Re: [xsl] XSLT 4 xsl:with

["Pieter Lamers pieter.lamers@xxxxxxxxxxxx" <xsl-list-service@xxxxxxxxxxxxxxxxxxxxxx>]

Hi Gerrit,

Thanks for your response, the <xsl:sequence> afterthought you brougtht 
up is finally dawning upon me. I did initially read through the thread, 
and I also think it is better to use existing vocabulary and to fill the 
gaps in the paradigm, rather than to cook up new verbs. It's just that I 
/also/ like <xsl:with>, which feels completely familiar to me, even 
though it does not exist yet.

@Michael Kay, I am happy to see that the discussion you intended to 
provoke with your article and presentation at XML Prague is actually 
happening! I was a bit worried that there would not be a single place 
where this would going to be, and now there could be two. (for the 
unfamiliar: 
https://www.xml.com/news/2020-04-slack-workspace-for-the-xml-community/)

Best
Pieter

On 19/05/2020 19:55, Imsieke, Gerrit, le-tex gerrit.imsieke@xxxxxxxxx wrote:
> Of course such an xsl:with instruction won't be penalized by HOAXCoQS 
> [1], unlike xsl:for-each, which reeks of XSLT 1 habits most of the time.
> xsl:with won't get brownie points though; I'd consider it "neutral".
>
> But read my previous messages in this thread. Instead of xsl:with I 
> propose to use a @context-item attribute (an @xsl:context-item 
> attribute on non-xsl elements). If you really need the kind of wrapper 
> that xsl:with provides, you can still write
>
> <xsl:sequence context-item="$context">
> B  <xsl:variable .../><!-- @select evaluated in the context of $context 
> -->
> B  <foo>...</foo><!-- any expression within is also evaluated in the 
> context of $context -->
> B  <bar>...</bar>
> </xsl:sequence>
>
> but most of the time the XSLT code will become more compact because 
> the context can be set by an attribute on almost any element, not with 
> an instruction.
>
> -- Gerrit
>
> [1] https://github.com/sydb/HOAXCoQS
>
> On 19.05.2020 19:14, Pieter Lamers pieter.lamers@xxxxxxxxxxxx wrote:
>> I think <xsl:with select="..."> is an elegant rewrite of for-each. It 
>> does make me wonder whether it would earn brownie points or penalties 
>> in Gerrit's highly opiniated book.
>>
>>
>> On 19/05/2020 18:52, Liam R. E. Quin liam@xxxxxxxxxxxxxxxx wrote:
>>> On Tue, 2020-05-19 at 08:54 +0000, Pieter Masereeuw pieter@xxxxxxxxxxxx
>>>> Because such a for-each does not repeat, I always add a comment like
>>>> <!-- Merely sets the context, does not repeat -->.
>>> I tend to comment these too, although not with the repeat part - be
>>> careful, too, as in English a less-informed reader might get an
>>> incorrect impression reinforced that for-each is a loop, rather than a
>>> mapping operator.
>>>
>>> Yes, an xsl:with-context-item select=.... would be clearer. Wkether
>>> it's worth increasing the size of the language for it is a judgment
>>> call that's a little harder, i think.
>>>
>>> Best,
>>>
>>> Liam
>>>
> 

-- 
Pieter Lamers
John Benjamins Publishing Company
Postal Address: P.O. Box 36224, 1020 ME AMSTERDAM, The Netherlands
Visiting Address: Klaprozenweg 75G, 1033 NN AMSTERDAM, The Netherlands
Warehouse: Kelvinstraat 11-13, 1446 TK PURMEREND, The Netherlands
tel: +31 20 630 4747
web: www.benjamins.com