Re: humanize.ordinal is very English-specific

2010-03-26 Thread Shai Berger
On Friday 26 March 2010, Russell Keith-Magee wrote: > However, if the OP (or anyone else) can propose a simple, low maintenance > solution that will allow ordinal to support more languages than it does > currently, I'm happy to entertain it. > Ok, I have a 2-part proposal: 1) As I've hinted befor

Re: Re: humanize.ordinal is very English-specific

2010-03-25 Thread Russell Keith-Magee
On Fri, Mar 26, 2010 at 10:57 AM, Sergiy Kuzmenko wrote: > ... just to clarify: >> There appears to be some confusion here. The pluralize filter *already >> exists*. It has existed since Django was open sourced. if you have >> complex pluralization rules to support, it wont work, but in those >> c

Re: Re: humanize.ordinal is very English-specific

2010-03-25 Thread Sergiy Kuzmenko
... just to clarify: > There appears to be some confusion here. The pluralize filter *already > exists*. It has existed since Django was open sourced. if you have > complex pluralization rules to support, it wont work, but in those > cases, you can use {% blocktrans %} in templates, or calls to > u

Re: Re: humanize.ordinal is very English-specific

2010-03-25 Thread Sergiy Kuzmenko
Hm, I thought I was specific: use the same approach that GNU's ngettext does! And apologies for spinning this conversation off the original subject. It appears to me that in GNU's ngettext the problem of pluralization is solved, and what is more, ngettext has the same interface as Django's ungette

Re: Re: humanize.ordinal is very English-specific

2010-03-24 Thread Russell Keith-Magee
On Thu, Mar 25, 2010 at 8:32 AM, Sergiy Kuzmenko wrote: >> If you have any suggestions for how we could expose ungettext better >> in templates, I'm all ears. However, given the complexity of >> pluralization rules and pluralization usage in the general case, it's >> not a simple task. Suggestions

Re: Re: humanize.ordinal is very English-specific

2010-03-24 Thread Sergiy Kuzmenko
> If you have any suggestions for how we could expose ungettext better > in templates, I'm all ears. However, given the complexity of > pluralization rules and pluralization usage in the general case, it's > not a simple task. Suggestions welcome. My suggestion is to make pluralization compatible

Re: Re: humanize.ordinal is very English-specific

2010-03-24 Thread Russell Keith-Magee
On Wed, Mar 24, 2010 at 10:31 PM, wrote: > Regarding Russ' comment on pluralization: > >>  Option 1: Spend a lifetime trying to make a single tag that can >> accommodate every possible pluralization rule scheme >> >>  Option 2: Provide a simple implementation that works for a lot of >> cases, and

Re: Re: humanize.ordinal is very English-specific

2010-03-24 Thread s . kuzmenko
Regarding Russ' comment on pluralization: Option 1: Spend a lifetime trying to make a single tag that can accommodate every possible pluralization rule scheme Option 2: Provide a simple implementation that works for a lot of cases, and encourage others to write specific template tags for thei

Re: humanize.ordinal is very English-specific

2010-03-23 Thread Russell Keith-Magee
On Wed, Mar 24, 2010 at 4:03 AM, Shai Berger wrote: > Hi, > > I've been doing some translation work and ran into the humanize.ordinal > template tag; I found it extremely English-specific. It works by choosing > suffixes ("st", "nd", "rd", "th") to add to numeric representation (turning > e.g. 22

Re: humanize.ordinal is very English-specific

2010-03-23 Thread Shai Berger
Hi again, On Tuesday 23 March 2010, s.kuzmenko wrote (I'm assuming you intended this to go to the group, rather than just me): > That's a hard one. In some languages those suffixes will depend on the noun > they refer to. To elaborate further your French example, the equivalent > of "1st" would