Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Thomas Broyer
    @tbroyer
    And iterating from that, adding Gradle Module Metadata (similar to what's been proposed for Guava, except we're not interested in the variants here: https://github.com/google/guava/pull/3683/files) implies releasing a new version (it shouldn't stop us doing a first RC without them though)
    Colin Alworth
    @niloc132
    aside from a few extra lines in the bom for the extra relocations, is there a downside for 4? It seems like getting it out as a snapshot or RC with the ability to test mixes 2.7/2.8/2.9 dependencies/projects would be a benefit, rather than adding it later?
    i readily admit to not understanding the implications of 5
    Thomas Broyer
    @tbroyer
    🤔 which "few extra lines in the bom for the extra relocations"? I'm diffing 3 and 4 the only changes are the addition of org.gwtproject 2.9.0 version (and related change in maven-metadata)
    Colin Alworth
    @niloc132
    then i misunderstood the summary (tried to avoid reading the code but stick to the summary and conclusion, to not get caught up in the weeds beyond "does it solve the problems we anticipated")

    https://github.com/tbroyer/gwt-relocation-tests#experiment-4 says

    ...with an additional relocation from org.gwtproject to com.google.gwt for version 2.9.0

    which seems like "extra relocations", no?

    oh i see, not extra relocations, but extra releases
    Thomas Broyer
    @tbroyer
    Yes, those are "new" org.gwtproject:gwt:2.9.0, org.gwtproject:gwt-user:2.9.0 and org.gwtproject:gwt-dev:2.9.0 artifacts; that relocates to their com.google.gwt alter egos
    Colin Alworth
    @niloc132
    so, a library that supports 2.9.0 and isnt ready for 2.10.0 could still move groupid to make the future path simpler
    s/library/project/
    Thomas Broyer
    @tbroyer

    Yes, that's what I was trying to say in the early conclusions with:

    Using the org.gwtproject:gwt:2.9.0 BOM helps stay on GWT 2.9.0 without risking mixed com.google.gwt and org.gwtproject dependencies, by automatically downgrading the latter (to 2.9.0, which relocates to the former).

    Colin Alworth
    @niloc132
    thanks, sorry for my misunderstanding - i had interpreted it to say that if you used o.g:2.10 it would be able to point out c.g.g:2.9 and mark them as incompatible
    Thomas Broyer
    @tbroyer
    Note that such a project would use the org.gwtproject:gwt BOM, but the com.google.gwt:gwt-user and com.google.gwt:gwt-dev libraries
    Colin Alworth
    @niloc132
    right, and when it updates the o.g bom, would correctly move c.g.g -> o.g libraries
    Thomas Broyer
    @tbroyer
    yes (through relocations), but the project itself should then depend on org.gwtproject:gwt-user directly (rather than relying on the relocation)
    Colin Alworth
    @niloc132
    could i ask you to post such a summary on the mailing list, as a way to bring the topic up again, and we can make these changes in the repo in anticipation of a release? i also notice that you have a htmlunit/jetty/asm patch that you might want to merge, i have a draft, doesn't make sense for us to conflict on that (but i'll certainly assist in testing your commit if you'd like to proceed)
    Thomas Broyer
    @tbroyer
    So, first step: use org.gwtproject:gwt:2.9.0 BOM; then when updating: update to org.gwtproject:gwt:2.10.0 and change direct dependencies from com.google.gwt to org.gwtproject.
    Colin Alworth
    @niloc132
    migration to org.gwtproject, dealing with degraded debugging experience in chrome, updated dependencies seems like a good 2.10 to aim for - i also have that remotewebdriver runstyle that could be worth putting into gwt proper
    Thomas Broyer
    @tbroyer
    Well, I do not have an htmlunit/jetty/asm patch actually. If you're referring to the note in the https://github.com/tbroyer/gwt-relocation-tests#experiment-5, this project uses empty JARs, and I only modified the POMs 🤷
    Colin Alworth
    @niloc132
    oh got it
    in that case i'll pick that back up again, so that as we merge these other three sets of changes we can also get that done
    i only had it done locally, and it was for a now out-of-date jetty and htmlunit (and i didn't directly poke asm iirc, at least not for its own sake)
    Rocco De Angelis
    @rdeangelis83
    Hi All, if I remember correctly there was script to package the gwt-sdk build result for maven?
    Colin Alworth
    @niloc132
    @rdeangelis83 please see maven/push-gwt.sh
    it pushes to your local .m2/repository by default
    Rocco De Angelis
    @rdeangelis83
    Hi @niloc132 , I have try patch. But it seems that "constructInvocation(function.getSourceInfo(), "Object.defineProperty", function, prop, descriptor).makeStmt();" expect an indexedMethod
    Colin Alworth
    @niloc132
    is this a compile time failure?
    i thought i had understood from related code that this could take a dotted reference and break it down
          JsStatement createGwtOnLoadFunctionCall =
              constructInvocation("ModuleUtils.addInitFunctions", arguments).makeStmt();
    for example
    but perhaps that is used as a key (irritating how sometimes strings must be individual js identifiers, and sometimes not...)
    Rocco De Angelis
    @rdeangelis83
    Unable to locate index method: Object.defineProperty. Yep during the compile time.
    I will try to fix it tomorrow morning.
    Colin Alworth
    @niloc132
    interestingly, the code you referenced says it is Object.defineProperties, but actually is Runtime.defineProperties
    perhaps these are java references and i missed the point - i was trying to come up with the js name reference
    Rocco De Angelis
    @rdeangelis83
    Yahhh I have seen that. :)
    Colin Alworth
    @niloc132
    i'll have a patch for you shortly - locally i did a bit of other cleanup, ill try to run this today and make sure it goes
    Runtime.defineProperties is a helper that does some extra work, in gwt
    i don't think we actually want that, based on the chromium link you gave me
    but it would obfuscate better
    if i get it running, i'll put a patch up on gwt-review for you to try
    Rocco De Angelis
    @rdeangelis83
    @niloc132 THX!
    Short question: What is J2Wasm?
    Thomas Broyer
    @tbroyer
    An experiment to output Web Assembly rather than Closure-style JS.
    Rocco De Angelis
    @rdeangelis83
    WOW
    Colin Alworth
    @niloc132
    (there is also a kotlin backend now too - ostensibly your java can be passed through j2cl to emit kotlin sources)
    @rdeangelis83 i'm sorry, i didnt get around to finishing something, but i'll put up a gist if you want to try what i did sketch out - i think the structure is right, but i didn't get the chance to validate it
    1 reply
    Rocco De Angelis
    @rdeangelis83
    @niloc132 THX 🙏🏻 I will give it tomorrow a try
    Miroslav Pokorny
    @mP1
    always wondered how much of blazor output is actually wasm, especially considering the question of GC.