1*e6d6c189SCody Peter Mello#! /bin/sh 2*e6d6c189SCody Peter Mello# From arnold@f7.net Sun Apr 22 20:15:25 2007 3*e6d6c189SCody Peter Mello# Date: Thu, 19 Apr 2007 17:09:02 +0300 4*e6d6c189SCody Peter Mello# From: Pekka Pessi <Pekka.Pessi@nokia.com> 5*e6d6c189SCody Peter Mello# X-Face: #V(jdpv[lI!TNUU=2*oh:="#suS*ponXW"yr6G;~L}<xZn_2^0)V{jqdc4y}@2b]ffd}SY# 6*e6d6c189SCody Peter Mello# :9||1pew85O,WjiYA"6C7bW^zt^+.{b#B{lEE+4$9lrXL(55g}dU>uZ\JfD\"IG#G{j`hZI;=DmT\H 7*e6d6c189SCody Peter Mello# pfDMyJ`i=:M;BM3R.`[>P^ER8+]i 8*e6d6c189SCody Peter Mello# Subject: UTF-8 locale and \n in regexps 9*e6d6c189SCody Peter Mello# To: bug-gawk@gnu.org 10*e6d6c189SCody Peter Mello# Cc: Pekka.Pessi@nokia.com 11*e6d6c189SCody Peter Mello# Message-id: <pvlkgoh2wx.fsf@nokia.com> 12*e6d6c189SCody Peter Mello# MIME-version: 1.0 13*e6d6c189SCody Peter Mello# Content-type: multipart/mixed; boundary="=-=-=" 14*e6d6c189SCody Peter Mello# 15*e6d6c189SCody Peter Mello# --=-=-= 16*e6d6c189SCody Peter Mello# 17*e6d6c189SCody Peter Mello# Hello, 18*e6d6c189SCody Peter Mello# 19*e6d6c189SCody Peter Mello# It looks like regexp with \n in [^] behaves badly if locale has 20*e6d6c189SCody Peter Mello# an UTF-8 ctype. 21*e6d6c189SCody Peter Mello# 22*e6d6c189SCody Peter Mello# It looks like if there is \n and an range without \n, like /\n[^x\n]foo/, 23*e6d6c189SCody Peter Mello# and first \n ends an even-numbered line within the string, regexp 24*e6d6c189SCody Peter Mello# does not match. 25*e6d6c189SCody Peter Mello# 26*e6d6c189SCody Peter Mello# Please see the attached script for an demonstration. 27*e6d6c189SCody Peter Mello# 28*e6d6c189SCody Peter Mello# --Pekka Pessi 29*e6d6c189SCody Peter Mello# 30*e6d6c189SCody Peter Mello# 31*e6d6c189SCody Peter Mello# --=-=-= 32*e6d6c189SCody Peter Mello# Content-Disposition: inline; filename=gawk-test 33*e6d6c189SCody Peter Mello# 34*e6d6c189SCody Peter Mello#! /bin/sh 35*e6d6c189SCody Peter Mello 36*e6d6c189SCody Peter Melloif [ -z "$AWK" ]; then 37*e6d6c189SCody Peter Mello printf '$AWK must be set\n' >&2 38*e6d6c189SCody Peter Mello exit 1 39*e6d6c189SCody Peter Mellofi 40*e6d6c189SCody Peter Mello 41*e6d6c189SCody Peter Mello# April 2010: Remove UNKNOWN, causes spurious failures on some systems 42*e6d6c189SCody Peter Mellofor LC_ALL in C POSIX en_US.ISO8859-1 en_US.UTF-8 #UNKNOWN 43*e6d6c189SCody Peter Mellodo 44*e6d6c189SCody Peter Melloexport LC_ALL 45*e6d6c189SCody Peter Mellocat <<EOF | 46*e6d6c189SCody Peter Melloline1 47*e6d6c189SCody Peter Melloline2 48*e6d6c189SCody Peter Melloline3 49*e6d6c189SCody Peter Melloline4 50*e6d6c189SCody Peter Melloline5 51*e6d6c189SCody Peter Melloline6 52*e6d6c189SCody Peter Melloline7 53*e6d6c189SCody Peter Melloline8 54*e6d6c189SCody Peter Melloline9 55*e6d6c189SCody Peter MelloEOF 56*e6d6c189SCody Peter Mello$AWK ' 57*e6d6c189SCody Peter MelloBEGIN { RS="\0"; } 58*e6d6c189SCody Peter Mello{ 59*e6d6c189SCody Peter Mello if (match($0, /\n[^2\n]*2/)) { got2=1; } else { print "no match 2"; } 60*e6d6c189SCody Peter Mello if (match($0, /\n[^3\n]*3/)) { got3=1; } else { print "no match 3"; } 61*e6d6c189SCody Peter Mello if (match($0, /\n[^4\n]*4/)) { got4=1; } else { print "no match 4"; } 62*e6d6c189SCody Peter Mello if (match($0, /\n[^5\t]*5/)) { got5=1; } else { print "no match 5"; } 63*e6d6c189SCody Peter Mello if (match($0, /\n[^6\n]*6/)) { got6=1; } else { print "no match 6"; } 64*e6d6c189SCody Peter Mello if (match($0, /\n[a-z]*7\n/)){ got7=1; } else { print "no match 7"; } 65*e6d6c189SCody Peter Mello if (match($0, /\n[^8\n]*8/)) { got8=1; } else { print "no match 8"; } 66*e6d6c189SCody Peter Mello if (match($0, /8.[^9\n]+9/)) { got9=1; } else { print "no match 9"; } 67*e6d6c189SCody Peter Mello} 68*e6d6c189SCody Peter Mello 69*e6d6c189SCody Peter MelloEND { exit(!(got2 && got3 && got4 && got5 && got6 && got7 && got8 && got9)); } 70*e6d6c189SCody Peter Mello' || { 71*e6d6c189SCody Peter Mello echo LC_ALL=$LC_ALL FAILED 72*e6d6c189SCody Peter Mello exit 1 73*e6d6c189SCody Peter Mello} 74*e6d6c189SCody Peter Melloecho LC_ALL=$LC_ALL passed 75*e6d6c189SCody Peter Mellodone 76*e6d6c189SCody Peter Mello# 77*e6d6c189SCody Peter Mello# --=-=-=-- 78*e6d6c189SCody Peter Mello# 79