FWD » Core Development » Conversion Tools

06/24/2026

10:06 AM Feature #6320: sharded schemata

Yes, using a per-module namespace would be a way to solve the issue I noted in #6320-5. Greg Shah

10:03 AM Feature #6320: sharded schemata

Greg Shah wrote:
> > This means that apps/modules need to be able to be converted in pieces and executed at runtime ... Octavian Adrian Gavril

07:08 AM Feature #7178: move the buffer scoping calculations to parse time

I've wrapped up the FOR EACH WHERE/BY, dataset with relation-fields, lots of unqualified fields/tables tests.
I'm ... Dănuț Filimon

06/23/2026

03:53 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

Constantin, have you had a chance to look at the approach mentioned above? Let me know what you think. Octavian Adrian Gavril

06/22/2026

07:11 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

I've committed the build changes as rev *16622*. @./gradlew all@ should work now. Paula Păstrăguș

06/19/2026

06:39 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

10804b was rebased, new rev is *16621*. Paula Păstrăguș

06:08 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Next, I'll rabase the branch, setup new big application, redo the conversion, and solve the remaining parser issues. ... Paula Păstrăguș

05:04 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Paula Păstrăguș wrote:
> One testcase is still failing: @AdaptiveQuery.initialize@ now receives @WHOLE-INDEX@ as a p... Paula Păstrăguș

06/18/2026

08:10 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

*Update*: LATEST Hotel GUI performance comparison between ANTLR2 and ANTLR4. (Please note that all conversions were e... Paula Păstrăguș

04:38 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

For Hotel GUI, the total time needed to generate the tokens with the ANTLR2 based lexer and convert them to ANTLR4 co... Paula Păstrăguș

08:02 AM Feature #7178: move the buffer scoping calculations to parse time

The tests were developed on the examples from #7178-2.
This is a recap of the tests that were completed or have a ... Dănuț Filimon

06/17/2026

09:24 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

The approach seems to work! 🥳 I was able to get the same AST for @standard.df@. However, there are still a few remain... Paula Păstrăguș

06/16/2026

06:43 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

I am moving to an alternative approach for the ANTLR4 lexer performance issue (based on the idea mentioned by Greg in... Paula Păstrăguș

06/15/2026

02:59 PM Feature #10804: create replacement AST implementation for use with ANTLR 4

I added temporary lexer profiling in order to measure the token generation cost per token type, aggregated across all... Paula Păstrăguș

12:26 PM Feature #11540: rework conversion back-end to add optional support for output to C#

This is long awaited feature. The conversion architecture was built with thin in mind. When I first read about it, I ... Ovidiu Maxiniuc

09:41 AM Feature #11540: rework conversion back-end to add optional support for output to C#

There are some customers that prefer C# to Java.
The core of both languages have nearly identical syntax and C# ... Greg Shah

09:21 AM Feature #11540 (New): rework conversion back-end to add optional support for output to C#

Greg Shah

06/12/2026

08:42 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Before moving forward, another project conversion progressed further than in previous attempts, it reached the *serve... Paula Păstrăguș

08:35 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Another slowdown source is the @evaluateExpr@ method: under ANTLR2, Hotel GUI spends *5.3 seconds* in it, whereas und... Paula Păstrăguș

06/11/2026

08:16 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

10804b is at rev *16584* now (here you can find today's parser improvements). Paula Păstrăguș

08:03 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

I also attached a log file in my previous post. It includes the total time ANTLR4 spends in prediction mode for each ... Paula Păstrăguș

08:00 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Hmm, for Hotel GUI, @createShadowNodes@ takes around @0.2@ seconds in total, which seems reasonable from my POV. (we ... Paula Păstrăguș

07:51 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

In this 2X difference for the front end, 40% comes from the ANTLR4 code and 60% from something else. The other major... Greg Shah

07:45 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

*Update*: Hotel GUI performance comparison between ANTLR2 and ANTLR4.
|_. Metric |_. ANTLR2 |_. ANTLR4 |
| Total ... Paula Păstrăguș

06/10/2026

02:35 PM Feature #11528: implement COMPILE stmt

Although this is a 4GL language statement, possibly best considered part of Base Language, from our perspective it ha... Greg Shah

02:22 PM Feature #11528 (New): implement COMPILE stmt

Greg Shah

10:12 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

10804b is at rev *16583* now (the latest). Paula Păstrăguș

10:04 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Greg, it is not 3x slower anymore, it is only around 2x now, and I think I can still push the boundaries:) Paula Păstrăguș

10:00 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

*Update*: the current Hotel GUI parse time is *49 seconds*. I think I can still improve this further through parser c... Paula Păstrăguș

09:20 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Right, I remember having this error from the very beginning, basically two years ago:-)
Anyway, I think a fresh ch... Paula Păstrăguș

09:16 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

> It is not valid, a dot is missing here:
Hmmm. Something is wrong here with the project. You are working with H... Greg Shah

07:11 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Greg Shah wrote:
> ANTLRv2 and ANTLRv4 both have their loops and optional alternatives set to be greedy by default. ... Paula Păstrăguș

07:06 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Paula Păstrăguș wrote:
> Greg Shah wrote:
> > > This suggests that we may also need some special tricks for cases w... Paula Păstrăguș

07:06 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Greg Shah wrote:
> > This suggests that we may also need some special tricks for cases where a @no viable alternativ... Paula Păstrăguș

06:54 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Paula Păstrăguș wrote:
> Another important thing I noticed is that the ANTLR4 prediction algorithm does not seem to ... Greg Shah

06:49 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

> One bottleneck I identified is the @toolbar.p@ file. It throws an NPE caused by a @no viable alternative@, which is... Greg Shah

06:23 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Greg, is a @()*@ construct in ANTLR2 greedy or non-greedy by default? I only found three @options@ blocks with @greed... Paula Păstrăguș

06:11 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

Another important thing I noticed is that the ANTLR4 prediction algorithm does not seem to provide a reliable way to ... Paula Păstrăguș

05:44 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

*Update*: Hotel GUI parse time has now dropped to *1 minute and 8 seconds*, with a total conversion time of *5 minute... Paula Păstrăguș

06/09/2026

02:31 PM Feature #10804: create replacement AST implementation for use with ANTLR 4

> Greg, I remember you suggested something related to this at some point, but I do not recall the exact details. Coul... Greg Shah

10:22 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

The lexer doesn't appear to be the bottleneck. Using @tokens.fill()@ and summing the elapsed times across all Hotel_G... Paula Păstrăguș

09:42 AM Feature #10804: create replacement AST implementation for use with ANTLR 4

I think it is time to document some findings and updates here.
At this point, I think I have stabilized the Progre... Paula Păstrăguș

12:48 PM Support #6409: generate "railroad style" syntax diagrams for the progress parser

h1. Analysis of Approach
h2. Objective
Provide high-quality, readable "railroad" syntax diagrams for the full r... Greg Shah

06/05/2026

12:08 PM Feature #6320: sharded schemata

Using TRPL for import is overkill since the tree structure is so trivial. So, yes, I agree it should move away. Greg Shah

11:36 AM Feature #6320: sharded schemata

Greg Shah wrote:
> If the details are in annotations in the DMOs, do we need the P2O documents? Annotations seems c... Ovidiu Maxiniuc

11:08 AM Feature #6320: sharded schemata

I think it is best to remove the @.dict@ files too. Greg Shah

11:00 AM Feature #6320: sharded schemata

Greg Shah wrote:
> If the details are in annotations in the DMOs, do we need the P2O documents? Annotations seems c... Constantin Asofiei

10:54 AM Feature #6320: sharded schemata

If the details are in annotations in the DMOs, do we need the P2O documents? Annotations seems cleaner. Greg Shah

10:53 AM Feature #6320: sharded schemata

Greg Shah wrote:
> Octavian Adrian Gavril wrote:
> > Understood, that makes perfect sense for production discipline... Constantin Asofiei

10:37 AM Feature #6320: sharded schemata

Octavian Adrian Gavril wrote:
> Understood, that makes perfect sense for production discipline. But does this setup ... Greg Shah

10:12 AM Feature #6320: sharded schemata

Understood, that makes perfect sense for production discipline. But does this setup still fundamentally confirm the O... Octavian Adrian Gavril

09:57 AM Feature #6320: sharded schemata

At production, the customer will expect a merged database not to run against a shared. This means there is an assump... Greg Shah

09:46 AM Feature #6320: sharded schemata

I did some testing in an OE environment and want to see if I got it right regarding how this feature works.
I set up... Octavian Adrian Gavril

04:26 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

Octavian Adrian Gavril wrote:
> This is the fix:
> [...]
>
> I will include this patch in the next revision of *... Octavian Adrian Gavril

03:21 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

I found a fix for the issue described in my previous note.
In @annotations/method_defs.rules@, we decide whether t... Octavian Adrian Gavril

06/04/2026

09:21 AM Feature #6320 (WIP): sharded schemata

Octavian Adrian Gavril

09:20 AM Feature #11404 (WIP): Implement annotation-based resolution for database dependencies.

Octavian Adrian Gavril

08:14 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

I've fixed the regression documented above by skipping collision checks for parent constructor names.
While testin... Octavian Adrian Gavril

03:16 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

Okay, it seems that @calculateUniqueJavaMethodName@ is searching for parent constructors as well. Since I disallowed ... Octavian Adrian Gavril

01:54 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

Unfortunately, I'm seeing a few new differences, so I might have regressed something regarding constructor names in t... Octavian Adrian Gavril

06/03/2026

11:51 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

I refactored the constructor generation in *rev16588*. These changes include updating the constructor name in @legacy... Octavian Adrian Gavril

06/02/2026

10:32 AM Feature #9638: resolve OO dependencies during parsing using annotations in already converted classes

Greg Shah wrote:
> In the meeting today, Octavian clarified that this info would be stored in the cvt database and i... Octavian Adrian Gavril