86435 - Rules for XUL templates should allow inequalities

Status: RESOLVED FIXED

(Core Graveyard :: RDF, enhancement, P3)

Description

[Roger D]

The current XUL templating mechanism only allows rules that match exactly (give 
or take the presence of variables specified in bindings). This makes it 
difficult or impossible to write rules that match over a range of values (e.g. 
for values of an integer literal n in the range 0 <= n <= 60 - unless you are 
really willing to write 61 separate <rule> clauses). It would be nice to have a 
way of specifying inequalities in XUL template rule matches. A possible 
implementation would be something along the following lines:

<rule>
  <content uri="?uri" />
  <triple subject="?uri"
          predicate="http://home.netscape.com/NC-rdf#n"
          object="?n" />
  <comparison variable1="?n" comparator=">=" variable2="0" />
  <comparison variable1="?n" comparator="<=" variable2="60" />
</rule>

A more flexible and powerful alternative would be to allow a Javascript function 
to be called, taking an arbitrary number of values and returning a true or false 
value: the rule does not match if false is returned, and can match (assuming all 
other conditions apply) if true is returned:

<rule>
  <content uri="?uri" />
  <triple subject="?uri"
          predicate="http://home.netscape.com/NC-rdf#n"
          object="?n" />
  <test script="mytest(?n)" />
</rule>

Obviously for the template to behave well the script should behave in a 
consistant fashion and not use pseudorandom numbers or changing global variables 
to determine its answer.

Comment 1

[Chris Waterson]

Sounds like a great idea.

Comment 2

[Chris Waterson]

*** Bug 81508 has been marked as a duplicate of this bug. ***

Comment 3

[Asa Dotzler [:asa]]

Bugs targeted at mozilla1.0 without the mozilla1.0 keyword moved to mozilla1.0.1 
(you can query for this string to delete spam or retrieve the list of bugs I've 
moved)

Comment 4

[benc]

tever is not RDF QA anymore

Comment 5

[Axel Hecht]

waterson left the building

Comment 6

[Olli Pettay [:smaug][[e-mail]]]

We could maybe use the expression syntax from
http://www.w3.org/Submission/2004/SUBM-RDQL-20040109/
So something like <expression test="?n &gt;= 0"/>

Comment 7

[Neil Deakin]

The patch in bug 285631 provides this feature.

Comment 8

[Neil Deakin]

Comparisons are now implemented via:

<conditions>
  <where subject="?name" rel="before" value="Fred"/>
</conditions>