Activity
From 03/15/2019 to 04/13/2019
04/13/2019
- 06:11 PM Feature #3809: ProDataSet support
- Ovidiu, now that branch 3750b has been merged to trunk 11302, I have rebased branch 3809a (currently at revision 1130...
04/12/2019
- 06:15 PM Feature #1587: implement full support for word indexes
- It should be noted that error handling for the UDF version of CONTAINS support is not quite right. Any error in the s...
- 06:07 PM Feature #3809: ProDataSet support
- Greg Shah wrote:
> I'm not suggesting to use variable types. The idea is to use the @tempIdx@ value in @SymbolResol... - 01:29 PM Feature #3809: ProDataSet support
- Constantin Asofiei wrote:
> Ovidiu Maxiniuc wrote:
> > > Also, do you have tests for shared datasets? I see no cha... - 01:25 PM Feature #3809: ProDataSet support
- Ovidiu, please add conversion and stub support for BUFFER:DATA-SOURCE attribute.
- 01:22 PM Feature #3809: ProDataSet support
- Ovidiu Maxiniuc wrote:
> > Also, do you have tests for shared datasets? I see no changes in SharedVariableManager r... - 01:20 PM Feature #3809: ProDataSet support
- Constantin Asofiei wrote:
> Hmm... related to OO, we need tests at least for:
> * DATASET and DATASET-HANDLE as met... - 01:15 PM Feature #3809: ProDataSet support
- Ovidiu Maxiniuc wrote:
> Constantin Asofiei wrote:
> > Ovidiu, do your testcases cover DATASET and DATASET-HANDLE a... - 01:03 PM Feature #3809: ProDataSet support
- Constantin Asofiei wrote:
> Ovidiu, do your testcases cover DATASET and DATASET-HANDLE as OO method arguments?
I ... - 12:59 PM Feature #3809: ProDataSet support
- Ovidiu, do your testcases cover DATASET and DATASET-HANDLE as OO method arguments?
- 12:31 PM Feature #3809: ProDataSet support
- I'm not suggesting to use variable types. The idea is to use the @tempIdx@ value in @SymbolResolver@ (see @addWrappe...
- 12:17 PM Feature #3809: ProDataSet support
- Greg Shah wrote:
> Code Review Task Branch 3809b Revision 11605
>
> 1. @annotations/datasets.rules@ implements li...
04/11/2019
- 03:00 PM Feature #3809: ProDataSet support
- > Is it OK to add move within RESERVED boundaries?
That is still for keywords. Instead, please put them inside th... - 02:34 PM Feature #3809: ProDataSet support
- Greg Shah wrote:
> Code Review Task Branch 3809b Revision 11605
Thank you for the review. It helps me fix some issu... - 11:26 AM Feature #3809: ProDataSet support
- Code Review Task Branch 3809b Revision 11605
1. @annotations/datasets.rules@ implements linking of dataset/datasou... - 10:31 AM Feature #3809: ProDataSet support
- Branch 3809b was updated. Current revision is (LE) 11605 . It contains fixes for regression introduces the last two d...
04/10/2019
- 02:43 AM Feature #3809: ProDataSet support
- Yes, there are two ambiguity issues.
I haven't added @options { generateAmbigWarnings = false; }@ to hide them yet...
04/09/2019
- 06:42 PM Feature #3809: ProDataSet support
- Ovidiu, antlr reports a lot of errors when I build 3809b/11596. Is that expected?
04/08/2019
- 06:12 PM Feature #4036: catch database UDF support up to current built-in function support level
- We have added support for many 4GL built-in functions over the past years. We need to review which of these can be us...
- 06:09 PM Feature #4036 (New): catch database UDF support up to current built-in function support level
- 05:27 PM Bug #4035: newly created record is not flushed at the right time
- The following test case illustrates the problem.
Constantin Asofiei wrote:
> Good news, I have a recreate; I thin... - 05:09 PM Bug #4035 (Closed): newly created record is not flushed at the right time
- 12:24 PM Feature #3809: ProDataSet support
- ...
- 11:41 AM Feature #3809: ProDataSet support
- Greg, yes, these makes sense, but I am trying to understand if a @dataset-handle@ parameter can receive a generic @ha...
- 11:33 AM Feature #3809: ProDataSet support
- Please see #3751-492 for some interesting details about @TABLE@ and @TABLE-HANDLE@ parameters for methods. It is my ...
- 11:04 AM Feature #3809: ProDataSet support
- Ovidiu Maxiniuc wrote:
> For the moment the @DATASET-HANDLE@ is not recognized as special data type. I am adding now... - 10:31 AM Feature #3809: ProDataSet support
- Consider the following ABL code:...
- 06:03 AM Feature #3809: ProDataSet support
- Ovidiu, please post a small snapshot of how DATASET and DATASET-HANDLE parameters will convert.
Also, I assume tha...
04/05/2019
- 04:26 PM Feature #3809: ProDataSet support
- Revision 11591.
Two important changes in converted code:
* the static definition of @DataSet@ was using buffer f... - 01:37 PM Feature #3809: ProDataSet support
- Revision 11587.
All provided files in Test Suite converted and compiled successfully. I updated the project and wi... - 11:16 PM Feature #3809: ProDataSet support
- Eric Faulhaber wrote:
> Please consider the recent refactoring I did with BUFFER-COPY and BUFFER-COMPARE.
BTW, I'...
04/04/2019
- 05:02 PM Feature #3809: ProDataSet support
- [Originally sent via email, but I wanted to document here...]:
Please consider the recent refactoring I did with B... - 04:59 PM Feature #3809: ProDataSet support
- Revision 11584.
Added missing method declarations. Added attributes and methods from existing interfaces. Fixed so...
04/03/2019
- 03:44 PM Feature #3809: ProDataSet support
- Revision 11583.
The project converts and generated sources compiles without errors.
I am going tomorrow to check ... - 12:21 PM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Constantin Asofiei wrote:
> The table should have more than one record :-)
Yes, it should be a "real" table with ... - 10:05 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- My suggestion was to test without the runtime change. Just write the conditional around the @FIND@ in the converted ...
- 09:42 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- We are talking about an WHERE clause which is known that can always match on an unique record. If the lambda express...
- 09:20 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Constantin Asofiei wrote:
> > How do we determine this efficiently? Within the current session is not too bad, but a... - 09:09 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Constantin Asofiei wrote:
> No, the lock doesn't matter.
My point is that holding a lock on the record is the only ... - 08:52 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Regarding Greg's #4033-2 - the idea here is to omit the conversion changes now, and just test a runtime change (which...
- 08:50 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Eric Faulhaber wrote:
> Is the assumption that the current session holds a lock on the record between FIND [ FIRST ]... - 08:47 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Eric Faulhaber wrote:
> > * The current record in the buffer meets the unique criteria.
> How do we determine this ... - 08:23 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- Is the assumption that the current session holds a lock on the record between FIND [ FIRST ] calls?
If NO-LOCK, wo... - 08:18 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- This is an interesting idea, but the details of the execution present some questions.
Greg Shah wrote:
> * The cu... - 08:06 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- The table should have more than one record :-)
- 07:17 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- One idea to test this:
1. Write a 4GL testcases that does 10000 of the same @FIND@ in a loop using a unique index ... - 07:08 AM Feature #4033: avoid query for FIND unique when the current record in the buffer would match the ...
- In discussions with a customer, it was noted that in the 4GL a @FIND@ (or @FIND FIRST@ etc) is very fast when the sam...
- 06:49 AM Feature #4033 (WIP): avoid query for FIND unique when the current record in the buffer would matc...
04/02/2019
- 03:16 PM Feature #3809: ProDataSet support
- Revision 11577.
Added missing parent-id-relation option and resolved conflicts with other variables.
Compiling th... - 10:52 PM Feature #4030: improve CompoundQuery optimizer
- Some time ago, I implemented a runtime optimizer for @CompoundQuery@, which would analyze the components of the compo...
- 10:36 PM Feature #4030 (New): improve CompoundQuery optimizer
04/01/2019
- 06:40 PM Feature #4024: PostgreSQL shared memory JDBC driver
- This concept is a more outside-the-box corollary to the idea behind #4012. In a system where lots of round trips to t...
- 06:22 PM Feature #4024 (New): PostgreSQL shared memory JDBC driver
- 06:20 PM Feature #4023: PostgreSQL query planning optimization
- To date, we have not made a concentrated effort to tune PostgreSQL's query planner for FWD's use, beyond some basics:...
- 06:12 PM Feature #4023 (New): PostgreSQL query planning optimization
- 06:10 PM Feature #4022: upgrade to a newer PostgreSQL release
- At the time of this writing, the most recent production PostgreSQL release is 11.2. We are currently working primaril...
- 05:48 PM Feature #4022 (WIP): upgrade to a newer PostgreSQL release
- 03:37 PM Feature #4021: better use of session-related resources such as PreparedStatements and caches
- At the moment, this task is somewhat speculative, in that I'm not sure how well we are amortizing the cost of setting...
- 03:22 PM Feature #4021 (New): better use of session-related resources such as PreparedStatements and caches
- 03:19 PM Feature #4020: reduce work Hibernate does flushing session
- Each time Hibernate does something that relies on the state of the database to be consistent with changes in the stat...
- 02:46 PM Feature #4020 (Rejected): reduce work Hibernate does flushing session
- 02:51 PM Feature #3809: ProDataSet support
- Committed r11576 of branch 3809b.
At this moment all the sources provided convert. However, the code is not 100% c... - 11:35 AM Feature #4019: rework dirty database implementation to use persistent database
- This task is meant to address two issues with the current dirty database implementation:
* it is limited, in that ... - 11:06 AM Feature #4019 (New): rework dirty database implementation to use persistent database
- 11:02 AM Feature #4018: possibly move temp-table support into persistent database
- We already do a form of optimization for @CompoundQuery@, where we attempt to determine whether a multi-table query w...
- 10:55 AM Feature #4018: possibly move temp-table support into persistent database
- In the early days of FWD, temp-table support was implemented in the persistent database (e.g., PostgreSQL). To avoid ...
- 10:25 AM Feature #4018 (New): possibly move temp-table support into persistent database
- 11:52 PM Feature #4012 (Review): replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as ...
- Greg, I think we can close this.
- 11:50 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Greg Shah wrote:
> What about PGCONF changes? Are AF_UNIX sockets enabled by default?
They are, at least in the ... - 11:43 PM Feature #4016: replace portions of Hibernate which are non-performant
- We maintain a lot of information in FWD's persistence layer which is redundant with information maintained by Hiberna...
- 11:35 PM Feature #4016 (Closed): replace portions of Hibernate which are non-performant
- 10:35 PM Feature #4015: upgrade to a newer Hibernate release
- As [[FWD:Patching_and_Building_Customized_3rd_Party_Libraries#Hibernate|documented here]], we currently are at a cust...
- 10:14 PM Feature #4015 (Rejected): upgrade to a newer Hibernate release
03/29/2019
- 07:17 AM Feature #3809: ProDataSet support
- I did the refactoring as you suggested. I think it's a good solution.
I am working now on the conversion of stati...
03/28/2019
- 02:04 AM Feature #3809: ProDataSet support
- I've previously created a class called @DataSource@ for use with XML and JSON serialization. However, this name alrea...
03/27/2019
- 10:12 AM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- > Don't we also need to add code (jar(s) and a native library) to the system to make this work?
I see you've made ... - 07:25 AM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- > No other change is needed.
Don't we also need to add code (jar(s) and a native library) to the system to make th... - 12:24 AM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Greg Shah wrote:
> https://github.com/fiken/junixsocket
>
> This last one has a PostgresqlAFUNIXSocketFactory tha...
03/26/2019
- 04:35 PM Feature #4011: database/persistence layer performance improvements
- This issue is for tracking ideas about improving the performance of the database and the persistence framework in FWD...
- 03:03 PM Feature #4011 (Closed): database/persistence layer performance improvements
- 04:14 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Using shared memory, we could eliminate the 4 kernel space transitions (write() goes down into the kernel and returns...
- 04:13 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Early implementations of the Progress Database Server implemented connectivity using shared memory. This design, com...
- 04:08 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Eric Faulhaber wrote:
> Hynek, what did your @pg_hba.conf@ file look like to get as far as you did for this attempt?... - 03:36 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Hynek, what did your @pg_hba.conf@ file look like to get as far as you did for this attempt?
- 03:07 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Hynek's results, copied from #3992:
Hynek Cihlar wrote:
> Eric Faulhaber wrote:
> > Hynek or Constantin, could o... - 03:05 PM Feature #4012: replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as PostgreSQ...
- Copied from #3992, from Eric:
This may be worth a look: https://impossibl.github.io/pgjdbc-ng/
The guy behind t... - 03:04 PM Feature #4012 (Review): replacing TCP/IP (AF_INET) sockets with Unix Domain Sockets (AF_UNIX) as ...
03/22/2019
- 05:01 PM Feature #3809: ProDataSet support
- Eric Faulhaber wrote:
> My intention was to rebase 3809a when 3750b was merged to trunk, because I didn't think we h... - 04:57 PM Feature #3809: ProDataSet support
- My intention was to rebase 3809a when 3750b was merged to trunk, because I didn't think we had dependencies on 3750b ...
- 04:54 PM Feature #3809: ProDataSet support
- Ovidiu Maxiniuc wrote:
> I will commit my changes for ProDataSet conversion to 3809a.
I added conversion support ... - 03:29 PM Feature #3809: ProDataSet support
- Here it is, attached.
I will commit my changes for ProDataSet conversion to 3809a. - 03:21 PM Feature #3809: ProDataSet support
- Constantin Asofiei wrote:
> Greg/Eric, can we postpone committing Ovidiu's changes until Monday?
Yes.
Ovidiu, ... - 03:16 PM Feature #3809: ProDataSet support
- Greg/Eric, can we postpone committing Ovidiu's changes until Monday?
- 08:35 AM Feature #3809: ProDataSet support
- Eric Faulhaber wrote:
> Ovidiu, how safe are these changes?
>
> Greg, Constantin, is it ok to commit this to 3750... - 08:29 AM Feature #3809: ProDataSet support
- Eric Faulhaber wrote:
> Ovidiu, how safe are these changes?
They are only token declarations. However, since the to... - 08:25 AM Feature #3809: ProDataSet support
- Ovidiu, how safe are these changes?
Greg, Constantin, is it ok to commit this to 3750b, or are we only allowing pe... - 08:22 AM Feature #3809: ProDataSet support
- Eric Faulhaber wrote:
> As of 3750b/11512, the parser is missing token-types for the following dataset-related featu...
03/21/2019
- 03:08 PM Feature #3809: ProDataSet support
- As of 3750b/11512, the parser is missing token-types for the following dataset-related features:...
03/20/2019
- 06:29 PM Feature #3809: ProDataSet support
- These are procedures defined in @dataset/log_hlp.i@ or functions defined in @dataset/help/log_hlp.i@. I'm not sure wh...
- 06:20 PM Feature #3809: ProDataSet support
- I tried to convert the provided testcases an I encountered some unknown token: @assert-true@, @assert-err@, @assert-n...
- 02:38 PM Feature #3809: ProDataSet support
- I have created task branch 3809a for this work. It is based on trunk 11301, but I think that's ok. I don't know of an...
- 12:40 PM Feature #3809 (WIP): ProDataSet support
Also available in: Atom