FWD » Core Development » Database

12/03/2019

06:57 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

What I'm trying to figure out is whether it is safe to move the FIND's where clause to the REPEAT PRESELECT level, ei... Eric Faulhaber

03:52 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

Roger Borrello wrote:
> Eric Faulhaber wrote:
> > Which construct controls sorting? That is, if the REPEAT PRESELEC... Roger Borrello

03:10 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

Eric Faulhaber wrote:
> Which construct controls sorting? That is, if the REPEAT PRESELECT specifies one index and t... Roger Borrello

02:53 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

Which construct controls sorting? That is, if the REPEAT PRESELECT specifies one index and the inner FIND a different... Eric Faulhaber

10:16 AM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

This is by far the most prevalent use of this syntax in customer code (@PRESELECT@ doesn't filter anything, and lets ... Roger Borrello

09:47 AM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

Eric Faulhaber wrote:
> We could add an API which accepts a client-side where clause in the form of a lambda, to be ... Roger Borrello

06:09 AM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

So, the problem is that we don't anticipate the where clause for the inner FIND statement. In every other case we've ... Eric Faulhaber

12/02/2019

03:37 PM Bug #4422: h2.sortNullsHigh in page-store mode

Eric Faulhaber wrote:
> Are you saying it isn't working in the version of H2 we're using now, but it _is_ working in... Constantin Asofiei

03:28 PM Bug #4422: h2.sortNullsHigh in page-store mode

Are you saying it isn't working in the version of H2 we're using now, but it _is_ working in 1.4.200? Eric Faulhaber

09:55 AM Bug #4422: h2.sortNullsHigh in page-store mode

Constantin Asofiei wrote:
> H2 1.4.200 doesn't have support for sortNullsHigh.
Scratch that, sortNullsHigh in page-... Constantin Asofiei

09:41 AM Bug #4422: h2.sortNullsHigh in page-store mode

H2 1.4.200 doesn't have support for sortNullsHigh. Constantin Asofiei

09:21 AM Bug #4421: blob/clob access with H2

H2 1.4.200 with page-store still has the LOB bugs. Constantin Asofiei

11/26/2019

01:44 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

@database_access.rules@ is one of the nastier persistence rule sets, because it tries to handle so many related but s... Eric Faulhaber

11/25/2019

04:35 AM Bug #4421: blob/clob access with H2

Constantin Asofiei wrote:
> ... then the abend is seen, but only if the LOB size exceeds the H2's BLOCK_LENGTH (2000... Constantin Asofiei

11/24/2019

12:32 PM Bug #4421: blob/clob access with H2

Eric, my conclusion is this: LOB in page-store is broken, only in MVStore is working - if I remove the @mv_store=fal... Constantin Asofiei

08:42 AM Bug #4421: blob/clob access with H2

Eric Faulhaber wrote:
> You can invoke @session.getInterceptor()@, which will return an instance of @ChangeBroker$Se... Constantin Asofiei

11/22/2019

05:03 PM Bug #4421: blob/clob access with H2

Constantin Asofiei wrote:
> @BlobUserType@ and @ClobUserType@ need to be changed, so that the LOB access is done in ... Eric Faulhaber

03:53 PM Bug #4421: blob/clob access with H2

Even after I added the @:_temp@ exclusion (so that only dirty and meta will use the stream lob APIs), it still fails.... Constantin Asofiei

03:39 PM Bug #4421: blob/clob access with H2

Ovidiu, your solution is working, but I need one more criteria, to distinguish between session-local tables and physi... Constantin Asofiei

02:44 PM Bug #4421: blob/clob access with H2

Constantin,
As a temporary workaround until a better solution is found, try to use @session.connection().getMetaData... Ovidiu Maxiniuc

01:43 PM Bug #4421: blob/clob access with H2

For reference, this is the kind of error H2 throws, when using the LOB setters:... Constantin Asofiei

09:45 AM Bug #4421: blob/clob access with H2

The Hibernate dialect has a method named @useInputStreamToInsertBlob@ - the implementation for H2 returns true, while... Constantin Asofiei

09:39 AM Bug #4421 (WIP): blob/clob access with H2

Constantin Asofiei

04:17 PM Bug #4426: Cannot find parent when processing REPEAT PRESELECT

h1. General
Expression execution error when there is a @preselect@ qualifier on @repeat@.
*20191231 Update: Cus... Roger Borrello

04:05 PM Bug #4426 (New): Cannot find parent when processing REPEAT PRESELECT

Roger Borrello

12:59 PM Bug #4422: h2.sortNullsHigh in page-store mode

Eric Faulhaber wrote:
> For our development/testing use of H2, I think it is ok to switch to the MV_STORE storage en... Constantin Asofiei

12:55 PM Bug #4422: h2.sortNullsHigh in page-store mode

For our development/testing use of H2, I think it is ok to switch to the MV_STORE storage engine when using H2 as the... Eric Faulhaber

10:32 AM Bug #4422: h2.sortNullsHigh in page-store mode

With #3961, we disabled MVCC and moved to page-store. But with page-store, the h2.sortNullsHigh is not working - thi... Constantin Asofiei

10:29 AM Bug #4422 (New): h2.sortNullsHigh in page-store mode

Constantin Asofiei

12:56 PM Bug #3961: "Unexpected code path" failure when using H2 in MVCC mode

Eric Faulhaber wrote:
> Please refresh my memory on this issue. Is there a production use case where this is a probl... Constantin Asofiei

12:43 PM Bug #3961: "Unexpected code path" failure when using H2 in MVCC mode

Please refresh my memory on this issue. Is there a production use case where this is a problem?
We do not support ... Eric Faulhaber

11/19/2019

10:35 AM Feature #3574: finish implementation of temp-table XML support

Methods:
* read-xmlschema() method
** file | memptr | handle | longchar %{color:#6080FF}OM: only file and longchar%... Ovidiu Maxiniuc

09:24 AM Feature #3574: finish implementation of temp-table XML support

How much of this task is complete as of branch 3809e? Greg Shah

09:42 AM Bug #4200: verify and fix all temp-table options which must result in different DMO implementatio...

Another part I needed to add is BEFORE-TABLE - if this doesn't match between two tables, then we must emit two differ... Constantin Asofiei

09:19 AM Feature #4418: improve/fix temp-table naming approach

As much as possible, I prefer for the temp-table names to retain the name given in the business logic definition. It... Greg Shah

09:17 AM Feature #4418: improve/fix temp-table naming approach

Ovidiu made changes to temp-table naming in #3324 which have been merged to trunk. However, there remains some issue... Greg Shah

09:15 AM Feature #4418 (New): improve/fix temp-table naming approach

Greg Shah

11/14/2019

12:47 PM Feature #3815 (WIP): more temp-table options

> Code page is properly resolved and set, but data it is not actually converted: when a @longchar/clob@ value is ass... Greg Shah

11/13/2019

05:58 PM Bug #3651 (Closed): regression in server startup when _connect metadata is active and a database ...

Greg Shah wrote:
> I see that 3791c was arched as merged on 20191025. What revision of trunk includes this change?
... Eric Faulhaber

05:26 PM Bug #3651: regression in server startup when _connect metadata is active and a database is loaded...

I see that 3791c was arched as merged on 20191025. What revision of trunk includes this change?
Can we close this? Greg Shah

04:17 PM Bug #3960 (Closed): localedef problems in Ubuntu 18.04

Task branch 4069a has been merged to trunk as revision 11339. Greg Shah

12:05 PM Feature #3759 (Closed): setup, test and document PostgreSQL on Windows

Greg Shah

11/12/2019

08:28 AM Bug #4404 (New): ConditionException inside full transaction FOR EACH loop

There is a bug when a ... Adrian Lungu

11/11/2019

04:23 PM Feature #4400: add /* UUID */ in where clause comment for every 4GL query so that SQL logging can...

The idea:
* In early annotations, we calculate a UUID for every 4GL query and store it as an annotation.
* In con... Greg Shah

04:19 PM Feature #4400 (New): add /* UUID */ in where clause comment for every 4GL query so that SQL loggi...

Greg Shah

02:06 PM Feature #4399: support session level "schema" switching

When thinking about a multi-tenant approach, there are 3 possible approaches:
* database instance per tenant
** a... Greg Shah

01:56 PM Feature #4399 (New): support session level "schema" switching

Greg Shah

03:58 AM Feature #3809: ProDataSet support

With 3809e revision 11373 all the ChUI regressions should be fixed. I started another run to make sure the changes do... Hynek Cihlar

11/10/2019

10:24 AM Support #4080: AWS Aurora support

I spoke with Amazon Principal Database Engineer Jim M. We discussed a range of items which I am documenting here.
... Greg Shah

11/09/2019

10:56 AM Feature #4397: add database attrs, methods and options

The following work is needed:
* @BUFFER:LAST-BATCH@ attribute
** Although this is related to PDS, it has no conve... Greg Shah

10:40 AM Feature #4397 (Closed): add database attrs, methods and options

Greg Shah

11/08/2019

04:34 AM Feature #3809: ProDataSet support

I see 53 failed tests in my last ChUI regression run of 3809e. I'm looking at them. Hynek Cihlar

11/06/2019

12:20 PM Feature #2156: datetime-tz and hibernate

Eric, you are right. I misread Stanislav's entry. There is no double-conversion with the use @TIMESTAMP WITH TZ@.
* ... Ovidiu Maxiniuc

12:04 PM Feature #2156: datetime-tz and hibernate

Ovidiu Maxiniuc wrote:
> Stanislav Lomany wrote:
> > Isn't storing data in TIMESTAMP WITHOUT TZ in UTC solve our pr... Eric Faulhaber

11:15 AM Feature #2156: datetime-tz and hibernate

Stanislav Lomany wrote:
> Isn't storing data in TIMESTAMP WITHOUT TZ in UTC solve our problems and has an extra plus... Ovidiu Maxiniuc

08:32 AM Feature #2156: datetime-tz and hibernate

Isn't storing data in TIMESTAMP WITHOUT TZ in UTC solve our problems and has an extra plus of avoiding double convers... Stanislav Lomany

08:25 AM Feature #2156: datetime-tz and hibernate

Stanislav Lomany wrote:
> Please confirm that my understanding how @TIMESTAMP WITH TIME ZONE@ will work in note 55 i... Eric Faulhaber

08:16 AM Feature #2156: datetime-tz and hibernate

Please confirm that my understanding how @TIMESTAMP WITH TIME ZONE@ will work in note 55 is correct and it is what we... Stanislav Lomany

08:09 AM Feature #2156: datetime-tz and hibernate

What is the status of this task? Greg Shah

09:16 AM Feature #3758: misc database features

Ovidiu Maxiniuc wrote:
> Indeed, the @DBTASKID@ should be marked as having full implementation. Where do I commit th... Greg Shah

06:25 AM Support #3871: determine how to change codepages/locales during import

I used this simple java code to generate dump files from @dump_file_template.txt@ Sergey Ivanovskiy

11/04/2019

08:34 AM Feature #3758: misc database features

Indeed, the @DBTASKID@ should be marked as having full implementation. Where do I commit this? Ovidiu Maxiniuc