AWK: Spickzettel

Textverarbeitung in der Kommandozeile mit Mustern und Aktionen

Ausführen: Datei skript.awk speichern und mit awk -f skript.awk datei.txt ausführen; einfache Programme direkt als awk '{print $1}' datei.txt.

awk 'BEGIN { print "Hallo" }'              # Programm direkt, ohne Eingabe
awk '{ print $1 }' datei.txt               # erste Spalte jeder Zeile
awk -f skript.awk datei.txt                # Programm aus Datei
awk -F, '{ print $2 }' daten.csv           # Feldtrenner Komma
echo "a b c" | awk '{ print $2 }'          # aus einer Pipe

BEGIN {
    print 7 + 2, 7 - 2, 7 * 2, 7 / 2, 7 % 3, 2 ^ 10, 2 ** 3
    a = 5; a += 3; a *= 2; print a
    i = 1; i++; ++i; print i
    print (3 < 5), (3 == 3), (3 != 3), ("a" < "b")
    print (1 && 0), (1 || 0), !1
    print "Text" " " "wird" " " "verkettet"
    s = "abc"; print s s
    print 1 " " 2
    print 10 / 3
    printf "%.2f\n", 10 / 3
    print int(3.99), int(-3.99)
    print sqrt(16), exp(0), log(1), sin(0), cos(0)
    print length("hallo")
    print 1e3, 0x10, 010
}

BEGIN {
    alter["Mia"] = 17
    alter["Tom"] = 25
    alter["Zoe"] = 31
    print alter["Tom"], length(alter)
    if ("Mia" in alter) print "Mia vorhanden"
    if (!("Ben" in alter)) print "Ben fehlt"
    delete alter["Tom"]
    n = asorti(alter, namen)
    for (i = 1; i <= n; i++) printf "%s=%d ", namen[i], alter[namen[i]]
    print ""

    zahlen[1] = "eins"; zahlen[2] = "zwei"; zahlen[3] = "drei"
    for (i = 1; i <= length(zahlen); i++) printf "%s ", zahlen[i]
    print ""

    # mehrdimensional: Schlüssel mit SUBSEP
    m[1,2] = "x"; m[2,1] = "y"
    PROCINFO["sorted_in"] = "@ind_str_asc"
    for (k in m) { split(k, t, SUBSEP); printf "(%d,%d)=%s ", t[1], t[2], m[k] }
    print ""
    delete m
    print length(m)
}

printf 'Mia 17 Berlin\nTom 25 Hamburg\nZoe 31 Koeln\nBen 12 Berlin\n' | awk '
    NR == 1 { print "Erste Zeile:", $0 }
    $2 > 20 { print $1, "ist über 20" }
    /Berlin/ { berlin++ }
    END { print NR, "Zeilen,", berlin, "aus Berlin" }
'

#!/usr/bin/awk -f
# Bericht über Verkäufe
BEGIN {
    FS = ","
    daten = "Nord,100\nSued,250\nNord,50\nWest,75\nSued,25"
    n = split(daten, zeilen, "\n")
    for (i = 1; i <= n; i++) {
        split(zeilen[i], f, ",")
        summe[f[1]] += f[2]
        anzahl[f[1]]++
    }
    PROCINFO["sorted_in"] = "@val_num_desc"
    for (r in summe)
        printf "%-6s Summe %4d  Anzahl %d  Mittel %.1f\n", r, summe[r], anzahl[r], summe[r] / anzahl[r]
}

BEGIN {
    t = "Datum: 2024-03-15, Mail: mia@example.com, Tel: +49 30 12345"
    print (t ~ /[0-9]{4}-[0-9]{2}-[0-9]{2}/)
    print (t ~ /^Datum/), (t ~ /Tel$/), (t ~ /Tel: \+49/)
    muster = "[a-z]+@[a-z]+\\.com"
    print (t ~ muster)
    print match(t, /[a-z]+@[a-z]+\.[a-z]+/), RSTART, RLENGTH
    print substr(t, RSTART, RLENGTH)
    n = gsub(/[0-9]/, "#", t)
    print n, t
}

cat > /tmp/awk-access.log <<'LOG'
10.0.0.1 - - [15/Mar/2024:10:00:01] "GET /index.html HTTP/1.1" 200 5120
10.0.0.2 - - [15/Mar/2024:10:00:05] "GET /bild.png HTTP/1.1" 200 20480
10.0.0.1 - - [15/Mar/2024:10:01:10] "GET /fehlt HTTP/1.1" 404 512
10.0.0.3 - - [15/Mar/2024:10:02:00] "POST /login HTTP/1.1" 302 0
10.0.0.2 - - [15/Mar/2024:10:03:30] "GET /index.html HTTP/1.1" 200 5120
LOG
awk '
{
    status[$9]++
    bytes += $10
    pfad[$7]++
    ip[$1]++
}
END {
    PROCINFO["sorted_in"] = "@ind_str_asc"
    print "Statuscodes:"
    for (s in status) printf "  %s: %d\n", s, status[s]
    PROCINFO["sorted_in"] = "@val_num_desc"
    print "Meistbesucht:"
    for (p in pfad) { printf "  %s (%d)\n", p, pfad[p]; if (++n >= 2) break }
    printf "Übertragen: %.1f KB in %d Anfragen\n", bytes / 1024, NR
    printf "Verschiedene IPs: %d\n", length(ip)
}' /tmp/awk-access.log
rm /tmp/awk-access.log

BEGIN {
    srand(1)
    zufall = rand()
    print (zufall >= 0 && zufall < 1)
    print ENVIRON["HOME"] != ""
    print "Zeichen:", sprintf("%c%c%c", 76, 111, 76)
    cmd = "echo hallo"
    cmd | getline ausgabe
    close(cmd)
    print "Befehl lieferte:", ausgabe
    print system("exit 3")
    printf "%s %s\n", "a", "b"
    printf "%5.1f%%\n", 99.5
}