Substitution from specify lines of a look up table - awk

Question

I'm trying to get a script to automate some tasks using the GAMESS package, from which I'd hope to extrapolate to more complex cases later. Alas it would seem my Unix programming skills are not up to par.
I have a general GAMESS input file 'ion.inp'of the form:
$CONTRL SCFTYP=<tag4> ICHARG=<tag5> MULT=<tag6> ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
<tag1> energy
Dnh 2
<tag2> <tag3> .0 .0 .0
$END
And I have (as a MWE) a look up table for the parameters of 'ion.inp' like 'table.dat', where the <tag#> are taken from each line of the table.
<tag1> | <tag2> | <tag3> | <tag4> | <tag5> | <tag6>
Hidrogen | H | 1.0 | ROHF | 0 | 2
Hidrogen cation | H | 1.0 | RHF | 1 | 1
For portability, I'd like to get a solution using POSIX sh, sed or awk, but after some trials (using sh or sed, I'm not familiar with awk at all, even though I know it is a potential solution in this case) I couldn't get it to work.
The file 'ion.inp' can be edited in place because it will be run inside a sh loop. I already got everything else working, except for this supposedly simple substitution.
Any help would be much appreciated!

Answer 1

Here is an example using sed and awk. The script is named script.awk. Everything is output on stdout but you could redirect it to files using > inside the AWK script. See second solution below.
The idea is to drive the process using awk and the table.dat file which contains data to make the substitutions and then for each batch of substitutions (each line of the file), we use sed to perform the actual substitutions once we have each tag and its value.
BEGIN { FS = "\\s*\\|\\s*" } changes the field separator to use "optional spaces followed by | then followed by optional spaces". That means $1, $2, ... will give us the values for the tag numbered 1, 2, ...
NR == 1 { next } is used to skip the first line which is useless since tags are ordered from 1 without any gap. If it was not the case, we would have to adapt the AWK script.
{ ... } for each line we build the sed command and execute it. The output of sed becomes the output of awk for that specific line.
BEGIN { FS = "\\s*\\|\\s*" }
NR == 1 { next }
{
s = ""
for (i=1; i <= NF; i++)
s = s sprintf(";s/%s/%s/g", "<tag" i ">", $i)
system("sed '" s "' ion.inp")
}
$ cat table.dat
<tag1> | <tag2> | <tag3> | <tag4> | <tag5> | <tag6>
Hidrogen | H | 1.0 | ROHF | 0 | 2
Hidrogen cation | H | 1.0 | RHF | 1 | 1
$ cat ion.inp
$CONTRL SCFTYP=<tag4> ICHARG=<tag5> MULT=<tag6> ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
<tag1> energy
Dnh 2
<tag2> <tag3> .0 .0 .0
$END
$ awk -f script.awk table.dat
$CONTRL SCFTYP=ROHF ICHARG=0 MULT=2 ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
Hidrogen energy
Dnh 2
H 1.0 .0 .0 .0
$END
$CONTRL SCFTYP=RHF ICHARG=1 MULT=1 ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
Hidrogen cation energy
Dnh 2
H 1.0 .0 .0 .0
$END
Redirecting to files with command awk -f script2.awk table.dat. The script script2.awk is:
BEGIN { FS = "\\s*\\|\\s*" }
NR == 1 { next }
{
s = ""
for (i=1; i <= NF; i++)
s = s sprintf(";s/%s/%s/g", "<tag" i ">", $i)
system("sed '" s "' ion.inp > " sprintf("output%02d.txt", NR-1))
}
$ cat output01.txt
$CONTRL SCFTYP=ROHF ICHARG=0 MULT=2 ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
Hidrogen energy
Dnh 2
H 1.0 .0 .0 .0
$END
$ cat output02.txt
$CONTRL SCFTYP=RHF ICHARG=1 MULT=1 ISPHER=1 NPRINT=-5 $END
$BASIS GBASIS=<tag> $END
$DATA
Hidrogen cation energy
Dnh 2
H 1.0 .0 .0 .0
$END