[Pkg-isocodes-devel] iso-codes 2.2-1: Please update the PO translation for the package iso-codes

[Alastair McKinstry]

I missed that one in the email lists.

Agreed, thats a good example of an incompatible upgrade, and 3.0 is  
justified.

Regards
Alastair

On 29 May 2008, at 13:47, Tobias Toedter wrote:

> On Thursday 29 May 2008 13:10:19 Alastair McKinstry wrote:
>> Can we avoid using the 3.0 version number? I would like to reserve  
>> the
>> major version number
>> to signify incompatible changes (that require upgrade/rewrite of iso-
>> codes reading software).
>> I'd prefer to see v2.111 than v120.1
>
> Well, the version has not been released yet, so it's not too late.  
> However, I
> did increase the major version, because in my opinion, we do have an
> incompatible change in our DTD / XML file for ISO 4217.
>
> Previously, the DTD said that the attribute "numeric_code" was  
> #REQUIRED. In
> the XML data, the entries for UIC-Franc and Gold-Franc did not  
> contain a
> numeric entry for those attributes, but three spaces "   " instead,  
> which is
> certainly not correct. In the only "official" source[1] I have  
> access to, the
> numeric value for those entries is specified as "Nil", which I would  
> not
> regard as empty value, but as a special value like NULL in SQL.  
> Therefore,
> I've changed the DTD to make the attribute numeric_code #IMPLIED, so  
> that it
> can be missing. Afterwards, I removed the attributes from UIC-Franc  
> and
> Gold-Franc. To me, this seems to be incompatible, because programs  
> using the
> XML file can no longer rely on numeric_code being always present.
>
> If we want to avoid version 3.0 at all, our option would be to  
> change the DTD
> back to #REQUIRED and either use the word "Nil" as value for the  
> attribute or
> use an empty string "".
>
> Originally, this has been reported as bug #310826 on Alioth.
>
> [1]
> <http://www.iso.org/iso/support/faqs/faqs_widely_used_standards/widely_used_standards_other/currency_codes/currency_codes_list-1.htm 
> >
>
> Regards,
> Tobias
>
> -- 
> Tobias Toedter   | Don't you just hate it when there's not enough  
> room to fin
> Hamburg, Germany |
> _______________________________________________
> Pkg-isocodes-devel mailing list
> Pkg-isocodes-devel at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/pkg-isocodes-devel

Regards,
Alastair

-- 
Alastair McKinstry  , <alastair at sceal.ie>     http://blog.sceal.ie

Anyone who believes exponential growth can go on forever in a finite  
world
is either a madman or an economist - Kenneth Boulter, Economist.