Is there hopefully a way to search for an exact string in Man Pages? I know if I want to search for anything containing -c I can just do this.
/-c
How would I search for "-c"? I want only "-c" to show up. So I tried this.
/"-c"
It took me literally and looked for the quotes also.
did you try grep ???
You can use ^ for beginning of line, but apparently not $, at least not on my ksh.
/^-c
You can grep in the shell history, but there are control characters in there, so look at it with "cat -vte" or "od -bc" first.
I forgot to say I'm talking about in the man pages. When you go to the help pages it says something similar to this.
Man pages are displayed on $PAGER, so you get your choice. The default is more or less. I use pg when i can get it. However, pager regex rules are a bit kinky.
Beyond what the more or less man page says about pattern searching and moving forward and, if you have a file input, backward:
The man pages have teletype print artifacts in them like "_\ba" for underscored "a" and "c\bc" for bold "c" (\b is backspace to overstrike). You can write a sed routine to flush out the .\b or you can use a web man page.
sed '
s/\(.\)\^H\1/\1/g
s/\(.\)\^H\1/\1/g
s/\(.\)\^H\1/\1/g
s/\(.\)\^H_/\1/g
s/_\^H\(.\)/\1/g
s/.\^H\(.\)/\1/g
'|sed '
/^$/{
:l
$b
N
s/^\n$//
t l
}
'
or in C:
$ cat mysrc/manclean.c
#include <stdio.h>
static void p_putchar( int c )
{
if ( c != EOF
&& EOF == putchar( c ) )
{
if ( ferror( stdout ) )
{
perror( "stdout" );
exit( 1 );
}
exit( 0 );
}
}
main()
{
static int c ;
static int c2 ;
static int c3 ;
static int ct = 0 ;
do
{
switch( c = getchar() )
{
case EOF:
if ( ferror( stdin ) )
{
perror( "stdin" );
}
continue ;
case '\b':
if ( ct < 2 )
{
ct = 0 ;
continue ;
}
c2 = c3 ;
ct = 1 ;
continue ;
case '\n':
if ( ct > 1
&& c3 == '\n'
&& c2 == '\n' )
{
ct = 2 ;
continue ;
}
/* intentional fall through */
default:
break ;
}
switch( ++ct )
{
case 1:
c2 = c ;
continue ;
case 2:
break ;
default:
p_putchar( c3 );
break ;
}
c3 = c2 ;
c2 = c ;
} while ( c != EOF );
switch( ct )
{
case 0:
break ;
case 1:
p_putchar( c2 );
default:
p_putchar( c3 );
p_putchar( c2 );
break ;
}
exit( 0 );
}
you can use \b
when viewing man, type any of these
/\b-c\b
/\b-c
/-c\b
What does the \b do? It doesn't seem like it is doing anything.
Someone figured this out for me and it works.
"/ -c "
(without the quotes but notice the space before and after the -c)
This also does the trick.
/[^a-z]-c[^a-z]
i used to use
man any_ninja_command | less
since less highlight the pattern you search for, which is much more confortable
by the way, if
/-c
didn't work, did you try
/\-c
?
\b is backspace, part of bold or underline for teletype printers in nroff/troff, which is the "word processor" for man pages.
I'll sleep a bit less ignorant tonight, tya DGnitPick:)
I think in this case it stands for "word boundary" not backspace.
Oki dude, thanx !
My bad:
Regex Tutorial - \b Word Boundaries
For me, the pattern '-..c' works, because of the overtyping boldness encoded by nroff/troff in the man page, see man grep example below:
$ man grep
grep(1) grep(1)
NAME
grep, egrep, fgrep - search a file for a pattern
SYNOPSIS
Plain call with pattern
grep [-E|-F] [-c|-l|-q] [-bhinsvx] pattern [file ...]
.
.
.
-c Only a count of matching lines is printed.
.
.
.
.
===========================================================
$ man grep | cat -v
g^Hg^Hg^Hgr^Hr^Hr^Hre^He^He^Hep^Hp^Hp^Hp(^H(^H(^H(1^H1^H1^H1)^H)^H)^H) g^Hg^Hg^Hgr^Hr^Hr^Hre^He^He^Hep^Hp^Hp^Hp(^H(^H(^H(1^H1^H1^H1)^H)^H)^H)
N^HN^HN^HNA^HA^HA^HAM^HM^HM^HME^HE^HE^HE
grep, egrep, fgrep - search a file for a pattern
S^HS^HS^HSY^HY^HY^HYN^HN^HN^HNO^HO^HO^HOP^HP^HP^HPS^HS^HS^HSI^HI^HI^HIS^HS^HS^HS
P^HP^HP^HPl^Hl^Hl^Hla^Ha^Ha^Hai^Hi^Hi^Hin^Hn^Hn^Hn c^Hc^Hc^Hca^Ha^Ha^Hal^Hl^Hl^Hll^Hl^Hl^Hl w^Hw^Hw^Hwi^Hi^Hi^Hit^Ht^Ht^Hth^Hh^Hh^Hh p^Hp^Hp^Hpa^Ha^Ha^Hat^Ht^Ht^Htt^Ht^Ht^Hte^He^He^Her^Hr^Hr^Hrn^Hn^Hn^Hn
g^Hg^Hg^Hgr^Hr^Hr^Hre^He^He^Hep^Hp^Hp^Hp [-^H-^H-^H-E^HE^HE^HE|-^H-^H-^H-F^HF^HF^HF] [-^H-^H-^H-c^Hc^Hc^Hc|-^H-^H-^H-l^Hl^Hl^Hl|-^H-^H-^H-q^Hq^Hq^Hq] [-^H-^H-^H-b^Hb^Hb^Hbh^Hh^Hh^Hhi^Hi^Hi^Hin^Hn^Hn^Hns^Hs^Hs^Hsv^Hv^Hv^Hvx^Hx^Hx^Hx] _^Hp_^Ha_^Ht_^Ht_^He_^Hr_^Hn [_^Hf_^Hi_^Hl_^He ...]
.
.
.
-^H-^H-^H-c^Hc^Hc^Hc Only a count of matching lines is printed.
.
.
.
.
But it also works if you just use less on a regular plain text file and use word boundaries.