RFR: 8307246 : Printing: banded raster path doesn't account for device offset values [v10]

[Prasanta Sadhukhan]

On Fri, 9 Feb 2024 09:47:25 GMT, vtstydev <duke at openjdk.org> wrote:

>>> > Revert the changes to the test which limit the pages printed out.
>>> > The test MUST print all orientations and MUST always print both opaque and alpha
>>> > 95% of the point of this test is to ensure consistency across all cases and if it doesn't test them it is mostly pointless.
>>> 
>>> @prrace May I revert everything in the test that @prsadhuk requested here: [#17030 (comment)](https://github.com/openjdk/jdk/pull/17030#issuecomment-1920510837) or you prefer that I change only how acting test by default?
>> 
>> I mean by default, when run by a testing framework, the test should print all those pages.
>> This is a manual test. It will be run rarely so when it does get run it ought to be thorough.
>> Manual doesn't mean someone typing "java AlphaPrintingOffsets", it means when jtreg is told to run the "@key manual" tests.
>> Any test options would only come in to play only if a developer wants to run the test directly by hand and chooses them.
>> FWIW I doubt they'll be useful but no harm leaving them but I'm also fine with complete reversion.
>> 
>> PS those options also complicate the instructions to the user !
>
> @prrace I would like to backport this path to jdk8. May I ask you to add labels into JBS ?

Hi @vtstydev It seems this bugfix has caused a regression [JDK-8372335](https://bugs.openjdk.org/browse/JDK-8372335) whereby the corner of the printout is cut-off as seen below...Will you be able to take a look?

<img width="3024" height="4032" alt="image" src="https://github.com/user-attachments/assets/36d4c605-5e1b-42a3-b38a-b927ce10ea7e" />

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

PR Comment: https://git.openjdk.org/jdk/pull/17030#issuecomment-3569610098