9651 scalar current date #10978

BaurzhanSakhariev · 2021-02-01T08:26:01Z

Summary of the changes / Why this improves CrateDB

Added date type
Added curdate scalar function returning current date in UTC

Checklist

Added an entry in CHANGES.txt for user facing changes
Updated documentation & sql_features table for user facing changes
Touched code is covered by tests
CLA is signed
This does not contain breaking changes, or if it does:
- It is released within a major release
- It is recorded in CHANGES.txt
- It was marked as deprecated in an earlier release if possible
- You've thought about the consequences and other components are adapted
  (E.g. AdminUI)

crate-jenkins · 2021-02-01T08:26:03Z

Can one of the admins verify this patch?

infoverload · 2021-02-01T09:59:59Z

docs/general/builtins/scalar.rst

+``CURRENT_DATE``
+----------------
+
+The ``CURRENT_DATE`` expression returns the date in UTC timezone at the time the SQL statement was handled. Clock


Thanks for your contribution!

Would you mind wrapping this to 79 chars per line?

mfussenegger

Thank you for your contribution.

There are a couple of parts that are missing to add a new type, so I've left a couple of comments.

We'd probably also need to follow up and support the date time in some artihmetic functions. E.g. to support something like current_date + '1 day'::interval. But I think that could be done in a follow up PR if you're interested.

docs/appendices/release-notes/unreleased.rst

server/src/main/java/io/crate/types/DateType.java

mfussenegger · 2021-02-01T11:04:50Z

server/src/main/java/io/crate/types/DateType.java

+
+    public static final int ID = 23;
+    public static final DateType INSTANCE = new DateType();
+    public static final int DATE_SIZE = (int) RamUsageEstimator.shallowSizeOfInstance(Long.class); // Value is written/read long number representing epoch milliseconds.


I think LocalDate is bigger than just a Long - this would have to be adjusted if we stick with LocalDate

Done. I thought I'm estimating size of the data, written to StreamOutput, somehow missed a hint coming from RamUsageEstimator class name)

Upd - As you suggested, I'm using now Long as internal date, reverted DATE_SIZE to Long size

server/src/main/java/io/crate/types/DateType.java

server/src/main/java/io/crate/types/DataTypes.java

nomicode

thanks for the PR!

I want to add a quick note to follow up on @mfussenegger's comment. last week, I added a TIP to the CURRENT_TIMESTAMP subsection about CURRENT_TIMESTAMP - '1 day'::interval (see de7cd71)

as the CURRENT_DATE subsection is being added directly below this, it might be a good idea to add a short note saying either that CURRENT_DATE - '1 day'::interval is supported or is not supported. might be overkill though?

docs/appendices/release-notes/unreleased.rst

BaurzhanSakhariev

thanks for the PR!

I want to add a quick note to follow up on @mfussenegger's comment. last week, I added a TIP to the CURRENT_TIMESTAMP subsection about CURRENT_TIMESTAMP - '1 day'::interval (see de7cd71)

as the CURRENT_DATE subsection is being added directly below this, it might be a good idea to add a short note saying either that CURRENT_DATE - '1 day'::interval is supported or is not supported. might be overkill though?

Thanks for the comments! Regarding upper case - done, but I used actual name CURDATE - I added big comment below with all my concern and one of them is that CURRENT_DATE name was failing tests while CURDATE passed them.

TIPS - I think everything related to +/- intreval has to be done in a separate PR in order to keep the current small.

I would like to highlight potential issues with the PR. I think it's fine for now but worth mentioning anyway.

Initially I used "current_date" name for the new scalar function. My test were failing and it took a while before I tried to use different name and same code worked with name "curdate".

As far as I understand, "current_time" is reserved word (saw it in SqlBaseParser) and probably having a new function with that name causes parsing failure.

Also, I saw builder.append("current_date"); in ExpressionFormatter, the only git commit for that line shows "add the presto sql parser".

While usage of curdate passes test, I found it confusing that current_date is a new function but I cannot use that name.

I haven't updated method visitCurrentTime in ExpressionAnalyzer class.It has switch (node.getType()) and that type enum already has DATE - looks like it was a placeholder for the future and I do have to add
case DATE:
funcName = CurrentTimeFunction.NAME;
break; but I decided to clarify that.
DataTypes class has 2 maps:
TYPE_IDS_TO_MAPPINGS for mapping type to ES type and I mapped new type to "date" - same as existing types TIMESTAMPZ and TIMESTAMP.

However, there is also a reversed map MAPPING_NAMES_TO_TYPES - where ES type is mapped to Crate type.
We cannot have 2 "date" keys and that's why I didn't an entry here. I think mapping ES date to Crate TIMESTAMPZ makes sense as TIMESTAMPZ is the most specific among DATE, TIMESTAMP, TIMESTAMPZ types.

I believe it's fine but I have to explicitly highlight it since it looks like a place where new type might leave a trace but it was not done.

New types' ID field value - I spent some time trying to figure out a pattern or registry of values as I had a feeling that value must be unique throughout all types but the only approach was looking over all type files and checking their ids and keeping some unused number. There are some maps with IDS but one cannot use them as a quick ID-dictionary since they are given not as numbers but in Type.ID form. I hope my approach is fine and I haven't missed some file or even script for assigning those ids.

BaurzhanSakhariev · 2021-02-01T12:27:26Z

server/src/main/java/io/crate/types/DateType.java

+
+    public static final int ID = 23;
+    public static final DateType INSTANCE = new DateType();
+    public static final int DATE_SIZE = (int) RamUsageEstimator.shallowSizeOfInstance(Long.class); // Value is written/read long number representing epoch milliseconds.


Done. I thought I'm estimating size of the data, written to StreamOutput, somehow missed a hint coming from RamUsageEstimator class name)

docs/appendices/release-notes/unreleased.rst

BaurzhanSakhariev · 2021-02-01T12:30:57Z

docs/general/builtins/scalar.rst

+``CURRENT_DATE``
+----------------
+
+The ``CURRENT_DATE`` expression returns the date in UTC timezone at the time the SQL statement was handled. Clock


seut · 2021-02-02T10:18:39Z

@BaurzhanSakhariev

I haven't updated method visitCurrentTime in ExpressionAnalyzer class.It has switch (node.getType()) and that type enum already has DATE - looks like it was a placeholder for the future and I do have to add
case DATE:
funcName = CurrentTimeFunction.NAME;
break; but I decided to clarify that.

Yes we'd have to add here a new case for DATE which must then result in the related current_date function name.

DataTypes class has 2 maps:
TYPE_IDS_TO_MAPPINGS for mapping type to ES type and I mapped new type to "date" - same as existing types TIMESTAMPZ and TIMESTAMP.

However, there is also a reversed map MAPPING_NAMES_TO_TYPES - where ES type is mapped to Crate type.
We cannot have 2 "date" keys and that's why I didn't an entry here. I think mapping ES date to Crate TIMESTAMPZ makes sense as TIMESTAMPZ is the most specific among DATE, TIMESTAMP, TIMESTAMPZ types.

Yes good spot. As the underlaying field mappers (of ES) do not make any difference between DATE and TIMESTAMP, we can not simply map SQL DATE to a storage field type. Because of that I suggest to not support storing SQL DATE column types for now, we'd need to think more about a working solution here, if any.
To prevent that, the entry here https://github.com/crate/crate/pull/10978/files#diff-63dbc4faae4044b73e5c10f6ecf640d41a16284a9615d40e0c59e2ef1517219eR445 must be removed.

New types' ID field value - I spent some time trying to figure out a pattern or registry of values as I had a feeling that value must be unique throughout all types but the only approach was looking over all type files and checking their ids and keeping some unused number. There are some maps with IDS but one cannot use them as a quick ID-dictionary since they are given not as numbers but in Type.ID form. I hope my approach is fine and I haven't missed some file or even script for assigning those ids.

Yes there is no real patter, it should just not be used already. So your chosen ID looks fine.

BaurzhanSakhariev · 2021-02-02T10:34:35Z

Thanks for the comment, added case to switch and removed map entry.

server/src/main/java/io/crate/protocols/postgres/types/PGTypes.java

server/src/main/resources/sql_features.tsv

seut

Thanks for your efforts. I've added another round of suggestions/comments.

docs/general/builtins/scalar.rst

server/src/main/java/io/crate/protocols/postgres/types/DateType.java

server/src/main/java/io/crate/types/DateType.java

server/src/test/java/io/crate/expression/scalar/CurrentDateFunctionTest.java

server/src/test/java/io/crate/protocols/postgres/types/DateTypeTest.java

server/src/test/java/io/crate/testing/DataTypeTesting.java

server/src/test/java/io/crate/types/DateTypeTest.java

server/src/main/java/io/crate/types/DateType.java

server/src/test/java/io/crate/types/DateTypeTest.java

seut

Looks really good already, I've added final documentation suggestions.

docs/general/builtins/scalar.rst

docs/appendices/release-notes/unreleased.rst

docs/general/builtins/scalar.rst

server/src/main/resources/sql_features.tsv

seut · 2021-02-16T08:51:55Z

retest this please

seut · 2021-02-16T11:27:25Z

@BaurzhanSakhariev Thanks a lot for addressing all comments.
Unfortunately the :server:checkstyleMain task complains now and fails, also the itest task fails because of the newly added Date and Array(Date) PG types, see line 163, in docs/interfaces/postgres.rst.
Could you have a look on these? Thanks!

btw. some hdfs related test also fails cause it uses an old lib version which isn't online anymore. We've fixed this already on master, so rebasing your branch with the latest master will should solve that.

BaurzhanSakhariev · 2021-02-16T11:34:07Z

I'm checking locally, some tests are failing with timeout (looks like I need to configure environment) for example InternalTestCluster.assertAfterTest but couldn't find non-environmental failure locally.

Thanks for pointing out the specific file (I couldn't see it inGitHub report as I need to log in to Jenkins) I will take a look for that specific case.

BaurzhanSakhariev · 2021-02-19T11:56:47Z

I fixed server:checkstyleMain and updated postgres.rst with 2 entries for the new date and corresponding array type.

However, neither itest nor app:run are not working on my machine. I updated postgres.rst by manually filling all columns and ordering by OID.

My OS is 64-bit Windows 10 and problem with itest is that blackbox/bootstrap.sh uses venv/bin/pip whereas for Windows bin should be replaced by Scripts.

I tried that and also changed commandLine "$projectDir/bin/test-docs" to commandLine "sh", "$projectDir/bin/test-docs" but still faced an issue with "Failed to import test module test_docs" - probably test-docs script also has to be adjusted for Windows.

Regarding app:run

it fails with

Exception in thread "main" java.lang.ExceptionInInitializerError
  at java.base/jdk.internal.icu.text.NormalizerBase$NFKDMode.getNormalizer2(NormalizerBase.java:305)
  at java.base/jdk.internal.icu.text.NormalizerBase.normalize(NormalizerBase.java:457)
  at java.base/jdk.internal.icu.text.NormalizerBase.normalize(NormalizerBase.java:461)
  at java.base/java.text.Normalizer.normalize(Normalizer.java:159)
  at java.base/sun.security.x509.AVA.toRFC2253CanonicalString(AVA.java:987)
  at java.base/sun.security.x509.RDN.toRFC2253StringInternal(RDN.java:437)
  at java.base/sun.security.x509.RDN.toRFC2253String(RDN.java:417)
  at java.base/sun.security.x509.X500Name.getRFC2253CanonicalName(X500Name.java:724)
  at java.base/sun.security.x509.X500Name.equals(X500Name.java:416)
  at java.base/sun.security.pkcs.PKCS7.getCertificate(PKCS7.java:694)
  at java.base/sun.security.pkcs.SignerInfo.getCertificate(SignerInfo.java:253)
  at java.base/sun.security.pkcs.SignerInfo.verify(SignerInfo.java:401)
  at java.base/sun.security.pkcs.PKCS7.verify(PKCS7.java:578)
  at java.base/sun.security.pkcs.PKCS7.verify(PKCS7.java:595)
  at java.base/sun.security.pkcs.SignerInfo.getTimestamp(SignerInfo.java:566)
  at java.base/sun.security.pkcs.SignerInfo.verify(SignerInfo.java:324)
  at java.base/sun.security.pkcs.PKCS7.verify(PKCS7.java:578)
  at java.base/sun.security.pkcs.PKCS7.verify(PKCS7.java:595)
  at java.base/sun.security.util.SignatureFileVerifier.processImpl(SignatureFileVerifier.java:293)
  at java.base/sun.security.util.SignatureFileVerifier.process(SignatureFileVerifier.java:269)
  at java.base/java.util.jar.JarVerifier.processEntry(JarVerifier.java:316)
  at java.base/java.util.jar.JarVerifier.update(JarVerifier.java:230)
  at java.base/java.util.jar.JarFile.initializeVerifier(JarFile.java:749)
  at java.base/java.util.jar.JarFile.ensureInitialization(JarFile.java:1016)
  at java.base/java.util.jar.JavaUtilJarAccessImpl.ensureInitialization(JavaUtilJarAccessImpl.java:72)
  at java.base/jdk.internal.loader.URLClassPath$JarLoader$2.getManifest(URLClassPath.java:873)
  at java.base/jdk.internal.loader.BuiltinClassLoader.defineClass(BuiltinClassLoader.java:811)
  at java.base/jdk.internal.loader.BuiltinClassLoader.findClassOnClassPathOrNull(BuiltinClassLoader.java:723)
  at java.base/jdk.internal.loader.BuiltinClassLoader.loadClassOrNull(BuiltinClassLoader.java:646)
  at java.base/jdk.internal.loader.BuiltinClassLoader.loadClass(BuiltinClassLoader.java:604)
  at java.base/jdk.internal.loader.ClassLoaders$AppClassLoader.loadClass(ClassLoaders.java:168)
  at java.base/java.lang.ClassLoader.loadClass(ClassLoader.java:522)
  at io.crate.types.DataTypeXContentExtension.getXContentWriters(DataTypeXContentExtension.java:40)
  at org.elasticsearch.common.xcontent.XContentBuilder.<clinit>(XContentBuilder.java:119)
  at org.elasticsearch.common.settings.Setting.arrayToParsableString(Setting.java:1308)
  at org.elasticsearch.common.settings.Setting$ListSetting.lambda$new$0(Setting.java:1333)
  at org.elasticsearch.common.settings.Setting.<init>(Setting.java:151)
  at org.elasticsearch.common.settings.Setting$ListSetting.<init>(Setting.java:1330)
  at org.elasticsearch.common.settings.Setting.listSetting(Setting.java:1282)
  at org.elasticsearch.common.settings.Setting.listSetting(Setting.java:1250)
  at org.elasticsearch.env.Environment.<clinit>(Environment.java:55)
  at org.elasticsearch.node.InternalSettingsPreparer.prepareEnvironment(InternalSettingsPreparer.java:66)
  at io.crate.bootstrap.CrateDB.createEnv(CrateDB.java:112)
  at org.elasticsearch.cli.EnvironmentAwareCommand.execute(EnvironmentAwareCommand.java:82)
  at org.elasticsearch.cli.Command.mainWithoutErrorHandling(Command.java:124)
  at org.elasticsearch.cli.Command.main(Command.java:90)
  at io.crate.bootstrap.CrateDB.main(CrateDB.java:91)
  at io.crate.bootstrap.CrateDB.main(CrateDB.java:84)
Caused by: java.lang.NullPointerException: Cannot invoke "java.io.InputStream.available()" because "is" is null
  at java.base/jdk.internal.icu.impl.ICUBinary.getRequiredData(ICUBinary.java:94)
  at java.base/jdk.internal.icu.impl.NormalizerImpl.load(NormalizerImpl.java:431)
  at java.base/jdk.internal.icu.impl.Norm2AllModes$Norm2AllModesSingleton.<init>(Norm2AllModes.java:274)
  at java.base/jdk.internal.icu.impl.Norm2AllModes$NFKCSingleton.<clinit>(Norm2AllModes.java:290)
  at java.base/jdk.internal.icu.impl.Norm2AllModes.getNFKCInstance(Norm2AllModes.java:263)
  at java.base/jdk.internal.icu.text.Normalizer2.getNFKDInstance(Normalizer2.java:123)
  at java.base/jdk.internal.icu.text.NormalizerBase$NFKDModeImpl.<clinit>(NormalizerBase.java:181)
  ... 48 more

FAILURE: Build failed with an exception.

* What went wrong:
Execution failed for task ':app:run'.
> Process 'command 'C:\Users\Baur\.gradle\caches\transforms-3\00cf94b11d7d07694077c263b7fcc56e\transformed\windows-15.0.2-x64.zip/bin/java'' finished with non-zero exit value 1

I've been investigating and trying to fix both commands failure - if I manage to make itest related scripts cross-platform I will create another PR for that.

It would be nice if we can trigger remaining jenkins/pr_tests check to see how my manual additions to the postgres.rst will behave in a proper environment.

amotl · 2021-02-19T12:46:48Z

Hi @BaurzhanSakhariev,

thanks for your efforts. Regarding Windows runtime support, CrateDB should work - you might have observed that the main test suite is also being invoked on GitHub Actions. While we occasionally have to adjust things in this space regarding flakyness (e.g. #11045), it works in general.

Regarding blackbox testing, I believe it is using cr8 under the hood. In order to bring that to Windows (see mfussenegger/cr8#327), I already started mfussenegger/cr8#337 but haven't been able to complete this yet.

This is just to give you a heads-up which kinds of patches would have to be applied as a minimum to make it work on Windows. The state of the onion is that, by using those patches, it should already work reasonably but currently fails on test layer teardown on GHA/CI due to some nasty deadlocking issue where I haven't been able to investigate further.

So, while it will probably be a good idea to continue day-to-day development of CrateDB on Linux or macOS, I am happy to hear you also know your way around Windows and hope that you will be available to improve Windows support for all the test harnesses we are running as well as Windows support in general. See, for example, #10516 and #6349.

Additionally, you might want to inspect the GHA recipe to learn about respective arguments to Gradle:

crate/.github/workflows/main.yml

Line 28 in 9389bb2

    
           arguments: :server:test -Dtests.crate.run-windows-incompatible=${{ matrix.os == 'ubuntu-latest' }}

It would be nice if we can trigger remaining jenkins/pr_tests check.

After rebasing your branch on top of current master, you might be able to trigger that on your own by adding a comment saying retest this please, which will act as a trigger command to Jenkins.

With kind regards,
Andreas.

seut · 2021-02-19T13:03:59Z

retest this please

BaurzhanSakhariev · 2021-03-10T16:32:52Z

retest this please

seut

Thanks for your changes. I've added some comments mostly related to documentation.

docs/general/ddl/data-types.rst

seut · 2021-03-11T08:25:13Z

docs/general/ddl/data-types.rst

+Date also accepts a ``bigint`` representing UTC milliseconds since
+the epoch.
+
+::
+
+    cr> select 1615248000000::date as cd;
+   +---------------+
+    |            cd |
+    +---------------+
+    | 1615248000000 |
+    +---------------+
+    SELECT 1 row in set (... sec)


I'd remove this example if you take my suggestion above which already mention the support of numeric->date casts.

Done. I removed string casting example select '2021-03-09'::date as well as it's also mentioned in suggestion.

I think keeping the string cast example would still be valuable (also this may be the common use case).

docs/general/ddl/data-types.rst

server/src/main/java/io/crate/expression/scalar/CurrentDateFunction.java

BaurzhanSakhariev · 2021-03-12T09:12:13Z

retest this please

seut · 2021-03-12T11:41:58Z

retest this please

mfussenegger

Left some more coments.
Overall seems like it's almost ready 👍

docs/general/ddl/data-types.rst

server/src/main/java/io/crate/expression/scalar/CurrentDateFunction.java

mfussenegger · 2021-03-15T09:24:52Z

server/src/main/java/io/crate/protocols/postgres/types/DateType.java

+        LocalDate dt = LocalDate.parse(s, ISO_FORMATTER);
+        return dt.atStartOfDay(ZoneOffset.UTC).toInstant().toEpochMilli();


PostgreSQL supports more text formats as input:

https://www.postgresql.org/docs/current/datatype-datetime.html

I think we either have to extend this, or make it explicit in the documentation that the text format support is not compatible with all PostgreSQL clients.

I updated data-types.rst and added TODO to DateType

BaurzhanSakhariev · 2021-03-16T10:34:23Z

retest this please

mfussenegger · 2021-03-22T08:37:02Z

ok to test

mfussenegger

Nice work 👍

seut

👍 Great!

BaurzhanSakhariev requested a review from a team February 1, 2021 08:26

infoverload reviewed Feb 1, 2021

View reviewed changes

mfussenegger reviewed Feb 1, 2021

View reviewed changes

nomicode reviewed Feb 1, 2021

View reviewed changes

docs/appendices/release-notes/unreleased.rst Outdated Show resolved Hide resolved

BaurzhanSakhariev commented Feb 1, 2021

View reviewed changes

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 2 times, most recently from 1546ef3 to 711c0ba Compare February 2, 2021 09:26

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch from 711c0ba to 3f7bda7 Compare February 2, 2021 10:33

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch from 3f7bda7 to f363c3d Compare February 2, 2021 10:43

seut reviewed Feb 2, 2021

View reviewed changes

server/src/main/java/io/crate/protocols/postgres/types/PGTypes.java Outdated Show resolved Hide resolved

server/src/main/java/io/crate/protocols/postgres/types/PGTypes.java Outdated Show resolved Hide resolved

server/src/main/resources/sql_features.tsv Outdated Show resolved Hide resolved

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 2 times, most recently from 056fe96 to 89572b8 Compare February 3, 2021 07:58

seut reviewed Feb 9, 2021

View reviewed changes

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 3 times, most recently from 87a1690 to 2426dc5 Compare February 10, 2021 15:24

seut reviewed Feb 11, 2021

View reviewed changes

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 2 times, most recently from 9c96c30 to f90d8ff Compare February 12, 2021 14:07

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 2 times, most recently from ebde802 to 7c417bf Compare February 19, 2021 10:34

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 6 times, most recently from 04e27ef to 9532635 Compare March 10, 2021 15:39

seut reviewed Mar 11, 2021

View reviewed changes

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch from d1214aa to 936bb83 Compare March 12, 2021 08:32

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch from 936bb83 to a36ae04 Compare March 12, 2021 11:38

seut requested a review from mfussenegger March 12, 2021 11:41

mfussenegger reviewed Mar 15, 2021

View reviewed changes

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch 4 times, most recently from 4046b55 to c6788c8 Compare March 16, 2021 09:51

seut requested a review from mfussenegger March 19, 2021 08:40

mfussenegger approved these changes Mar 22, 2021

View reviewed changes

mfussenegger requested a review from seut March 22, 2021 09:07

seut approved these changes Mar 22, 2021

View reviewed changes

seut added the ready-to-merge Let Mergify merge the PR once approved and checks pass label Mar 22, 2021

9651: Implement current_date Scalar Function

1c70e58

BaurzhanSakhariev force-pushed the 9651-scalar-current-date branch from 396e910 to 1c70e58 Compare March 22, 2021 11:50

Merge branch 'master' into 9651-scalar-current-date

9bb8201

mergify bot merged commit 10cbed3 into crate:master Mar 22, 2021

mfussenegger mentioned this pull request Mar 22, 2021

Add current_date expression #9651

Closed

		LocalDate dt = LocalDate.parse(s, ISO_FORMATTER);
		return dt.atStartOfDay(ZoneOffset.UTC).toInstant().toEpochMilli();

9651 scalar current date #10978

9651 scalar current date #10978

Conversation

BaurzhanSakhariev commented Feb 1, 2021

Summary of the changes / Why this improves CrateDB

Checklist

crate-jenkins commented Feb 1, 2021

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mfussenegger left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

BaurzhanSakhariev Feb 1, 2021 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nomicode left a comment

Choose a reason for hiding this comment

BaurzhanSakhariev left a comment

Choose a reason for hiding this comment

BaurzhanSakhariev Feb 1, 2021 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

seut commented Feb 2, 2021

BaurzhanSakhariev commented Feb 2, 2021 • edited

seut left a comment

Choose a reason for hiding this comment

seut left a comment

Choose a reason for hiding this comment

seut commented Feb 16, 2021

seut commented Feb 16, 2021

BaurzhanSakhariev commented Feb 16, 2021

BaurzhanSakhariev commented Feb 19, 2021

amotl commented Feb 19, 2021 • edited

seut commented Feb 19, 2021

BaurzhanSakhariev commented Mar 10, 2021

seut left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

BaurzhanSakhariev commented Mar 12, 2021

seut commented Mar 12, 2021

mfussenegger left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

BaurzhanSakhariev commented Mar 16, 2021

mfussenegger commented Mar 22, 2021

mfussenegger left a comment

Choose a reason for hiding this comment

seut left a comment

Choose a reason for hiding this comment

BaurzhanSakhariev Feb 1, 2021 •

edited

BaurzhanSakhariev Feb 1, 2021 •

edited

BaurzhanSakhariev commented Feb 2, 2021 •

edited

amotl commented Feb 19, 2021 •

edited