tiny naive but extendable html/rss/xml parser in AWK

xml.awk

#!/usr/bin/env -S awk -f 
BEGIN { RS = "<" }
NR > 1 {
    # Find the end of the opening/closing tag
    idx = index($0, ">")
    if (idx > 0) {
        v = substr($0, 1, idx-1)    # Everything inside < ... >
        txt = substr($0, idx+1)     # Everything after >
        
        if (v ~ /^\//) {
            if (p > 0) { process(s_n[p], stack[p], s_v[p]); p--; }
        } else if (v !~ /(\/|^\?|^!)/) {
            # OPENING TAG
            # Separate tag name from attribute string using first whitespace
            match(v, /[ \t\n\r]/)
            if (RSTART) {
                s_n[++p] = substr(v, 1, RSTART-1)
                stack[p] = substr(v, RSTART)   # Keep the whole attribute string
            } else {
                s_n[++p] = v
                stack[p] = ""
            }
            s_v[p] = txt
        }
    }
}

function process(tag, attr_str, val,    attr_map, n, i, b, pair) {
    # 1. Extract attributes: key="value":
    # Using a more inclusive regex for the attribute values
    n = patsplit(attr_str, b, /[a-zA-Z0-9_-]+=[^ \t\n\r>]+/)
    
    for (i = 1; i <= n; i++) {
        split(b[i], pair, "=")
        gsub(/^["']|["']$/, "", pair[2])
        attr_map[pair[1]] = pair[2]
    }

    # 2. Clean up value text
    gsub(/^[ \t\n\r]+|[ \t\n\r]+$/, "", val)

    # 3. Output
    printf "Tag: %s", tag
    if (val) printf " | Text: %s", val
    for (k in attr_map) printf " | %s: %s", k, attr_map[k]
    print ""
}