[
https://issues.apache.org/jira/browse/VFS-771?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17108058#comment-17108058
]
Rémi Villé commented on VFS-771:
--------------------------------
I don't know what would be the good solution.
Maybe adding an option to FileSystemOptions to not set FileSystemObject.attr to
null during FileSystemObject.refresh() (attr == null causes the call to
getStat() during file.getType() of AbstractFileObject#traverse())
> SFTP multiple get bad performances if copyFrom() called after findFiles()
> -------------------------------------------------------------------------
>
> Key: VFS-771
> URL: https://issues.apache.org/jira/browse/VFS-771
> Project: Commons VFS
> Issue Type: Bug
> Affects Versions: 2.4.1, 2.6.0
> Reporter: Rémi Villé
> Priority: Major
>
> Similar to VFS-698: if you call sftpFile.findFiles(selector) before
> localFile.copyFrom(sftpFile, selector) the second call reset file stats in
> the cache (in .AbstractFileSystem#resolveFile() =>
> {code:java}
> if
> (getFileSystemManager().getCacheStrategy().equals(CacheStrategy.ON_RESOLVE)) {
> file.refresh();
> }
> {code}
> )
> Then the stats are retrieved one by one which result in poor performances.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
