FWD » Core Development » Database

07/22/2024

02:57 PM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru, about the other issue related to leaking slave buffers. The issue is that a buffer can be defined, but wi... Constantin Asofiei

02:30 PM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I found the reason of (some of) the @A different DMO instance of type ... is already bound@ at least in a scenario wh... Constantin Asofiei

08:12 AM Feature #7960: Allow query logs to be collected in a H2 database

Review of 7960a:
* Small issues with @P2JLoggingDatabaseHelper@:
** history entry @Module@ is not the same as the ... Alexandru Lungu

07/12/2024

04:32 PM Feature #7020 (Internal Test): always use "expanded" extent fields

After debugging I was able to identify one cause of the regressions being incorrect hydration. This defect was hidden... Ovidiu Maxiniuc

09:21 AM Feature #7020: always use "expanded" extent fields

This is OK. As you can see from the #7020-30 table and the considerations below, the current @normalized@ and the new... Ovidiu Maxiniuc

09:04 AM Feature #7020: always use "expanded" extent fields

Ovidiu, the original concern with denormalized extents (or 'expanded extents') was because the .p2o (and the DMOs) we... Constantin Asofiei

07/11/2024

10:35 AM Bug #8952: Maximum function call inside a dynamic query is broken

It works just fine for me with following changes:... Ovidiu Maxiniuc

10:01 AM Bug #8952: Maximum function call inside a dynamic query is broken

Ovidiu Maxiniuc wrote:
> No, make the @com.goldencode.expr.UnknownType@ public. It is needed by @Function.isAssignme... Artur Școlnic

09:16 AM Bug #8952: Maximum function call inside a dynamic query is broken

No, make the @com.goldencode.expr.UnknownType@ public. It is needed by @Function.isAssignmentCompatible()@.
Also, p... Ovidiu Maxiniuc

09:06 AM Bug #8952: Maximum function call inside a dynamic query is broken

I created a new @UnknownType@ class in persist package. The conversion was successful, but at runtime I get the error... Artur Școlnic

08:45 AM Bug #8952: Maximum function call inside a dynamic query is broken

I am trying to port the @matchSignature@ method to @RuntimeJastInterpreter@, but @UnknownType@ is package-private. Artur Școlnic

06:25 AM Bug #8952: Maximum function call inside a dynamic query is broken

I probably have some additional changes in my working environment. Pasting the big code chunk before @return false;@ ... Ovidiu Maxiniuc

02:39 AM Bug #8952: Maximum function call inside a dynamic query is broken

I tried porting these changes, but the conversion fails. What branch is this from? Artur Școlnic

09:32 AM Feature #7020: always use "expanded" extent fields

Nice summary!
> # @normalized@. This is the old solution. In this case the properties are accessed either in bulk ... Greg Shah

08:10 AM Feature #7020: always use "expanded" extent fields

I have to make a distinction about the way the mapping of the ABL extent fields in converted code. There are 3 possib... Ovidiu Maxiniuc

07/10/2024

10:25 AM Bug #8952: Maximum function call inside a dynamic query is broken

Artur Școlnic wrote:
> I may be wrong, but isn't @BaseDataType maximum(BaseDataType... args)@ exactly for this type ... Ovidiu Maxiniuc

10:20 AM Bug #8952: Maximum function call inside a dynamic query is broken

Try with this patch:... Ovidiu Maxiniuc

10:20 AM Bug #8952: Maximum function call inside a dynamic query is broken

Ovidiu Maxiniuc wrote:
> This NPE is cause by lack of support for variable parameter list method in FWD. The list of... Artur Școlnic

10:15 AM Bug #8952: Maximum function call inside a dynamic query is broken

Ovidiu Maxiniuc wrote:
> This NPE is cause by lack of support for variable parameter list method in FWD. The list of... Artur Școlnic

10:14 AM Bug #8952: Maximum function call inside a dynamic query is broken

Investigating the issue led to several findings:
* The function call does not to be dynamic, a static call of the fu... Artur Școlnic

09:37 AM Bug #8952: Maximum function call inside a dynamic query is broken

This NPE is cause by lack of support for variable parameter list method in FWD. The list of current arguments is comp... Ovidiu Maxiniuc

08:59 AM Bug #8952: Maximum function call inside a dynamic query is broken

The stack trace is... Artur Școlnic

08:58 AM Bug #8952 (WIP): Maximum function call inside a dynamic query is broken

The issue is derived from #8912-86. Artur Școlnic

08:57 AM Bug #8952: Maximum function call inside a dynamic query is broken

... Artur Școlnic

08:55 AM Bug #8952 (WIP): Maximum function call inside a dynamic query is broken

Artur Școlnic

08:51 AM Bug #8593 (Test): H2 performance improvement for table 'truncate by multiplex'

Greg Shah

08:48 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Branch 8593d was merged to trunk rev 15331 and archived. Constantin Asofiei

08:40 AM Bug #8593 (Merge Pending): H2 performance improvement for table 'truncate by multiplex'

You can merge now. Greg Shah

08:37 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Greg Shah wrote:
> Is this ready/safe to merge?
Yes. Constantin Asofiei

08:25 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Is this ready/safe to merge? Greg Shah

08:20 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Constantin Asofiei wrote:
> Rebased 8593d from trunk rev 15330 - new rev 15336. I need to run ETF and some other ap... Constantin Asofiei

03:13 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Rebased 8593d from trunk rev 15330 - new rev 15336. I need to run ETF and some other app testing. Constantin Asofiei

07/09/2024

10:15 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I am still facing an infinite loop, but this is due to #8945 which seems unrelated.
The issue from #8593-120 is to b... Alexandru Lungu

08:01 AM Bug #8593 (Internal Test): H2 performance improvement for table 'truncate by multiplex'

New issue in the unit tests. I am facing a @Requested to persist DMO of type@ kind of exception. This is because:
* ... Alexandru Lungu

07/08/2024

08:11 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> Committed 8593d/rev. 15317 including changes to remove the out-of-scope buffer from the sla... Constantin Asofiei

07:45 AM Bug #8593 (Review): H2 performance improvement for table 'truncate by multiplex'

Committed 8593d/rev. 15317 including changes to remove the out-of-scope buffer from the slave collection. This is eno... Alexandru Lungu

06:55 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I think the finally should be a sub-block of the block it guards, and execute as part of that, not sequentially, afte... Ovidiu Maxiniuc

07/05/2024

10:37 AM Feature #8931 (New): enable a logical database name to be specified in database configuration/adm...

Eric Faulhaber

07:25 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I'm not saying that we don't need to fix the finalizable problem. The non-default buffers need to be cleaned from @d... Constantin Asofiei

07:05 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Constantin Asofiei wrote:
> Alexa, the @TemporaryBuffer$Context.domains@ never gets cleaned, also - the salve buffer... Alexandru Lungu

06:51 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alex, the @TemporaryBuffer$Context.domains@ never gets cleaned, also - the salve buffers for a DMO do not get removed... Constantin Asofiei

06:08 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

The fix is not obvious. The *major* issue is that the scope of the buffer is closed so that it release, commits, etc.... Alexandru Lungu

07/04/2024

07:47 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> I think this is related to @finally@ block, as the scope of the buffer is not managed prope... Dănuț Filimon

07:36 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Found the issue:... Alexandru Lungu

07/03/2024

03:41 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

> Does emptyDataset - clearBuffers fall under any of these scenarios, or is there another scenario I have missed?
... Alexandru Lungu

07/02/2024

04:22 PM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> Progress:
>
> The culprit buffer has its record deleted by an @emptyDataset@ - @clearBuf... Eric Faulhaber

10:14 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> However, the buffer is not registered with the @ChangeBroker@ (why? this is still to be inv... Constantin Asofiei

10:12 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Progress:
The culprit buffer has its record deleted by an @emptyDataset@ - @clearBuffers@. However, the buffer is ... Alexandru Lungu

08:27 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I don't have the very exact scenario this is causing this, but I am trying to find one. To post some insights:
@Se... Alexandru Lungu

04:00 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

ChUI regression tests passed completely.
The unit-tests are way better (in the sense that the found problem was fixe... Alexandru Lungu

07/01/2024

08:29 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Constantin Asofiei wrote:
> Alexandru Lungu wrote:
> > Is this always true? Can't the 4GL code simply clear the bef... Alexandru Lungu

08:27 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> Is this always true? Can't the 4GL code simply clear the before table like @EMPTY TEMP-TABL... Constantin Asofiei

08:24 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

> The 'after-table' buffers have their records deleted first; if there are referenced records in these after buffers,... Alexandru Lungu

08:13 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> * For @TemporaryBuffer@: isn't the flag you are changing a guard to avoid spoiling the afte... Constantin Asofiei

08:10 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Review of 8593d/rev. 15316.
* I am OK with the changes in @Session@
* For @TemporaryBuffer@: isn't the flag you a... Alexandru Lungu

07:18 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Thanks, the recreate is this:... Constantin Asofiei

07:02 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Also, to clarify things, due to this error regarding before buffers, @DirectAccessHelper.killMultiplex@ is called to ... Alexandru Lungu

06:56 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

TBD: also done ChUI regression testing and there are 14 regressions, while only 1 was expected. Can't tell the reason... Alexandru Lungu

06:51 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

... Alexandru Lungu

06:45 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Do you have a FWD stacktrace when this is happening? Constantin Asofiei

05:31 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Constantin, the exception is triggered by a:
@You may not delete BEFORE-TABLE <name> record. Use ACCEPT-ROW-CHANGES... Alexandru Lungu

04:43 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I found another issue with this test:... Constantin Asofiei

08:28 AM Bug #8878: Jumble sort dictates a full preselect

Artur Școlnic wrote:
> I think Constantin already did that and there were no issues.
If you meant #8878-12 (about... Alexandru Lungu

08:23 AM Bug #8878: Jumble sort dictates a full preselect

I think Constantin already did that and there were no issues. Artur Școlnic

04:30 AM Bug #8878: Jumble sort dictates a full preselect

Artur, I think 8878a should be conversion tested:
* Re-convert a large customer application (create a source basel... Alexandru Lungu

06/28/2024

04:59 PM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

I have some bad news for 8593d:
* After 3 unit test modules that went smooth, the server entered (from the looks of ... Alexandru Lungu

06:51 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Done, 8593d_h2 was merged to rev 49 (and uploaded to artifacts) and 8593d rev 15315 updates build.gradle to 1.49 Constantin Asofiei

06:10 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

> ChUI regression testing.
I have a stable system with 6667i done just recently. I suppose I can move with 6667g s... Alexandru Lungu

05:17 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> For the example provided: I checked your test in 4GL and indeed, @EMPTY TEMP-TABLE@ (simila... Constantin Asofiei

04:59 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Got it!
For the example provided: I checked your test in 4GL and indeed, @EMPTY TEMP-TABLE@ (similar to empty data... Alexandru Lungu

04:00 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru Lungu wrote:
> *Review of 8593d/rev. 15314:*
>
> * @BufferImpl@: The @addDirtyBatchBuffer@ addition is ... Constantin Asofiei

03:56 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

*Review of 8593d/rev. 15314:*
* @BufferImpl@: The @addDirtyBatchBuffer@ addition is a nice catch, but I would like... Alexandru Lungu

03:22 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru: please review. Constantin Asofiei

01:38 AM Bug #8908: Remove CAST operator, if not needed

In order to remove the CAST operator completely, we would have to:
* confirm for sure whether it is needed, now th... Artur Școlnic

01:37 AM Bug #8908 (New): Remove CAST operator, if not needed

Artur Școlnic

06/27/2024

11:21 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

8593d passed ETF testing. Constantin Asofiei

09:06 AM Bug #8593: H2 performance improvement for table 'truncate by multiplex'

Alexandru, I've rebased 8593d from trunk rev 15311 . In 8593d/15314 there are several changes:
* during FILL will a... Constantin Asofiei

06/26/2024

02:50 AM Bug #8878 (Internal Test): Jumble sort dictates a full preselect

Should I do some more testing for 8878a? Artur Școlnic

06/25/2024

02:50 AM Bug #7924 (Internal Test): add support for creating an empty database

Ovidiu Maxiniuc wrote:
> Review of 7924a, rev.15307-15311.
> I saw no problem in the new code except for:
>
> * ... Dănuț Filimon