RFR(JDK12/JAXP/java.xml) 8207760: SAXException: Invalid UTF-16 surrogate detected: d83c ?

[Daniel Fuchs]

Hi Joe,

On 13/09/2018 00:25, Lance Andersen wrote:
> Hi Joe,
> 
> The change  seems reasonable

Agreed. However the following condition in ToStream::handleEscaping
is a bit cryptic:

1155             if ((ihs && (i + 1 < end)) || (ils && i != 0)) {
1156                 i++ ; // process two input characters
1157             }

could the comment be fleshed out to explain it?

I suspect that: `(ihs && (i + 1 < end))` means that
`writeUTF16Surrogate(c, ch, i, end);` has written the two surrogate, in
which case i should be incremented in order to skip the low surrogate
which has just been written.

I am not sure what `(ils && i != 0)` means, though...

best regards

-- daniel

> 
>> On Sep 12, 2018, at 2:11 PM, Joe Wang <huizhe.wang at oracle.com> wrote:
>>
>> Hi,
>>
>> Please review a patch for a situation where a surrogate pair is at the edge of a buffer. What the existing impl did was to report it as an error. This patch fixes it by caching the high surrogate and prints it out along with the low surrogate. Similar issue exists also in the CDATA section and is fixed in this patch. The CDATA impl had a couple of bugs where an indent could be written inside the CDATA and an unicode character written in between two CDATA sections. Both are fixed in this patch.
>>
>> JBS: https://bugs.openjdk.java.net/browse/JDK-8207760
>> webrevs: http://cr.openjdk.java.net/~joehw/jdk12/8207760/webrev/
>>
>> Thanks,
>> Joe
>>
>>
> 
>   <http://oracle.com/us/design/oracle-email-sig-198324.gif>
>   <http://oracle.com/us/design/oracle-email-sig-198324.gif> <http://oracle.com/us/design/oracle-email-sig-198324.gif>
>   <http://oracle.com/us/design/oracle-email-sig-198324.gif>Lance Andersen| Principal Member of Technical Staff | +1.781.442.2037
> Oracle Java Engineering
> 1 Network Drive
> Burlington, MA 01803
> Lance.Andersen at oracle.com <mailto:Lance.Andersen at oracle.com>
> 
> 
>