Hi Cody, Minor rectification: pomegranate needs an XDBC server, it doesn't rely on the Docs appserver.
Pity that pomegranate has no install-script for Windows environment. That makes booster much easier to integrate into Windows environments. :-) Not sure, but it doesn't seem to utilize its own deploy capabilities to deploy itself either. By the way, Cody, have you considered copying the booster xqy into Admin instead of Docs? The Admin always has to be there, but the Docs can have been removed as you might be working with a production environment, where there is no need for the Docs appserver. Kind regards, Geert -----Oorspronkelijk bericht----- Van: [email protected] [mailto:[email protected]] Namens [email protected] Verzonden: woensdag 26 januari 2011 0:48 Aan: [email protected] Onderwerp: Re: [MarkLogic Dev General] [ann] Booster - MarkLogic automation tool Thanks Eric! I have looked at Pomegranate. It's a neat project and impressive in scope. Whereas Pomegranate is a wholistic, configuration management application, Booster is a much smaller bootstraping utility. Given a freshly installed MarkLogic cluster, how do you perform an initial application installation in an automated fashion? Pomegranate handles this bootstrapping problem by leveraging the default "Docs" app server on port 8000 and some custom xquery: https://github.com/dranderson/Pomegranate/blob/master/System/common/install-actions.sh https://github.com/dranderson/Pomegranate/blob/master/System/common/bootstrap-pomegranate.xqy XQDebug encourages a similar approach: http://code.google.com/p/xqdebug/source/browse/trunk/README.txt http://code.google.com/p/xqdebug/source/browse/trunk/install/install.xqy Booster provides a self contained and semi-flexible solution to this bootstrapping gap. I work with a group where the linux toolset is common and there's a fondness for the unix philosophy of small things loosely joined. This utility approach fits in well with our systems. I wrote a little bit more about this here: http://code.google.com/p/booster-xqy/wiki/WhyBooster thanks, cody On Tue, 25 Jan 2011, Eric Bloch wrote: > Hey Cody, > > This looks useful, perhaps overlapping to some degree with the Don's > Pomegranate<https://github.com/dranderson/Pomegranate/wiki/> project. Did > you look > at that one at all? > > -Eric > > > On Jan 25, 2011, at 11:29 AM, <[email protected]<mailto:[email protected]>> > <[email protected]<mailto:[email protected]>> wrote: > > > Booster is a thin XQuery wrapper around MarkLogic Server api methods. It > aims to provide automation friendly access to MarkLogic admin tasks, using > http requests and tools like curl. It's a building block for automated > provisioning, targeted at system administration and QA automation. > > > Here's a quick example: > # create a new database > curl --digest -u"admin:pass" \ > "http://server:8001/booster.xqy?\ > action=database-create\ > &database-name=MyNewDatabase\ > &schema-db-name=Schemas&security-db-name=Security > > > You can learn more here: > http://code.google.com/p/booster-xqy/ > > > Community feedback, feature requests, and participation are welcome. I'd > be especially appreciative of code reviews and advice on more idiomatic > xquery. > > thanks, > cody > _______________________________________________ > General mailing list > [email protected] > http://developer.marklogic.com/mailman/listinfo/general > > _______________________________________________ General mailing list [email protected] http://developer.marklogic.com/mailman/listinfo/general _______________________________________________ General mailing list [email protected] http://developer.marklogic.com/mailman/listinfo/general
