Shell Programming and Scripting

Hello,
I am working on an Urdu to Hindi dictionary which has the following structure:

Code:

a=b
a=c
n=d
n=q

and so on.
i.e. Headword separated from gloss by a

Code:

=

I am giving below a live sample

Code:

بتا=बता
بتا=बित्ता
بتا=बुत्ता
بتان=बतान
بتان=बितान
بتانا=बिताना

I need the following output:

Code:

a=b/c
n=d/q

i.e. if the headword on the left hand side is repeated and maps to two different glosses on the right hand side, the similar headwords be reduced to one and the glosses that map to the head word be mapped to the head-word and each gloss be separated by a

Code:

/ (slash sign)

I work under windows and a perl or awk script would help.
At present I am using a macro to merge the data but this is huge and a script would help.
Many thanks

Try:

Code:

perl -a -F"=" -nle 'push @{$x{$F[0]}},$F[1];END{for $i (keys %x) {print "$i=". join "/", @{$x{$i}}}}' file

Many thanks for your help. The script worked just fine.

Code:

awk -F = '{a[$1]=(a[$1]=="")?$2:a[$1]"/"$2}
        END{for (i in a) print i FS a[i]}' infile