Wouldn't it be possible for the API plugin to check whether or not the base plugin had already been applied? Op 12 apr. 2012 01:18 schreef "Eric Berry" <[email protected]> het volgende:
> Hi Adam, and Peter, > Thank you both for the quick response. > > Adam, guess we'll have to stick with the "apply from" scripts that have > the extra code to check to see if the plugin is installed. > > Peter, we're using what I think you'd call binary plugins at Chegg. We > have an artifactory repository that hosts the plugin's artifacts, which our > developers use by way of a a script: > > [code src=" > http://internal.host/gradle-plugins/installation/chegg-corp.groovy";] > buildscript { > repositories { > mavenRepo url: "http://internal.host/artifactory/repo"; > } > dependencies { > classpath "com.chegg.corp.gradle:chegg-plugin:12.2" > } > } > // Check to make sure CheggCorpPlugin isn't already added. > boolean hasCorp = project.plugins.find { plugin -> > "${plugin.getClass()}" == "${com.chegg.corp.CheggCorpPlugin}" > } > if (!hasCorp) { > project.apply(plugin: com.chegg.corp.CheggCorpPlugin) > } > [/code] > > This is fine except for the extra junk at the end where I need to check to > see if the plugin is already installed. > > I did some testing and I found that if you use the plugin ID's (eg. apply > plugin: "chegg-corp") gradle gracefully handles the plugin being applied > multiple times, but because of this bug: > http://issues.gradle.org/browse/GRADLE-1517 > > We are forced to use the plugin's class, which doesn't seem to work the > same way. > > The problem here is that we have one plugin called "chegg-corp", and > another "chegg-corp-api" which also applies the "chegg-corp" plugin. Some > of our projects use the subprojects closure to apply the "chegg-corp" > plugin, and then have an "api" module that applies the "chegg-corp-api" > plugin. Gradle complains about the "chegg-corp" plugin already being > installed without the check. So I have to do the check in multiple places - > once in the "apply from" scripts, and anywhere else that one plugin applies > another. > > If #1517 could be fixed, that would reduce a lot of boiler plate for our > plugins and our "apply from" scripts. I've already voted for the bug, but > would really appreciate it if that would could be fixed for 1.0, or at > least make it so applying the plugin via class works the same way as > applying it via ID. > > -- > Learn from the past. Live in the present. Plan for the future. > Blog: http://eric-berry.blogspot.com > jEdit <http://www.jedit.org> - Programmer's Text Editor > Bazaar <http://bazaar.canonical.com> - Version Control for Humans >
