[
https://issues.apache.org/jira/browse/CODEC-325?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17914708#comment-17914708
]
Julian Reschke edited comment on CODEC-325 at 1/20/25 2:57 PM:
---------------------------------------------------------------
I could use "setEncodeTable()", but then my code needs to specify the actual
table.
So yes, the ability to get the hex table would work as well. I would just avoid
the term "clone" in the message signature, as how it's generated is of no
concern for the caller.
That said, to keep things simple, I would *prefer* just a boolean switch (that
would choose the encode table internally, just like the old constructor). After
all, [https://datatracker.ietf.org/doc/html/rfc4648] specifies exactly two
encodings for Base32, so the ability to specify the whole table really sounds
like an edge case to me (but maybe I'm missing something here...).
was (Author: reschke):
I could use "setEncodingTable()", but then my code needs to specify the actual
table.
So yes, the ability to get the hex table would work as well. I would just avoid
the term "clone" in the message signature, as how it's generated is of no
concern for the caller.
> Base32 builder: make it easier to use the hex encoding table
> ------------------------------------------------------------
>
> Key: CODEC-325
> URL: https://issues.apache.org/jira/browse/CODEC-325
> Project: Commons Codec
> Issue Type: Wish
> Reporter: Julian Reschke
> Priority: Minor
>
> In our (Jackrabbit) quest to remove Guava dependencies, we started using the
> commons-codec Base32 class.
> I first started with the Builder, but noted that using hexEncoding is a bit
> ... complicated. Other than with the constructor:
>
> https://commons.apache.org/proper/commons-codec/apidocs/org/apache/commons/codec/binary/Base32.html#%3Cinit%3E(boolean)
> we need to specify the actual encoding table (a nice feature to have), but
> use one of the two known tables requires actually copying the table over.
> It would be nice if there either was a shortcut (using a boolean as before),
> or if the two "known" tables could be made visible to the caller (either
> using public constants, or some convenience methods).
> Best regards, Julian
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
