OK; here's where I'm ending the week:
toInstallOrUpdate = allMetadataRepos.query(
QueryUtil.createQuery(traverseQuery,
new
String[]{"com.example.rcpmain.ui.utils.feature.feature.group",
"com.example.rcpmain.ui.feature.feature.group"}),
monitor).toUnmodifiableSet();
where traverseQuery is:
final String traverseQuery = "$0.traverse(parent |
parent.requirements.collect(rc | select(iu |iu ~= rc)).flatten())";
Yields:
java.lang.IllegalArgumentException: Cannot convert a java.lang.String into
an iterator
I'm puzzled and would appreciate a steer about how to use the API. Thanks
in advance.
Regards,
Dave
On Fri, Dec 10, 2010 at 3:21 PM, David Orme
<[email protected]>wrote:
> Ahhh; I see you've already answered my question, but cleverly disguised as
> an explanation.
>
> As a UI guy, I'm used to posting questions and answers in the form of terse
> code snippets to the SWT newsgroup, so apologies while I adjust to the
> customs of the natives around here. ;)
>
> I have some questions/clarifications regarding this that I'll ask, as much
> for myself as for anyone else who searches the list archives later looking
> for this information:
>
> On Fri, Dec 10, 2010 at 8:35 AM, Thomas Hallgren <[email protected]> wrote:
>
>> It expects a collection (i.e. your set of roots) as it's first argument.
>> You then apply it to your IQueryable.
>>
>
> OK, this makes sense, but with two questions as caveats: The signature of
> a Query with parameters is:
>
> public static IQuery<IInstallableUnit> createQuery(String expression,
> Object... parameters) {
>
> 1) Do I literally pass a collection as the 0th vararg element or is each
> vararg element supposed to be a separate element of the collection? If a
> collection, can I just pass any arbitrary Iterable?
>
> 2) What is the data type of the parameter? A string which is the
> <featureID>.feature.group identifier? Something else?
>
>
>> In the expression, 'parent' is the variable holding each item from the
>> roots. The expression compares all iu's from your IQueryable with the
>> requirements of the 'parent' and this results in new instances that will
>> become 'parent' as the query traverses recursively.
>>
>
> Yes, this makes perfect sense. Thanks much for the detailed explanation.
>
> If I work out an answer (through trial and error experimentation) before
> someone answers, I'll post it here. Otherwise, if someone can get to it
> first, I'd appreciate a quick steer. Thanks in advance and...
>
>
> Regards,
>
> Dave Orme
>
>
>
_______________________________________________
p2-dev mailing list
[email protected]
https://dev.eclipse.org/mailman/listinfo/p2-dev
