This is a not uncommon issue raised on the Users list, so I'm happy to see you addressing it, Jason. The general suggestion until now has been "run Archiva or similar" or "run cron every hour to set -r on all artifacts in your repo" but a "native Maven solution" will be much nicer.
Wayne On 10/19/07, Brett Porter <[EMAIL PROTECTED]> wrote: > I think that's a reasonable assumption. The worst cases I can think of: > - it deploys again if the metadata is missing or is missing the > version, no worse than now > - it doesn't deploy again if the metadata thinks it is there and it > isn't (which we can state clearly with a message so there'll be no > doubt how to fix). > Works for me. > > Thanks! > > - Brett > > On 19/10/2007, at 4:02 PM, Jason van Zyl wrote: > > > > > On 18 Oct 07, at 10:40 PM 18 Oct 07, Brett Porter wrote: > > > >> rock on. > >> > >> How are you going to implement? I'm presuming you also don't want > >> to download the previous artifact, so either will trust the > >> metadata, or maybe have a wagon.exists() method check that can > >> HEAD the file on HTTP and (whatever else) on the others? > >> > > > > I am assuming the metadata has not been diddled and use it if it is > > there. I also look for the checksum and compare them because in my > > case the source versioning is sometime wrong and it is actually > > something new so I issue a message in this case (my tool bumps the > > version automatically) but if they are the same then it's > > definitely a no no. I'm assuming people are not idiots and so they > > use the deploy plugin to populate their repositories so I'm not > > using wagon.exists() but that can't hurt either. The case is almost > > always a corporate environment so they have used the deployment > > plugin. > > > >> - Brett > >> > >> On 19/10/2007, at 3:28 PM, Jason van Zyl wrote: > >> > >>> I've been working on some release/deployment tools lately for a > >>> client and I modified their deployments so that you could not > >>> accidentally release the same version of an artifact more then > >>> once. I can stop this on the server side only when running a > >>> repository manager, but we have some simple Apache servers that > >>> are DAV enables and someone release an artifact with the same > >>> version by mistake and wreaked havoc for 4 hours. > >>> > >>> I think this one is pretty obvious, I'm not going to write a > >>> proposal I'm just going to fix it because it's not very bright > >>> that we allow the re-release of something that already has been. > >>> I'll leave a option to redeploy if you so choose in the event the > >>> driver knows what he's doing. > >>> > >>> Thanks, > >>> > >>> Jason > >>> > >>> ---------------------------------------------------------- > >>> Jason van Zyl > >>> Founder, Apache Maven > >>> jason at sonatype dot com > >>> ---------------------------------------------------------- > >>> > >>> > >>> > >>> > >>> -------------------------------------------------------------------- > >>> - > >>> To unsubscribe, e-mail: [EMAIL PROTECTED] > >>> For additional commands, e-mail: [EMAIL PROTECTED] > >> > >> -- > >> Brett Porter - [EMAIL PROTECTED] > >> Blog: http://www.devzuz.org/blogs/bporter/ > >> > >> --------------------------------------------------------------------- > >> To unsubscribe, e-mail: [EMAIL PROTECTED] > >> For additional commands, e-mail: [EMAIL PROTECTED] > >> > > > > Thanks, > > > > Jason > > > > ---------------------------------------------------------- > > Jason van Zyl > > Founder, Apache Maven > > jason at sonatype dot com > > ---------------------------------------------------------- > > > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > -- > Brett Porter - [EMAIL PROTECTED] > Blog: http://www.devzuz.org/blogs/bporter/ > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
