FWD » Core Development » Database

05/26/2021

08:23 PM Support #5183: database collation

I've reviewed the output of the @cp_coll.p@ program from several environments:
* en_US / 8859-1
** Linux
** Wind... Eric Faulhaber

05/25/2021

03:53 PM Feature #3930: allow session level control over the database(s) which are auto-connected

The functionality is implemented and tested with the test app. Will test with the real customer app.
Committed to ... Igor Skornyakov

05/21/2021

01:39 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> That being said, I'm not sure that code looks right now...what if we had two @database@ keys... Igor Skornyakov

01:33 PM Feature #3930: allow session level control over the database(s) which are auto-connected

That being said, I'm not sure that code looks right now...what if we had two @database@ keys mapped to the same @Sess... Eric Faulhaber

01:13 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> The current @deregisterDatabase@ code is just about cleaning up local data structures when a... Igor Skornyakov

01:12 PM Feature #3930: allow session level control over the database(s) which are auto-connected

The current @deregisterDatabase@ code is just about cleaning up local data structures when a database connection ends... Eric Faulhaber

01:05 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
>
> > As far as I understand the same logic is used for local data... Igor Skornyakov

01:02 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> As far as I understand the same logic is used for local databases as well. Should I fix i... Eric Faulhaber

12:56 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> IIRC, for remote connections, we create one virtual session between server A and server B, r... Igor Skornyakov

12:40 PM Feature #3930: allow session level control over the database(s) which are auto-connected

IIRC, for remote connections, we create one virtual session between server A and server B, regardless of how many use... Eric Faulhaber

06:51 AM Feature #3930: allow session level control over the database(s) which are auto-connected

I'm confused with the following code in the @ConnectionManager.disconnectImmediately()@:... Igor Skornyakov

03:04 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Constantin Asofiei wrote:
> Igor, as I mentioned - Agents can reset their context (for State-reset mode) - in this c... Igor Skornyakov

02:01 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor, as I mentioned - Agents can reset their context (for State-reset mode) - in this case, I think these databases ... Constantin Asofiei

12:46 PM Support #5183 (WIP): database collation

Eric Faulhaber

05/20/2021

05:29 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Initialization of the transient auto-connected databases in @StandardServer.standardEntry()@ (before @startupProc@ p... Igor Skornyakov

02:19 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Constantin: How do we avoid the @StandardServer.standardEntry()@ connection logic when it is call... Constantin Asofiei

02:17 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Constantin, where would the disconnect occur? I don't think @ContextLocal.cleanup@ would wor... Constantin Asofiei

02:00 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Constantin: How do we avoid the @StandardServer.standardEntry()@ connection logic when it is called for an appserver ... Greg Shah

02:00 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> But to your question: Constantin was answering in the context of appserver agents. For non... Eric Faulhaber

01:59 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> But to your question: Constantin was answering in the context of appserver agents. For non-a... Igor Skornyakov

01:57 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Eric Faulhaber wrote:
> > Igor Skornyakov wrote:
> >
> > > Eric: is it correct that the... Eric Faulhaber

01:51 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
>
> > Eric: is it correct that the entry point mentioned by Consta... Igor Skornyakov

01:48 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Eric: is it correct that the entry point mentioned by Constantin is the only one where th... Eric Faulhaber

01:29 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Constantin Asofiei wrote:
> Greg Shah wrote:
> > Constantin: Is there a different entry point for appserver agents?... Igor Skornyakov

01:27 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor, I would follow Greg's advice here instead of my earlier comments on the @SessionListen... Igor Skornyakov

01:26 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> I don't object to the non-database changes in @BaseSession@. But I wonder if we need such a thin... Igor Skornyakov

01:24 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Constantin: Is there a different entry point for appserver agents?
In case of appserver Agents, ... Constantin Asofiei

01:21 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Instead of looking at the session level stuff (which is generic), please look at the places in ... Eric Faulhaber

01:18 PM Feature #3930: allow session level control over the database(s) which are auto-connected

> How can we avoid any changes at least in the @RouterSessionManager@ if the databases are supposed to be connected o... Greg Shah

01:18 PM Feature #3930: allow session level control over the database(s) which are auto-connected

The way I would approach this would be to:
# Take a working, test configuration which currently connects a databas... Eric Faulhaber

01:11 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> I think you were on the right track with using the @SessionListener@ interface. If we find b... Igor Skornyakov

01:10 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Greg Shah wrote:
> > There should not be any changes in the @com.goldencode.p2j.net@ packa... Eric Faulhaber

12:54 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> There should not be any changes in the @com.goldencode.p2j.net@ package. It is not appropriate t... Igor Skornyakov

12:43 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> There should not be any changes in the @com.goldencode.p2j.net@ package. It is not appropriate t... Igor Skornyakov

12:42 PM Feature #3930: allow session level control over the database(s) which are auto-connected

There should not be any changes in the @com.goldencode.p2j.net@ package. It is not appropriate to have database-spec... Greg Shah

12:28 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> It is hard to know what is going on without seeing the changes you've made. If they are most... Igor Skornyakov

12:21 PM Feature #3930: allow session level control over the database(s) which are auto-connected

It is hard to know what is going on without seeing the changes you've made. If they are mostly isolated to @DatabaseM... Eric Faulhaber

11:05 AM Feature #3930: allow session level control over the database(s) which are auto-connected

I've implemented the session-local databases' activation in the @SessionListener.initialize()@ method of the @Session... Igor Skornyakov

09:49 PM Feature #3930: allow session level control over the database(s) which are auto-connected

From Igor via email:
---
At this moment I've implemented most of the functionality except timed meta-database d... Eric Faulhaber

05/19/2021

01:00 PM Feature #3930: allow session level control over the database(s) which are auto-connected

I agree. The disabled code related to Hibernate can be dropped at this moment.
Ovidiu Maxiniuc

12:23 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Eric Faulhaber wrote:
> > These were left over from the shift from Hibernate. I think at t... Eric Faulhaber

12:15 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
> > When I'm trying to connect to a database in the session start (i... Igor Skornyakov

12:15 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> These were left over from the shift from Hibernate. I think at this point, they can be remov... Igor Skornyakov

12:08 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> When I'm trying to connect to a database in the session start (in my case it was a connecti... Eric Faulhaber

12:06 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> There is a lot of commented code in the @DatabaseManager@. This makes reading and understan... Eric Faulhaber

05/17/2021

06:47 PM Bug #5377: H2 might block in certain conditions

The investigations from #4479 proved that some methods were not correctly handling locking on its internal @meta@ tab... Ovidiu Maxiniuc

06:38 PM Bug #5377 (New): H2 might block in certain conditions

Ovidiu Maxiniuc

06:14 PM Bug #5376: Reintroduce UDFs for handling special date-related operators

The investigation from #5325 lead to conclusion that in the current revision FWD lacks support for some @+@ / @-@ ope... Ovidiu Maxiniuc

05:28 PM Bug #5376 (New): Reintroduce UDFs for handling special date-related operators

Ovidiu Maxiniuc

05/15/2021

06:19 PM Feature #3930: allow session level control over the database(s) which are auto-connected

When I'm trying to connect to a database in the session start (in my case it was a connection from the application se... Igor Skornyakov

05/14/2021

06:13 PM Feature #3930: allow session level control over the database(s) which are auto-connected

There is a lot of commented code in the @DatabaseManager@. This makes reading and understanding the code difficult. D... Igor Skornyakov

05/12/2021

03:18 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Since we are rewriting this logic now, I think we should plan for this eventuality. I tried ... Igor Skornyakov

03:15 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> At this moment we initialize the meta-database for all configured local databases regardl... Eric Faulhaber

02:36 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> The concept of "transient" databases is old and is not really something I care to preserve. ... Igor Skornyakov

02:21 PM Feature #3930: allow session level control over the database(s) which are auto-connected

The concept of "transient" databases is old and is not really something I care to preserve. It was developed back whe... Eric Faulhaber

01:24 PM Feature #3930: allow session level control over the database(s) which are auto-connected

I'm trying to understand the logic around reference counting with @DatabseManager.transientDatabases@. It looks prett... Igor Skornyakov

04:23 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Actually, we really do implement this same limit. Our parser will not properly parse database re... Igor Skornyakov

05/11/2021

06:57 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Actually, we really do implement this same limit. Our parser will not properly parse database references unless the ... Greg Shah

06:30 PM Feature #3930: allow session level control over the database(s) which are auto-connected

> > BTW: how do we support this restriction? It looks weird to me.
> I do not think we do. FWD connects the database... Greg Shah

05:59 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Ovidiu Maxiniuc wrote:
> Sorry, I am afraid I was not very clear here. What I intended to say is that FWD does not s... Igor Skornyakov

05:30 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Ovidiu Maxiniuc wrote:
> > Igor Skornyakov wrote:
> > > BTW: how do we support this restr... Ovidiu Maxiniuc

05:21 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric/Greg.
At this moment we initialize the meta-database for all configured local databases regardless of the @lo... Igor Skornyakov

05:06 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Ovidiu Maxiniuc wrote:
> Igor Skornyakov wrote:
> > BTW: how do we support this restriction? It looks weird to me.
... Igor Skornyakov

05:02 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> BTW: how do we support this restriction? It looks weird to me.
I do not think we do. FWD c... Ovidiu Maxiniuc

04:37 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Ovidiu Maxiniuc wrote:
> Just as a remainder, although you are probably already aware of this: you cannot use the da... Igor Skornyakov

04:34 PM Feature #3930: allow session level control over the database(s) which are auto-connected

These things happen. I did this myself and not even once.
Just as a remainder, although you are probably already a... Ovidiu Maxiniuc

04:25 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Ovidiu Maxiniuc wrote:
> > I was looking at the last line of exception you posted. It says... Igor Skornyakov

04:16 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Ovidiu Maxiniuc wrote:
> I was looking at the last line of exception you posted. It says: @org.h2.jdbc.JdbcSQLSyntax... Igor Skornyakov

04:15 PM Feature #3930: allow session level control over the database(s) which are auto-connected

I was looking at the last line of exception you posted. It says: @org.h2.jdbc.JdbcSQLSyntaxErrorException@.
Yet, H2 ... Ovidiu Maxiniuc

04:09 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Ovidiu Maxiniuc wrote:
> Igor,
> Does the database already exist? Because if there is no physical file at the mome... Igor Skornyakov

04:05 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor,
Does the database already exist? Because if there is no physical file at the moment the connection is made, H... Ovidiu Maxiniuc

03:44 PM Feature #3930: allow session level control over the database(s) which are auto-connected

It seems that at this moment a @CONNECT@ statement support for the local databases is broken.
For example, if the @h... Igor Skornyakov

06:34 PM Bug #4945 (Feedback): Optimize indexes for H2 dialect

My changes were committed to a newly created bzr repository located at @devsrv01:/opt/secure/code/p2j_repo/fwd-h2/@, ... Ovidiu Maxiniuc

03:10 PM Bug #5306: AdaptiveQuery sets a found record into the buffer twice in dynamic mode

I have investigated this issue but I cannot reproduce it. I took the isolated testcase from #3814-406, converted and ... Ovidiu Maxiniuc

05/10/2021

03:40 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> > Actually, my question was not about the name but if you're planning to implement auto-conn... Igor Skornyakov

03:38 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Actually, my question was not about the name but if you're planning to implement auto-con... Eric Faulhaber

03:30 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Greg Shah wrote:
>
> > Change the name of the directory node in the new configuration to ... Igor Skornyakov

03:27 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Change the name of the directory node in the new configuration to make it less confusing.
I ... Eric Faulhaber

03:22 PM Feature #3930: allow session level control over the database(s) which are auto-connected

At this time, we don't care about this "auto-connect" 4GL feature at all. It has nothing to do with our work and can... Greg Shah

02:44 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> This looks not exactly the same we've discussed here. Is this difference intentional?
... Eric Faulhaber

02:18 PM Feature #3930: allow session level control over the database(s) which are auto-connected

As I can see in the OpenEdge documentation (https://docs.progress.com/bundle/openedge-programmimg-interfaces-117/page... Igor Skornyakov

09:15 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Greg Shah wrote:
> > No. In my opinion, this is not a very risky change. If you disagree... Greg Shah

08:28 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> No. In my opinion, this is not a very risky change. If you disagree, please help us understand ... Igor Skornyakov

08:08 AM Feature #3930: allow session level control over the database(s) which are auto-connected

No. In my opinion, this is not a very risky change. If you disagree, please help us understand what parts are risky. Greg Shah

08:02 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Don't get hung up on this aspect of things. The helper methods for lookup in the directory will ... Igor Skornyakov

07:54 AM Feature #3930: allow session level control over the database(s) which are auto-connected

> But in your examples, we have only one configuration.
His example is the equivalent of the current approach whi... Greg Shah

02:25 PM Feature #5355: _Usertablestat metadata improvements

Eric Faulhaber wrote:
> Support for this VST was mostly implemented in #3814, however a few things remain to be addr... Igor Skornyakov

01:43 PM Feature #5355: _Usertablestat metadata improvements

Regarding the temp-table support, there are cases where conversion optimizes 4GL code which deletes temp-table record... Eric Faulhaber

01:32 PM Feature #5355: _Usertablestat metadata improvements

Support for this VST was mostly implemented in #3814, however a few things remain to be addressed:
* add update co... Eric Faulhaber

01:25 PM Feature #5355 (WIP): _Usertablestat metadata improvements

Eric Faulhaber

01:34 PM Feature #3814 (Closed): more schema metadata

Any remaining work related to the @_Usertablestat@ VST will be managed in #5355. Eric Faulhaber

05/09/2021

03:56 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
>
> > My question was about the session attribute the selection of... Igor Skornyakov

05/08/2021

06:06 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> My question was about the session attribute the selection of the connection profile shoul... Eric Faulhaber

02:26 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
> It is up to the person who configures the directory for a given pr... Igor Skornyakov

02:08 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Igor Skornyakov wrote:
> Eric, thank you very much for the detailed explanations.
You're welcome.
> However,... Eric Faulhaber

09:45 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Let's take a step back and focus on the actual goal of this task, which is not so much about... Igor Skornyakov

05/07/2021

05:59 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Let's take a step back and focus on the actual goal of this task, which is not so much about partitioned applications... Eric Faulhaber

01:40 PM Feature #3930: allow session level control over the database(s) which are auto-connected

It is difficult to follow the discussion in #2475. It seems that it is mostly about conversion.
Here is my underst... Igor Skornyakov

01:01 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Eric Faulhaber wrote:
> Any work done so far on this concept is in #2475. There is no other documentation per se.
>... Igor Skornyakov

12:49 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Any work done so far on this concept is in #2475. There is no other documentation per se.
The requirements which l... Eric Faulhaber

12:34 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> It is the same thing we have now. I'm just referring to the *concept* that it is a logical and n... Igor Skornyakov

12:20 PM Feature #3930: allow session level control over the database(s) which are auto-connected

It is the same thing we have now. I'm just referring to the *concept* that it is a logical and not physical session. Greg Shah

12:06 PM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> Your work is about the logical database session, which is different.
Sorry, what is a _logical... Igor Skornyakov

12:00 PM Feature #3930: allow session level control over the database(s) which are auto-connected

> I understood that in FWD the CONNECT does not mean establishing the JDBC connection to the relational database,
... Greg Shah

11:27 AM Feature #3930: allow session level control over the database(s) which are auto-connected

Greg Shah wrote:
> To implement this, we must separate database auto-connect from server startup. Then we must allo... Igor Skornyakov

02:05 PM Bug #5322 (WIP): FqlToSqlConverter fails to identify property

Exactly. Ovidiu Maxiniuc

11:51 AM Bug #5322: FqlToSqlConverter fails to identify property

Ovidiu, I understand this logs a warning, but it is not detrimental, correct? If so, let's lower the priority. Eric Faulhaber

02:45 AM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Code review 3814a/12395:
>
> I don't understand why you renamed the original @RecordBuffe... Igor Skornyakov

05/06/2021

07:58 PM Feature #3814: more schema metadata

Code review 3814a/12395:
I don't understand why you renamed the original @RecordBuffer@ method @delete@ to @_delet... Eric Faulhaber

04:09 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> See the call hierarchy for @RecordBuffer.delete(Supplier<logical> valexp, Supplier<character... Igor Skornyakov

03:02 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Igor Skornyakov wrote:
>
> > I cannot find a better place for updating the 'delete' count... Igor Skornyakov

02:56 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> I cannot find a better place for updating the 'delete' counter (see #3814-413).
The de... Eric Faulhaber

02:33 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> I believe a good solution is to:
>
> * remove all flushing logic from the various @P2JQue... Igor Skornyakov

05/05/2021

08:21 PM Bug #5305 (Closed): dynamic function detection in RuntimeJastInterpreter

Fixed in 3821c/12370. Confirmed the fix in a customer scenario. Eric Faulhaber

05/04/2021

07:52 PM Bug #5305 (WIP): dynamic function detection in RuntimeJastInterpreter

Eric Faulhaber

07:41 PM Bug #5322: FqlToSqlConverter fails to identify property

There are known warnings printed by @FqlToSqlConverter@ into log about @Property '<identifier>' not found to extract ... Ovidiu Maxiniuc

07:40 PM Bug #5322 (WIP): FqlToSqlConverter fails to identify property

Ovidiu Maxiniuc

05/03/2021

07:16 PM Bug #5305: dynamic function detection in RuntimeJastInterpreter

Eric Faulhaber wrote:
> Is it safe for me to modify the @DYNAMIC_CALL@ condition to check for any of these variants?... Ovidiu Maxiniuc

03:39 PM Bug #5305: dynamic function detection in RuntimeJastInterpreter

Ovidiu, while recreating a separate issue, I am getting many messages like this in the terminal:... Eric Faulhaber

03:30 PM Bug #5305 (Closed): dynamic function detection in RuntimeJastInterpreter

Eric Faulhaber

06:01 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> > For now, let's get the flushing part of this settled and we'll consider this task compl... Eric Faulhaber

05:16 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> They may do that for performance considerations, and we may need to do something similar if ... Igor Skornyakov

04:51 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> Eric Faulhaber wrote:
> > My point is: isn't this already working for temp-tables, without... Eric Faulhaber

04:06 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> My point is: isn't this already working for temp-tables, without further implementation? Is ... Igor Skornyakov

03:55 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> Well, I planned to ignore the statistics updates for temp-tables but it looks that I forg... Eric Faulhaber

03:50 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> AFAICT, we make no distinction between persistent and temp-tables in the current implementat... Igor Skornyakov

03:27 PM Feature #3814: more schema metadata

AFAICT, we make no distinction between persistent and temp-tables in the current implementation. So, I am just wonder... Eric Faulhaber

03:17 PM Feature #3814: more schema metadata

Greg Shah wrote:
> > I do not think that it is a big deal. But I need to analyze how 4GL supports counters for the t... Igor Skornyakov

03:11 PM Feature #3814: more schema metadata

> I do not think that it is a big deal. But I need to analyze how 4GL supports counters for the temp-tables first.
... Greg Shah

03:08 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> > Unfortunatelly, FqlToSQLConverter is not used in the @AdaptiveQuery@ and @RandomAccessQuer... Igor Skornyakov

02:53 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> Unfortunatelly, FqlToSQLConverter is not used in the @AdaptiveQuery@ and @RandomAccessQue... Eric Faulhaber

06:17 AM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> > Updating the counter in the @BufferImpl.deleteRecord@ works for the DELETE statement as w... Igor Skornyakov

05:17 AM Feature #3814: more schema metadata

Code review 3821c/12341:
Igor Skornyakov wrote:
> Updating the counter in the @BufferImpl.deleteRecord@ works ... Eric Faulhaber

04:48 PM Bug #5307: AdaptiveQuery goes into dynamic mode too aggressively

While implementing #3814, we used the following code in a test case:... Eric Faulhaber

03:58 PM Bug #5307 (WIP): AdaptiveQuery goes into dynamic mode too aggressively

Eric Faulhaber

03:53 PM Bug #5306: AdaptiveQuery sets a found record into the buffer twice in dynamic mode

When @AdaptiveQuery@ finds a record in dynamic mode, it sets it into @RecordBuffer@ via @setRecord@ twice. This was d... Eric Faulhaber

03:44 PM Bug #5306 (Closed): AdaptiveQuery sets a found record into the buffer twice in dynamic mode

Eric Faulhaber

04/30/2021

04:23 PM Feature #3814: more schema metadata

> Code review 3814a rev 12304-12307, part 2.
>
> *Functional:*
>
> I understand your term for persisting the st... Igor Skornyakov

07:25 AM Feature #3814: more schema metadata

I've noticed one more discrepancy between 4GL and FWD.
If a record is updated twice w/o commit like this:... Igor Skornyakov

03:51 AM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> The root cause of the double-counting of the retrieve statistic is that we are setting the r... Igor Skornyakov

04/29/2021

06:36 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> Eric Faulhaber wrote:
> > This observation is not directly to the problem at hand, but it ... Eric Faulhaber

06:09 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> This observation is not directly to the problem at hand, but it is curious that there isn't ... Igor Skornyakov

05:34 PM Feature #3814: more schema metadata

This observation is not directly to the problem at hand, but it is curious that there isn't anything I see about your... Eric Faulhaber

05:12 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Please post the original 4GL code and the converted code (not the whole program, just the sn... Igor Skornyakov

04:58 PM Feature #3814: more schema metadata

Please post the original 4GL code and the converted code (not the whole program, just the snippets relevant to this q... Eric Faulhaber

03:54 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Is the same record being set into the buffer in both calls to @RecordBuffer.setRecord@?
I... Igor Skornyakov

03:51 PM Feature #3814: more schema metadata

Is the same record being set into the buffer in both calls to @RecordBuffer.setRecord@? Eric Faulhaber

03:43 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Sorry I forgot to respond to this yesterday. Are you sure these code paths are not mutually ... Igor Skornyakov

03:38 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> Indeed, @RecordBuffer.setRecord@ looks to be a better place. However I've encountered a p... Eric Faulhaber

03:00 PM Feature #3814: more schema metadata

@GET-CURRENT@ has an interesting effect on the @_UserTableStat_read@ counter.
* The combination @GET-CURRENT@/@DEL... Igor Skornyakov

04/28/2021

12:57 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Code review 3814a rev 12304-12307, part 1.
> *Update*
>
> Does the VST count each indi... Igor Skornyakov

06:31 AM Feature #3814: more schema metadata

I've noticed what looks like incompatibility.
In 4GL if inside a transaction block two records are inserted where ... Igor Skornyakov

04/27/2021

04:55 PM Feature #3814: more schema metadata

It looks that if a trigger rejects CREATE/DELETE/FIND/WRITE operation the corresponding @_UserTableStat@ counter is n... Igor Skornyakov

12:46 PM Bug #5289: Misc. problems with triggers

In addition.
The REPLICATION triggers are not invoked in FWD. See the same test as above. Just add the following tr... Igor Skornyakov

07:20 AM Bug #5289: Misc. problems with triggers

How to reproduce:
Just run @stat1.p@ with our standard "customer" table and two triggers (see @sftp://<userid>@xfer... Igor Skornyakov

04/26/2021

05:14 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Code review 3814a rev 12304-12307, part 2.
>
> *Functional:*
>
> I understand your ter... Igor Skornyakov

05:01 PM Feature #3814: more schema metadata

It seems we have enough trigger-related items to warrant a separate issue. Please open a normal priority bug issue in... Eric Faulhaber

04:58 PM Feature #3814: more schema metadata

Code review 3814a rev 12304-12307, part 2.
*Functional:*
I understand your term for persisting the state from t... Eric Faulhaber

04:56 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> In some situations, the @FIND@ rejected by the trigger causes NPE in the @PreselectQuery@ l... Igor Skornyakov

04:36 PM Feature #3814: more schema metadata

In some situations, the @FIND@ rejected by the trigger causes NPE in the @PreselectQuery@ line 5630 (@results@ field ... Igor Skornyakov

03:59 PM Feature #3814: more schema metadata

Eric Faulhaber wrote:
> Does the 4GL return @false/no@ from this BUFFER-CREATE method call? If so, then that looks l... Igor Skornyakov

03:58 PM Feature #3814: more schema metadata

Igor Skornyakov wrote:
> After rebasing 3814a to 3821c/12334 the trigger is working but contrary to 4GL behavior the... Igor Skornyakov

03:53 PM Feature #3814: more schema metadata

Does the 4GL return @false/no@ from this BUFFER-CREATE method call? If so, then that looks like a bug in our create t... Eric Faulhaber

03:32 PM Feature #3814: more schema metadata

After rebasing 3814a to 3821c/12334 the trigger is working but contrary to 4GL behavior the converted @RETURN ERROR@ ... Igor Skornyakov

03:13 PM Feature #3814: more schema metadata

Constantin Asofiei wrote:
> I've tested with 3821c:
> * added the @TABLE-TRIGGER "CREATE" NO-OVERRIDE PROCEDURE "cr... Igor Skornyakov

03:06 PM Feature #3814: more schema metadata

Constantin Asofiei wrote:
> I've tested with 3821c:
> * added the @TABLE-TRIGGER "CREATE" NO-OVERRIDE PROCEDURE "cr... Igor Skornyakov

03:01 PM Feature #3814: more schema metadata

I've tested with 3821c:
* added the @TABLE-TRIGGER "CREATE" NO-OVERRIDE PROCEDURE "create-trg.p" CRC "30042" @ line ... Constantin Asofiei

02:51 PM Feature #3814: more schema metadata

Here is the result of the table conversion:... Igor Skornyakov

02:45 PM Feature #3814: more schema metadata

Constantin Asofiei wrote:
> Please post here the create-trg.p program and the program which tests this trigger.
>
... Igor Skornyakov

02:42 PM Feature #3814: more schema metadata

Please post here the create-trg.p program and the program which tests this trigger.
I assume you are *not* running... Constantin Asofiei

02:41 PM Feature #3814: more schema metadata

Add the path inside your trigger definition in the @.df@ otherwise it cannot be found in the propath since the propat... Greg Shah

02:36 PM Feature #3814: more schema metadata

Greg Shah wrote:
> OK, so your code should not just be dropped in the root directory of the project. It should be i... Igor Skornyakov

02:28 PM Feature #3814: more schema metadata

OK, so your code should not just be dropped in the root directory of the project. It should be in some structured pa... Greg Shah

02:24 PM Feature #3814: more schema metadata

Greg Shah wrote:
> Those are both conversion-time values and they have no meaning at runtime. What project are you ... Igor Skornyakov

02:22 PM Feature #3814: more schema metadata

Those are both conversion-time values and they have no meaning at runtime. What project are you using? Greg Shah

02:20 PM Feature #3814: more schema metadata

Constantin Asofiei wrote:
> What is the PROPATH in directory.xml? That one is used at runtime.
>
> Is @create-trg... Igor Skornyakov

02:11 PM Feature #3814: more schema metadata

What is the PROPATH in directory.xml? That one is used at runtime.
Is @create-trg.p@ in some folder? If so, add th... Constantin Asofiei

02:07 PM Feature #3814: more schema metadata

Greg Shah wrote:
> It must be found in the propath.
Greg.
In my environment the @PROPATH@ is defined in @p2j.cfg... Igor Skornyakov

01:50 PM Feature #3814: more schema metadata

It must be found in the propath. Greg Shah

01:21 PM Feature #3814: more schema metadata

I've added a trigger to the test table:... Igor Skornyakov

05:13 PM Bug #5289: Misc. problems with triggers

I've found a number of issues with triggers' support in FWD:
* In 4GL if the @CREATE@ trigger rejects the operation ... Igor Skornyakov

05:05 PM Bug #5289 (WIP): Misc. problems with triggers

Igor Skornyakov

05:06 PM Feature #4175 (Closed): add support for -1 (single user) option for CONNECT statement

I did some additional code cleanup in 3821c/12333. Considering #5279, this issue can be closed. Eric Faulhaber

09:04 AM Feature #4175: add support for -1 (single user) option for CONNECT statement

Eric Faulhaber wrote:
> Warning added, functionally looks fine, but please follow coding standards when formatting (... Igor Skornyakov

08:21 AM Feature #4175: add support for -1 (single user) option for CONNECT statement

Code review 3821c/12313:
Warning added, functionally looks fine, but please follow coding standards when formattin... Eric Faulhaber