Two things about this: a) ArrayList, for all intents and purposes should be considered depreciated from .NET 2.0 on. If you want an ArrayList, then use List<object>.
b) SupportClass isn't very ".NET". Nested types like this don't really make
sense from a .NET perspective. If anything, a namespace should be created and
these items added to that namespace. This is a type with functionality
completely unto itself, with no relation to the nesting type other than it is
used in a utility sense. Nested types should have some relation to the code
that they are nested in. Namespaces, in the .NET point of view, are the
better logical container.
- Nick
-----Original Message-----
From: George Aroush (JIRA) [mailto:[email protected]]
Sent: Tuesday, November 10, 2009 12:24 AM
To: [email protected]
Subject: [jira] Commented: (LUCENENET-218)
Lucene.Net.Util.TestAttributeSource.TestCloneAttributes
[
https://issues.apache.org/jira/browse/LUCENENET-218?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12775297#action_12775297
]
George Aroush commented on LUCENENET-218:
-----------------------------------------
Like you pointed out the order is important, so we have to do the same.
Instead of writing a mimic of LinkedHashMap in C#, why not -- maybe at the
cost of performance, but let leave that for the profiler to tell us -- use
ArrayLIst?
If you write a LinkedHasMap, it should go into SupportClass.cpp
> Lucene.Net.Util.TestAttributeSource.TestCloneAttributes
> -------------------------------------------------------
>
> Key: LUCENENET-218
> URL: https://issues.apache.org/jira/browse/LUCENENET-218
> Project: Lucene.Net
> Issue Type: Bug
> Reporter: Michael Garski
>
> The test is failing - I'm working on it :)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
smime.p7s
Description: S/MIME cryptographic signature
