Hi Tom,
Your example is excellent. If you or anyone else develop other collations,
esp. for Greek and Latin, please add them to the new DCwiki page I set up:
http://wiki.digitalclassicist.org/Collations_for_Ancient_Languages_in_XSLT_
and_XQuery
And if I've not captured the language of the problem well, please revise
the body of the article.
Thanks!
jk
--
Joel Kalvesmaki
Editor in Byzantine Studies
Dumbarton Oaks
1703 32nd St. NW
Washington, DC 20007
(202) 339-6435
On 2/20/14 9:41 AM, "Tom Elliott" <[log in to unmask]> wrote:
>Joel:
>
>Gah! I had just assumed greek worked, having recently done it this way
>with Syriac and had no problems.
>
>There is a saxonic way to do a custom collation in XSLT; we used it to
>produce sorts that deal with extended Latin characters in an intuitive
>fashion. Viz:
>
>definition:
> https://github.com/srophe/srp-places-app/blob/master/xsl/collations.xsl#L
>8
>
>application:
> https://github.com/srophe/srp-places-app/blob/master/xsl/placepage.xsl#L3
>34
>
>No reason it couldn¡¯t be done for Greek in short order, especially given
>the code you already have.
>
>Tom
>
>Tom Elliott, Ph.D.
>Associate Director for Digital Programs and Senior Research Scholar
>Institute for the Study of the Ancient World (NYU)
>http://isaw.nyu.edu/people/staff/tom-elliott
>
>
>
>On Feb 20, 2014, at 8:35 AM, Kalvesmaki, Joel <[log in to unmask]>
>wrote:
>
>> Hi Jens:
>> I'm using XQuery/XSLT to create alphabetized concordances of linguistic
>> tokens from sequences generated both as isolated text() nodes and as
>> sequences generated by applying fn:tokenize() to longer strings. My
>> results are the same independent of the method (i.e., node vs.
>>sequence).
>> I'm not certain I understand your question, so I'm not certain I've
>> answered it.
>>
>> Hi Tom:
>> Yes, I've tried @lang="(gr|grc|gr-AN|grc-AN)", and to no effect.
>>
>> Interestingly, Michael Kay, founder of Saxonica--or at least someone
>>with
>> his name--in a post in 2004 implies that grc collation was not
>> accommodated by his or any other engine:
>> https://www.stylusstudio.com/xsllist/200404/post20790.html
>>
>> I know Saxon allows customized collations, but I wanted to see if anyone
>> on this list knew if one had already been written.
>>
>> Best wishes,
>>
>> jk
>> --
>> Joel Kalvesmaki
>> Editor in Byzantine Studies
>> Dumbarton Oaks
>> 1703 32nd St. NW
>> Washington, DC 20007
>> (202) 339-6435
>>
>>
>>
>>
>>
>> On 2/20/14 5:11 AM, "Jens ¨ªstergaard Petersen" <[log in to unmask]>
>>wrote:
>>
>>> Hi Joel,
>>>
>>> You started out by writing that you were making a transform ¡© could I
>>>ask
>>> which XQuery functions you use in the transform which require a special
>>> ancient Greek collation? An (identity) transform can be used for many
>>> things, but if you are making an XML > HTML transform, for instance,
>>> collation would not normally be an issue, since it does not involve
>>> sorting or string comparison.
>>>
>>> Jens
|