NOT EXISTS "trial_values" (
	trial_value_id INTEGER NOT NULL,
	trial_id INTEGER NOT NULL,
	objective INTEGER NOT NULL,
	value FLOAT NOT NULL,
	PRIMARY KEY (trial_value_id),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id),
	UNIQUE (trial_id, objective)
);
sqlite> .schema trial_params
CREATE TABLE IF NOT EXISTS "trial_params" (
	param_id INTEGER NOT NULL,
	trial_id INTEGER,
	param_name VARCHAR(512),
	param_value FLOAT,
	distribution_json TEXT,
	PRIMARY KEY (param_id),
	FOREIGN KEY(trial_id) REFERENCES trials (trial_id),
	UNIQUE (trial_id, param_name)
);

PostgreSQL

In PostgreSQL, real is a 4 bytes floating point data type and double precision is a 8 bytes floating point data type. SQLAlchemy's Float field actually uses double precision data type by default. so the precision does not change before or after this PR.

Before running migration

optuna=# \d trial_params
                                            Table "public.trial_params"
      Column       |          Type          | Collation | Nullable |                    Default
-------------------+------------------------+-----------+----------+------------------------------------------------
 param_id          | integer                |           | not null | nextval('trial_params_param_id_seq'::regclass)
 trial_id          | integer                |           |          |
 param_name        | character varying(512) |           |          |
 param_value       | double precision       |           |          |
 distribution_json | text                   |           |          |
Indexes:
    "trial_params_pkey" PRIMARY KEY, btree (param_id)
    "trial_params_trial_id_param_name_key" UNIQUE CONSTRAINT, btree (trial_id, param_name)
Foreign-key constraints:
    "trial_params_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id)

optuna=# \d trial_values
                                           Table "public.trial_values"
     Column     |       Type       | Collation | Nullable |                       Default
----------------+------------------+-----------+----------+------------------------------------------------------
 trial_value_id | integer          |           | not null | nextval('trial_values_trial_value_id_seq'::regclass)
 trial_id       | integer          |           | not null |
 objective      | integer          |           | not null |
 value          | double precision |           | not null |
Indexes:
    "trial_values_pkey" PRIMARY KEY, btree (trial_value_id)
    "trial_values_trial_id_objective_key" UNIQUE CONSTRAINT, btree (trial_id, objective)
Foreign-key constraints:
    "trial_values_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id)

optuna=# \d trial_intermediate_values
                                                        Table "public.trial_intermediate_values"
           Column            |       Type       | Collation | Nullable |                                    Default
-----------------------------+------------------+-----------+----------+--------------------------------------------------------------------------------
 trial_intermediate_value_id | integer          |           | not null | nextval('trial_intermediate_values_trial_intermediate_value_id_seq'::regclass)
 trial_id                    | integer          |           | not null |
 step                        | integer          |           | not null |
 intermediate_value          | double precision |           | not null |
Indexes:
    "trial_intermediate_values_pkey" PRIMARY KEY, btree (trial_intermediate_value_id)
    "trial_intermediate_values_trial_id_step_key" UNIQUE CONSTRAINT, btree (trial_id, step)
Foreign-key constraints:
    "trial_intermediate_values_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id

After running migration

optuna=# \d trial_params
                                            Table "public.trial_params"
      Column       |          Type          | Collation | Nullable |                    Default
-------------------+------------------------+-----------+----------+------------------------------------------------
 param_id          | integer                |           | not null | nextval('trial_params_param_id_seq'::regclass)
 trial_id          | integer                |           |          |
 param_name        | character varying(512) |           |          |
 param_value       | double precision       |           |          |
 distribution_json | text                   |           |          |
Indexes:
    "trial_params_pkey" PRIMARY KEY, btree (param_id)
    "trial_params_trial_id_param_name_key" UNIQUE CONSTRAINT, btree (trial_id, param_name)
Foreign-key constraints:
    "trial_params_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id)

optuna=# \d trial_intermediate_values
                                                        Table "public.trial_intermediate_values"
           Column            |       Type       | Collation | Nullable |                                    Default
-----------------------------+------------------+-----------+----------+--------------------------------------------------------------------------------
 trial_intermediate_value_id | integer          |           | not null | nextval('trial_intermediate_values_trial_intermediate_value_id_seq'::regclass)
 trial_id                    | integer          |           | not null |
 step                        | integer          |           | not null |
 intermediate_value          | double precision |           |          |
Indexes:
    "trial_intermediate_values_pkey" PRIMARY KEY, btree (trial_intermediate_value_id)
    "trial_intermediate_values_trial_id_step_key" UNIQUE CONSTRAINT, btree (trial_id, step)
Foreign-key constraints:
    "trial_intermediate_values_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id)

optuna=# \d trial_values
                                           Table "public.trial_values"
     Column     |       Type       | Collation | Nullable |                       Default
----------------+------------------+-----------+----------+------------------------------------------------------
 trial_value_id | integer          |           | not null | nextval('trial_values_trial_value_id_seq'::regclass)
 trial_id       | integer          |           | not null |
 objective      | integer          |           | not null |
 value          | double precision |           | not null |
Indexes:
    "trial_values_pkey" PRIMARY KEY, btree (trial_value_id)
    "trial_values_trial_id_objective_key" UNIQUE CONSTRAINT, btree (trial_id, objective)
Foreign-key constraints:
    "trial_values_trial_id_fkey" FOREIGN KEY (trial_id) REFERENCES trials(trial_id)

MySQL

In MySQL, we can specify precision of floating point number like FLOAT(p). According to a MySQL documentation, a precision from 24 to 53 results in an 8-byte double-precision DOUBLE column. 53 seems to be reasonable precision number.

I confirmed that the precision of each columns are changed from FLOAT to DOUBLE. Please see the following schemas for details.

Before running migration

mysql> desc trial_intermediate_values;
+-----------------------------+-------+------+-----+---------+----------------+
| Field                       | Type  | Null | Key | Default | Extra          |
+-----------------------------+-------+------+-----+---------+----------------+
| trial_intermediate_value_id | int   | NO   | PRI | NULL    | auto_increment |
| trial_id                    | int   | NO   | MUL | NULL    |                |
| step                        | int   | NO   |     | NULL    |                |
| intermediate_value          | float | NO   |     | NULL    |                |
+-----------------------------+-------+------+-----+---------+----------------+
4 rows in set (0.04 sec)

mysql> desc trial_values;
+----------------+-------+------+-----+---------+----------------+
| Field          | Type  | Null | Key | Default | Extra          |
+----------------+-------+------+-----+---------+----------------+
| trial_value_id | int   | NO   | PRI | NULL    | auto_increment |
| trial_id       | int   | NO   | MUL | NULL    |                |
| objective      | int   | NO   |     | NULL    |                |
| value          | float | NO   |     | NULL    |                |
+----------------+-------+------+-----+---------+----------------+
4 rows in set (0.00 sec)

mysql> desc trial_params;
+-------------------+--------------+------+-----+---------+----------------+
| Field             | Type         | Null | Key | Default | Extra          |
+-------------------+--------------+------+-----+---------+----------------+
| param_id          | int          | NO   | PRI | NULL    | auto_increment |
| trial_id          | int          | YES  | MUL | NULL    |                |
| param_name        | varchar(512) | YES  |     | NULL    |                |
| param_value       | float        | YES  |     | NULL    |                |
| distribution_json | text         | YES  |     | NULL    |                |
+-------------------+--------------+------+-----+---------+----------------+
5 rows in set (0.03 sec)

After running migration

mysql> desc trial_intermediate_values;
+-----------------------------+--------+------+-----+---------+----------------+
| Field                       | Type   | Null | Key | Default | Extra          |
+-----------------------------+--------+------+-----+---------+----------------+
| trial_intermediate_value_id | int    | NO   | PRI | NULL    | auto_increment |
| trial_id                    | int    | NO   | MUL | NULL    |                |
| step                        | int    | NO   |     | NULL    |                |
| intermediate_value          | double | YES  |     | NULL    |                |
+-----------------------------+--------+------+-----+---------+----------------+
4 rows in set (0.01 sec)

mysql> desc trial_values;
+----------------+--------+------+-----+---------+----------------+
| Field          | Type   | Null | Key | Default | Extra          |
+----------------+--------+------+-----+---------+----------------+
| trial_value_id | int    | NO   | PRI | NULL    | auto_increment |
| trial_id       | int    | NO   | MUL | NULL    |                |
| objective      | int    | NO   |     | NULL    |                |
| value          | double | NO   |     | NULL    |                |
+----------------+--------+------+-----+---------+----------------+
4 rows in set (0.00 sec)

mysql> desc trial_params;
+-------------------+--------------+------+-----+---------+----------------+
| Field             | Type         | Null | Key | Default | Extra          |
+-------------------+--------------+------+-----+---------+----------------+
| param_id          | int          | NO   | PRI | NULL    | auto_increment |
| trial_id          | int          | YES  | MUL | NULL    |                |
| param_name        | varchar(512) | YES  |     | NULL    |                |
| param_value       | double       | YES  |     | NULL    |                |
| distribution_json | text         | YES  |     | NULL    |                |
+-------------------+--------------+------+-----+---------+----------------+
5 rows in set (0.01 sec)

optuna/storages/_rdb/alembic/versions/v3.0.0.b.py

Co-authored-by: Masashi Shibata <c-bata@users.noreply.github.com>

contramundum53 · 2022-05-10T09:15:18Z

optuna/storages/_rdb/alembic/versions/v3.0.0.b.py

+    with op.batch_alter_table("trial_params") as batch_op:
+        batch_op.alter_column(
+            "param_value",
+            type_=Float(precision=FLOAT_PRECISION),
+            existing_nullable=True,
+        )


Do you need param_value to be nullable?

I'm sorry but I couldn't understand your question. I didn't change the nullable setting of param_value, and I set existing_nullable=True.
Do you mean we need to replace existing_nullable=True with nullable=True?

contramundum53 · 2022-05-11T02:36:49Z

Otherwise this looks good to me. I couldn't test this change on mssql, but for other databases I confirmed that the migration works as intended.

github-actions bot added the optuna.storages Related to the `optuna.storages` submodule. This is automatically labeled by github-actions. label Feb 22, 2022

github-actions bot added the stale Exempt from stale bot labeling. label Mar 8, 2022

github-actions bot removed the stale Exempt from stale bot labeling. label Mar 10, 2022

toshihikoyanase added the enhancement Change that does not break compatibility and not affect public interfaces, but improves performance. label Mar 11, 2022

himkt self-assigned this Mar 14, 2022

github-actions bot added the stale Exempt from stale bot labeling. label Mar 28, 2022

github-actions bot removed the stale Exempt from stale bot labeling. label Apr 12, 2022

toshihikoyanase added 2 commits April 27, 2022 07:21

Set precision=53 according to IEEE 754.

3f6c6fb

Define float precision constant.

8ea3644

toshihikoyanase force-pushed the set-precision-of-float-in-rdb branch from bc74f1f to 8ea3644 Compare April 27, 2022 07:22

toshihikoyanase added 3 commits April 27, 2022 16:52

Add a migration script and update corresponding tests.

e4a6fa5

Apply black

e1e8d4b

Update import of optuna.storages._rdb.models

f9a8f13

toshihikoyanase commented Apr 27, 2022

View reviewed changes

himkt removed their assignment Apr 27, 2022

Remove duplicated nullable option.

0086320

toshihikoyanase marked this pull request as ready for review April 28, 2022 07:43

toshihikoyanase assigned c-bata and contramundum53 Apr 28, 2022

c-bata approved these changes May 10, 2022

View reviewed changes

optuna/storages/_rdb/alembic/versions/v3.0.0.b.py Outdated Show resolved Hide resolved

Update optuna/storages/_rdb/alembic/versions/v3.0.0.b.py

f29ef89

Co-authored-by: Masashi Shibata <c-bata@users.noreply.github.com>

contramundum53 reviewed May 10, 2022

View reviewed changes

contramundum53 merged commit c23e17c into optuna:master May 13, 2022

c-bata mentioned this pull request May 16, 2022

Add intermediate_value_type column to represent inf/-inf on RDBStorage. #3564

Merged

knshnb added this to the v3.0.0-b1 milestone Jun 3, 2022

toshihikoyanase deleted the set-precision-of-float-in-rdb branch February 17, 2025 07:13

not522 mentioned this pull request Apr 11, 2025

Add float precision tests for storages #6040

Merged

Uh oh!

Conversation

toshihikoyanase commented Feb 22, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Motivation

Description of the changes

Data types of the other RDB

SQLite3

PostgreSQL

TODO

Uh oh!

github-actions bot commented Mar 8, 2022

Uh oh!

toshihikoyanase commented Mar 10, 2022

Uh oh!

codecov-commenter commented Mar 10, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

github-actions bot commented Mar 28, 2022

Uh oh!

himkt commented Apr 12, 2022

Uh oh!

toshihikoyanase commented Apr 14, 2022

Uh oh!

toshihikoyanase Apr 27, 2022

Choose a reason for hiding this comment

Uh oh!

toshihikoyanase commented Apr 27, 2022

Uh oh!

toshihikoyanase commented Apr 28, 2022

Uh oh!

toshihikoyanase commented Apr 28, 2022

Uh oh!

xadrianzetx commented May 7, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

c-bata commented May 8, 2022

Uh oh!

xadrianzetx commented May 8, 2022

Uh oh!

c-bata left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

SQLite3

PostgreSQL

MySQL

Uh oh!

Uh oh!

contramundum53 May 10, 2022

Choose a reason for hiding this comment

Uh oh!

toshihikoyanase May 11, 2022

Choose a reason for hiding this comment

Uh oh!

contramundum53 commented May 11, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

toshihikoyanase commented Feb 22, 2022 •

edited

Loading

codecov-commenter commented Mar 10, 2022 •

edited

Loading

xadrianzetx commented May 7, 2022 •

edited

Loading

c-bata left a comment •

edited

Loading

contramundum53 commented May 11, 2022 •

edited

Loading