#1 Yep #2 Dependencies are: Drools - Good old Drools we already have org.treblereel.gwt.yaml.mapper - Under Apache 2.0 and under this community umbrella and already used by kogito-runtimes editors. jshell - From Oracle, a completely new dependency introduction
Alternatives, or options that the user can select, for jshell are likely FEEL and MVEL, both already in use by the community. This depends on the direction we want to take this. #3+4 This one is a tricky one to answer. Right now the features include Decision Tables and Literal Expressions, but depending on the rising needs and community approval, new features would be added by demand and by vote. If we feel voting is necessary. Current driver is Red Hat's need to have simpler method for writing Decision Logic. Once YaRD is put into use, then there are the same requirements to maintain and support it as there is for the XLS decision tables. Documentation, as I understand, should follow with any commit that adds a new feature into the code base. If this is not the case, then we might have to have a proposal about that :-) The implementation itself is a YAML interface to write Drools code. Giving easier, more manageable ( but also more liming ) methods for Logic authoring. One big driver is the ability to write Decision Tables with just text and the benefits of Decision Tables are readability and indebt validation ( rule overlaps, missing logic gaps and so on ). Even when YaRD has a specific name, on a technical level it is far closer to the XLS Decision Tables than it is to DMN. Not saying it can not grow to be as much as DMN, but there is only so much one file can handle. However it does not make sense to call it YAML Decision tables, since it is a bit more and has the potential to go further, it is and will be as the name says: Yet Another Rule Definition. Toni On 2024/03/05 13:11:58 Gabriele Cardosi wrote: > Hi Toni, > to recap, it seems there are the following points of discussion (I add a > couple) > > 1. timing (i.e. not before the 10.0 release, and I think we all agree > on that) > 2. transitory dependencies > 3. long-term maintenance plan > 4. community contribution to it (e.g. there is a new DRL refactoring > initiative, and the guys after it are dedicating time and effort to also > document, explain, and simplify the code to facilitate external > contribution) > > > > Il giorno mar 5 mar 2024 alle ore 13:24 Toni Rikkola <rikk...@apache.org> > ha scritto: > > > Sure I can start a vote once the 10.0.0 is done. > > > > I will introduce the dependencies in that vote. > > > > Toni > > > > On 2024/03/05 04:31:42 Jan Šťastný wrote: > > > Hello Toni and others, > > > I'd just add to this that we should also evaluate what 3rd party > > libraries > > > we're bringing in with this addition. If they're allowed in ASF and if > > they > > > are well maintained and addressed for vulnerabilities. (Parsing libraries > > > seems to be prone to these issues more than others.) > > > > > > So I was wondering if such an addition (if/when it happens) would be run > > > through voting where we could safe-check the concerns I stated above, > > > rather than a PR where it's sufficient to get 2 approving reviews from > > > committers. > > > > > > Regards > > > Jan > > > > > > Dne po 4. 3. 2024 19:40 uživatel Toni Rikkola <rikk...@apache.org> > > napsal: > > > > > > > Sure. I am working on this alone for now, so a reasonable delay is not > > an > > > > issue. > > > > > > > > Tonino > > > > > > > > On 2024/03/04 15:26:01 Alex Porcelli wrote: > > > > > Toni, > > > > > > > > > > Although you mentioned that it's quite isolated, I feel more > > > > > comfortable if we could delay to introduce it to the codebase after > > > > > the 10.0.0 release - in the meantime maybe use a feature branch > > > > > directly in the Apache repository. > > > > > > > > > > On Mon, Mar 4, 2024 at 10:15 AM Toni Rikkola <rikk...@apache.org> > > wrote: > > > > > > > > > > > > Future goals can include having YaRD next to SonataFlow. Allowing > > easy > > > > access to Decision Tables and other Decision logic that is not easy to > > > > implement with a flow structure and of course the other way around. > > > > > > > > > > > > However there are also other targets. One method is to use the YaRD > > > > files as scorecards. Based on the inputs the end result gives a % of > > > > success. > > > > > > > > > > > > YaRD can also be used to aggregate several REST calls into a one > > more > > > > preferable format. > > > > > > > > > > > > Toni > > > > > > > > > > > > On 2024/03/04 14:27:07 Yeser Amer wrote: > > > > > > > Hi Tonino! > > > > > > > > > > > > > > Thanks for sharing a brief description of YaRD here. > > > > > > > I agree with you that having the editor in kie-tools (apache) and > > > > the engine in another area (kie?) is not ideal. > > > > > > > > > > > > > > Can I ask where the code is currently located? > > > > > > > Since you said the feature is "experimental", can I ask what the > > > > plans are for YaRD? > > > > > > > I think clarifying this point can help us make the best decision. > > > > > > > > > > > > > > > > > > > > > On 2024/03/04 13:53:07 Alex Porcelli wrote: > > > > > > > > Toni, > > > > > > > > > > > > > > > > As in previous discussion related to the new Drools parser, my > > > > concern > > > > > > > > is how much this might affect the 10.0.0 release. > > > > > > > > > > > > > > > > I personally am not a big fan of YAML, in a recent demo I got > > in > > > > > > > > trouble because of 2 spaces :D - but I understand that this > > might > > > > look > > > > > > > > appealing to k8s audience :) > > > > > > > > > > > > > > > > On Mon, Mar 4, 2024 at 8:23 AM Toni Rikkola <t...@rikkola.net> > > > > wrote: > > > > > > > > > > > > > > > > > > Hello, > > > > > > > > > > > > > > > > > > I am proposing we add YaRD into the kie-drools repository. > > YaRD > > > > is used > > > > > > > > > to describe Decisions, Rules and Declarative logic in the > > YAML > > > > format. > > > > > > > > > At the current state it supports Decision Tables and Literal > > > > > > > > > Expressions. > > > > > > > > > > > > > > > > > > Example of a Decision Table is in attached file > > > > > > > > > "yard-dtable-example.txt". ( In .txt for easier viewing. ) > > > > > > > > > A more complex example is in attached "costs.txt". > > > > > > > > > > > > > > > > > > The YAML structure and the used scripting language is not > > set in > > > > stone > > > > > > > > > and the feature at the moment is experimental. > > > > > > > > > > > > > > > > > > Under the YAML cover it runs Drools, in more detail the > > Drools > > > > Rule > > > > > > > > > Units DSL. > > > > > > > > > > > > > > > > > > Due to the move to Apache we are in a weird situation where > > we > > > > already > > > > > > > > > have an editor for YaRD in kie-tools without a backend. The > > > > existing > > > > > > > > > editor has a simple autocomplete with support for Validation > > > > that finds > > > > > > > > > redundancy and subsumption from the Decision Tables. To > > follow > > > > the > > > > > > > > > Apache community rules and to show visibility I am now > > asking for > > > > > > > > > approval to add the backend in. > > > > > > > > > > > > > > > > > > Thank you > > > > > > > > > Toni Rikkola > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org > > For additional commands, e-mail: dev-h...@kie.apache.org > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@kie.apache.org For additional commands, e-mail: dev-h...@kie.apache.org
