Re: Another request for obliterate...

On Tuesday 12 April 2005 22:55, Steve Greenland wrote:

> > In fact, it's a very common practice to do what we're doing, allowing
> > the underlying SCM to snapshot each and every build.
>
> Actually, I don't think it is. Most "best practices" documents advise
> storing only things that cannot be regenerated. I've never seen it cause
> anything but problems in the long term.

Unfortunately, sometimes this "best practice" is not applicable.

This happens when:

a) generated files require tools that not all people who checkout have

b) generated files are long to reproduce, (in my case, refined protein
structure coordinates ;)

c) generated files are subject of peer review (in case of software project --
outputs of compiler optimiser, if the project develops that optimiser ;). In
our case -- refined protein structures)

d) generated files are input to tha manual reconstruction (in my case --
refined protein structure coordinates ;)

Thus sometimes storing generated files may become inevitable.

-- 
Saulius Gražulis
VisuomeninÄ— organizacija "Atviras Kodas Lietuvai"
P.Vileišio g. 18
LT-10306 Vilnius
Lietuva (Lithuania)
tel/fax:      (+370-5)-210 40 05
mobilus:      (+370-684)-49802, (+370-614)-36366