Shell Programming and Scripting

hello,

i have a little problem, i want match the complete field ($1 or $2) with a complete word in another variable.

example:
i have a file with either one or two words per lane:

Code:

 
hsa-mir-4449 
hsa-mir-4707 
hsa-mir-4707* hsa-mir-4707
novelMiR_3551 novelMiR_3563
novelMiR_4330 novelMiR_1963
hsa-mir-625 
novelMiR_3660 hsa-mir-625
hsa-mir-4707* 
hsa-mir-4707 hsa-mir-4707*
novelMiR_440 novelMiR_1963
hsa-mir-470

and want search each word in the file with each line if this word is present.


my try to use match, i.e.

Code:

if (match($1,i) || match($2,i))

with $1,$2 both words per lane and i the complete lane, was not possible, because 'hsa-mir-470' also matches 'hsa-mir-4707'. how can i match only the complete field with complete words in the variable?

can i tell awk that only the complete $1 should match a complete word in i.
something like

Code:

/^$1$/ ~ i

or

Code:

match(\<$1\>,i)

- both not working...

Not sure what you are after.. How about:

Code:

if ( $1==i || $2==i )

no, in i are both words, that means:

Code:

 
 hsa-mir-4707 == hsa-mir-4707* hsa-mir-4707

which is always false...

i want match the first word (which is a complete field) with a complete word in i...

You could try:

Code:

split(i,F); for(j in F) if ( $1==F[j] || $2==F[j] )

this should work - i'll try...