[tei-council] Jenkins vs. SVN

[Sebastian Rahtz]

On 1 Jul 2011, at 21:09, Piotr Bański wrote:

> and Jenkins and, in that context, about the tcw20 doc [1]. I understand
> that Jenkins takes the SVN source[2] and builds from it, and this is why
> the content of Jenkins workspace[3] is different from the SVN -- is that
> correct?
yes indeed. it has a persistent working area.

> The tcw20 document tells me to validate the modified source against the
> p5odds.rnc schema. But that schema is missing from the SVN and present
> in the Jenkins listing, so I assume that it is derived in the process of
> building a distribution from the source. Is that correct, please?
it is indeed

> And does that mean that I can't modify the Guidelines without having
> previously run 'make' on the freshly updated SVN checkout, because if I
> do and mess something up, I won't be able to build, and without that, I
> won't be able to validate my modified source? I'm sensing a Catch-22
> here

Technically true, yes.  So what you do is build p5odds.rnc ("make oddschema")
before you start working, then its available to you for checking your work. 
If someone else commits something that stops p5odds.rnc from building,
then we're all messed up.

so there's a theoretical problem here, but my opinion is that its not likely
to be serious holdup.
--
Sebastian Rahtz      
Head of Information and Support Group, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Sólo le pido a Dios
que el futuro no me sea indiferente