[C++-sig] Re: Boost.python suggestion + pyste bug

[Patrick Hartling]

David Abrahams wrote:
> "Niall Douglas" <s_sourceforge at nedprod.com> writes:
> 
> 
>>On 10 Aug 2003 at 20:18, Nicodemus wrote:
>>
>>
>>>Now, if you are saying that bases<> are always empty, I guess the
>>>problem is that you're using AllFromHeader, which is broken in the
>>>current CVS. Try changing from AllFromHeader to Class calls, that
>>>should make the problem go away.
>>
>>Ah, but if you remember, my pyste file self-generates itself. This 
>>means it will adapt to changes in the base library almost 
>>automatically.
>>
>>BTW Nicodemus, you have said you have placed your recent bug fixes in 
>>CVS. Last night I pulled a fresh copy off the CVS on boost-
>>consulting.com and there had been no changes to pyste whatsoever.
>>
>>What's going on? Is your CVS different from the sourceforge 
>>developers one?
> 
> 
> The boost-consulting mirror is updated nightly by downloading and
> unpacking the SF CVS tarball, but it's hard to say whether it's
> updated before or after SF makes its nightly tarball.  So the mirror
> could be as much as 2 days out-of-date.  The SF tarball is always 1
> day out-of-date.  I think I'll change the mirror job to happen in the
> morning so we're more likely to be up-to-date.

The tarball may be more out of date than just 1 or 2 days.  In my 
experience with downloading CVS snapshots, they are often many days 
behind.  For example, I have a cron job that tries to download three 
repository tarballs every night using wget(1) for tape backup purposes, 
and it won't download the file unless the remote copy is different from 
the local copy.  One of those repository tarballs has not been updated 
since July 25 in spite of numerous commits I and other people have made 
in the last 11 days (most of the committers were out of town between July 
26 and August 1).  I may have something screwed up with my wget(1) 
options, but I'm inclined to believe that it's been at least 11 days 
since the tarball was re-rolled by SourceForge.

My suspicion is that the SourceForge folks have a script that iterates 
over their /cvsroot directory and tars up the repositories one by one. 
As the number of repositories grows and as the size of each active 
repository grows, this process would slow down and make the tarballs get 
more and more out of date.

  -Patrick


-- 
Patrick L. Hartling                     | Research Assistant, VRAC
patrick at vrac.iastate.edu                | 2624 Howe Hall: 1.515.294.4916
http://www.137.org/patrick/             | http://www.vrac.iastate.edu/