Search and Replace

UNIX for Beginners Q & A UNIX for Dummies Questions & Answers
1

I am trying to search and replace using two different files with strict search rules.
One file contain some data and the other file contain some numbers as shown below.
DataFile.txt

>L1_T1
text data...
text data..
>L1_T1
text data...
text data..
>L1_T1
text data...
text data..
.....
.....
>L1_T2
text data...
text data..
>L1_T2
text data...
text data..
...
... upto

>L7_T8
text data...
text data..
>L7_T8
text data...
text data..

data header ranges from L1_T1 .. L1_T8 .. upto L7_T8 (56 headers altogether)

NumberFile.txt

1
3
4
7
8
75
86
...
... 
4025
4030
4032

I want to replace the headers(L*_T*) of DataFile using the numbers from the NumberFile. However, the condition is each header has a particular range to take the numbers and one number can't be used twice. For Example:
L1_T1 can take those available numbers within the range 1 - 72
L1_T2 can take those available numbers within the range 73 - 144
L1_T3 can take those available numbers within the range 144 - 216
....
L1_T8 can take those available numbers within the range 505 - 576
....
...
L7_T8 can take those available numbers within the range 3961 - 4032
(Increment of 72 for each increasing Header)

Example Output for the above data can be:

>1
text data...
text data..
>3
text data...
text data..
>7
text data...
text data..
.....
.....
>75
text data...
text data..
>86
text data...
text data..
...
... 

>4030
text data...
text data..
>4032
text data...
text data..

Thanks for any help.

2

You already have ordered your datafile & number file ?
so in fact you just need to do the replacement step, not the checking rule step , it that correct ?

---------- Post updated at 12:31 PM ---------- Previous update was at 12:23 PM ----------
If so maybe you can try something like

awk 'NR==FNR{A[++i]=$1}/^>/{sub($1,">"A[++j],$0)}NR!=FNR' numberfile datafile

This code just do the replacement (by line) of the >Lx_Tx found by >(entrie found in the numberfile)

3

Thanks for your reply.
Both the files are ordered. Thats right.
However, there are more numbers than records in the datafile.
For example, there are three L1_T1 headers, however, for this the numbers are 1,3,4,7 and 8.
So any three numbers can be used as long as they are within the range 1-72.
The code you provided converts:
>L1_T2 to >7
>L1_T2 to >8
which should be
>L1_T2 to >75
>L1_T2 to >86
(range for L1_T2 is 73 - 144)

also
L7_T7 to >4025
L7_T7 to >4030
(range for L7_T7 is 3961 - 4032)

4

Try this

awk 'NR==FNR{A[++i]=$1}/^>/{x=$1;gsub("[^0-9]*","",x);L=substr(x,1,1);T=substr(x,2,1);s=((8*L+T-9)*72+1);do{r=0;y=++j;if (s<=A[y] && A[y]<=s+71) r=sub($1,">"A[y],$0)}while(!r&&A[y]<s)}NR!=FNR' numberf dataf

Use nawk instead of awk if you are on a SunOS / Solaris plateform :

# cat numberf
1
3
4
7
8
75
86
560
2048
4025
4030
4032
# cat dataf
>L1_T1
text data...
text data..
>L1_T1
text data...
text data..
>L1_T1
text data...
text data..
.....
.....
>L1_T2
text data...
text data..
>L1_T2
text data...
text data..
...
... upto

>L7_T8
text data...
text data..
>L7_T8
text data...
text data..
# nawk 'NR==FNR{A[++i]=$1}/^>/{x=$1;gsub("[^0-9]*","",x);L=substr(x,1,1);T=substr(x,2,1);s=((8*L+T-9)*72+1);do{r=0;y=++j;if (s<=A[y] && A[y]<=s+71) r=sub($1,">"A[y],$0)}while(!r&&A[y]<s)}NR!=FNR' numberf dataf
>1
text data...
text data..
>3
text data...
text data..
>4
text data...
text data..
.....
.....
>75
text data...
text data..
>86
text data...
text data..
...
... upto

>4025
text data...
text data..
>4030
text data...
text data..
5

Gr8.Working Fine.Thanks.