The addition of stdckdint.h to CREATED_HEADERS in Makefile.in leads to
an error when installing loadable builtins on platforms that provide
the header:
install: cannot stat '/tmp/bash/stdckdint.h': No such file or directory
make[2]: *** [Makefile:903: install-headers] Error 1
make[2]: Leaving dire
Use `\fP` to restore previous font in places where it adequately
returns to roman face rather than the explicit `\fR` (which can clobber
the "previous font selection" datum in the formatter).
---
doc/bash.1 | 30 +++---
1 file changed, 15 insertions(+), 15 deletions(-)
dif
Double quotes aren't any easier to use with AT&T troff just because you
tuck them inside a string definition. (The _name_ of the string doesn't
matter; that gets interpolated before "quote removal", if you will.)
https://www.gnu.org/software/groff/manual/groff.html.node/Calling-Macros.html
---
d
While font mounting position 3 is usually assigned a bold typeface, it
seems a bit fragile to not explicitly ask for bold. The concept of
mounting positions is also not something man page authors typically need
to think about.
---
doc/bash.1 | 6 +++---
1 file changed, 3 insertions(+), 3 deletion
This prevents lines in the displayed struct definition from oversetting
on DWB troff.
Diff of rendering (DWB only):
@@ -10150,11 +10150,11 @@
The state of the History library is encapsulated into a sin-
gle structure:
-/*
- * structure for
Use a *roff minus sign for minus signs and the `-` character in email
addresses.
---
lib/readline/doc/history.3 | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/lib/readline/doc/history.3 b/lib/readline/doc/history.3
index 30b4d7be..40b85b20 100644
--- a/lib/readline/doc
This unfilled block of text was oversetting the line on DWB nroff, which
uses a line length of 65n (just like Seventh Edition Unix nroff).
Recast comments to occupy less space.
Also bracket this inline code display with `EX`/`EE` macros; this sets
the code in a monospace/constant-width font on pl
When formatting for typesetters, a blank input line produces 1 vee (one
line height, basically) of vertical space, but the default
inter-paragraph distance in man(7) is 0.4v.
Fixes numerous style warnings from "nroff -rCHECKSTYLE=3 -man" in groff
1.23.0 of the following form.
an.tmac:lib/readline
Replace no-op font selection escape sequence with what appears to have
been intended.
---
lib/readline/doc/history.3 | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/readline/doc/history.3 b/lib/readline/doc/history.3
index 40b85b20..7b6ab73b 100644
--- a/lib/readline/doc/hi
Set a period in the intended face (roman, not bold).
Stop breaking a line in the middle of a sentence without motivation.
Diff of rendering:
(groff Git, 1.23.0, 1.22.4; mandoc; Heirloom Doctools nroff)
- G Apply the following “s”
- or “&” modifier once to each word in the ev
Now that we're using a correct caret/circumflex/hat glyph on
modern typesetters (and terminals with a Unicode repertoire), drop the
thicket of partial-line motion and type size-altering escape sequences,
certain to frighten and confuse any unfrozen caveman page authors.
I expect this to slightly r
The following syntax is killing the page on AT&T troff. You get the
header and footer and nothing else--just an ocean of blank space.
.de Fn1
With traditional roff eyeballs, this is interpreted the same as this:
.deFn1
or:
.de Fn 1
...which may at last make the problem clear. Identifiers ar
When formatting for typesetters, a blank input line produces 1 vee (one
line height, basically) of vertical space, but the default
inter-paragraph distance in man(7) is 0.4v. Second, the package always
puts inter-paragraph spacing before a (sub)section heading anyway.
Third, forcing vertical space
Diff of rendering (ignoring differences in character set, default line
length, and a font selection affecting only a period):
(groff Git, 1.23.0, 1.22.4; mandoc; Heirloom Doctools nroff; DWB nroff)
- expansion, word splitting, pathname expansion, and quote removal .
+ expansion, word spli
On 3/26/24 9:02 AM, Grisha Levit wrote:
When \# or \! is decoded in a ${var@P} expansion embedded in a prompt
string, the value can be off by one from that of the same sequence
embedded directly in the prompt string.
Thanks for the report and patch.
Chet
--
``The lyf so short, the craft so lo
On 3/26/24 9:49 AM, Grisha Levit wrote:
Having funsubs in PS2 seems to cause input tokens that are split across
lines to be partially replaced with the last token of the last funsub
to be parsed while expanding the prompt string.
$ PS2='${ : ; }> '
$ printf '[%s]\n' ab\
> cd
El 27/03/24 a les 18:51, Chet Ramey ha escrit:
On 3/26/24 1:33 PM, Jordi Ferrer wrote:
Just in case for people running old servers:
This simple code cause a memory leak in bash:
#!/bin/bash
#
Array=("S" "E")
while [ 1 ]; do
17 matches
Mail list logo
