On 11 October 2013 23:04, Simon Toedt <[email protected]> wrote: > On Fri, Oct 11, 2013 at 12:15 AM, Lionel Cons <[email protected]> > wrote: >> On 10 October 2013 14:28, Simon Toedt <[email protected]> wrote: >>> On Thu, Oct 10, 2013 at 12:22 PM, Lionel Cons <[email protected]> >>> wrote: >>>> On 9 October 2013 17:46, Glenn Fowler <[email protected]> wrote: >>>>> >>>>> the AT&T Software Technology ast alpha 2013-10-10 source release >>>>> has been posted to the download site >>>>> http://www.research.att.com/sw/download/alpha/ >>>>> the package names and md5 checksums are >>>>> INIT 63bd36aef7833ab08b8c3b381e412c8a >>>>> ast-base 1c26b780b153d9e35cea978ccbfdcea1 >>>>> ast-open 9221f5411ff69665186a7d328445c75a >>>>> ast-gpl c29ca45ed1fd62ed336d178c83c418c0 >>>>> ast-ksh 612793fe34837a0d409d6cf360dcb01c >>>>> ast-jmake 3a4975ca9fd9e7d6f4749ae7f9cf8238 >>>>> the md5 sums should match the ones listed on the download page >>>>> >>>>> the last day at at&t for { dgk gsf } is 2013-10-10 >>>>> this will be the last at&t post by us >>>>> for the immediate future ek will be handling ast opensource posts >>>>> { dgk gsf } will forward changes to ek for future posts >>>>> this is not an ideal situation -- there are plans to remedy this >>>>> with { git bugzilla } or equivalents -- but first we have to land >>>>> on our feet outside of at&t >>>>> >>>>> for completeness the alpha download page contains some packages >>>>> that haven't been posted in a while >>>>> >>>>> changes since 2013-09-26 >>>>> >>>>> :::::::: bgp :::::::: >>>>> >>>>> 13-10-10 bgp-mrt.c: fix STATE_BGP_TABLE to parse all attr() before nlri() >>>>> >>>>> :::::::: ksh93 :::::::: >>>>> >>>>> 13-10-08 --- Release ksh93v- --- >>>>> 13-10-08 The shell arithmetic now recognized suffices f,F, l, and L for >>>>> floating point constants. >>>>> 13-10-07 The shell now prints an error message on standard error when a >>>>> job specified with %job does not exist. >>>>> 13-10-04 A bug in the value of i.MAX inside an arithmetic expression >>>>> when the >>>>> variable i is unsigned long has been fixed. >>>>> >>>>> :::::::: libast :::::::: >>>>> >>>>> 13-09-27 path/pathcanon.c: fix buffer clobber bug between pathdev() and >>>>> fgetcwd() >>>>> 13-09-27 path/pathopen.c: handle easy O_XATTR case for /dev/fd open() >>>>> semantics >>>>> >>>>> :::::::: mailx :::::::: >>>>> >>>>> 13-10-01 mailx: add autosign to auto sign new messages >>>>> >>>>> :::::::: std :::::::: >>>>> >>>>> 13-09-26 split.c: fix opetget() docs >>>> >>>> There are a lot of patches missing which have been submitted in the >>>> past (e.g. getconf XATTR_EXISTS or getconf -f to deal with probing >>>> file descriptors). >>>> >>>> Now that this has happened *very* often - this is not the first time >>>> I've been observing this - is there a reason for this or just a row of >>>> oversights? From Jan - Aug 2013 less than 30% of the patches submitted >>>> have been included in ast. >>> >>> Well, that is pretty much the reason why a lot of people like Debian >>> or Suse gave up pushing patches to upstream ('ATT'). Patches are >>> submitted but deliberately ignored. >> >> No no, I don't think its done deliberately, > > But It happens so often that its IMHO either a symptom of total > disorganization or total ignorance. Sure, accidents happen, but that > option is only valid if something is a rare occasion. This issue isn't > rare, it's a habit.
Well, at the end we have to talk to David and Glenn to see how this can be fixed in the future, after SAT and UWIN have moved to astopen.org Lionel _______________________________________________ ast-developers mailing list [email protected] http://lists.research.att.com/mailman/listinfo/ast-developers
