| Age | Commit message (Collapse) | Author |
|---|
|
which hostkeys are already recorded in known_hosts. This avoids
hostkey warnings when connecting to servers with new ECDSA keys
that are preferred by default; with markus@
|
|
to vi.
Based on a diff from martynas@, previously requested by a couple of
other people.
|
|
report/fix Joachim Schipper; ok djm@
|
|
it is time to remove the .Sp, .Vb, and .Ve kludge
that was added to the man(7) library to build Perl manuals.
|
|
to MANDOCERR_LINESCOPE, i.e. from FATAL to ERROR.
That is, when an element next line scope is empty except for one
or more NSCOPED macros (like in .B .br .br .B), throw away the
empty element including its NSCOPED children such that the parser
can prod on.
While here, improve the error message and tell the user which element
has been thrown away.
|
|
Of course, we don't want to encourage low-level physical markup,
but pod2man(1) writes such requests, so Perl manuals contain them,
and some Xenocara and lots and lots of ports manuals use them as well.
In base and Xenocara, this will reduce mandoc -Tlint ERROR noise;
in ports, it will improve rendering of many manuals.
|
|
requests .am, .ami, .am1, .dei, and .rm for a long time.
Since ignoring them can (rarely) cause information loss and serious
misformatting, throw an ERROR: NOT IMPLEMENTED when finding them.
Implementing them would not be too difficult, but they are so rare
in practice that i can find better use for my time right now.
In this context,
- Put the string "NOT IMPLEMENTED" into two other error messages
as well, to distinguish them from those caused by broken input.
- Print the string "unknown macro" once, not twice in the error message
associated with MANDOCERR_MACRO, and begin printing the buffer at the
point where the unknown macro really is, not at the start of line.
|
|
Ignoring these can neither cause information loss nor serious
formatting issues. As they are frequently used by pod2man(1),
this considerably reduces ERROR noise from mandoc -Tlint
for the Perl manuals.
|
|
such that, after some more improvements, we will eventually
be able to install it.
jmc@ agrees with the plan.
|
|
1) When using a user-defined string of length 0 as a macro,
do not access memory before the start of the string (segfault).
2) When beginning to define a user-defined macro, initialize
the string representing the macro to the empty string,
not to the NULL pointer, such that, in case the macro turns
out to not have any content, like in
.de IX
..
the macro will be defined and empty instead of undefined.
This avoids large numbers of bogus mandoc ERROR messages about
undefined macros (which are actually defined and empty), in
particular in man(7) code generated from pod2man(1), for example
in Perl and OpenSSL.
|
|
Extend and improve .ds documentation.
Improve .ig and .nr documentation.
Add minimal explanations for .am, .ami, .am1, and .dei.
Start improving the distinction of requests and macros.
Add myself as an author.
|
|
This file still needs lots of love before we can consider installing it.
|
|
transfers, e.g. to enable agent forwarding which is particularly useful
in this case; bz#1837 ok dtucker@
|
|
The .de1 instructions is a GNU extension not found in traditional roff and
not even in old groff, defined as "define a macro that will be executed with
traditional roff compatibility mode switched off during macro execution".
Since we ran into it in the wild, we have been parsing and ignoring it for
a long time. Now that we have proper .de support, we can as well use the
contents, even though we don't implement compatibility mode at all.
|
|
such that arguments can contain blank characters.
Also support escaping of quote characters by doubling them.
For example, the argument "a""b c." resolves to: a"b c.
|
|
This fixes various Xenocara manuals.
Do not define your own macros in new manuals, though:
this code exists purely to cope with existing and old stuff.
Like in both traditional and GNU roff, the .de and .ds (define string)
roff instructions share the same string table, so one can abuse strings
as macros and vice versa. This implementation supports multi-line
user-defined macros and user-defined macros taking up to 9 arguments.
Project started near the end of p2k10, now mature for production,
but there is still room for future improvements in various respects.
|
|
ok markus deraadt dtucker
|
|
in debug mode (channel_close_fds is called transitively from the session
code post-fork); bz#1719, ok dtucker
|
|
channel; report bz#1842 from jchadima AT redhat.com; ok dtucker@
|
|
global options.strict_modes
|
|
be and I can't find it, but the flag itself is a useless optimisation
that only applies to automatic-resize windows, so just dispose of it
entirely.
Fixes problems reported by Nicholas Riley.
|
|
buffers rather than directly on files. This will make a few things
easier to do in the future; ok markus@
|
|
feedback and ok markus@
|
|
|
|
tedu agreed with an idea, tested by Luis Useche and me; ok deraadt
|
|
|
|
Now that we have dynamically sized buffers there may not be room
for the NUL until after realloc.
|
|
|
|
|
|
same line.
|
|
the returned count and set and end pointer instead. we no longer
need to allocate a dummy array when diffing against a non-existent
directory so eliminate slurpdir() and call scandir() directly.
OK krw@
|
|
outputs anymore, otherwise they can't be cleaned up by dev_close()
|
|
|
|
when the active window is closed and freed. Reported by sthen@.
|
|
hardcoding lowdelay/throughput.
bz#1733 patch from philipp AT redfish-solutions.com; ok markus@ deraadt@
|
|
|
|
using lint.
|
|
|
|
|
|
these have been around for years by this time. ok markus
|
|
instead. This makes slurpdir() just a thin wrapper around scandir(3).
OK schwarze@
|
|
|
|
dynamic range of 24-bit hardware can be used. Offline processing like
mixing and channel mapping is done with 24-bit precision too.
Since most hardware and audio programs use 16-bit precision, 24-bit
processing is a small waste of CPU time in many cases, so 24-bit
processing is not desirable and this is disabled by default.
Furthermore it's available on i386, amd64 and sparc64 only. To
experiment with it:
make COPTS="-DADATA_BITS=24"
|
|
|
|
than s16 (using hardcoded constants). Currently s16 is the native format,
so no behavior change.
|
|
updates instead of showing absolut figures. Makes it easier to
estimate the "livelockyness" of a system.
OK dlg@
|
|
|
|
audio samples and cleanup ADATA_XXX macros. This allows easilly
switching to 24 bit fixed point arithmetic by simply redefining
the adata_t typedef to int and updating ADATA_XXX macros. No
object change.
|
|
|
|
- Tn for "YACC" is silly
- since ex(1) and vi(1) are the same page, let's not list them both in SEE ALSO
|
