DataMuseum.dk

top - metrics - download

    Length: 4404 (0x1134)
    Types: TextFile
    Notes: UNIX file
    Names: »format.n«

└─⟦26887b7e0⟧ Bits:30009717 Comet 32 harddisk image
    └─⟦28c352965⟧ »/a« UNIX Filesystem
        └─⟦this⟧ »usr/man/mann/format.n« 

TextFile

.TH FORMAT 1 
.SH NAME
left, right, mid, index, rindex, ins, del, con, len
.br
.in +10
\-  String operations available in shell
.in
.SH SYNOPSIS
.B left
string count
.sp
.B right
string count
.sp
.B mid
string position count
.sp
.B index
string target
.sp
.B rindex
string target
.sp
.B ins
string insertion_string position
.sp
.B del
string position count
.sp
.B con
string1 string2
.sp
.B len
string
.SH DESCRIPTION
.I Left
returns a substring of string made up of the leftmost count characters.
The returned string is terminated by '\\0' and is followed by
a newline character.
If the value passed in count is greater than the number of characters in
the string then the entire string is returned, it is not padded.
.PP
.I Right 
is similar to left in that it returns a substring of the string
argument, it returns the rightmost count characters from the string.
.PP
.I Mid
is similar to the application of
.I right
then
.I left
as it returns a substring of the string argument that begins at
character position and contains count characters.
If the start position of the substring is greater than the length of
the string then nothing is returned, if the count plus the start
position are greater 
than the length of the string then the right portion of the string,
starting at character `position' is returned. No padding is
carried out.
.PP
.I Index
is used to return the leftmost occurrence of a string or character
within the target string. Index considers the first character in a
string to be at character position zero.
.PP
.I Rindex
is exactly the same as index except that it finds the rightmost
occurrence of the character or string.
.PP
.I Ins
is used to insert one string or character at a particular position
within another string.
If the position for the insertion is greater than the length of the
target string then the insertion is made at the end of the target string,
ie. the two strings are concatenated.
.PP
.I Del
is used to remove a string of specified length from the string
starting at the character specified in `position'.
If the start position is greater than the length of the string then
no alteration occurs and the original string is returned intact, if
however the start position falls within the string but the
start position plus the count of the characters to be removed is
greater than the length of the string then the leftmost characters
up to the start position are returned (as in left).
.PP
.I Con
stands for concatenate and will return the two string arguments passed to
it as a single string.
.PP
.I Len
will return the length of the string argument passed to it.
.SH "RETURN CODES"
.ta 4 20
.nf
	CODE	MEANING
.sp
.ta 6 14
	0	SUCCESSFUL
	1	TARGET STRING NOT FOUND (Index and rindex)
	2	INVALID COMMAND FORMAT
.fi
.SH EXAMPLES
.PP
The following programs were hacked together to show how the programs
given above can be used from within shell scripts, no claims are made
for them.
.sp 2
.in +15
tim
.sp
mid "`date`" 11 8
.in
.PP
This program will only work if date returns output in the format:
.sp
.in +15
Sat Jan  5 15:15:46 GMT 1985
.sp
.in
Notice that mid can be used here and cut can not, cut would treat the
double space between Jan and 5 as a separator followed by a field, however
if the date was Jan 15 then the double space is reduced to a single
space so altering the number of fields in the output.
This highlights the main use of these programs, they can be used when
the number of characters present is known, cut is better when the
number of fields are known.
.sp
.in +15
scanner
.sp 2
.nf
if test `left $1 1` = "-"
then
  length=`len $1`
  i=1
  while test $i -lt 6
  do
    echo `mid $1 $i 1` is set
    i=`expr $i + 1`
  done
else
  echo No flags set in first argument
fi
.in
.fi
.PP
It must be said that this program is hideously slow, but try something
like  scanner -hello, this should demonstrate its workings (and its speed).
.sp 2
.in +15
alarm
.sp
.nf
if test $# != 2
then
  echo "Usage: $0 Hour Minute"
else
  time=$1:$2:
  da=`date`
  dt=`mid "$da" 11 6`
  if test $dt != $time
  then
    (sleep 50; alarm $1 $2) &
  else
    echo 
    echo Alarm call
    echo 
  fi
fi
.in
.fi
.PP
This final program is simply an alarm clock that runs in background,
time is specified using the 24 hour clock and the program is accurate
to 30 seconds (on average, worst case 50 seconds).
.SH AUTHOR
Craig Wylie, University of Stirling (cww@sgcs).