These files haven't been updated in more than fifteen years and their
number of translated strings has dropped to below twenty percent.
Also, prevent them from coming back when next syncing with the TP.
The documentation (https://www.haiku-os.org/docs/api/libroot.html)
says that BSD extensions are enabled by default, but somehow this
does not seem to be the case on my system. With _DEFAULT_SOURCE
defined, though, src/files.c now compiles fine.
When safe_tempfile() returns a valid filename, it should also
return a valid open stream.
This fixes https://savannah.gnu.org/bugs/?61064.
Bug existed since version 1.3.8, commit 5e068c60.
Opening a file whose name starts with a plus is not a problem at all.
Instead describe how to get the Alt+Up keystroke to scroll on a Linux
console (on Debian and friends).
This addresses https://savannah.gnu.org/bugs/?57294.
Any error message will already have been written to the terminal by
statusline(). Also, "Buffer not written to" looked too much like
"Buffer written to"; one had to look closely to see the difference.
Make sure that when multiple nanos with nameless buffers die,
each of them tries saving those buffers to different files.
This addresses https://savannah.gnu.org/bugs/?60902.
Reported-by: Julian Rosen <julianrosen@gmail.com>
Having a local variable that gets assigned to just once looks rather
poor. To me it makes more sense to have two separate exit points: one
for failure somewhere in the middle, and one for success at the end.
This way it is completely clear which value gets returned when.
And make the wording for the --historylog option similar.
This fixes https://savannah.gnu.org/bugs/?60868.
Bug existed since version 2.3.0, commit 8e2a9307, since the
saving and restoring of the cursor position was introduced.
When something that is called from die() calls die() again,
there is nothing we can do any more but give up.
This would have prevented the spiralling as
seen in https://savannah.gnu.org/bugs/?60853.
This mitigates a problem seen sometimes among the Red Hat crash reports:
somehow wgetch() keeps returning ERR, as if it were in nodelay mode.
(I've seen this happen once after nano came out of suspension.)
Using a much larger error-count value gives the user a few seconds
to type something and hopefully get out of the error condition.
Calling die() could trigger another call of emergency_save(), which
could fail and cause another call to statusline(), which would call
die() again... Spiralling until stack space runs out.
This fixes https://savannah.gnu.org/bugs/?60853.
Bug existed since version 5.1, commit 7942dab0.
Original report was in https://bugzilla.redhat.com/1976410.
Reported-by: Souptik Dutta Roy <duttaroy.souptik@gmail.com>
Removals:
- All eblits were dropped from Gentoo years ago.
- HDEPEND was never fully approved, superseded by BDEPEND.
- Keywords 'dohard' and 'dosed' have been banned since EAPI 4:
https://projects.gentoo.org/pms/5/pms.html#x1-130001r8
- Several arches are dead. For the currently supported list:
https://gitweb.gentoo.org/repo/gentoo.git/tree/profiles/arch.list
Additions:
- Two eapply functions replace the epatch functions, but epatch
is still supported until EAPI 6 is banned.
- Modern build systems (ninja and qmake5) have new functions.
- EAPI 7 added more builtins: dostrip, get_libdir, ver_*
- Highlight the cross-compiler boolean function. (There are too
many toolchain functions to add them all, but this one is very
common and has a long name, so highlighting it is useful for
catching typos early.)
The EAPI documentation:
https://projects.gentoo.org/pms/8/pms.html#x1-590007
An extensive write-up on EAPI 7:
https://mgorny.pl/articles/the-ultimate-guide-to-eapi-7.html
Signed-off-by: David Michael <fedora.dm0@gmail.com>
Add IDEPEND from EAPI 8, and also add more EAPI 7 roots so that all
of ROOT, EROOT, SYSROOT, ESYSROOT, and BROOT are supported.
Signed-off-by: David Michael <fedora.dm0@gmail.com>
Benno Schulenberg