使用awk处理多文件：统计区间内点数的语法问题求助

Hey there! Let's walk through your AWK script issue step by step—this is a super common snag for new AWK users, so you’re not alone here.

The Problem Context #

You’re trying to count how many points (from the points file) fall into each interval defined in the windows file. Your first script threw a syntax error near unexpected token 'i', but after tweaking it, you got the expected output. Let’s break down why the error happened and how the corrected code works.

Common Cause of the Initial Syntax Error #

That error almost always comes from forgetting to wrap your entire AWK code in single quotes when running it in the shell. Without quotes, the shell tries to interpret AWK variables like i as shell variables, which breaks the script’s syntax. For example, writing:

awk NR==FNR{ name[NR]=$1 pos[NR]=$2 next } { for(i in name){ ... } } points windows

Instead of wrapping the code in single quotes causes the shell to misparse i, leading to the "unexpected token" error.

The Corrected Working Code #

Here’s the properly formatted script that delivers the right results:

NR==FNR{ 
    name[NR] = $1 
    pos[NR] = $2 
    next 
} 
{ 
    for(i in name){ 
        if(name[i] == $1 && pos[i] >= $2 && pos[i] <= $3){ 
            sum[FNR] += 1 
        } 
    } 
} 
END { 
    for(i = 1; i <= FNR; i++){ 
        print sum[i]+0  # Ensures we print 0 for intervals with no matching points
    } 
}

Run it with this shell command (note the single quotes wrapping the AWK code):

awk 'NR==FNR{ name[NR]=$1; pos[NR]=$2; next } { for(i in name){ if(name[i]==$1 && pos[i]>=$2 && pos[i]<=$3){ sum[FNR]++ } } } END{ for(i=1; i<=FNR; i++){ print sum[i]+0 } }' points windows > output

How This Code Works #

Let’s break down each section to make it clear:

1. Reading the points file:

   • NR==FNR is a standard AWK trick to detect when we’re processing the first input file (points). NR is the total number of lines read so far, and FNR is the line number within the current file—they’re equal only for the first file.
   • We store each point’s chromosome name in the name array and its position in the pos array, using the line number as the index.
   • next skips the rest of the code, so this block only runs for lines in points.

2. Processing the windows file:

   • For each interval (line in windows), we loop through every point stored in our arrays.
   • We check two conditions: the point’s chromosome matches the interval’s chromosome, and the point’s position falls between the interval’s start ($2) and end ($3).
   • If both conditions are true, we increment the counter sum[FNR]—FNR here is the line number of the current interval, so each interval gets its own unique counter.

3. Output the results:

   • The END block runs after all files are processed. We loop through the sum array from 1 to the total number of intervals (which is FNR at the end of processing windows).
   • sum[i]+0 ensures that even if an interval has no matching points (and sum[i] is uninitialized), we print 0 instead of nothing.

Test Data & Expected Output #

Your test files:
points

chr1 52
chr1 65
chr2 120
chr2 101
chr2 160
chr3 20
chr4 50

windows

chr1 0 100
chr1 100 200
chr2 0 100
chr2 100 200
chr3 0 100
chr3 100 200
chr4 0 100
chr5 0 100
chr6 0 100
chr6 100 200
chr7 0 100
chr8 0 100

The output file will contain:

2
0
0
3
1
0
1
0
0
0
0
0

This aligns with counting points per interval: 2 points in chr1 0-100, 3 in chr2 100-200, 1 in chr3 0-100, 1 in chr4 0-100, and 0 for all other intervals.

内容的提问来源于stack exchange，提问作者gpapprank