According to Gabriel Gerhardsson <[EMAIL PROTECTED]>:
> For example, I tar the dictionary and distribute it that way in the rpm.
> Only when the rpm is installed, I untar it. Is this the right way of
> doing it? or is rpms compressed automaticly?
> And I wouldn't mind if anyone would suggest some better wordings in the
> %description and such. I did just put something there.
>
> Btw, I put up a test rpm here:
> www.cs.umu.se/~dva99ggn/abiword-dictionary-se-0.0.1-1.noarch.rpm
This is not a noarch, as currently hash files are endian dependent and
each arch has is own endian.
On the other, I think it is time to say what I think about spell checking
dictionnaries.
First, the AbiWord package should depend on abispell package. Each
abispell package has its own name with the language in it (say
abispell-fr-FR) and each provide abispell. So that we always have
one of the abispell package, but can have several. No language is
required. Just one.
Any idea ?
Hub
