[
https://issues.apache.org/jira/browse/WICKET-7042?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17710791#comment-17710791
]
ASF GitHub Bot commented on WICKET-7042:
----------------------------------------
theigl commented on code in PR #573:
URL: https://github.com/apache/wicket/pull/573#discussion_r1162428244
##########
wicket-core/src/main/java/org/apache/wicket/response/StringResponse.java:
##########
@@ -30,6 +30,8 @@
public class StringResponse extends Response
{
+ private static final int DEFAULT_INITIAL_CAPACITY = 128;
Review Comment:
> This is a problem indeed!
> We can improve StringResponse in 10.x and introduce LazyStringResponse in
9.x as deprecated.
@martin-g: I can do that. But in what order should I do it? Should I merge
the PRs as they are now, then backport them and then deprecate
`LazyStringResponse` in 9.x and unify the responses in 10x? Or should I unify
them now and create a separate PR for backporting? What do you usually do in
such a case?
> Improve sizing of StringResponse when writing scripts in PartialPageResponse
> ----------------------------------------------------------------------------
>
> Key: WICKET-7042
> URL: https://issues.apache.org/jira/browse/WICKET-7042
> Project: Wicket
> Issue Type: Improvement
> Components: wicket-core
> Affects Versions: 9.12.0
> Reporter: Thomas Heigl
> Assignee: Thomas Heigl
> Priority: Major
>
> When rendering scripts in PartialPageResponse, an AppendingStringBuilder with
> initial capacity 1024 is created. This buffer is then written to a
> StringResponse that uses a buffer with an initial capacity of 128 causing
> unnecessary allocations.
> We should create the StringResponse with enough capacity to hold the scripts.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
