[ 
https://issues.apache.org/jira/browse/COMPRESS-667?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17821859#comment-17821859
 ] 

Gary D. Gregory commented on COMPRESS-667:
------------------------------------------

{quote}The current Builder implementation appears to not have convenience 
methods for the most common construction scenarios e.g. construction from a 
File or Path.
{quote}
It does, look for all callers of {{ZipFile.builder()}} and you'll find plenty 
of tests using different input types. For example:

ZipFileTest.testZipWithShortBeginningGarbage():
{code:java}
try (ZipFile zipFile = ZipFile.builder().setPath(path).get()) {
 ...
}
{code}
Constructors on builders just duplicates part of the problems of constructors 
on a domain class; there is always one more "convenience" constructor to add, 
and then you're back to ZipFile's fourteen constructors with no end in sight.

> ZipFile Builder lacks convenience constructors for File/Path
> ------------------------------------------------------------
>
>                 Key: COMPRESS-667
>                 URL: https://issues.apache.org/jira/browse/COMPRESS-667
>             Project: Commons Compress
>          Issue Type: Improvement
>    Affects Versions: 1.26.0
>            Reporter: Daniel Lowe
>            Priority: Major
>
> The current Builder implementation appears to not have convenience methods 
> for the most common construction scenarios e.g. construction from a File or 
> Path.
> As 
> {color:#000000}{color:#000000}ZipFile.{color}{color:#000000}builder{color}{color:#000000}().get()
>  {color}{color}isn't valid, personally I think the Builder constructor itself 
> should require a 
> {color:#000000}{color:#000000}SeekableByteChannel/{color}{color}File/Path{color:#000000}{color:#000000}
>  {color}{color}
> I tend to agree with Elliotte's comments 
> ([https://lists.apache.org/thread/5vtbbs18d4rntn8tpnpsvo0g9h6svc30),] and 
> would also see de-deprecating the one argument  constructors as a valid 
> solution



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to