[Biopython-dev] MySQL (BioSQL)

[Tiago Antao]

By my mistake I answered Peter without Ccing the list. Just for
reference all the content is below [I just add that I will do the
BioSQL corrections as discussed]

 On Fri, 2 May 2014 15:00:37 +0100
Peter Cock <p.j.a.cock at googlemail.com> wrote:

> [Did you mean to go off list? You can include this in any public
> reply]
> 
> On Fri, May 2, 2014 at 2:55 PM, Tiago Antao <tra at popgen.net> wrote:
> > On Fri, 2 May 2014 14:35:41 +0100
> > Peter Cock <p.j.a.cock at googlemail.com> wrote:
> >
> >> It sounds like our BioSQL adapter ought to handle turning
> >> unicode strings into bytes for this field - in most cases there
> >> will be no strange characters when reading GenBank files,
> >> but some test cases are probably going to be needed.
> >
> > I will revert the SQL changes and patch this out in this way.
> 
> Great; or rewrite the branch if you prefer.
> 
> >> > In Jython is does not apply. On the others, I would be inclined
> >> > to ask buildbot ;)
> >>
> >> Heh... but for that to work we'll need to install this new
> >> dependency on the Windows buildslaves, so a manual
> >> test wouldn't be much extra work anyway ;)
> >
> > In my case (the windows 7 buildbot) does not even has mysql
> > installed (it is a very partial test). Indeed one of the reasons I
> > embarked on the docker thing (for Linux at least), was to assure a
> > complete testing environment could be produced easily.
> 
> Yes indeed. Automated Windows VMs would be another cool trick...
> but for now I'm glad to have any Windows buildslaves for testing
> since I've not worked directly on a Windows machine for many
> years now.
> 
> In fact I'm not 100% sure if *any* of the buildslaves have got
> PostgreSQL or MySQL working... a useful job for a volunteer
> would be to scan the recent build logs to spot things we're
> missing across the slaves.
> 
> Cheers,
> 
> Peter