We always have option of using apache-royale as package name. On Mon, Oct 30, 2017 at 11:32 AM, Harbs <[email protected]> wrote:
> It’s a shame that “royale” seems to already be taken on npm. > > I would vote for two packages: > > 1. To install *everything* (i.e. swf, js, node, etc. and future targets > when/if we add them): > npm install apache-royale -g > > 2. To install js-only: > npm install apache-royale-js -g > > If we see a demand for further packages (i.e. compiler only), we can add > them as additional packages later. > > Harbs > > > On Oct 30, 2017, at 8:23 PM, OmPrakash Muppirala <[email protected]> > wrote: > > > > So, "npm install" downloads a tarball from npmjs.org. The package > usually > > contains the code we want others to use. It also contains a > "package.json" > > file which specify all its dependencies. These dependencies (and their > > sub-dependencies) are all downloaded from npmjs.org as part of "npm > > install". > > > > There are options to run custom scripts before and after the npm install. > > In the case of FlexJS, we run a script afterwards that simply downloads > our > > non-npmjs.org dependencies (royale sdk, fonts, flash player, air, etc.) > and > > puts them in the correct places. > > > > So, our options are: > > > > 1. Publish two different packages on npmjs.org: jsonly and js+swf. We > > need to figure out the names of these packages, since they are unique > > identifiers on npmjs's registry. > > > > Then the command the users would run would look like: > > npm install royale-jsonly -g > > npm install royale-js-and-swf -g > > > > 2. Publish only the jsonly package. > > Then the command the users would run would look like: > > npm install royale-jsonly -g > > > > 3. Possibly, we can figure out a way to optionally download swf support. > > This way, by default the jsonly is downoaded and unzipped. Then we could > > (possibly) look at the args or have the user run another command that > > downloads the swf support. > > > > Then the command the users would run would (possibly) look like: > > npm install royale -- -include-swf-support -g > > > > (or) > > npm install royale-jsonly -g > > and then > > ./update-royale-include-swf-support > > > > In all three cases, we can definitely run a script that alters xml > configs, > > etc. to suit our needs. > > > > Hope that helps. > > > > Thanks, > > Om > > > > On Mon, Oct 30, 2017 at 9:20 AM, Alex Harui <[email protected]> > > wrote: > > > >> Om, > >> > >> Can you explain to us what our options are? Essentially, the JS-only > >> package will be a subset of a package that can output both SWF and JS > and > >> will probably have slightly different default settings in, for example, > a > >> frameworks/royale-config.xml file. > >> > >> It is looking like we can create a zip package for JS-only that will > work > >> in Moonshine and VSCode, but to fully make it work in Flash Builder (and > >> maybe some other IDEs) you will need to run a script of some sort that > >> fixes up some FB launch configurations that convert Flex projects to > >> Royale projects. > >> > >> The current plan for a "FlexJS" package that has SWF support (for users > >> that want use SWF for testing or as a migration step) will require that > >> users unzip a package and run an Ant script to bring down Adobe > >> dependencies. I'm thinking we won't use the Flex installer. > >> > >> I'm still working through why one of our users isn't getting code > >> completion working in FB and the answer there may affect packaging as > well. > >> > >> I don't know NPM well enough to have an opinion on, if we distribute two > >> packages (flexjs-with-swf-support and js-only), whether NPM allows us to > >> have two different packages or whether it is better to structure NPM > >> releases as js-only package and a swf-support-add-on package. > >> > >> I also don't know if the NPM install should run a script that fixes up > >> those launch configs. Maybe it is better to continue to leave them as > "FB > >> users have to run this additional Ant script" or something like that. > I'm > >> not sure how important FB still is to our ease-of-migration story. > >> > >> Maybe showing us what folks would have to type on the command line might > >> help us form opinions. > >> > >> Thoughts? > >> -Alex > >> > >> > >> On 10/30/17, 4:36 AM, "[email protected] on behalf of Carlos > Rovira" > >> <[email protected] on behalf of [email protected]> > >> wrote: > >> > >>> Hi Om, > >>> > >>> I think that would be great! > >>> > >>> If we end having multiple products as Alex suggested, I think we should > >>> have as well multiple NPM installs. > >>> So for me is ok to sync products we deliver with NPM installations > flavors > >>> > >>> Thanks > >>> > >>> > >>> > >>> 2017-10-30 10:58 GMT+01:00 Yishay Weiss <[email protected]>: > >>> > >>>> You’re likely to do most of the maintenance work, so it’s up to you… > As > >>>> far as users go there are some users writing client code in AIR and > >>>> server > >>>> code in node (in fact I’m involved in such a project right now). So I > >>>> wouldn’t make sweeping assumptions. > >>>> > >>>> > >>>> > >>>> ________________________________ > >>>> From: [email protected] <[email protected]> on behalf of OmPrakash > >>>> Muppirala <[email protected]> > >>>> Sent: Monday, October 30, 2017 10:21:37 AM > >>>> To: [email protected] > >>>> Subject: Re: Publishing royale to npm > >>>> > >>>> On Mon, Oct 30, 2017 at 1:19 AM, Harbs <[email protected]> wrote: > >>>> > >>>>> Why not publish both versions? > >>>>> > >>>> > >>>> It looks like the js only is going to be just a zip file. That makes > >>>> for > >>>> easy maintenance. > >>>> The swf version has a bunch of dependencies to be downloaded. > >>>> > >>>> Not a big deal, just thinking out loud if we really need to publish > two > >>>> different packages that might lead to confusion. > >>>> > >>>> I'm open to both, though. > >>>> > >>>> Thanks, > >>>> Om > >>>> > >>>> > >>>>> > >>>>>> On Oct 30, 2017, at 10:15 AM, OmPrakash Muppirala < > >>>> [email protected]> > >>>>> wrote: > >>>>>> > >>>>>> I was wondering if we should publish the apache.royale-jsonly verson > >>>> via > >>>>>> npm instead of the full version with swf support. > >>>>>> After all, users coming in vial npm would most likely not expect swf > >>>>>> support. > >>>>>> > >>>>>> Any thoughts on this proposal? > >>>>>> > >>>>>> Thanks, > >>>>>> Om > >>>>> > >>>>> > >>>> > >>> > >>> > >>> > >>> -- > >>> > >>> <https://na01.safelinks.protection.outlook.com/?url= > >> http%3A%2F%2Fwww.codeo > >>> scopic.com&data=02%7C01%7C%7C5f3b122f189e4e0f119b08d51f8a > >> 81b0%7Cfa7b1b5a7b > >>> 34438794aed2c178decee1%7C0%7C0%7C636449602097009881& > >> sdata=wZgQd0X2xX6ed8y0 > >>> t4O87r66gMlVy%2F8aHqtpwnq8O6w%3D&reserved=0> > >>> > >>> Carlos Rovira > >>> > >>> Director General > >>> > >>> M: +34 607 22 60 05 > >>> > >>> https://na01.safelinks.protection.outlook.com/?url= > >> http%3A%2F%2Fwww.codeos > >>> copic.com&data=02%7C01%7C%7C5f3b122f189e4e0f119b08d51f8a > >> 81b0%7Cfa7b1b5a7b3 > >>> 4438794aed2c178decee1%7C0%7C0%7C636449602097009881& > >> sdata=wZgQd0X2xX6ed8y0t > >>> 4O87r66gMlVy%2F8aHqtpwnq8O6w%3D&reserved=0 > >>> > >>> > >>> Conocenos Avant2 en 1 minuto! > >>> <https://na01.safelinks.protection.outlook.com/?url= > >> https%3A%2F%2Favant2.e > >>> s%2F%23video&data=02%7C01%7C%7C5f3b122f189e4e0f119b08d51f8a > >> 81b0%7Cfa7b1b5a > >>> 7b34438794aed2c178decee1%7C0%7C0%7C636449602097009881& > >> sdata=JK22xVqobAGGnZ > >>> b8laWESXHS3NA5nLdscBYTEHml7Pk%3D&reserved=0> > >>> > >>> > >>> Este mensaje se dirige exclusivamente a su destinatario y puede > contener > >>> información privilegiada o confidencial. Si ha recibido este mensaje > por > >>> error, le rogamos que nos lo comunique inmediatamente por esta misma > vía y > >>> proceda a su destrucción. > >>> > >>> De la vigente Ley Orgánica de Protección de Datos (15/1999), le > >>> comunicamos > >>> que sus datos forman parte de un fichero cuyo responsable es > CODEOSCOPIC > >>> S.A. La finalidad de dicho tratamiento es facilitar la prestación del > >>> servicio o información solicitados, teniendo usted derecho de acceso, > >>> rectificación, cancelación y oposición de sus datos dirigiéndose a > >>> nuestras > >>> oficinas c/ Paseo de la Habana 9-11, 28036, Madrid con la documentación > >>> necesaria. > >> > >> > >
