Hi Chauncey,
Yes, generally this looks like a good solution. And a search on the
Internet suggests that the _JAVA_AWT_WM_NONREPARENTING variable is
pretty much a standard now. We'll still need to undergo an internal API
approval process (CCC) to adopt this new variable name, but I don't
expect any obstacles there.
Would you like to contribute a patch for this issue? You will need to
file a new bug at http://bugs.java.com/ , prepare a patch, test it, and
then send it to this mailing list for a review.
However, before we can accept fixes from you, you will need to have an
OCA signed. Please find more details here:
http://www.oracle.com/technetwork/community/oca-486395.html
--
best regards,
Anthony
On 9/9/2014 9:18 PM, Chauncey Goss wrote:
Wow, outlook decided to try to send this as HTML rather than plain text
and in the process screwed up the entire text. Apologies again for the triple
post,
I promise I'll never do this again.
AWT currently includes a (very incomplete) hard-coded list of
non-reparenting window managers. To work around this, many distributions
patch openjdk to add support for this to be controlled by an environment
variable (_JAVA_AWT_WM_NONREPARENTING), see for instance
(https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=508650 or
https://projects.archlinux.org/svntogit/packages.git/plain/trunk/openjdk7_nonreparenting-wm.diff?h=packages/java7-openjdk)
A more minimal implementation than the above is to add an extra clause
to isNonReparentingWM in XWM.java something along the lines of
(XWM.getWMID() == XWM.OTHER_WM &&
XToolkit.getEnv("_JAVA_AWT_WM_NONREPARENTING") != null) which avoids the
need to define a new WM.
This change seems like it would have minimal side effects but greatly
improve usability (a quick google search for java tiling window manager
reveals a lot of pain surrounding this issue).
Does this seem reasonable? Sorry if I'm breaking any protocols or am
missing prior discussions, this is my first time looking at the OpenJDK
source.
