On 27/Jan/2010 20:23, Mark Hindess wrote: > In message <4b6062d9.6050...@gmail.com>, Tim Ellison writes: >> On 26/Jan/2010 19:30, Mark Hindess wrote: >>> +1 votes from Catherine, Nathan, Tim, Oliver, and myself. No other >>> votes, so the vote passes. >>> >>> Thanks for testing the artifacts and voting. >>> >>> I've uploaded binaries for the tested platforms to: >>> >>> http://www.apache.org/dist/harmony/milestones/M12a/ >>> >>> so that they will get sync'd to the mirrors in preparation for an >>> announcement tomorrow. Feel free to give them a test run. >> Cool, thanks. >> >> The src archives are great, happy with the vote etc. The binaries I >> tried on Windows work fine too. >> >> However, the src jars in the Windows binaries (i.e. those in >> jre/lib/boot/*-src.jar) contain files with the wrong EOL. Text files >> not in the jars are ok. I went back and checked on M12 and the EOLs >> look ok everywhere there. >> >> I guess it is a bit pedantic, but I think the src JARs on Windows should >> contain the correct EOLs, as we have done previously. >> >> So I'm left wondering why things turned out different this time? I'll >> produce them on my Windows box if you want. > > Yes please. Not sure why they turned out like this.
Sorry, it appears that WinZip(?) is playing tricks on me again. I recreated the artifacts and checked them on my machine, the EOL looks ok on windows. I uploaded them to the server, then did a download test of the same file, the EOLs look wrong now ;-/ The md5 were the same, so the tools must be 'helping' me somehow. >> p.s. You probably noticed I updated the staging site, please take a look >> and check it works as expected. > > Will do. I'll make the changes live now. Regards, Tim
