[
https://issues.apache.org/jira/browse/SOLR-10503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16052278#comment-16052278
]
Steve Rowe commented on SOLR-10503:
-----------------------------------
Thanks for the review, [~hossman].
In general, I assumed that CurrencyField would go away eventually (in 8.0 I
assume), and if CurrencyPointField and CurrencyField are tangled in various
ways, then removing CurrencyField would either leave unnecessary abstractions
in place, or require post-removal simplification.
bq. why copy so much of AbstractCurrencyFieldTest into a new
CurrencyPointFieldTest instead of just refactoring/extending it?
In a few places, AbstractCurrencyFieldTest directly refers to CurrencyField or
its members. Making it work with both would require both to implement/extend a
common thing. See above.
{quote}
bq. "I had to pull some top-level classes out of CurrencyField"
in general, maybe instead of refactoring these various inner classes into top
level classes,
{quote}
Those top-level classes were not inner classes; they were, as I said,
top-level. In general though I see your point.
bq. CurrencPointField still seems to suffer from SOLR-10502?
Yes, I forgot about it, thanks for the reminder.
bq. In SOLR-1050 I hypothosized changing to use docValues by default on the
subfields and letting the options on the "parent" field override the
indexed/docValues options on the subfields – I still think that's viable and
would not be opposed to it, but In hindsight I think a better model would be to
follow in the example of LatLonField & PointType and allow/force the user to
configure a codeStrSuffix and amountLongSuffix that must have corrisponding
dynamicFields defined in the schema.
I like it. LatLonField & PointType also include an alternative to the
subFieldSuffix (via AbstractSubTypeFieldType): subFieldType. I'm inclined not
to support both, and the suffix approach allows users more freedom. Making
them optional seems like a good idea to me.
I'll work on refactoring the patch.
> CurrencyField should be changed from TrieLongField to LongPointField for
> underlying raw-polyfield
> -------------------------------------------------------------------------------------------------
>
> Key: SOLR-10503
> URL: https://issues.apache.org/jira/browse/SOLR-10503
> Project: Solr
> Issue Type: Sub-task
> Security Level: Public(Default Security Level. Issues are Public)
> Reporter: Hoss Man
> Attachments: SOLR-10503.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
