Codereview request for 8006295: Base64.Decoder.wrap(java.io.InputStream) returns InputStream which throws unspecified IOException on attempt to decode invalid Base64 byte stream
xueming.shen at oracle.com
Thu Jan 31 22:55:18 UTC 2013
Because the error is not really triggered by an io operation when
reading from the underlying input stream. Ideally it should be a
more specific "InvalidedBase64EncodingException" exception in
this kinda of situation (like we we do for the decoder/encoder in
charset), but it is considered too heavy in Base64 class. Maybe
the word "obviously" in my email is not appropriate, it may be
more appropriate worded as "IAE is more appropriate than
IOE for ...".
On 01/31/2013 02:46 PM, Louis Wasserman wrote:
> In what way is this "obviously" inappropriate? (I don't object, I just don't follow.)
> On Thu, Jan 31, 2013 at 2:31 PM, Lance Andersen - Oracle <Lance.Andersen at oracle.com <mailto:Lance.Andersen at oracle.com>> wrote:
> On Jan 31, 2013, at 5:15 PM, Xueming Shen wrote:
> > Hi,
> > Obviously ioe is not an appropriate exception for invalid base64 bytes,
> > and it's inconsistent with the rest of decode methods, as the submitter
> > suggested.
> > The change is to explicitly specify (as other decoding methods do) that
> > IAE will be thrown if the input stream contains invalid base64 bytes. The
> > impl is also updated according.
> > Please help review.
> > http://cr.openjdk.java.net/~sherman/8006295/webrev/ <http://cr.openjdk.java.net/%7Esherman/8006295/webrev/>
> > Thanks
> > -Sherman
> Lance Andersen| Principal Member of Technical Staff | +1.781.442.2037 <tel:%2B1.781.442.2037>
> Oracle Java Engineering
> 1 Network Drive
> Burlington, MA 01803
> Lance.Andersen at oracle.com <mailto:Lance.Andersen at oracle.com>
> Louis Wasserman
More information about the core-libs-dev