sum1 # sum1 - print column sums
sum1 # input: rows of numbers
sum1 # output: sum of each column
sum1 # missing entries are treated as zeros
sum1
sum1 { for (i = 1; i <= NF; i++)
sum1 sum[i] += $i
sum1 if (NF > maxfld)
sum1 maxfld = NF
sum1 }
sum1 END { for (i = 1; i <= maxfld; i++) {
sum1 printf("%g", sum[i])
sum1 if (i < maxfld)
sum1 printf("\t")
sum1 else
sum1 printf("\n")
sum1 }
sum1 }
sum2 # sum2 - print column sums
sum2 # check that each line has the same number of fields
sum2 # as line one
sum2
sum2 NR==1 { nfld = NF }
sum2 { for (i = 1; i <= NF; i++)
sum2 sum[i] += $i
sum2 if (NF != nfld)
sum2 print "line " NR " has " NF " entries, not " nfld
sum2 }
sum2 END { for (i = 1; i <= nfld; i++)
sum2 printf("%g%s", sum[i], i < nfld ? "\t" : "\n")
sum2 }
sum3 # sum3 - print sums of numeric columns
sum3 # input: rows of integers and strings
sum3 # output: sums of numeric columns
sum3 # assumes every line has same layout
sum3
sum3 NR==1 { nfld = NF
sum3 for (i = 1; i <= NF; i++)
sum3 numcol[i] = isnum($i)
sum3 }
sum3
sum3 { for (i = 1; i <= NF; i++)
sum3 if (numcol[i])
sum3 sum[i] += $i
sum3 }
sum3
sum3 END { for (i = 1; i <= nfld; i++) {
sum3 if (numcol[i])
sum3 printf("%g", sum[i])
sum3 else
sum3 printf("--")
sum3 printf(i < nfld ? "\t" : "\n")
sum3 }
sum3 }
sum3
sum3 function isnum(n) { return n ~ /^[+-]?[0-9]+$/ }
3-4.ans { total[$1] += $2 }
3-4.ans END { for (x in total) print x, total[x] | "sort" }
percent # percent
percent # input: a column of nonnegative numbers
percent # output: each number and its percentage of the total
percent
percent { x[NR] = $1; sum += $1 }
percent
percent END { if (sum != 0)
percent for (i = 1; i <= NR; i++)
percent printf("%10.2f %5.1f\n", x[i], 100*x[i]/sum)
percent }
histogram # histogram
histogram # input: numbers between 0 and 100
histogram # output: histogram of deciles
histogram
histogram { x[int($1/10)]++ }
histogram
histogram END { for (i = 0; i < 10; i++)
histogram printf(" %2d - %2d: %3d %s\n",
histogram 10*i, 10*i+9, x[i], rep(x[i],"*"))
histogram printf("100: %3d %s\n", x[10], rep(x[10],"*"))
histogram }
histogram
histogram function rep(n,s, t) { # return string of n s's
histogram while (n-- > 0)
histogram t = t s
histogram return t
histogram }
hist.sh awk '
hist.sh # generate random integers
hist.sh BEGIN { for (i = 1; i <= 200; i++)
hist.sh print int(101*rand())
hist.sh }
hist.sh ' |
hist.sh awk -f histogram
histans1.awk
histans1.awk { x[int($1/10)]++ }
histans1.awk END { max = MAXSTARS = 25
histans1.awk for (i = 0; i <= 10; i++)
histans1.awk if (x[i] > max)
histans1.awk max = x[i]
histans1.awk for (i = 0; i <= 10; i++)
histans1.awk y[i] = x[i]/max * MAXSTARS
histans1.awk for (i = 0; i < 10; i++)
histans1.awk printf(" %2d - %2d: %3d %s\n",
histans1.awk 10*i, 10*i+9, x[i], rep(y[i],"*"))
histans1.awk printf("100: %3d %s\n", x[10], rep(y[10],"*"))
histans1.awk }
histans1.awk
histans1.awk function rep(n,s, t) { # return string of n s's
histans1.awk while (n-- > 0)
histans1.awk t = t s
histans1.awk return t
histans1.awk }
sumcomma # sumcomma - add up numbers containing commas
sumcomma
sumcomma { gsub(/,/, ""); sum += $0 }
sumcomma END { print sum }
addcomma # addcomma - put commas in numbers
addcomma # input: a number per line
addcomma # output: the input number followed by
addcomma # the number with commas and two decimal places
addcomma
addcomma { printf("%-12s %20s\n", $0, addcomma($0)) }
addcomma
addcomma function addcomma(x, num) {
addcomma if (x < 0)
addcomma return "-" addcomma(-x)
addcomma num = sprintf("%.2f", x) # num is dddddd.dd
addcomma while (num ~ /[0-9][0-9][0-9][0-9]/)
addcomma sub(/[0-9][0-9][0-9][,.]/, ",&", num)
addcomma return num
addcomma }
addcomma.ans /^[+-]?[0-9][0-9]?[0-9]?(,[0-9][0-9][0-9])*$/ {
addcomma.ans gsub(/,/, "")
addcomma.ans sum += $0
addcomma.ans next
addcomma.ans }
addcomma.ans { print "bad format:", $0 }
addcomma.ans END { print sum }
addcomma.ans2 /^[+-]?[0-9][0-9]?[0-9]?(,[0-9][0-9][0-9])*([.][0-9]*)?$/ {
addcomma.ans2 gsub(/,/, "")
addcomma.ans2 sum += $0
addcomma.ans2 next
addcomma.ans2 }
addcomma.ans2 { print "bad format:", $0}
addcomma.ans2 END { print sum }
datecvt # date convert - convert mmddyy into yymmdd in $1
datecvt
datecvt { $1 = substr($1,5,2) substr($1,1,2) substr($1,3,2); print }
date.data 013042 mary's birthday
date.data 032772 mark's birthday
date.data 052470 anniversary
date.data 061209 mother's birthday
date.data 110175 elizabeth's birthday
daynum function daynum(y, m, d, days, i, n) { # 1 == Jan 1, 1901
daynum split("31 28 31 30 31 30 31 31 30 31 30 31", days)
daynum # 365 days a year, plus one for each leap year
daynum n = (y-1901) * 365 + int((y-1901)/4)
daynum if (y % 4 == 0) # leap year from 1901 to 2099
daynum days[2]++
daynum for (i = 1; i < m; i++)
daynum n += days[i]
daynum return n + d
daynum }
daynum { print daynum($1, $2, $3) }
nm.output file.o:
nm.output 00000c80 T _addroot
nm.output 00000b30 T _checkdev
nm.output 00000a3c T _checkdupl
nm.output U _chown
nm.output U _client
nm.output U _close
nm.output funmount.o:
nm.output 00000000 T _funmount
nm.output U cerror
nm.format # nm.format - add filename to each nm output line
nm.format
nm.format NF == 1 { file = $1 }
nm.format NF == 2 { print file, $1, $2 }
nm.format NF == 3 { print file, $2, $3 }
prchecks # prchecks - print formatted checks
prchecks # input: number \t amount \t payee
prchecks # output: eight lines of text for preprinted check forms
prchecks
prchecks BEGIN {
prchecks FS = "\t"
prchecks dashes = sp45 = sprintf("%45s", " ")
prchecks gsub(/ /, "-", dashes) # to protect the payee
prchecks "date" | getline date # get today's date
prchecks split(date, d, " ")
prchecks date = d[2] " " d[3] ", " d[6]
prchecks initnum() # set up tables for number conversion
prchecks }
prchecks NF != 3 || $2 >= 1000000 { # illegal data
prchecks printf("\nline %d illegal:\n%s\n\nVOID\nVOID\n\n\n", NR, $0)
prchecks next # no check printed
prchecks }
prchecks { printf("\n") # nothing on line 1
prchecks printf("%s%s\n", sp45, $1) # number, indented 45 spaces
prchecks printf("%s%s\n", sp45, date) # date, indented 45 spaces
prchecks amt = sprintf("%.2f", $2) # formatted amount
prchecks printf("Pay to %45.45s $%s\n", $3 dashes, amt) # line 4
prchecks printf("the sum of %s\n", numtowords(amt)) # line 5
prchecks printf("\n\n\n") # lines 6, 7 and 8
prchecks }
prchecks
prchecks function numtowords(n, cents, dols) { # n has 2 decimal places
prchecks cents = substr(n, length(n)-1, 2)
prchecks dols = substr(n, 1, length(n)-3)
prchecks if (dols == 0)
prchecks return "zero dollars and " cents " cents exactly"
prchecks return intowords(dols) " dollars and " cents " cents exactly"
prchecks }
prchecks
prchecks function intowords(n) {
prchecks n = int(n)
prchecks if (n >= 1000)
prchecks return intowords(n/1000) " thousand " intowords(n%1000)
prchecks if (n >= 100)
prchecks return intowords(n/100) " hundred " intowords(n%100)
prchecks if (n >= 20)
prchecks return tens[int(n/10)] " " intowords(n%10)
prchecks return nums[n]
prchecks }
prchecks
prchecks function initnum() {
prchecks split("one two three four five six seven eight nine " \
prchecks "ten eleven twelve thirteen fourteen fifteen " \
prchecks "sixteen seventeen eighteen nineteen", nums, " ")
prchecks split("ten twenty thirty forty fifty sixty " \
prchecks "seventy eighty ninety", tens, " ")
prchecks }
checkfix.ans # prchecks - print formatted checks
checkfix.ans # input: number \t amount \t payee
checkfix.ans # output: eight lines of text for preprinted check forms
checkfix.ans
checkfix.ans BEGIN {
checkfix.ans FS = "\t"
checkfix.ans dashes = sp45 = sprintf("%45s", " ")
checkfix.ans gsub(/ /, "-", dashes) # to protect the payee
checkfix.ans "date" | getline date # get today's date
checkfix.ans split(date, d, " ")
checkfix.ans date = d[2] ". " d[3] ", " d[6]
checkfix.ans initnum() # set up tables for number conversion
checkfix.ans }
checkfix.ans NF != 3 {
checkfix.ans printf("\nrec %d has %d fields:\n|%s|\n\nVOID\nVOID\n\n\n",
checkfix.ans NR, NF, $0)
checkfix.ans next
checkfix.ans }
checkfix.ans { printf("\n") # nothing on line 1
checkfix.ans printf("%s%s\n", sp45, $1) # number, indented 45 spaces
checkfix.ans printf("%s%s\n", sp45, date) # date, indented 45 spaces
checkfix.ans amt = sprintf("%.2f", $2) # formatted amount
checkfix.ans printf("Pay to %45.45s $%s\n", $3 dashes, amt) # line 4
checkfix.ans printf("the sum of %s\n", numtowords(amt)) # line 5
checkfix.ans printf("\n\n\n") # lines 6, 7 and 8
checkfix.ans }
checkfix.ans function numtowords(n, cents, dols, s) { # n has 2 decimal places
checkfix.ans cents = substr(n, length(n)-1, 2)
checkfix.ans dols = substr(n, 1, length(n)-3)
checkfix.ans if (dols == 0)
checkfix.ans s = "zero dollars and " cents " cents exactly"
checkfix.ans else
checkfix.ans s = intowords(dols) " dollars and " cents " cents exactly"
checkfix.ans sub(/^one dollars/, "one dollar", s)
checkfix.ans gsub(/ +/, " ", s)
checkfix.ans return s
checkfix.ans }
checkfix.ans function intowords(n) {
checkfix.ans n = int(n)
checkfix.ans if (n >= 1000000)
checkfix.ans return("VOID")
checkfix.ans if (n >= 1000)
checkfix.ans return intowords(n/1000) " thousand " intowords(n%1000)
checkfix.ans if (n >= 100)
checkfix.ans return intowords(n/100) " hundred " intowords(n%100)
checkfix.ans if (n >= 20)
checkfix.ans return tens[int(n/10)] " " intowords(n%10)
checkfix.ans return nums[n]
checkfix.ans }
checkfix.ans function initnum() {
checkfix.ans split("one two three four five six seven eight nine " \
checkfix.ans "ten eleven twelve thirteen fourteen fifteen " \
checkfix.ans "sixteen seventeen eighteen nineteen", nums, " ")
checkfix.ans split("ten twenty thirty forty fifty sixty " \
checkfix.ans "seventy eighty ninety", tens, " ")
checkfix.ans }
colcheck # colcheck - check consistency of columns
colcheck # input: rows of numbers and strings
colcheck # output: lines whose format differs from first line
colcheck
colcheck NR == 1 {
colcheck nfld = NF
colcheck for (i = 1; i <= NF; i++)
colcheck type[i] = isnum($i)
colcheck }
colcheck { if (NF != nfld)
colcheck printf("line %d has %d fields instead of %d\n",
colcheck NR, NF, nfld)
colcheck for (i = 1; i <= NF; i++)
colcheck if (isnum($i) != type[i])
colcheck printf("field %d in line %d differs from line 1\n",
colcheck i, NR)
colcheck }
colcheck
colcheck function isnum(n) { return n ~ /^[+-]?[0-9]+$/ }
p12check # p12check - check input for alternating .P1/.P2 delimiters
p12check
p12check /^\.P1/ { if (p != 0)
p12check print ".P1 after .P1, line", NR
p12check p = 1
p12check }
p12check /^\.P2/ { if (p != 1)
p12check print ".P2 with no preceding .P1, line", NR
p12check p = 0
p12check }
p12check END { if (p != 0) print "missing .P2 at end" }
delim.ans BEGIN {
delim.ans expects["aa"] = "bb"
delim.ans expects["cc"] = "dd"
delim.ans expects["ee"] = "ff"
delim.ans }
delim.ans /^(aa|cc|ee)/ {
delim.ans if (p != "")
delim.ans print "line", NR, ": expected " p
delim.ans p = expects[substr($0, 1, 2)]
delim.ans }
delim.ans /^(bb|dd|ff)/ {
delim.ans x = substr($0, 1, 2)
delim.ans if (p != x) {
delim.ans print "line", NR, ": saw " x
delim.ans if (p)
delim.ans print ", expected", p
delim.ans }
delim.ans p = ""
delim.ans }
delim.ans END {
delim.ans if (p != "")
delim.ans print "at end, missing", p
delim.ans }
passwd # passwd - check password file
passwd
passwd BEGIN {
passwd FS = ":" }
passwd NF != 7 {
passwd printf("line %d, does not have 7 fields: %s\n", NR, $0) }
passwd $1 ~ /[^A-Za-z0-9]/ {
passwd printf("line %d, nonalphanumeric user id: %s\n", NR, $0) }
passwd $2 == "" {
passwd printf("line %d, no password: %s\n", NR, $0) }
passwd $3 ~ /[^0-9]/ {
passwd printf("line %d, nonnumeric user id: %s\n", NR, $0) }
passwd $4 ~ /[^0-9]/ {
passwd printf("line %d, nonnumeric group id: %s\n", NR, $0) }
passwd $6 !~ /^\// {
passwd printf("line %d, invalid login directory: %s\n", NR, $0) }
checkgen.data NF != 7 does not have 7 fields
checkgen.data $1 ~ /[^A-Za-z0-9]/ nonalphanumeric user id
checkgen.data $2 == "" no password
checkgen # checkgen - generate data-checking program
checkgen # input: expressions of the form: pattern tabs message
checkgen # output: program to print message when pattern matches
checkgen
checkgen BEGIN { FS = "\t+" }
checkgen { printf("%s {\n\tprintf(\"line %%d, %s: %%s\\n\",NR,$0) }\n",
checkgen $1, $2)
checkgen }
valid.ans BEGIN { FS = "\t" }
valid.ans /^=/ { print substr($0, 2); next }
valid.ans { printf("%s {\n\tprintf(\"line %%d, %s: %%s\\n\",NR,$0) }\n",
valid.ans $1, $2)
valid.ans }
compat # compat - check if awk program uses new built-in names
compat
compat BEGIN { asplit("close system atan2 sin cos rand srand " \
compat "match sub gsub", fcns)
compat asplit("ARGC ARGV FNR RSTART RLENGTH SUBSEP", vars)
compat asplit("do delete function return", keys)
compat }
compat
compat { line = $0 }
compat
compat /"/ { gsub(/"([^"]|\\")*"/, "", line) } # remove strings,
compat /\// { gsub(/\/([^\/]|\\\/)+\//, "", line) } # reg exprs,
compat /#/ { sub(/#.*/, "", line) } # and comments
compat
compat { n = split(line, x, "[^A-Za-z0-9_]+") # into words
compat for (i = 1; i <= n; i++) {
compat if (x[i] in fcns)
compat warn(x[i] " is now a built-in function")
compat if (x[i] in vars)
compat warn(x[i] " is now a built-in variable")
compat if (x[i] in keys)
compat warn(x[i] " is now a keyword")
compat }
compat }
compat
compat function asplit(str, arr) { # make an assoc array from str
compat n = split(str, temp)
compat for (i = 1; i <= n; i++)
compat arr[temp[i]]++
compat return n
compat }
compat
compat function warn(s) {
compat sub(/^[ \t]*/, "")
compat printf("file %s, line %d: %s\n\t%s\n", FILENAME, FNR, s, $0)
compat }
bundle # bundle - combine multiple files into one
bundle
bundle { print FILENAME, $0 }
unbundle # unbundle - unpack a bundle into separate files
unbundle
unbundle $1 != prev { close(prev); prev = $1 }
unbundle { print substr($0, index($0, " ") + 1) >$1 }
addr.1 Adam Smith
addr.1 1234 Wall St., Apt. 5C
addr.1 New York, NY 10021
addr.1 212 555-4321
addr.1
addr.1 David W. Copperfield
addr.1 221 Dickens Lane
addr.1 Monterey, CA 93940
addr.1 408 555-0041
addr.1 work phone 408 555-6532
addr.1 Mary, birthday January 30
addr.1
addr.1 Canadian Consulate
addr.1 555 Fifth Ave
addr.1 New York, NY
addr.1 212 586-2400
ny1.awk BEGIN { RS = "" }
ny1.awk /New York/
ny2.awk BEGIN { RS = ""; ORS = "\n\n" }
ny2.awk /New York/
smith.awk BEGIN { RS = ""; FS = "\n" }
smith.awk $1 ~ /Smith$/ { print $1, $4 } # name, phone
msort.sh # pipeline to sort address list by last names
msort.sh
msort.sh awk '
msort.sh BEGIN { RS = ""; FS = "\n" }
msort.sh { printf("%s!!#", x[split($1, x, " ")])
msort.sh for (i = 1; i <= NF; i++)
msort.sh printf("%s%s", $i, i < NF ? "!!#" : "\n")
msort.sh }
msort.sh ' |
msort.sh sort |
msort.sh awk '
msort.sh BEGIN { FS = "!!#" }
msort.sh { for (i = 2; i <= NF; i++)
msort.sh printf("%s\n", $i)
msort.sh printf("\n")
msort.sh }
msort.sh '
addr.2 accountant
addr.2 Adam Smith
addr.2 1234 Wall St., Apt. 5C
addr.2 New York, NY 10021
addr.2
addr.2 doctor - ophthalmologist
addr.2 Dr. Will Seymour
addr.2 798 Maple Blvd.
addr.2 Berkeley Heights, NJ 07922
addr.2
addr.2 lawyer
addr.2 David W. Copperfield
addr.2 221 Dickens Lane
addr.2 Monterey, CA 93940
addr.2
addr.2 doctor - pediatrician
addr.2 Dr. Susan Mark
addr.2 600 Mountain Avenue
addr.2 Murray Hill, NJ 07974
doctors1.awk /^doctor/, /^$/
doctors2.awk /^doctor/ { p = 1; next }
doctors2.awk p == 1
doctors2.awk /^$/ { p = 0; next }
checks.data check 1021
checks.data to Champagne Unlimited
checks.data amount 123.10
checks.data date 1/1/87
checks.data
checks.data deposit
checks.data amount 500.00
checks.data date 1/1/87
checks.data
checks.data check 1022
checks.data date 1/2/87
checks.data amount 45.10
checks.data to Getwell Drug Store
checks.data tax medical
checks.data
checks.data check 1023
checks.data amount 125.00
checks.data to International Travel
checks.data date 1/3/87
checks.data
checks.data amount 50.00
checks.data to Carnegie Hall
checks.data date 1/3/87
checks.data check 1024
checks.data tax charitable contribution
checks.data
checks.data to American Express
checks.data check 1025
checks.data amount 75.75
checks.data date 1/5/87
check1 # check1 - print total deposits and checks
check1
check1 /^check/ { ck = 1; next }
check1 /^deposit/ { dep = 1; next }
check1 /^amount/ { amt = $2; next }
check1 /^$/ { addup() }
check1
check1 END { addup()
check1 printf("deposits $%.2f, checks $%.2f\n",
check1 deposits, checks)
check1 }
check1
check1 function addup() {
check1 if (ck)
check1 checks += amt
check1 else if (dep)
check1 deposits += amt
check1 ck = dep = amt = 0
check1 }
check2 # check2 - print total deposits and checks
check2
check2 BEGIN { RS = ""; FS = "\n" }
check2 /(^|\n)deposit/ { deposits += field("amount"); next }
check2 /(^|\n)check/ { checks += field("amount"); next }
check2 END { printf("deposits $%.2f, checks $%.2f\n",
check2 deposits, checks)
check2 }
check2
check2 function field(name, i,f) {
check2 for (i = 1; i <= NF; i++) {
check2 split($i, f, "\t")
check2 if (f[1] == name)
check2 return f[2]
check2 }
check2 printf("error: no field %s in record\n%s\n", name, $0)
check2 }
check3 # check3 - print check information
check3
check3 BEGIN { RS = ""; FS = "\n" }
check3 /(^|\n)check/ {
check3 for (i = 1; i <= NF; i++) {
check3 split($i, f, "\t")
check3 val[f[1]] = f[2]
check3 }
check3 printf("%8s %5d %8s %s\n",
check3 val["date"],
check3 val["check"],
check3 sprintf("$%.2f", val["amount"]),
check3 val["to"])
check3 for (i in val)
check3 delete val[i]
check3 }