FWD » Core Development » Base Language

04/04/2024

05:26 PM Bug #8579 (Test): Unit and OEUnit conversion: ignore annotations from other framework

Greg Shah

05:21 PM Bug #8579: Unit and OEUnit conversion: ignore annotations from other framework

Greg Shah wrote:
> You can merge 8579a to trunk now.
Done. Rev. 15123. Vladimir Tsichevski

04:23 PM Bug #8579 (Merge Pending): Unit and OEUnit conversion: ignore annotations from other framework

You can merge 8579a to trunk now. Greg Shah

01:14 PM Bug #8579: Unit and OEUnit conversion: ignore annotations from other framework

Greg Shah wrote:
> Code Review Task Branch 8579a Revision 15121
>
> The change is good.
>
> What additional te... Vladimir Tsichevski

12:58 PM Bug #8579: Unit and OEUnit conversion: ignore annotations from other framework

Code Review Task Branch 8579a Revision 15121
The change is good.
What additional testing is needed? Greg Shah

11:20 AM Bug #8579: Unit and OEUnit conversion: ignore annotations from other framework

Greg Shah wrote:
> Code Review Task Branch 8579a Revision 15120
>
> I think the code would be simpler and easier ... Vladimir Tsichevski

12:12 PM Feature #6407: name_map.xml improvements

I did replace the @Set@ with @List@ in r15108.
Tomorrow I'm testing again in testcases. A good list of different p... Galya B

11:45 AM Feature #6407: name_map.xml improvements

@MultiClassLoader@:... Galya B

11:43 AM Feature #6407: name_map.xml improvements

So what I'm saying is that there is no classpath order in my humble opinion. Iterating HashMap is not ordered. Galya B

11:39 AM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> > > @MultiClassLoader@ finds the class by iterating @HashSet@.
> > Actually it's a collection, ... Galya B

11:32 AM Feature #6407: name_map.xml improvements

> > @MultiClassLoader@ finds the class by iterating @HashSet@.
> Actually it's a collection, backed by the @HashMap... Greg Shah

11:20 AM Feature #6407 (Review): name_map.xml improvements

6407a r15107 based on trunk r15101 ready for review: adds support for reading mappings from multiple name_map.xml fil... Galya B

09:57 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> @MultiClassLoader@ finds the class by iterating @HashSet@.
Actually it's a collection, backed by ... Galya B

09:42 AM Feature #6407: name_map.xml improvements

We can rework it to iterate jars alphabetically. Galya B

09:39 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> My understanding is that JVM keeps the same order of jars as defined in classpath (it do... Galya B

09:37 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> Greg Shah wrote:
> > Do we really need an explicit list of precedence order? Why not implicitly t... Constantin Asofiei

09:29 AM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> Do we really need an explicit list of precedence order? Why not implicitly take it from the clas... Galya B

09:28 AM Feature #6407: name_map.xml improvements

Do we really need an explicit list of precedence order? Why not implicitly take it from the classpath and honor the ... Greg Shah

09:21 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> From what you posted in #6407-56, only one concern: the granularity for the precedence o... Galya B

05:45 AM Feature #6407: name_map.xml improvements

From what you posted in #6407-56, only one concern: the granularity for the precedence order I think needs to be "thi... Constantin Asofiei

04:39 AM Feature #6407: name_map.xml improvements

P.S. The previous comment got updated. Galya B

04:36 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> And I think I understand now Greg's point about the .r and folder deploy in OpenEdge:
>... Galya B

04/03/2024

06:05 PM Bug #8579 (Internal Test): Unit and OEUnit conversion: ignore annotations from other framework

Code Review Task Branch 8579a Revision 15120
I think the code would be simpler and easier to read if you create a ... Greg Shah

03:11 PM Bug #8579 (Review): Unit and OEUnit conversion: ignore annotations from other framework

Fixed in 8579a rev. 15120. Please, review. Vladimir Tsichevski

03:10 PM Bug #8579 (WIP): Unit and OEUnit conversion: ignore annotations from other framework

Vladimir Tsichevski

02:51 PM Bug #8579: Unit and OEUnit conversion: ignore annotations from other framework

Currently, if a test 4gl source includes both ABLUnit annotations and OEUnit annotations, all these annotations are c... Vladimir Tsichevski

02:46 PM Bug #8579 (Test): Unit and OEUnit conversion: ignore annotations from other framework

Vladimir Tsichevski

03:08 PM Feature #6407: name_map.xml improvements

And I think I understand now Greg's point about the .r and folder deploy in OpenEdge:
* in FWD, you would convert th... Constantin Asofiei

02:34 PM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> > But, in practice, I think .pl archives are used more - and depending on the PROPATH, runtime co... Constantin Asofiei

02:29 PM Feature #6407: name_map.xml improvements

> But, in practice, I think .pl archives are used more - and depending on the PROPATH, runtime could use a .p from di... Greg Shah

02:26 PM Feature #6407: name_map.xml improvements

We had cases of the same OE qualified class name used in different modules. The qualified legacy name can't be chang... Constantin Asofiei

01:28 PM Feature #6407: name_map.xml improvements

To solve case 2. we need the precedence order to be fed to the jmv via startup args, so that the classloader can take... Galya B

01:09 PM Feature #6407: name_map.xml improvements

Galya B wrote:
> I speak about case 3: different maps for different files with different attributes. Obviously if yo... Galya B

01:08 PM Feature #6407: name_map.xml improvements

I speak about case 3: different maps for different files with different attributes, but same path. Obviously if you l... Galya B

01:02 PM Feature #6407: name_map.xml improvements

> This is not something we can decide even with configs and Constantin was right that it is a major issue.
Case 1 ... Greg Shah

12:58 PM Feature #6407: name_map.xml improvements

One pname corresponds to only one jname, because that's how conversion seems to work. The question is if the other at... Galya B

11:36 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> Other than that the jar can have a completely different package for the actual classes.
Actually n... Galya B

11:34 AM Feature #6407: name_map.xml improvements

If we need to load the jars in a specific order, we'll need explicit instructions, probably new config in directory. Galya B

11:32 AM Feature #6407: name_map.xml improvements

Case 2 is the same as java and can cause malfunctioning programs like in java, this is what I meant.
Case 1 is fine ... Galya B

11:20 AM Feature #6407: name_map.xml improvements

> In java this is the same as having multiple versions of the same dependency in the classpath.
Not exactly. In... Greg Shah

11:14 AM Feature #6407: name_map.xml improvements

> If you are saying that module M1 has a program foo/bar/F.p and module M2 has a program foo/bar/F.p. which are compl... Greg Shah

11:12 AM Feature #6407: name_map.xml improvements

This is actually more complicated than it looked at first. We're trying to combine different apps coming possibly fro... Galya B

11:03 AM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> Let's consider this from the perspective of how it works for this multi-app case in OE. If both ... Constantin Asofiei

10:55 AM Feature #6407: name_map.xml improvements

Originally (currently) if multiple paths are found for the same filename when resolving the name map, the paths end u... Galya B

10:45 AM Feature #6407: name_map.xml improvements

> The 'pname' in @class-mapping@ is the full path relative to the @abl/@ folder
Yes, this is why I called them rel... Greg Shah

10:30 AM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> > > And all of this just for a warning log? I would say it's more reasonable to log the warning a... Constantin Asofiei

10:26 AM Feature #6407: name_map.xml improvements

I don't have an actual example though. Not sure if the path is supposed to be the same in two different jars. Galya B

10:25 AM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> > > And all of this just for a warning log? I would say it's more reasonable to log the warning a... Galya B

10:23 AM Feature #6407: name_map.xml improvements

> > And all of this just for a warning log? I would say it's more reasonable to log the warning always when a second ... Greg Shah

06:16 AM Feature #6407: name_map.xml improvements

> REST services written directly in Java can be configured via the 'rest-service' element.
So these nodes are not ... Constantin Asofiei

06:06 AM Feature #6407: name_map.xml improvements

OK, then I need one more clarification. Since there is a separate container @rest-service@, where programs get resolv... Galya B

06:00 AM Feature #6407: name_map.xml improvements

Yes. Constantin Asofiei

05:57 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> Galya B wrote:
> > And all of this just for a warning log? I would say it's more reason... Galya B

05:57 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> And all of this just for a warning log? I would say it's more reasonable to log the warning always ... Constantin Asofiei

05:53 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> Galya B wrote:
> > Constantin, maybe you've missed why I'm assigned to the task, that i... Galya B

05:49 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> Constantin, maybe you've missed why I'm assigned to the task, that is in #6407-4.
OK, thanks for t... Constantin Asofiei

05:49 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> The point was to move to annotations/other structures inside the .java
Constantin, ... Galya B

05:43 AM Feature #6407: name_map.xml improvements

The point was to move to annotations/other structures inside the .java, here I mean all info from name_map.xml; ideal... Constantin Asofiei

05:33 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> I don't mean the annotations in the .class - I mean the state from name_map.xml must mat... Galya B

05:22 AM Feature #6407: name_map.xml improvements

I don't mean the annotations in the .class - I mean the state from name_map.xml must match if there is an existing en... Constantin Asofiei

03:55 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> * load them into @SourceNameMapper@, one by one, and address conflicts - if there are co... Galya B

05:26 AM Bug #8510: replace all 'new logical' in FWD runtime with logical.of or logical.UNKNOWN

8150a was merged to trunk rev 15116 and archived. Constantin Asofiei

05:26 AM Bug #8510: replace all 'new logical' in FWD runtime with logical.of or logical.UNKNOWN

Created task branch 8510a from trunk rev 15115. In rev 15116 I removed the @logical.of@ and @logical.UNKNOWN@ from R... Constantin Asofiei

04/01/2024

06:41 PM Bug #8154: refactor scopeable stacks

Code Review Task Branch 8154a Revision 15109
The changes look promising. There is clearly quite a bit more that c... Greg Shah

12:51 PM Bug #8154 (WIP): refactor scopeable stacks

Created task branch 8154a from trunk rev 15108.
Rev 15109 contains first pass at moving scoped stacks into (TopLev... Constantin Asofiei

04:00 AM Feature #6424: fill gaps in Java Open Client replacement (appserver client support)

Galya, I've placed the project under @devsrv01:/tmp/open_client_example.zip@. There is a small readme.txt in the pro... Constantin Asofiei

03/29/2024

10:57 AM Feature #6407: name_map.xml improvements

Galya B wrote:
> @SourceNameMapper.initMappingData@ seems to be executed only once for the lifespan of the runtime m... Constantin Asofiei

10:20 AM Feature #6407 (WIP): name_map.xml improvements

6407a created from trunk r15101. Galya B

10:03 AM Feature #6407: name_map.xml improvements

@SourceNameMapper.initMappingData@ seems to be executed only once for the lifespan of the runtime machine with a serv... Galya B

09:46 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> Also, SourceNameMapper will not load the .class and annotations when the entry in @name_... Galya B

09:45 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> Java has tools to read an archive.
If you mean the custom FWD's @JarClassLoader@, then ... Galya B

09:38 AM Feature #6407: name_map.xml improvements

You don't need to unzip the jar - Java has tools to read an archive. This just means we can't rely on @Class.getReso... Constantin Asofiei

09:37 AM Feature #6407: name_map.xml improvements

Why can't we use the same technique from @Utils.searchResourceJars()@? We don't need the propath part but we definit... Greg Shah

09:29 AM Feature #6407: name_map.xml improvements

That is of course if the check for the file listed by @jar tvf filename.jar@ gives positive. Galya B

09:27 AM Feature #6407: name_map.xml improvements

Constantin Asofiei wrote:
> Galya, what package would you use for looking into multiple jars? Look everywhere? The ... Galya B

08:41 AM Feature #6407: name_map.xml improvements

Galya, what package would you use for looking into multiple jars? Look everywhere? The problem with @name_map.xml@ i... Constantin Asofiei

08:31 AM Feature #6407: name_map.xml improvements

> @MultiClassLoader@ has a list of all jars in the classpath. It can look for @name_map.xml@ in each jar (with @jar t... Greg Shah

07:08 AM Feature #6407: name_map.xml improvements

@MultiClassLoader@ has a list of all jars in the classpath. It can look for @name_map.xml@ in each jar (with @jar tvf... Galya B

06:16 AM Feature #6431: implement support to allow CPINTERNAL set to UTF-8 and CP936

I know some work has happened to explore I18N testcases. Having proper testcases will be very important.
I want t... Greg Shah

03/28/2024

05:09 PM Feature #6407: name_map.xml improvements

> We had this discussion before - moving *everything* to annotations will mean the entire converted .class code needs... Greg Shah

03:58 PM Feature #6407: name_map.xml improvements

Greg Shah wrote:
> > another improvement would be to emit some kind of special constructs for the virtual functions/... Constantin Asofiei

03:50 PM Feature #6407: name_map.xml improvements

> another improvement would be to emit some kind of special constructs for the virtual functions/internal procedures ... Greg Shah

09:49 AM Feature #6417 (WIP): cleanup some known items in appserver support

Galya B

09:44 AM Feature #6417: cleanup some known items in appserver support

Test in OE for client type:
Procedure running in batch client:... Galya B

03/27/2024

07:46 PM Feature #6422: implement auditing support including AUDIT-POLICY and AUDIT-CONTROL

How can I access @TxWrapper$TransactionHelper.maybeActivateTxWrapper@? I currently have the following code (branch 64... Theodoros Theodorou

05:07 PM Feature #6410: implement additional built-in OO classes/interfaces

> It's probably useless to implement that as I don't expect those to be extended in the first place but the possibili... Greg Shah

05:06 PM Feature #6410: implement additional built-in OO classes/interfaces

Some notes from Marian:
> We had constantly issues with conversion when classes extended/implemented built-in cla... Greg Shah

03/26/2024

04:09 AM Feature #6410: implement additional built-in OO classes/interfaces

@Constantin, are there any changes in trunk pertaining conversion especially for classes extending/implementing skele... Marian Edu

03/25/2024

09:05 AM Feature #6417: cleanup some known items in appserver support

> I'm not sure if my reasoning is correct, but it looks like the same appserver agent can execute all types of appser... Greg Shah

06:40 AM Feature #6417: cleanup some known items in appserver support

Actually there several places where appserver tasks are executed in a different way. I've found these:
* @Agent.Agen... Galya B

06:32 AM Feature #6417: cleanup some known items in appserver support

I'm not sure if my reasoning is correct, but it looks like the same appserver agent can execute all types of appserve... Galya B

08:37 AM Feature #6826: reduce the BDT usage in FWD runtime

There are many cases where a BDT is used as a short-lived constant value. The most of case are @logical@ and @interge... Ovidiu Maxiniuc

03/24/2024

05:17 AM Bug #8510 (New): replace all 'new logical' in FWD runtime with logical.of or logical.UNKNOWN

Constantin Asofiei

03/22/2024

09:37 AM Feature #6424: fill gaps in Java Open Client replacement (appserver client support)

This will need a guide - how to use gradle cache to run and debug a Java client project.
Galya B

09:24 AM Feature #6424: fill gaps in Java Open Client replacement (appserver client support)

Nevermind, it was on devsrv01: @scp devsrv01:/tmp/gradle-skylab.zip .@ . Galya B

09:22 AM Feature #6424: fill gaps in Java Open Client replacement (appserver client support)

Constantin, where is the project on the shared drives? I remember you gave me a path, I can't find now. Galya B

03/21/2024

04:26 AM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Branch 7045f was created from trunk rev 15077 - in rev 15078 re-fixed the regression in 7045d/e. Merged to trunk rev... Constantin Asofiei

03/20/2024

01:00 PM Feature #7045 (Test): re-implement "normal" (non-abend) stack unwinding to avoid throwing an exce...

Branch 7045e was merged into trunk as rev. 15077 and archived. Constantin Asofiei

12:45 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

If you are confident it is safe, it can be merged now. Greg Shah

12:44 PM Feature #7045: re-implement "normal" (non-abend) stack unwinding to avoid throwing an exception

Greg, can this be merged? It affects a customer's app. Constantin Asofiei

07:34 AM Feature #7045 (Internal Test): re-implement "normal" (non-abend) stack unwinding to avoid throwin...

Code Review Task Branch 7045e Revision 15074
I'm OK with the change. Greg Shah

02:54 AM Feature #7045 (Review): re-implement "normal" (non-abend) stack unwinding to avoid throwing an ex...

Created task branch 7045e from trunk rev 15073. In rev 15074 fixes "RETURN statements from within a CASE statement a... Constantin Asofiei

03/19/2024

11:04 AM Feature #7045 (Test): re-implement "normal" (non-abend) stack unwinding to avoid throwing an exce...

Branch 7045d was merged into trunk as rev. 15068 and archived. Constantin Asofiei

11:02 AM Feature #7045 (Merge Pending): re-implement "normal" (non-abend) stack unwinding to avoid throwin...

Merging this now. Constantin Asofiei

03/18/2024

09:08 AM Feature #6417 (Review): cleanup some known items in appserver support

> * versioninfo fields are read-only, but writing them is a runtime error and not compile error in FWD.
* @session/l... Galya B