Hi George, these new xpbproxies are doing some not-nice things for me. I'm trying this latest with xterms and RetroOffice (an X build of NeoOffice) So I haven't even gotten as far as testing between X and Aqua. RetroOffice uses ^C/^V/^X (copy/paste/cut) within the application. Up to 2.3.1 or so, using the mouse to highlight then middle button to paste was unreliable, and pasting into another app with anything copied from RetroOfice was also unreliable whichever way it was copied. This was just within X, never mind trying to copy between X and Aqua. Now I at least have consistent behavior, but it isn't what I expect. First, ^C to copy no longer works at all in RetroOffice. Using cmd-C copies, but if it is something complex, like cells in a spreadsheet, it invokes the RetroOffice text import manager when pasting, which internal copy and paste should not do. It loses the formatting (as a consequence of invoking the import manager I presume). Actually, I had some italic text converted to plain but with the font changed to white, which took a while to figure out. Anyway, it plays havoc with normal spreadsheet moving groups of cells around sort of editing, which used to work fine. Secondly, using the mouse to highlight then middle button to paste is working only within the same app while the text stays highlighted. After the highlighting goes, the pasted text is whatever was previously yanked with cmd-C, wherever that happened. This is not nice, I am a highlight->middle button person by habit within X. Simple text copied with cmd-C in Aqua and pasted into RetroOffice using ^V or middle mouse works, as does cmd-C in RetroOffice and cmd-V in Aqua (FireFox in this case). But the cmd-C is necessary in X to get it to paste in Aqua, just highlighting text doesn't do it. I'm sure this has worked in the past (maybe as far in the past as Tiger, fink's xorg and autocutsel though). I have not changed the X11 Pasteboard preferences, they are all checked except the last (update Pasteboard immediately). Perhaps I need to adjust these preferences?? I also have focus follows mouse enabled, and I'm testing at the moment on a MacBook Air. Many thanks for the good work, having cut'n'paste work for things like images is way cool. I just don't happen to need it myself at the moment. -- Viv On Wed, 29 Oct 2008, George Peter Staplin wrote:
xpbproxy is a pasteboard proxy. It allows copying and pasting text and images between X11 and Mac applications.
*Please see the included README.txt for testing and installation instructions.*
Changes since the X11-2.3.2_beta2.pkg release:
1. PICT support has been added, so applications that set only the PICT clipboard type should now work with paste in X11 apps.
2. Copy and paste with Nedit works.
3. Copy and paste with Eterm works (see below).
Eterm has 3 bugs which have been worked around in the TARGETS property that it sends.
The first bug is that it sets the TARGETS property type to TARGETS instead of ATOM; which is not ICCCM compliant. The 2nd is that it mishandles the size of the TARGETS list. Thus, the Atoms are being sent as a series of 8-bit quantities, which will not work if say the server and client are running on different endian systems. The 3rd bug is that in some cases the list of TARGET atoms may be incorrect from Eterm, from truncated values. It's sending 16 bits of data for what should be 64 (2 Atoms). xpbproxy attempts to detect and work around the bugs.
The tarball may be found from the webpage here: http://people.freedesktop.org/~gstaplin/
or directly via this link: http://people.freedesktop.org/~gstaplin/xpbproxy_Oct-29-2008.tar.bz2
Please report problems to this xquartz-dev list or directly to me if you prefer via gstaplin at apple.com
Enjoy,
George -- http://people.freedesktop.org/~gstaplin/ _______________________________________________ Xquartz-dev mailing list Xquartz-dev@lists.macosforge.org http://lists.macosforge.org/mailman/listinfo.cgi/xquartz-dev
________________________________________________ Dr Viv Kendon http://quantum.leeds.ac.uk/~viv tel: +44 113 343 4864 Physics and Astronomy Quantum Information Group University of Leeds