Switch to JSR 310 Date and Time API #233

dragneelfps · 2021-05-28T22:13:46Z

Removes joda from encoding/decoding
common
mysql
postgres

continuation of #229

dragneelfps · 2021-05-28T22:15:54Z

Getting following in CI build.. :/

docker: Error response from daemon: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit.

oshai · 2021-05-30T06:51:21Z

Hi, thanks for the PR!
There was some interest for it in the past. For reference see #131 and #157.
Ideally, if we could have both implementation as a non-breaking change it would be preferred.
But I understand this might complicate things a lot.
Since this is a breaking change we will need to bump the major version when merged.

dragneelfps · 2021-05-30T18:19:57Z

I could try to have Joda Time supported as well. But the thing is, do we really want to continue supporting it?
I prefer the major version upgrade. But seems odd going from 1.1.7 to 2.0.0.
Suggestions on how to move forward with this please?

oshai · 2021-06-01T21:19:14Z

Bump version to 2.0.0 sounds reasonable to me.
Please rebase changes from master again, build should work now in CI.

oshai · 2021-06-02T21:20:32Z

Note that some tests are still failing on CI.

dragneelfps · 2021-06-05T08:17:42Z

I fixed most. Just having issues understanding logic behind

jasync-sql/postgresql-async/src/test/java/com/github/aysnc/sql/db/integration/TimeAndDateSpec.kt

Line 151 in 152cdef

assertThat(dateTime.millis).isGreaterThan(915779106000L)

can you help, @oshai ?

oshai · 2021-06-05T08:47:18Z

I fixed most. Just having issues understanding logic behind

jasync-sql/postgresql-async/src/test/java/com/github/aysnc/sql/db/integration/TimeAndDateSpec.kt

Line 151 in 152cdef

assertThat(dateTime.millis).isGreaterThan(915779106000L)

can you help, @oshai ?

I didn't wrote it originally, but from reading the code I think it checks the following:

set timestamp that includes mills/micros
check that the timestamp millis conversion is greater than the second before and smaller from the second after

There might be a problem either with the timezone or the millis if I have to guess.

dragneelfps · 2021-06-05T17:00:55Z

I noticed something, postgres is returning timestamps in UTC string, even when I insert it with non UTC offset.
The tests are expecting non UTC offset. So I changed a bit and is actually testing if the two timestamps are equal as OffsetDateTime
Is this correct approach?
I am guessing postgres returning UTC string is related to server settings (which I am not really changing). Is this correct assumption?

codecov · 2021-06-05T18:33:18Z

Codecov Report

Merging #233 (c8313ef) into master (a5a881f) will increase coverage by 0.14%.
The diff coverage is 64.28%.

@@             Coverage Diff              @@
##             master     #233      +/-   ##
============================================
+ Coverage     80.03%   80.17%   +0.14%     
- Complexity     1074     1079       +5     
============================================
  Files           265      266       +1     
  Lines          3916     3919       +3     
  Branches        522      523       +1     
============================================
+ Hits           3134     3142       +8     
+ Misses          534      531       -3     
+ Partials        248      246       -2

Impacted Files	Coverage Δ
.../src/main/java/com/github/jasync/sql/db/RowData.kt	`84.00% <ø> (ø)`
.../com/github/jasync/sql/db/column/SQLTimeEncoder.kt	`66.66% <0.00%> (ø)`
.../jasync/sql/db/mysql/binary/decoder/DateDecoder.kt	`100.00% <ø> (ø)`
...ync/sql/db/mysql/binary/encoder/CalendarEncoder.kt	`0.00% <0.00%> (ø)`
...ync/sql/db/mysql/binary/encoder/DateTimeEncoder.kt	`25.00% <0.00%> (ø)`
...ync/sql/db/mysql/binary/encoder/JavaDateEncoder.kt	`25.00% <0.00%> (ø)`
...nc/sql/db/mysql/binary/encoder/LocalTimeEncoder.kt	`0.00% <0.00%> (ø)`
.../db/mysql/binary/encoder/ReadableInstantEncoder.kt	`0.00% <0.00%> (ø)`
...sync/sql/db/mysql/binary/encoder/SQLDateEncoder.kt	`25.00% <0.00%> (ø)`
...sync/sql/db/mysql/binary/encoder/SQLTimeEncoder.kt	`25.00% <0.00%> (+5.00%)`	⬆️
... and 17 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a5a881f...c8313ef. Read the comment docs.

- Removes joda from encoding/decoding - Using Threeten for PeriodDuration

oshai · 2021-06-05T21:59:59Z

I noticed something, postgres is returning timestamps in UTC string, even when I insert it with non UTC offset.
The tests are expecting non UTC offset. So I changed a bit and is actually testing if the two timestamps are equal as OffsetDateTime
Is this correct approach?
I am guessing postgres returning UTC string is related to server settings (which I am not really changing). Is this correct assumption?

According to https://www.postgresql.org/docs/9.1/datatype-datetime.html 8.5.2. Date/Time Output:

The default is the ISO format
and it looks like:

ISO | ISO 8601/SQL standard | 1997-12-17 07:37:16-08

(without time zone)

Conversion on test makes sense (to see it's the same date as inserted). It's just not clear how this was changed from pervious impl.

I still need to review all changes, but it will take some time.

oshai

Thanks again for the PR.
I left some comments in code and have some more general things:

I saw that in some cases there is no test coverage: do you think it's an issue with the framework, or can you please add tests to cover those cases?
I saw in many places move from millis to nano. Is it because java 8 uses nano and joda/db use millis?

About the migration to 2.x: I am thinking if we can add a suggestion to users how to do that more easily? Do you have any suggestion?

db-async-common/src/main/java/com/github/jasync/sql/db/column/LocalDateTimeEncoderDecoder.kt

gradle.properties

mysql-async/src/main/java/com/github/jasync/sql/db/mysql/binary/encoder/DateTimeEncoder.kt

oshai · 2021-06-13T08:40:11Z

mysql-async/src/test/java/com/github/jasync/sql/db/mysql/ConnectionHelper.kt

@@ -66,8 +66,8 @@ open class ConnectionHelper : ContainerHelper() {
        """CREATE TEMPORARY TABLE posts (
       id INT NOT NULL AUTO_INCREMENT,
       created_at_date DATE not null,
-       created_at_datetime DATETIME not null,
-       created_at_timestamp TIMESTAMP not null,
+       created_at_datetime DATETIME(6) not null,


why was it changed?

dragneelfps · 2021-06-13T16:24:19Z

I saw that in some cases there is no test coverage: do you think it's an issue with the framework, or can you please add tests to cover those cases?

Can you point me to which cases are you talking about specifically?

I saw in many places move from millis to nano. Is it because java 8 uses nano and joda/db use millis?

Yes. Java uses nano, joda uses millis, db micro(at least in mysql)

oshai · 2021-06-13T20:02:34Z

Can you point me to which cases are you talking about specifically?

Look at LocalDateTimeEncoderDecoder:17 and see more examples in 'files changed' with the warning: Added line #Lxx was not covered by tests

oshai · 2021-06-24T20:20:48Z

@dragneelfps before merging there was still one answered question in the review about DATETIME(6).

In addition, will you have some time to improve test coverage?

github-actions · 2021-08-24T00:02:19Z

Stale pull request message

oshai · 2021-08-24T20:33:06Z

I will merge this PR soon. Will create 1.x branch before that.

oshai · 2021-08-30T19:38:22Z

Thanks for the PR!
I merged it and will release a new version and update docs soon.

dragneelfps force-pushed the master branch 2 times, most recently from a6c447a to 6ea3c8f Compare June 2, 2021 03:11

Switch to JSR 310 Date and Time API

b0aab75

- Removes joda from encoding/decoding - Using Threeten for PeriodDuration

dragneelfps force-pushed the master branch from 51a4f5e to b0aab75 Compare June 5, 2021 18:36

oshai reviewed Jun 13, 2021

View reviewed changes

Fix formatter

7c702c8

github-actions bot added the no-pr-activity label Aug 24, 2021

Merge branch 'master' into master

c8313ef

oshai merged commit 46ab065 into jasync-sql:master Sep 1, 2021

oshai removed the no-pr-activity label Sep 1, 2021

Uh oh!

Switch to JSR 310 Date and Time API #233

Switch to JSR 310 Date and Time API #233

Uh oh!

Conversation

dragneelfps commented May 28, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

dragneelfps commented May 28, 2021

Uh oh!

oshai commented May 30, 2021

Uh oh!

dragneelfps commented May 30, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oshai commented Jun 1, 2021

Uh oh!

oshai commented Jun 2, 2021

Uh oh!

dragneelfps commented Jun 5, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

oshai commented Jun 5, 2021

Uh oh!

dragneelfps commented Jun 5, 2021

Uh oh!

codecov bot commented Jun 5, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

oshai commented Jun 5, 2021

Uh oh!

oshai left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

oshai Jun 13, 2021

Choose a reason for hiding this comment

Uh oh!

dragneelfps commented Jun 13, 2021

Uh oh!

oshai commented Jun 13, 2021

Uh oh!

oshai commented Jun 24, 2021

Uh oh!

github-actions bot commented Aug 24, 2021

Uh oh!

oshai commented Aug 24, 2021

Uh oh!

oshai commented Aug 30, 2021

Uh oh!

Uh oh!

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

dragneelfps commented May 28, 2021 •

edited

Loading

dragneelfps commented May 30, 2021 •

edited

Loading

dragneelfps commented Jun 5, 2021 •

edited

Loading

codecov bot commented Jun 5, 2021 •

edited

Loading