FWD » Core Development » Base Language

02/14/2023

02:22 PM Feature #7122 (WIP): convert default getter/setter as plain Java methods

Constantin Asofiei

09:35 AM Feature #7122: convert default getter/setter as plain Java methods

I don't see any reason to have the @BlockManager.function/procedure@ wrappers for default scalar getter/setter Java m... Constantin Asofiei

09:34 AM Feature #7122 (WIP): convert default getter/setter as plain Java methods

Constantin Asofiei

09:59 AM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

I've used the attached program for testing cryptography.
May be it can be useful. Igor Skornyakov

08:57 AM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

I'm OK with closing this task. Before we do that, please do/consider the following:
* Gather all of your various ... Greg Shah

08:38 AM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

I tried many combinations with @AES_CBC_256@. Some of them can be found below. I keep getting identical results with ... Theodoros Theodorou

02/13/2023

02:26 PM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

As far as I rememeber after the Constantin's fix we are 100% binary compatible with 4GL with respect to the encryptio... Igor Skornyakov

02:21 PM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

From #5915-40:
>The problem is with @generate-pbe-key@, where @AES_CBC_256@ is used for this key (32 length). There... Constantin Asofiei

01:45 PM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

I have read some existing tests under @testcases/security/security_policy@ and I tried to understand them. I tried ma... Theodoros Theodorou

09:28 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar wrote:
> And some cleanup:
Great, I'll test this and include it in my larger changes. Constantin Asofiei

09:12 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

And some cleanup:... Hynek Cihlar

09:09 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Hynek, what times do you get for @StringHelper.hashCodeCaseInsensitive@ test? Please te... Hynek Cihlar

08:52 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek, what times do you get for @StringHelper.hashCodeCaseInsensitive@ test? Please test what happens if you use so... Constantin Asofiei

08:42 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

Greg Shah wrote:
> How many JMX instrumentation locations are there? I'd like to run some non-trivial application c... Constantin Asofiei

08:39 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

Try attached testcase in command-line. On devsrv01, the times are these:... Constantin Asofiei

08:17 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

How many JMX instrumentation locations are there? I'd like to run some non-trivial application code with and without... Greg Shah

08:06 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

Greg Shah wrote:
> > I think we need to add a JVM flag to enable or disable the JMX code, so for production mode it ... Constantin Asofiei

08:04 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

> I think we need to add a JVM flag to enable or disable the JMX code, so for production mode it will go via the dire... Greg Shah

02/12/2023

03:14 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

I've did more testing and profiling of lambda calls. Even a simple lambda call as we have in @ContextLocal.get(boole... Constantin Asofiei

02/10/2023

12:54 PM Feature #4065: server-side processing of client platform dependencies

> With respect to the anticipated use cases does it make sense to replace these with the managed file system resource... Greg Shah

11:39 AM Feature #4065: server-side processing of client platform dependencies

Greg, I found several places in the client UI where file IO is performed directly. An example is a button widget load... Hynek Cihlar

02/09/2023

01:49 PM Feature #4065: server-side processing of client platform dependencies

Hynek Cihlar wrote:
> Greg, do we want to interpolate the file paths with a set of known variables? Like logged in u... Greg Shah

01:06 PM Feature #4065: server-side processing of client platform dependencies

Greg, do we want to interpolate the file paths with a set of known variables? Like logged in user, application name, ... Hynek Cihlar

11:19 AM Feature #4065: server-side processing of client platform dependencies

Constantin Asofiei wrote:
> Hynek Cihlar wrote:
> > I created task branch 4065a based on trunk revision 14477.
> F... Hynek Cihlar

11:16 AM Feature #4065: server-side processing of client platform dependencies

Hynek Cihlar wrote:
> I created task branch 4065a based on trunk revision 14477.
FYI, trunk is on rev 14483. Constantin Asofiei

10:20 AM Feature #4065: server-side processing of client platform dependencies

> > > Should we pursue some kind of EXECUTE permissions on the OS-COMMAND statement?
> >
> > It is a really good i... Greg Shah

10:11 AM Feature #4065: server-side processing of client platform dependencies

Greg Shah wrote:
> > Should we pursue some kind of EXECUTE permissions on the OS-COMMAND statement?
>
> It is a r... Hynek Cihlar

09:49 AM Feature #4065: server-side processing of client platform dependencies

I think you can add this @EXECUTE@ extra bit into the @BitFlagsResource@ directory. Greg Shah

09:48 AM Feature #4065: server-side processing of client platform dependencies

> Except I think we should implicitly deny access when no permission rule is matched.
Agreed. We always should a... Greg Shah

08:55 AM Feature #4065: server-side processing of client platform dependencies

Hynek Cihlar wrote:
> Also we should introduce the EXECUTE permission.
I will rephrase this to a question. Shoul... Hynek Cihlar

08:49 AM Feature #4065: server-side processing of client platform dependencies

Greg Shah wrote:
> > The actual file-system security model will be based on the Linux file-system security model.
>... Hynek Cihlar

08:32 AM Feature #4065: server-side processing of client platform dependencies

> Current FWD security model doesn't have primitives for file-system resources, this will have to be extended.
> The... Greg Shah

08:04 AM Feature #4065: server-side processing of client platform dependencies

I created task branch 4065a based on trunk revision 14477. Hynek Cihlar

07:34 AM Feature #4065: server-side processing of client platform dependencies

I'm currently working on the file-system server-side processing. Here are a few notables:
* Current FWD security m... Hynek Cihlar

09:55 AM Feature #6421 (Feedback): finish GENERATE-RANDOM-KEY support

Theodoros Theodorou

04:01 AM Feature #6421: finish GENERATE-RANDOM-KEY support

> Did you add the unknown value test code?
I added this:... Theodoros Theodorou

09:54 AM Bug #6399 (WIP): ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter ...

Theodoros Theodorou

02/08/2023

02:08 PM Bug #7089 (WIP): sanitize p2j.oo method signature

I have the changes to solve the signature inconsistencies in @p2j.oo@, but I'll wait to create a branch after 6129c i... Constantin Asofiei

02:01 PM Feature #6818 (WIP): analyze the performance of the FWD open client proxy client support

I have changes force "schema marshall" to NONE in case of TABLE/DATASET parameters (although I need somehow to force ... Constantin Asofiei

01:06 PM Feature #6421: finish GENERATE-RANDOM-KEY support

Yes. At this time we don't branch testcases, so we have no trunk concept there. Instead of saying pushing to trunk ... Greg Shah

12:33 PM Feature #6421: finish GENERATE-RANDOM-KEY support

> > @sed ...@
>
> These changes should actually be the default in Hotel GUI. Please make the edits and commit the... Theodoros Theodorou

02/07/2023

11:51 AM Feature #6421: finish GENERATE-RANDOM-KEY support

> @rm -rf ~/projects/hotel_gui/abl/*@
I think you should not do this, in general. Try @ant clean.all@ instead.
... Greg Shah

10:33 AM Feature #6421: finish GENERATE-RANDOM-KEY support

I need help running the test with FWD. I tried some things but it didn't work. My project tree is:... Theodoros Theodorou

08:04 AM Feature #6421: finish GENERATE-RANDOM-KEY support

No, I am referring to the Progress 4GL concept of unknown value (which can be represented with the literal constant @... Greg Shah

02:33 AM Feature #6421: finish GENERATE-RANDOM-KEY support

By unknown value, you mean non-ASCII characters? I wasn't able to run the test with FWD yet. I will try to run it tod... Theodoros Theodorou

02/06/2023

09:56 AM Feature #6421: finish GENERATE-RANDOM-KEY support

I think it is good. Just add a test for unknown value. You can't do that one with a preprocessor define, so just ha... Greg Shah

09:30 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Please review the following test. The rest was developed and run on the AWS VM. I also included encryption/decryption... Theodoros Theodorou

02/02/2023

11:15 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Theodoros Theodorou wrote:
> Is this test below a good test? Should I store it at @testcases/security/security-polic... Igor Skornyakov

11:10 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Do you have test code for the error cases that I mentioned in #6421-16?
In your testcase #6421-21, you should avoi... Greg Shah

10:41 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Hello Igor and thank you for your review.
I replaced @getAlgKeyLen()@ with @SecurityPlicyManager.getSymmetricCiper... Theodoros Theodorou

09:58 AM Feature #6410: implement additional built-in OO classes/interfaces

> we can then keep OE11.7 as main target and OE12.2 'extensions/fixes' are only nice to have?
Correct. If there a... Greg Shah

09:41 AM Feature #6410: implement additional built-in OO classes/interfaces

Greg Shah wrote:
> I prefer not to have different FWD branches for different OE versions.
Understood, then only... Marian Edu

09:25 AM Feature #6410: implement additional built-in OO classes/interfaces

11.7 and 12.2 is fine.
> Now, this only has effect at compile time - for FWD will be when the conversion is execut... Greg Shah

03:12 AM Feature #6410: implement additional built-in OO classes/interfaces

Greg Shah wrote:
> Yes, please use @PROVERSION@ and preprocessing to conditionally implement logic by OE version.
... Marian Edu

02/01/2023

02:34 PM Feature #6649: improve the performance or SourceNameMapper runtime

Does @SourceNameMapper.convertName()@ often get called with the same @legacyProgName@ and @lookup@ function? If so, ... Greg Shah

02:34 PM Feature #6649: improve the performance or SourceNameMapper runtime

Code Review Task Branch 6129b Revision 14394
The changes look good. Greg Shah

01:26 PM Feature #6421: finish GENERATE-RANDOM-KEY support

Theodoros Theodorou wrote:
> I read the encryption algorithm by using @SecurityPolicyManager.getSymmetricEncryptionA... Igor Skornyakov

12:55 PM Feature #6421: finish GENERATE-RANDOM-KEY support

Greg Shah wrote:
> The original code in @SecurityOps.generateRandomKey()@ was forced to @AES_CBC_128@. Take a look... Igor Skornyakov

12:47 PM Feature #6421: finish GENERATE-RANDOM-KEY support

The original code in @SecurityOps.generateRandomKey()@ was forced to @AES_CBC_128@. Take a look at trunk. Greg Shah

12:45 PM Feature #6421: finish GENERATE-RANDOM-KEY support

Greg Shah wrote:
> Igor: Please review.
Sorry, I do not understand. As far as I remember and can see from the cod... Igor Skornyakov

12:26 PM Feature #6421: finish GENERATE-RANDOM-KEY support

> Do I need to checkout 6421a from the testcases project or use the main branch?
I'm not exactly sure what you are... Greg Shah

08:45 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Do I need to checkout 6421a from the testcases project or use the main branch?
Then I should include my @test-genr... Theodoros Theodorou

09:39 AM Bug #7089: sanitize p2j.oo method signature

In incremental conversion, it was found that a @RequestBuilder.put@ method's LegacySignature was inconsistent with th... Constantin Asofiei

09:27 AM Bug #7089 (WIP): sanitize p2j.oo method signature

Constantin Asofiei

08:52 AM Feature #6410 (WIP): implement additional built-in OO classes/interfaces

Marian Edu

01/31/2023

09:30 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Yes, except it would be useful to put slightly more detail in the comment. Something like this seems appropriate:
... Greg Shah

09:18 AM Feature #6421: finish GENERATE-RANDOM-KEY support

History entry means to add a comment in the beginning of the file?
Like the following?... Theodoros Theodorou

09:12 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Theodoros Theodorou wrote:
> How to write test cases?
[[p2j:Writing 4GL Testcases]]
[[p2j:Testing_Process#Proje... Greg Shah

09:03 AM Feature #6421: finish GENERATE-RANDOM-KEY support

How to write test cases? I thought that the p2j test cases do not run. Theodoros Theodorou

06:20 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Code Review Task Branch 6421a Revision 14484
1. @generateRandomKey()@ needs complete error handling. Please write... Greg Shah

05:41 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Igor: Please review. Greg Shah

02:06 AM Feature #6421 (Review): finish GENERATE-RANDOM-KEY support

I have committed my changes and the task is ready for review Theodoros Theodorou

02:05 AM Feature #6421: finish GENERATE-RANDOM-KEY support

I have created the branch 6421a Theodoros Theodorou

01/30/2023

10:37 AM Feature #6649: improve the performance or SourceNameMapper runtime

I checked in various performance changes in 6129b revision 14394. Please review. Hynek Cihlar

01/27/2023

01:18 PM Feature #7082: Improve converted call site names

I think we can provide this as an option, though if we do we will want to reduce some of the text (maybe remove the @... Greg Shah

12:57 PM Feature #7082: Improve converted call site names

I suggest to make call site names more readable. E.g. I have to click each of this sites to understand what this code... Stanislav Lomany

12:51 PM Feature #7082 (New): Improve converted call site names

Stanislav Lomany

08:04 AM Feature #3853: implement LOG-MANAGER runtime

I've found a workaround fix for batch writing all @stdout@ through @ErrorWriterBatch#displayError()@, which messed up... Galya B

01/26/2023

10:06 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

The use of @String.toLowerCase@ and @toUpperCase@ in @P2JField.computeHash - 90k@ and @Validation.setUniqueQueryPara... Hynek Cihlar

09:58 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

The use of @toLowerCase@ and @toUpperCase@ in @h2.Function.getSimpleValue - 178k@ is valid there. Hynek Cihlar

09:04 AM Feature #6421: finish GENERATE-RANDOM-KEY support

Create a task branch (6421a) and you can commit changes there. Greg Shah

08:52 AM Feature #6421: finish GENERATE-RANDOM-KEY support

I read the encryption algorithm by using @SecurityPolicyManager.getSymmetricEncryptionAlgorithm().toStringMessage()@.... Theodoros Theodorou

01/25/2023

04:13 PM Feature #6649: improve the performance or SourceNameMapper runtime

Greg Shah wrote:
> 3821c?
I meant 6129b. 3821c has burned into my cerebral cortex after the years working with th... Hynek Cihlar

04:08 PM Feature #6649: improve the performance or SourceNameMapper runtime

3821c? Greg Shah

03:48 PM Feature #6649: improve the performance or SourceNameMapper runtime

3821c revision 14382 introduces caching of resolved procedure internal entries. It speeds up tests in a large custome... Hynek Cihlar

01:06 PM Feature #3853: implement LOG-MANAGER runtime

I've found an unexpected behavior with batch execution. @stdout@ is displayed using @ErrorManager.displayError(lineOu... Galya B

07:37 AM Bug #7073 (New): Launching non-interactive program creates JNI core dump in process_linux.c due t...

Greg Shah

11:25 PM Bug #7073: Launching non-interactive program creates JNI core dump in process_linux.c due to ncurses

Greg Shah wrote:
> That failing code is not code that is checked in to trunk. It is specific to *your system*. You... Roger Borrello

07:29 PM Bug #7073 (Rejected): Launching non-interactive program creates JNI core dump in process_linux.c ...

That failing code is not code that is checked in to trunk. It is specific to *your system*. Your logging is broken. Greg Shah

01/24/2023

06:57 PM Bug #7073: Launching non-interactive program creates JNI core dump in process_linux.c due to ncurses

With logging added to @process_linux.c@ we see it occur in @ fprintf(fp, "pseudoTerminalWait: move cursor part 2... Roger Borrello

06:56 PM Bug #7073: Launching non-interactive program creates JNI core dump in process_linux.c due to ncurses

@uast/util/process_launch_helper.p@ can be used.
Using Swing, run the program, and pass in any command, like *ls*
... Roger Borrello

06:50 PM Bug #7073 (New): Launching non-interactive program creates JNI core dump in process_linux.c due t...

Roger Borrello

02:53 AM Feature #6649: improve the performance or SourceNameMapper runtime

Constantin Asofiei wrote:
> Hynek Cihlar wrote:
> > Constantin Asofiei wrote:
> > > Bellow are the backtraces for ... Hynek Cihlar

02:38 AM Feature #6649: improve the performance or SourceNameMapper runtime

Hynek Cihlar wrote:
> Constantin Asofiei wrote:
> > Bellow are the backtraces for @SourceNameMapper.getInternalEntr... Constantin Asofiei

02:37 AM Feature #6649: improve the performance or SourceNameMapper runtime

Constantin Asofiei wrote:
> Bellow are the backtraces for @SourceNameMapper.getInternalEntry@ - these are reduced af... Hynek Cihlar

02:24 AM Feature #6649: improve the performance or SourceNameMapper runtime

Bellow are the backtraces for @SourceNameMapper.getInternalEntry@ - these are reduced after the InvokeConfig call-sit... Constantin Asofiei

01/23/2023

12:02 PM Feature #6649: improve the performance or SourceNameMapper runtime

Constantin, I didn't find any major performance issues profiling the big customer application. Can you please share m... Hynek Cihlar

01/20/2023

09:48 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

I marked the issue at 100%. But there are still some left out places, those that were more complex to address (see #6... Hynek Cihlar

09:27 AM Feature #6819: refactor FWD proxy implementation to use ReflectASM instead of Java Method reflection

A resume of what was done at this task:
Experiments were done using lambda expression instead of Java reflection o... Constantin Asofiei

09:25 AM Support #4032: block/transaction processing context-local and other performance improvements

Changes mentioned in previous notes:
* cache the 'helper' instance at the WrappedResource, object and other cases, t... Constantin Asofiei

09:23 AM Feature #6827: reduce the memory footprint of WrappedResource implementations: replace BDTs with ...

Profiling and manual/automated code review was used to identify BDT and collection fields in WrappedResource implemen... Constantin Asofiei

09:22 AM Feature #6826 (Review): reduce the BDT usage in FWD runtime

Profiling and manual/automated code review was used to identify BDT field and local variable definitions. These were... Constantin Asofiei

09:21 AM Feature #6824: lazy initialization of collection class members

The fixes part of the fixes from previous notes were determined from profiling and manual/automated code review, used... Constantin Asofiei

09:20 AM Feature #6650: improve performance of Scopeable notification in TM.processScopeNotifications

After what was done in #6650-34, scope processing is still an expensive area, although the possibilities to improve i... Constantin Asofiei

08:51 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

I understand. My concern is not with your change, it is with my code in @convert/control_flow.rules@. It doesn't co... Greg Shah

08:34 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

The code in control_flow.rules considers the case where the 4GL @return.@ statement is the last one in the top-level ... Constantin Asofiei

08:30 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Code Review Task Branch 7045a Revision 14481
I don't object to the change, but I'm unsure if it will work quite ri... Greg Shah

03:25 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

The changes from 7045a are ported and committed to 6129b/14375. Constantin Asofiei

02:37 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

7045a/14481 added another condition to emit a real Java return: the 4GL return statement must not have any downstream... Constantin Asofiei

07:05 AM Feature #6649: improve the performance or SourceNameMapper runtime

6129b revision 14376 reimplements the case insensitive maps in @SourceNameMapper@ with @CaseInsensitiveHashMap@. Hynek Cihlar

01/19/2023

02:33 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Hynek Cihlar wrote:
> > ... 6129b revision 14369. And related fixes. Please review.
> ... Hynek Cihlar

02:20 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar wrote:
> ... 6129b revision 14369. And related fixes. Please review.
I see only one issue:
* TempTabl... Constantin Asofiei

11:21 AM Feature #6820 (Review): reduce String.toLowerCase and toUpperCase usage and String instances

I removed all the upper/lower case occurrences in the list in #6820-4 with the exception of the following list. These... Hynek Cihlar

06:48 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Eric Faulhaber wrote:
> Hynek Cihlar wrote:
>
> > I'm not sure we want to give up @AbstractHashedMap@. My prelimin... Hynek Cihlar

06:33 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

XmlImport.tablesByXmlName changed to @CaseInsensitiveHashMap@ and some other cases in 6129b revision 14369. And relat... Hynek Cihlar

02:28 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Code Review Task Branch 7045a Revision 14480
Weird. I don't understand the reason this works, but the code seems ... Greg Shah

02:11 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

The problem described in #7045-15 is fixed in 7045a rev 14480. Constantin Asofiei

08:11 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

There was a regression, in some cases like:... Constantin Asofiei

11:21 AM Feature #6649 (WIP): improve the performance or SourceNameMapper runtime

Hynek Cihlar

01/18/2023

03:49 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

In the large customer application I see an improvement between 100-200ms between 6129b revisions 14366 and 14368.
Hynek Cihlar

03:23 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

@XmlImport.tablesByXmlName@ can be switched to a @CaseInsensitiveHashMap@. The other cas remaining - from @TempTable... Constantin Asofiei

02:45 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Hynek Cihlar wrote:
> > Constantin Asofiei wrote:
> > > Unfortunately this introduces ... Hynek Cihlar

02:36 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Both commons-collections-3.2.2.jar and commons-collections4-4.4.jar are copied in @build... Hynek Cihlar

02:21 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar wrote:
> Constantin Asofiei wrote:
> > Unfortunately this introduces a visible slowdown in a large cus... Constantin Asofiei

02:06 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Unfortunately this introduces a visible slowdown in a large customer application perform... Hynek Cihlar

12:25 PM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Unfortunately this introduces a visible slowdown in a large customer application performance testing - ~200ms or so. Constantin Asofiei

11:12 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar wrote:
> Actually we explicitly declare dependency on @commons-collection-3.2.2@. I deliberately added ... Constantin Asofiei

11:09 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar wrote:
> I'm not sure we want to give up @AbstractHashedMap@. My preliminary testing shows it outperf... Eric Faulhaber

10:41 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Constantin Asofiei wrote:
> Hynek, you add a dependency on @commons-collections4@ ver @4.4@, but FWD already is (in)... Hynek Cihlar

10:16 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

Hynek, you add a dependency on @commons-collections4@ ver @4.4@, but FWD already is (in)directly dependent on commons... Constantin Asofiei

08:56 AM Feature #6820: reduce String.toLowerCase and toUpperCase usage and String instances

6129b revision 14367 resolves most of the cases reported in #6820-4.
Some cases were kept because the use of lower... Hynek Cihlar

03:27 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

OK, go ahead when it makes sense. Greg Shah

03:24 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Greg Shah wrote:
> Constantin: Do you want me to merge this into 6129b? It does require a reconversion, but I'd lik... Constantin Asofiei

02:43 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Constantin: Do you want me to merge this into 6129b? It does require a reconversion, but I'd like to get the perform... Greg Shah

08:39 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

I haven't ported this to 6129b but it should apply pretty easily. Please try it on the customer app to see the diffe... Greg Shah

08:35 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Greg Shah wrote:
> I use @wa.tm.getBlockType()@ to check the current block instead of looking at @nearest@ which wou... Constantin Asofiei

08:33 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

I use @wa.tm.getBlockType()@ to check the current block instead of looking at @nearest@ which would be @BlockType.FUN... Greg Shah

08:32 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

It does exit. I've tested it here. In this case the current block type is NOT @BlockType.FUNCTION@ so it will throw... Greg Shah

08:13 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

My point is about this code:... Constantin Asofiei

07:39 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

> Isn't @BlockManager.returnNormal@ supposed to throw the @ReturnUnwindException@? I mean, this code should distingui... Greg Shah

02:33 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Greg Shah wrote:
> Constantin: Please review 7045a revision 14479.
Isn't @BlockManager.returnNormal@ supposed to th... Constantin Asofiei

01/17/2023

06:28 PM Feature #7045 (Review): re-implement "normal" (non-abend) stack unwinding to avoid throwing an ex...

Constantin: Please review 7045a revision 14479. Greg Shah

06:26 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

In addition to user-defined functions, 2 additional callable program types which are efectively the equivalent of use... Greg Shah

09:24 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

I'm working on a simple approach for now. Rather than reimplementing all of the flow control for the cases where the... Greg Shah

09:13 AM Feature #3853: implement LOG-MANAGER runtime

+This diagram shows the current state of the call stack producing error logs (not a complete view):+
!clientlog-ol... Galya B

06:36 AM Feature #6821: java collection performance

As far as I can see, only @pruneSessionCache()@ issue is left here. Stanislav Lomany

06:35 AM Feature #6821: java collection performance

> You will see that the Scopeable scopestart/finished do lots of work - please check if it may be possible to condens... Stanislav Lomany

02:35 AM Feature #6817: replace INSTANTIATING-PROCEDURE and other handle<ExternalProgram> with the actual ...

Constantin Asofiei wrote:
> 6129b/14365 fixes a problem related to WidgetPool .
The change is good, thanks! Hynek Cihlar

01/16/2023

03:54 PM Feature #6824: lazy initialization of collection class members

Some other changes related to this are in 6129b/14366. Constantin Asofiei

03:54 PM Support #4032: block/transaction processing context-local and other performance improvements

Another batch of changes related to this is in 6129b/14366 Constantin Asofiei

03:53 PM Feature #6827: reduce the memory footprint of WrappedResource implementations: replace BDTs with ...

6129b/14366 contains a set of changes related to this, I don't think there's anything else remaining related to the w... Constantin Asofiei

03:53 PM Feature #6826: reduce the BDT usage in FWD runtime

6129b/14366 contains a set of changes related to this, mainly from what profiling a large customer app showed. Constantin Asofiei

03:51 PM Feature #6817 (Test): replace INSTANTIATING-PROCEDURE and other handle<ExternalProgram> with the ...

6129b/14365 fixes a problem related to WidgetPool . Constantin Asofiei

03:51 AM Feature #6817 (Review): replace INSTANTIATING-PROCEDURE and other handle<ExternalProgram> with th...

I checked in the changes in 6129b revision 14364. In addition to the attached 6817a.diff I also modified the followin... Hynek Cihlar

02:10 AM Feature #6817: replace INSTANTIATING-PROCEDURE and other handle<ExternalProgram> with the actual ...

Hynek, I don't see any issue with the patch. Also, the comments don't introduce anything new. Please add header ent... Constantin Asofiei

11:51 AM Bug #6399: ENCRYPT/DECRYPT built-in function results different than 4GL when IV is shorter than r...

The runtime implementation of this is in @SecurityOps.encrypt()/decrypt()@. There should be no conversion changes ne... Greg Shah

11:49 AM Feature #6421: finish GENERATE-RANDOM-KEY support

The runtime implementation of this is in @SecurityOps.generateRandomKey()@. There should be no conversion changes ne... Greg Shah

11:41 AM Feature #6421 (WIP): finish GENERATE-RANDOM-KEY support

Theodoros Theodorou

10:41 AM Feature #6821: java collection performance

Stanislav Lomany wrote:
> Performance overhead in function vs procedure is cause by throwing @ReturnUnwindException@... Greg Shah

08:24 AM Feature #6821: java collection performance

Performance overhead in function vs procedure is cause by throwing @ReturnUnwindException@. I think that in vast majo... Stanislav Lomany

10:29 AM Feature #7045 (Test): re-implement "normal" (non-abend) stack unwinding to avoid throwing an exce...

Greg Shah

03:52 AM Feature #6820 (WIP): reduce String.toLowerCase and toUpperCase usage and String instances

Hynek Cihlar