Shell Programming and Scripting

I am parsing a file and I get differnt results everytime.
Sometimes I get 12s sometimes I get 54m and sometime 3h..

v1=12s or v1=54m or v1=3h

12s - 12 seconds
54m - 54 minutes
3h - 3 hour

I have to write a script in such a way that it whenever v1 is in minutes,
I should strip "m" from it.

if v1=32s ignore, if v1=54m then set v1=54, if v1=3h then ignore.

Any help would be appreciated. thx.
Without awk would be great.

Code:

$ a=50m
$ echo ${a/m/}
50
$ a=50s
$ echo ${a/m/}
50s
$ a=50h
$ echo ${a/m/}
50h

#echo $a
99m
DR# echo ${a/m/}
ksh: ${a/m/}: bad substitution

Code:

a=50m
echo ${a%m}

It works. Thanks a lot.
I truly appreciate your help.

Regards,

Jay

---------- Post updated at 11:55 AM ---------- Previous update was at 11:45 AM ----------

One more question,
In my script sometimes value of variable test4 is empty.
because of that I get following.
##
ParseLog.sh[37]: 0.00%: more tokens expected
##

what can I do ?

Use

Code:

"$test4"

Suggestion did not work.

Can you provide me small script, with following info.

if $test value is empty or $test value is 12s( any numeric ) or $tet value is 32h ( any numeric instead of 32 ) , ignore.
If $test value is 15m ( any numeric ) I should do echo $(test%m).

your help would be highly appreciated. thx.