[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: lynx, ncurses, gpm



Glyn Kennington <glyn.kennington@hertford.oxford.ac.uk> wrote:



GK> I have found some ways that work, though.  For all the
GK> following, I'm using my hacked gpm ncurses compiled after
GK> a ./configure --with-shared --with-gpm --with-develop the
GK> untouched lynx binary, with LD_LIBRARY_PATH set to find the
GK> new libraries

GK> versions that work: ncurses-5.2.tar.gz ncurses-5.3.tar.gz

GK> These are both from http://ftp.gnu.org/pub/gnu/ncurses/ The
GK> delay-till-keypress still occurs immediately after an fg,
GK> but once a key has been pressed, subsequent mouse presses
GK> are processed immediately.  Also, with the original gpm,
GK> the exit-on-fg bug only happens in when the ^Z is issued
GK> immediately after a previous fg, with no key presses in
GK> between.

GK> versions that don't work: ncurses-5.2.20020112a-7 (obtained
GK> from apt-get source ncurses) ncurses-5.2.20020112a (without
GK> applying the debian a-7.diff.gz)

GK> These have the delay-till-keypress bug at all times.

GK> So it seems that something broke in the middle of the 5.2-
GK> releases, and was put right again for the 5.3 release.  Is
GK> that a correct understanding?

In case you did not change versions of gpm then possibly so.
the ncurses FAQ makes mention of a problem with certain
versions of gpm that has a bug in them that prevents libgpm
from linking properly to ncurses. This could be tested by doing
a 'nm libgpm.so' and see if wgetch is defined in it and not
listed as an external. This is according to some old notes I
have on this issue and thus the bug is supposedly with gpm and
not ncurses.

You can always file a bug with the Debian BTS. I believe that
Thomas Dickey the ncurses upstream maintainer is reading the BTS.




Regards,

Morten

-- 
"Anyone can do any amount of work provided it isn't the work he is supposed
 to be doing at that moment."                             (Robert Benchley)




Reply to: