gawk mismatch

[mslider]

I don't know why but this code below used to perfom crosstab query (see crosstab with AWK question)returns
mismatch (with printing duplicate lines) if I use gawk interpretor instead awk. My gawk version is: GNU Awk 3.1.4.

what's wrong ?

BEGIN {
FS=OFS=";"
}

{
if ( !($2 in cols) ) cols[$2]=++colsN
tag[$1 , $2] = $3
}

END {
# inverting the assoviative array - make it an indexed array
for ( i in cols)
cols[cols] = i

printf("tag%s", OFS)
for(i=1; i<= colsN; i++){
printf("%s%s", cols, (i<colsN) ? OFS : "\n")
}
for( i in tag) {
split(i, iA, SUBSEP)
printf("%s%s", iA[1], OFS)
for(i=1; i <= colsN; i++) {
idx=iA[1] SUBSEP cols
printf("%s%s", (idx in tag) ? tag[idx] : "0", (i<colsN) ? OFS : "\n")
if ( idx in tag) delete tag[idx]
}
}
}

 

[vgersh99]

it seems gawk is having problem deleting an array cell from within the loop iterating through the same array.

for now use futurelet's code which works correctly under both awk-s.

vlad
+----------------------------+
| #include<disclaimer.h> |
+----------------------------+

 

[PHV]

Have you tried to comment out this line ?
if ( idx in tag) delete tag[idx]

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[mslider]

this is my input file:

AAAAAAAAAA;GSM1;17
AAAAAAATCA;GSM1;1
AAAAAAATTT;GSM1;1
AAAAAACAAA;GSM1;1
AAAAAACTCC;GSM1;1
AAAAAATAAA;GSM1;1
AAAAACAAAA;GSM1;1
AAAAAAAAAA;GSM10419;54
AAAAAAAAAC;GSM10419;2
AAAAAAAAAG;GSM10419;1


with gawk and without comment out this line:
if ( idx in tag) delete tag[idx]

it returns:

tag;GSM1;GSM10419
AAAAAAAAAG;0;1
AAAAACAAAA;1;0
AAAAAAATTT;1;0
AAAAAAAAAA;17;54
AAAAAATAAA;1;0
AAAAAACTCC;1;0
AAAAAACAAA;1;0
AAAAAAAAAA;0;0
AAAAAAAAAC;0;2
AAAAAAATCA;1;0

and now if i comment out the line:

tag;GSM1;GSM10419
AAAAAAAAAG;0;1
AAAAACAAAA;1;0
AAAAAAATTT;1;0
AAAAAAAAAA;17;54
AAAAAATAAA;1;0
AAAAAACTCC;1;0
AAAAAACAAA;1;0
AAAAAAAAAA;17;54
AAAAAAAAAC;0;2
AAAAAAATCA;1;0