[
https://issues.apache.org/jira/browse/CSV-184?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15331240#comment-15331240
]
M. Steiger commented on CSV-184:
--------------------------------
I believe that passing a Map to the toMap() method makes sense, because it
allows users to pass any implementation and most importantly, reuse a Map
instance.
I see two shortcomings of the current implementation:
a) HashMap does not preserve insertion order
b) A new instance is created for every record
c) Iteration is slow (at least relatively speaking) since entries are unlinked
Imho, it could be improved by
a) using a LinkedHashMap
b) reusing the Map instance across the entire iteration (and protecting the
returned instance by Collections.unmodifiableMap )
c) Alternatively allow users to provide a Map instance
> CSVRecord toMap Method(LoadFactor for HashMap)
> ----------------------------------------------
>
> Key: CSV-184
> URL: https://issues.apache.org/jira/browse/CSV-184
> Project: Commons CSV
> Issue Type: Improvement
> Components: Parser
> Affects Versions: 1.4
> Reporter: Gaurav Agarwal
> Priority: Minor
>
> Set loadFactor for Hashmap Object to 1(Default : 0.75) as currently it will
> always resize the map when we are putting entries in it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
