[jdk8u-dev] RFR: 8212678: Windows IME related patch

[Takuya Kiriyama]

On Mon, 24 Nov 2025 08:13:49 GMT, Frederic Thevenet <fthevenet at openjdk.org> wrote:

>> Hi All,  
>> 
>> I would like to backport this bug fix (JDK-8212768) done in jdk13 to jdk8. It is basically a clean backport, but only the copyright of awt_InputMethod.cpp and awt_Component.cpp is not updated because the current source is newer.  
>> 
>> The following fixes are resolved.
>> 
>> **Issue A: Cannot display surrogate pair character on Windows floating IME window for passive component**  
>> It has been reproduced using the test program (SwingSet2) in a specific environment and verified that this fix resolves it. 
>> 
>> **Issue B: Control Windows IME status using InputSubset and UnicodeBlock for CJK countries**  
>> It has been reproduced using the test program (ImSubsetsTest.java) in a specific environment and verified that this fix resolves it. 
>> 
>> **Issue C: Check pre-edit string availability for Windows Chinese IME**  
>> It probably won't happen again in the new environment. I have verified that the behavior of the test program  (IMFTestA.java) remains unchanged after this fix.
>> 
>> JTREG tests were executed using the `-a` option (automatic tests only) for the test suites  java/awt/Component and java/awt/InputMethods. No failures were found.
>> 
>> The changes implemented in JDK-8212678 alone are not sufficient to fully resolve IME-related issues. To ensure a complete fix, the changes from JDK-8258805 must also be integrated.
>> I plan to create a separate pull request.
>> 
>> Thank you.
>
> [Disclaimer: I'm not a formal reviewer for the JDK Updates project]
> 
> I had a look this PRs which, as stated above is indeed a clean backport (aside from copyright year) and according to its JBS issue, it was in fact  back-ported to Oracle's 8u291 release back in 2021.
> Furthermore, I have checked that the entirety of the original content this in patch is still present "as is" in the current JDK, which presumes it's not been found to cause  issues since it was introduced back in JDK 13.
> I therefore feel it is safe to integrate.
> 
> Regards,
> Frederic

Hi, @fthevenet , @gnu-andrew.
Thank you for your review! I create approval request.

-------------

PR Comment: https://git.openjdk.org/jdk8u-dev/pull/706#issuecomment-3575184345