summaryrefslogtreecommitdiff
path: root/test/regex
diff options
context:
space:
mode:
authorPeter S. Mazinger <ps.m@gmx.net>2006-03-01 20:35:48 +0000
committerPeter S. Mazinger <ps.m@gmx.net>2006-03-01 20:35:48 +0000
commit2688d30ffbe025d026f426b0e4976f2fb50f1606 (patch)
tree7ae2bdd5228bb3cd581b1013e11b1d0c8c11226f /test/regex
parent1bf97fb207fc5e11875c32236efd6a611aa0926a (diff)
Add AT&T research regex testsuite
Diffstat (limited to 'test/regex')
-rw-r--r--test/regex/LICENSE72
-rw-r--r--test/regex/Makefile22
-rw-r--r--test/regex/basic.dat216
-rw-r--r--test/regex/categorize.dat62
-rw-r--r--test/regex/forcedassoc.dat30
-rw-r--r--test/regex/leftassoc.dat16
-rw-r--r--test/regex/nullsubexpr.dat73
-rw-r--r--test/regex/repetition.dat79
-rw-r--r--test/regex/rightassoc.dat16
-rw-r--r--test/regex/testregex.c2121
-rw-r--r--test/regex/testregexc.c1
-rw-r--r--test/regex/testregexf.c1
-rw-r--r--test/regex/testregexl.c1
-rw-r--r--test/regex/testregexn.c1
-rw-r--r--test/regex/testregexp.c1
-rw-r--r--test/regex/testregexr.c1
16 files changed, 2713 insertions, 0 deletions
diff --git a/test/regex/LICENSE b/test/regex/LICENSE
new file mode 100644
index 000000000..2e5acb9cd
--- /dev/null
+++ b/test/regex/LICENSE
@@ -0,0 +1,72 @@
+From gsf@research.att.com Wed Mar 1 20:30:54 2006
+Return-Path: <gsf@research.att.com>
+X-Original-To: mps@bridge.intra
+Delivered-To: mps@bridge.intra
+Received: from localhost (localhost [127.0.0.1])
+ by localhost (Postfix) with ESMTP id B8C814E4F
+ for <mps@bridge.intra>; Wed, 1 Mar 2006 20:30:53 +0100 (CET)
+Received: from mail.bridge.intra ([127.0.0.1])
+ by localhost (lnx.bridge.intra [127.0.0.1]) (amavisd-new, port 10024)
+ with LMTP id 05987-03 for <mps@bridge.intra>;
+ Wed, 1 Mar 2006 20:30:42 +0100 (CET)
+Received: from pop.gmx.net (localhost [127.0.0.1])
+ by mail.bridge.intra (Postfix) with ESMTP id C8C73794D
+ for <mps@bridge.intra>; Wed, 1 Mar 2006 20:30:38 +0100 (CET)
+X-Flags: 0000
+Delivered-To: GMX delivery to ps.m@gmx.net
+Received: (qmail invoked by alias); 01 Mar 2006 19:23:46 -0000
+Received: from mail-red.research.att.com (EHLO mail-white.research.att.com) [192.20.225.110]
+ by mx0.gmx.net (mx085) with SMTP; 01 Mar 2006 20:23:46 +0100
+Received: from raptor.research.att.com (raptor.research.att.com [135.207.23.32])
+ by mail-blue.research.att.com (Postfix) with ESMTP id B7929147CBB
+ for <ps.m@gmx.net>; Wed, 1 Mar 2006 14:23:45 -0500 (EST)
+Received: (from gsf@localhost)
+ by raptor.research.att.com (SGI-8.9.3p2/8.8.7) id OAA86112
+ for ps.m@gmx.net; Wed, 1 Mar 2006 14:23:45 -0500 (EST)
+Date: Wed, 1 Mar 2006 14:23:45 -0500 (EST)
+From: Glenn Fowler <gsf@research.att.com>
+Message-Id: <200603011923.OAA86112@raptor.research.att.com>
+Organization: AT&T Research
+X-Mailer: mailx (AT&T/BSD) 9.9 2005-04-21
+Mime-Version: 1.0
+Content-Type: text/plain; charset=us-ascii
+Content-Transfer-Encoding: 7bit
+References: <Pine.LNX.4.44.0603012011250.6386-100000@lnx.bridge.intra>
+To: mps@bridge.intra
+Subject: Re: testregex licensing question
+X-GMX-Antivirus: -1 (not scanned, may not use virus scanner)
+X-GMX-Antispam: 0 (Mail was not recognized as spam)
+X-GMX-UID: lJF3ZO9DeSEkJ2TcbHQhaXN1IGRvb0Ca
+X-Virus-Scanned: by amavisd-new at localhost
+Status: RO
+X-Status:
+X-Keywords:
+X-UID: 44736
+
+
+you may include it directly
+retain the testregex.c header comment
+it uses a very free license to maximize distribution
+you can copy that .c comment to any test data files you use
+using # comment style to be complete
+
+let me know how it works with your libc
+also pass on any new tests you cook up
+
+On Wed, 1 Mar 2006 20:15:02 +0100 (CET) Peter S. Mazinger wrote:
+> Hello Glenn!
+
+> I would want to add testregex.c and the related *.dat files to the uClibc
+> testsuite. uClibc is licensed under LGPL v2.1. I haven't found any
+> licensing related info on testregex.
+
+> Is it allowed to use the code there, or should I accomodate the testsuite
+> to download the needed files from the original site each time it is ran?
+
+> Thanks, Peter
+
+> --
+> Peter S. Mazinger <ps dot m at gmx dot net> ID: 0xA5F059F2
+> Key fingerprint = 92A4 31E1 56BC 3D5A 2D08 BB6E C389 975E A5F0 59F2
+
+
diff --git a/test/regex/Makefile b/test/regex/Makefile
new file mode 100644
index 000000000..827d54041
--- /dev/null
+++ b/test/regex/Makefile
@@ -0,0 +1,22 @@
+# uClibc regex tests
+# Licensed under the GNU Library General Public License, see COPYING.LIB
+
+TESTS_DISABLED := testregexn
+
+include ../Test.mak
+
+CFLAGS-OMIT_testregex.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexc.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexf.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexl.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexn.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexp.c := -D_GNU_SOURCE
+CFLAGS-OMIT_testregexr.c := -D_GNU_SOURCE
+
+OPTS_testregex := < basic.dat
+OPTS_testregexc := < categorize.dat
+OPTS_testregexf := < forcedassoc.dat
+OPTS_testregexl := < leftassoc.dat
+OPTS_testregexn := < nullsubexpr.dat
+OPTS_testregexp := < repetition.dat
+OPTS_testregexr := < rightassoc.dat
diff --git a/test/regex/basic.dat b/test/regex/basic.dat
new file mode 100644
index 000000000..5c50f37b4
--- /dev/null
+++ b/test/regex/basic.dat
@@ -0,0 +1,216 @@
+NOTE all standard compliant implementations should pass these : 2002-05-31
+
+BE abracadabra$ abracadabracadabra (7,18)
+BE a...b abababbb (2,7)
+BE XXXXXX ..XXXXXX (2,8)
+E \) () (1,2)
+BE a] a]a (0,2)
+B } } (0,1)
+E \} } (0,1)
+BE \] ] (0,1)
+B ] ] (0,1)
+E ] ] (0,1)
+B { { (0,1)
+B } } (0,1)
+BE ^a ax (0,1)
+BE \^a a^a (1,3)
+BE a\^ a^ (0,2)
+BE a$ aa (1,2)
+BE a\$ a$ (0,2)
+BE ^$ NULL (0,0)
+E $^ NULL (0,0)
+E a($) aa (1,2)(2,2)
+E a*(^a) aa (0,1)(0,1)
+E (..)*(...)* a (0,0)
+E (..)*(...)* abcd (0,4)(2,4)
+E (ab|a)(bc|c) abc (0,3)(0,2)(2,3)
+E (ab)c|abc abc (0,3)(0,2)
+E a{0}b ab (1,2)
+E (a*)(b?)(b+)b{3} aaabbbbbbb (0,10)(0,3)(3,4)(4,7)
+E (a*)(b{0,1})(b{1,})b{3} aaabbbbbbb (0,10)(0,3)(3,4)(4,7)
+E a{9876543210} NULL BADBR
+E ((a|a)|a) a (0,1)(0,1)(0,1)
+E (a*)(a|aa) aaaa (0,4)(0,3)(3,4)
+E a*(a.|aa) aaaa (0,4)(2,4)
+E a(b)|c(d)|a(e)f aef (0,3)(?,?)(?,?)(1,2)
+E (a|b)?.* b (0,1)(0,1)
+E (a|b)c|a(b|c) ac (0,2)(0,1)
+E (a|b)c|a(b|c) ab (0,2)(?,?)(1,2)
+E (a|b)*c|(a|ab)*c abc (0,3)(1,2)
+E (a|b)*c|(a|ab)*c xc (1,2)
+E (.a|.b).*|.*(.a|.b) xa (0,2)(0,2)
+E a?(ab|ba)ab abab (0,4)(0,2)
+E a?(ac{0}b|ba)ab abab (0,4)(0,2)
+E ab|abab abbabab (0,2)
+E aba|bab|bba baaabbbaba (5,8)
+E aba|bab baaabbbaba (6,9)
+E (aa|aaa)*|(a|aaaaa) aa (0,2)(0,2)
+E (a.|.a.)*|(a|.a...) aa (0,2)(0,2)
+E ab|a xabc (1,3)
+E ab|a xxabc (2,4)
+Ei (Ab|cD)* aBcD (0,4)(2,4)
+BE [^-] --a (2,3)
+BE [a-]* --a (0,3)
+BE [a-m-]* --amoma-- (0,4)
+E :::1:::0:|:::1:1:0: :::0:::1:::1:::0: (8,17)
+E :::1:::0:|:::1:1:1: :::0:::1:::1:::0: (8,17)
+{E [[:upper:]] A (0,1) [[<element>]] not supported
+E [[:lower:]]+ `az{ (1,3)
+E [[:upper:]]+ @AZ[ (1,3)
+BE [[-]] [[-]] (2,4)
+BE [[.NIL.]] NULL ECOLLATE
+BE [[=aleph=]] NULL ECOLLATE
+}
+BE$ \n \n (0,1)
+BEn$ \n \n (0,1)
+BE$ [^a] \n (0,1)
+BE$ \na \na (0,2)
+E (a)(b)(c) abc (0,3)(0,1)(1,2)(2,3)
+BE xxx xxx (0,3)
+E1 (^|[ (,;])((([Ff]eb[^ ]* *|0*2/|\* */?)0*[6-7]))([^0-9]|$) feb 6, (0,6)
+E1 (^|[ (,;])((([Ff]eb[^ ]* *|0*2/|\* */?)0*[6-7]))([^0-9]|$) 2/7 (0,3)
+E1 (^|[ (,;])((([Ff]eb[^ ]* *|0*2/|\* */?)0*[6-7]))([^0-9]|$) feb 1,Feb 6 (5,11)
+E3 ((((((((((((((((((((((((((((((x)))))))))))))))))))))))))))))) x (0,1)(0,1)(0,1)
+E3 ((((((((((((((((((((((((((((((x))))))))))))))))))))))))))))))* xx (0,2)(1,2)(1,2)
+E a?(ab|ba)* ababababababababababababababababababababababababababababababababababababababababa (0,81)(79,81)
+E abaa|abbaa|abbbaa|abbbbaa ababbabbbabbbabbbbabbbbaa (18,25)
+E abaa|abbaa|abbbaa|abbbbaa ababbabbbabbbabbbbabaa (18,22)
+E aaac|aabc|abac|abbc|baac|babc|bbac|bbbc baaabbbabac (7,11)
+BE$ .* \x01\xff (0,2)
+E aaaa|bbbb|cccc|ddddd|eeeeee|fffffff|gggg|hhhh|iiiii|jjjjj|kkkkk|llll XaaaXbbbXcccXdddXeeeXfffXgggXhhhXiiiXjjjXkkkXlllXcbaXaaaa (53,57)
+L aaaa\nbbbb\ncccc\nddddd\neeeeee\nfffffff\ngggg\nhhhh\niiiii\njjjjj\nkkkkk\nllll XaaaXbbbXcccXdddXeeeXfffXgggXhhhXiiiXjjjXkkkXlllXcbaXaaaa NOMATCH
+E a*a*a*a*a*b aaaaaaaaab (0,10)
+BE ^ NULL (0,0)
+BE $ NULL (0,0)
+BE ^$ NULL (0,0)
+BE ^a$ a (0,1)
+BE abc abc (0,3)
+BE abc xabcy (1,4)
+BE abc ababc (2,5)
+BE ab*c abc (0,3)
+BE ab*bc abc (0,3)
+BE ab*bc abbc (0,4)
+BE ab*bc abbbbc (0,6)
+E ab+bc abbc (0,4)
+E ab+bc abbbbc (0,6)
+E ab?bc abbc (0,4)
+E ab?bc abc (0,3)
+E ab?c abc (0,3)
+BE ^abc$ abc (0,3)
+BE ^abc abcc (0,3)
+BE abc$ aabc (1,4)
+BE ^ abc (0,0)
+BE $ abc (3,3)
+BE a.c abc (0,3)
+BE a.c axc (0,3)
+BE a.*c axyzc (0,5)
+BE a[bc]d abd (0,3)
+BE a[b-d]e ace (0,3)
+BE a[b-d] aac (1,3)
+BE a[-b] a- (0,2)
+BE a[b-] a- (0,2)
+BE a] a] (0,2)
+BE a[]]b a]b (0,3)
+BE a[^bc]d aed (0,3)
+BE a[^-b]c adc (0,3)
+BE a[^]b]c adc (0,3)
+E ab|cd abc (0,2)
+E ab|cd abcd (0,2)
+E a\(b a(b (0,3)
+E a\(*b ab (0,2)
+E a\(*b a((b (0,4)
+E ((a)) abc (0,1)(0,1)(0,1)
+E (a)b(c) abc (0,3)(0,1)(2,3)
+E a+b+c aabbabc (4,7)
+E a* aaa (0,3)
+E (a*)* - (0,0)(0,0)
+E (a*)+ - (0,0)(0,0)
+E (a*|b)* - (0,0)(0,0)
+E (a+|b)* ab (0,2)(1,2)
+E (a+|b)+ ab (0,2)(1,2)
+E (a+|b)? ab (0,1)(0,1)
+BE [^ab]* cde (0,3)
+E (^)* - (0,0)(0,0)
+BE a* NULL (0,0)
+E ([abc])*d abbbcd (0,6)(4,5)
+E ([abc])*bcd abcd (0,4)(0,1)
+E a|b|c|d|e e (0,1)
+E (a|b|c|d|e)f ef (0,2)(0,1)
+E ((a*|b))* - (0,0)(0,0)(0,0)
+BE abcd*efg abcdefg (0,7)
+BE ab* xabyabbbz (1,3)
+BE ab* xayabbbz (1,2)
+E (ab|cd)e abcde (2,5)(2,4)
+BE [abhgefdc]ij hij (0,3)
+E (a|b)c*d abcd (1,4)(1,2)
+E (ab|ab*)bc abc (0,3)(0,1)
+E a([bc]*)c* abc (0,3)(1,3)
+E a([bc]*)(c*d) abcd (0,4)(1,3)(3,4)
+E a([bc]+)(c*d) abcd (0,4)(1,3)(3,4)
+E a([bc]*)(c+d) abcd (0,4)(1,2)(2,4)
+E a[bcd]*dcdcde adcdcde (0,7)
+E (ab|a)b*c abc (0,3)(0,2)
+E ((a)(b)c)(d) abcd (0,4)(0,3)(0,1)(1,2)(3,4)
+BE [A-Za-z_][A-Za-z0-9_]* alpha (0,5)
+E ^a(bc+|b[eh])g|.h$ abh (1,3)
+E (bc+d$|ef*g.|h?i(j|k)) effgz (0,5)(0,5)
+E (bc+d$|ef*g.|h?i(j|k)) ij (0,2)(0,2)(1,2)
+E (bc+d$|ef*g.|h?i(j|k)) reffgz (1,6)(1,6)
+E (((((((((a))))))))) a (0,1)(0,1)(0,1)(0,1)(0,1)(0,1)(0,1)(0,1)(0,1)(0,1)
+BE multiple words multiple words yeah (0,14)
+E (.*)c(.*) abcde (0,5)(0,2)(3,5)
+BE abcd abcd (0,4)
+E a(bc)d abcd (0,4)(1,3)
+E a[-]?c ac (0,3)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Qaddafi (0,15)(?,?)(10,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Mo'ammar Gadhafi (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Kaddafi (0,15)(?,?)(10,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Qadhafi (0,15)(?,?)(10,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Gadafi (0,14)(?,?)(10,11)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Mu'ammar Qadafi (0,15)(?,?)(11,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Moamar Gaddafi (0,14)(?,?)(9,11)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Mu'ammar Qadhdhafi (0,18)(?,?)(13,15)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Khaddafi (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Ghaddafy (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Ghadafi (0,15)(?,?)(11,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Ghaddafi (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muamar Kaddafi (0,14)(?,?)(9,11)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Quathafi (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Muammar Gheddafi (0,16)(?,?)(11,13)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Moammar Khadafy (0,15)(?,?)(11,12)
+E M[ou]'?am+[ae]r .*([AEae]l[- ])?[GKQ]h?[aeu]+([dtz][dhz]?)+af[iy] Moammar Qudhafi (0,15)(?,?)(10,12)
+E a+(b|c)*d+ aabcdd (0,6)(3,4)
+E ^.+$ vivi (0,4)
+E ^(.+)$ vivi (0,4)(0,4)
+E ^([^!.]+).att.com!(.+)$ gryphon.att.com!eby (0,19)(0,7)(16,19)
+E ^([^!]+!)?([^!]+)$ bas (0,3)(?,?)(0,3)
+E ^([^!]+!)?([^!]+)$ bar!bas (0,7)(0,4)(4,7)
+E ^([^!]+!)?([^!]+)$ foo!bas (0,7)(0,4)(4,7)
+E ^.+!([^!]+!)([^!]+)$ foo!bar!bas (0,11)(4,8)(8,11)
+E ((foo)|(bar))!bas bar!bas (0,7)(0,3)(?,?)(0,3)
+E ((foo)|(bar))!bas foo!bar!bas (4,11)(4,7)(?,?)(4,7)
+E ((foo)|(bar))!bas foo!bas (0,7)(0,3)(0,3)
+E ((foo)|bar)!bas bar!bas (0,7)(0,3)
+E ((foo)|bar)!bas foo!bar!bas (4,11)(4,7)
+E ((foo)|bar)!bas foo!bas (0,7)(0,3)(0,3)
+E (foo|(bar))!bas bar!bas (0,7)(0,3)(0,3)
+E (foo|(bar))!bas foo!bar!bas (4,11)(4,7)(4,7)
+E (foo|(bar))!bas foo!bas (0,7)(0,3)
+E (foo|bar)!bas bar!bas (0,7)(0,3)
+E (foo|bar)!bas foo!bar!bas (4,11)(4,7)
+E (foo|bar)!bas foo!bas (0,7)(0,3)
+E ^(([^!]+!)?([^!]+)|.+!([^!]+!)([^!]+))$ foo!bar!bas (0,11)(0,11)(?,?)(?,?)(4,8)(8,11)
+E ^([^!]+!)?([^!]+)$|^.+!([^!]+!)([^!]+)$ bas (0,3)(?,?)(0,3)
+E ^([^!]+!)?([^!]+)$|^.+!([^!]+!)([^!]+)$ bar!bas (0,7)(0,4)(4,7)
+E ^([^!]+!)?([^!]+)$|^.+!([^!]+!)([^!]+)$ foo!bar!bas (0,11)(?,?)(?,?)(4,8)(8,11)
+E ^([^!]+!)?([^!]+)$|^.+!([^!]+!)([^!]+)$ foo!bas (0,7)(0,4)(4,7)
+E ^(([^!]+!)?([^!]+)|.+!([^!]+!)([^!]+))$ bas (0,3)(0,3)(?,?)(0,3)
+E ^(([^!]+!)?([^!]+)|.+!([^!]+!)([^!]+))$ bar!bas (0,7)(0,7)(0,4)(4,7)
+E ^(([^!]+!)?([^!]+)|.+!([^!]+!)([^!]+))$ foo!bar!bas (0,11)(0,11)(?,?)(?,?)(4,8)(8,11)
+E ^(([^!]+!)?([^!]+)|.+!([^!]+!)([^!]+))$ foo!bas (0,7)(0,7)(0,4)(4,7)
+E .*(/XXX).* /XXX (0,4)(0,4)
+E .*(\\XXX).* \XXX (0,4)(0,4)
+E \\XXX \XXX (0,4)
+E .*(/000).* /000 (0,4)(0,4)
+E .*(\\000).* \000 (0,4)(0,4)
+E \\000 \000 (0,4)
diff --git a/test/regex/categorize.dat b/test/regex/categorize.dat
new file mode 100644
index 000000000..d34851278
--- /dev/null
+++ b/test/regex/categorize.dat
@@ -0,0 +1,62 @@
+NOTE regex implementation categorization 2004-05-31
+
+?E aa* xaxaax (1,2) POSITION=leftmost
+; POSITION=bug
+
+?E (a*)(ab)*(b*) abc (0,2)(0,1)(?,?)(1,2) ASSOCIATIVITY=right
+|E (a*)(ab)*(b*) abc (0,2)(0,0)(0,2)(2,2) ASSOCIATIVITY=left
+; ASSOCIATIVITY=bug
+
+?E ((a*)(ab)*)((b*)(a*)) aba (0,3)(0,2)(0,0)(0,2)(2,3)(2,2)(2,3) SUBEXPRESSION=precedence
+|E ((a*)(ab)*)((b*)(a*)) aba (0,3)(0,1)(0,1)(?,?)(1,3)(1,2)(2,3) SUBEXPRESSION=grouping
+; SUBEXPRESSION=bug
+
+?E (...?.?)* xxxxxx (0,6)(4,6) REPEAT_LONGEST=first
+|E (...?.?)* xxxxxx (0,6)(2,6) REPEAT_LONGEST=last
+|E (...?.?)* xxxxxx OK REPEAT_LONGEST=unknown
+; REPEAT_LONGEST=bug
+
+?E (a|ab)(bc|c) abcabc (0,3)(0,2)(2,3) EXPECTED
+|E (a|ab)(bc|c) abcabc (0,3)(0,1)(1,3) BUG=alternation-order
+; BUG=alternation-order-UNKNOWN
+
+?E (aba|a*b)(aba|a*b) ababa (0,5)(0,2)(2,5) EXPECTED
+|E (aba|a*b)(aba|a*b) ababa (0,4)(0,3)(3,4) BUG=first-match
+; BUG=unknown-match
+
+?B a\(b\)*\1 a NOMATCH EXPECTED
+|B a\(b\)*\1 a (0,1) BUG=nomatch-match
+|B a\(b\)*\1 abab (0,2)(1,2) # BUG=repeat-any
+; BUG=nomatch-match-UNKNOWN
+
+?E (a*){2} xxxxx (0,0)(0,0) EXPECTED
+|E (a*){2} xxxxx (5,5)(5,5) BUG=range-null
+; BUG=range-null-UNKNOWN
+
+?B a\(b\)*\1 abab NOMATCH EXPECTED
+|B a\(b\)*\1 abab (0,1) # BUG=nomatch-match
+|B a\(b\)*\1 abab (0,2)(1,2) BUG=repeat-any
+; BUG=repeat-any-UNKNOWN
+
+?E (a*)* a (0,1)(0,1) EXPECTED
+|E (a*)* ax (0,1)(0,1) BUG=repeat-null-unknown
+|E (a*)* a (0,1)(1,1) BUG=repeat-null
+; BUG=repeat-null-UNKNOWN
+
+?E (aba|a*b)* ababa (0,5)(2,5) EXPECTED
+|E (aba|a*b)* ababa (0,5)(3,4) BUG=repeat-short
+|E (aba|a*b)* ababa (0,4)(3,4) # LENGTH=first
+; BUG=repeat-short-UNKNOWN
+
+?E (a(b)?)+ aba (0,3)(2,3) EXPECTED
+|E (a(b)?)+ aba (0,3)(2,3)(1,2) BUG=repeat-artifact
+; BUG=repeat-artifact-UNKNOWN
+
+?B \(a\(b\)*\)*\2 abab NOMATCH EXPECTED
+|B \(a\(b\)*\)*\2 abab (0,4)(2,3)(1,2) BUG=repeat-artifact-nomatch
+; BUG=repeat-artifact-nomatch-UNKNOWN
+
+?E (a?)((ab)?)(b?)a?(ab)?b? abab (0,4)(0,1)(1,1)(?,?)(1,2)(2,4) BUG=subexpression-first
+|E .*(.*) ab (0,2)(2,2) EXPECTED
+|E .*(.*) ab (0,2)(0,2) BUG=subexpression-first
+; BUG=subexpression-first-UNKNOWN
diff --git a/test/regex/forcedassoc.dat b/test/regex/forcedassoc.dat
new file mode 100644
index 000000000..39f3111da
--- /dev/null
+++ b/test/regex/forcedassoc.dat
@@ -0,0 +1,30 @@
+NOTE left-assoc:pass-all right-assoc:pass-all : 2002-04-29
+
+E (a|ab)(c|bcd) abcd (0,4)(0,1)(1,4)
+E (a|ab)(bcd|c) abcd (0,4)(0,1)(1,4)
+E (ab|a)(c|bcd) abcd (0,4)(0,1)(1,4)
+E (ab|a)(bcd|c) abcd (0,4)(0,1)(1,4)
+E ((a|ab)(c|bcd))(d*) abcd (0,4)(0,4)(0,1)(1,4)(4,4)
+E ((a|ab)(bcd|c))(d*) abcd (0,4)(0,4)(0,1)(1,4)(4,4)
+E ((ab|a)(c|bcd))(d*) abcd (0,4)(0,4)(0,1)(1,4)(4,4)
+E ((ab|a)(bcd|c))(d*) abcd (0,4)(0,4)(0,1)(1,4)(4,4)
+E (a|ab)((c|bcd)(d*)) abcd (0,4)(0,2)(2,4)(2,3)(3,4)
+E (a|ab)((bcd|c)(d*)) abcd (0,4)(0,2)(2,4)(2,3)(3,4)
+E (ab|a)((c|bcd)(d*)) abcd (0,4)(0,2)(2,4)(2,3)(3,4)
+E (ab|a)((bcd|c)(d*)) abcd (0,4)(0,2)(2,4)(2,3)(3,4)
+E (a*)(b|abc) abc (0,3)(0,0)(0,3)
+E (a*)(abc|b) abc (0,3)(0,0)(0,3)
+E ((a*)(b|abc))(c*) abc (0,3)(0,3)(0,0)(0,3)(3,3)
+E ((a*)(abc|b))(c*) abc (0,3)(0,3)(0,0)(0,3)(3,3)
+E (a*)((b|abc)(c*)) abc (0,3)(0,1)(1,3)(1,2)(2,3)
+E (a*)((abc|b)(c*)) abc (0,3)(0,1)(1,3)(1,2)(2,3)
+E (a*)(b|abc) abc (0,3)(0,0)(0,3)
+E (a*)(abc|b) abc (0,3)(0,0)(0,3)
+E ((a*)(b|abc))(c*) abc (0,3)(0,3)(0,0)(0,3)(3,3)
+E ((a*)(abc|b))(c*) abc (0,3)(0,3)(0,0)(0,3)(3,3)
+E (a*)((b|abc)(c*)) abc (0,3)(0,1)(1,3)(1,2)(2,3)
+E (a*)((abc|b)(c*)) abc (0,3)(0,1)(1,3)(1,2)(2,3)
+E (a|ab) ab (0,2)(0,2)
+E (ab|a) ab (0,2)(0,2)
+E (a|ab)(b*) ab (0,2)(0,2)(2,2)
+E (ab|a)(b*) ab (0,2)(0,2)(2,2)
diff --git a/test/regex/leftassoc.dat b/test/regex/leftassoc.dat
new file mode 100644
index 000000000..9c068c61b
--- /dev/null
+++ b/test/regex/leftassoc.dat
@@ -0,0 +1,16 @@
+NOTE left-assoc:pass-all right-assoc:pass-none : 2002-04-29
+
+E (a|ab)(c|bcd)(d*) abcd (0,4)(0,1)(1,4)(4,4)
+E (a|ab)(bcd|c)(d*) abcd (0,4)(0,1)(1,4)(4,4)
+E (ab|a)(c|bcd)(d*) abcd (0,4)(0,1)(1,4)(4,4)
+E (ab|a)(bcd|c)(d*) abcd (0,4)(0,1)(1,4)(4,4)
+
+E (a*)(b|abc)(c*) abc (0,3)(0,0)(0,3)(3,3)
+E (a*)(abc|b)(c*) abc (0,3)(0,0)(0,3)(3,3)
+E (a*)(b|abc)(c*) abc (0,3)(0,0)(0,3)(3,3)
+E (a*)(abc|b)(c*) abc (0,3)(0,0)(0,3)(3,3)
+
+E (a|ab)(c|bcd)(d|.*) abcd (0,4)(0,1)(1,4)(4,4)
+E (a|ab)(bcd|c)(d|.*) abcd (0,4)(0,1)(1,4)(4,4)
+E (ab|a)(c|bcd)(d|.*) abcd (0,4)(0,1)(1,4)(4,4)
+E (ab|a)(bcd|c)(d|.*) abcd (0,4)(0,1)(1,4)(4,4)
diff --git a/test/regex/nullsubexpr.dat b/test/regex/nullsubexpr.dat
new file mode 100644
index 000000000..c73d8f09e
--- /dev/null
+++ b/test/regex/nullsubexpr.dat
@@ -0,0 +1,73 @@
+NOTE null subexpression matches : 2002-06-06
+
+E (a*)* a (0,1)(0,1)
+E SAME x (0,0)(0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+E (a*)+ a (0,1)(0,1)
+E SAME x (0,0)(0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+E (a+)* a (0,1)(0,1)
+E SAME x (0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+E (a+)+ a (0,1)(0,1)
+E SAME x NOMATCH
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+
+E ([a]*)* a (0,1)(0,1)
+E SAME x (0,0)(0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+E ([a]*)+ a (0,1)(0,1)
+E SAME x (0,0)(0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaax (0,6)(0,6)
+E ([^b]*)* a (0,1)(0,1)
+E SAME b (0,0)(0,0)
+E SAME aaaaaa (0,6)(0,6)
+E SAME aaaaaab (0,6)(0,6)
+E ([ab]*)* a (0,1)(0,1)
+E SAME aaaaaa (0,6)(0,6)
+E SAME ababab (0,6)(0,6)
+E SAME bababa (0,6)(0,6)
+E SAME b (0,1)(0,1)
+E SAME bbbbbb (0,6)(0,6)
+E SAME aaaabcde (0,5)(0,5)
+E ([^a]*)* b (0,1)(0,1)
+E SAME bbbbbb (0,6)(0,6)
+E SAME aaaaaa (0,0)(0,0)
+E ([^ab]*)* ccccxx (0,6)(0,6)
+E SAME ababab (0,0)(0,0)
+
+E ((z)+|a)* zabcde (0,2)(1,2)
+
+{E a+? aaaaaa (0,1) no *? +? mimimal match ops
+E (a) aaa (0,1)(0,1)
+E (a*?) aaa (0,0)(0,0)
+E (a)*? aaa (0,0)
+E (a*?)*? aaa (0,0)
+}
+
+B \(a*\)*\(x\) x (0,1)(0,0)(0,1)
+B \(a*\)*\(x\) ax (0,2)(0,1)(1,2)
+B \(a*\)*\(x\) axa (0,2)(0,1)(1,2)
+B \(a*\)*\(x\)\(\1\) x (0,1)(0,0)(0,1)(1,1)
+B \(a*\)*\(x\)\(\1\) ax (0,2)(1,1)(1,2)(2,2)
+B \(a*\)*\(x\)\(\1\) axa (0,3)(0,1)(1,2)(2,3)
+B \(a*\)*\(x\)\(\1\)\(x\) axax (0,4)(0,1)(1,2)(2,3)(3,4)
+B \(a*\)*\(x\)\(\1\)\(x\) axxa (0,3)(1,1)(1,2)(2,2)(2,3)
+
+E (a*)*(x) x (0,1)(0,0)(0,1)
+E (a*)*(x) ax (0,2)(0,1)(1,2)
+E (a*)*(x) axa (0,2)(0,1)(1,2)
+
+E (a*)+(x) x (0,1)(0,0)(0,1)
+E (a*)+(x) ax (0,2)(0,1)(1,2)
+E (a*)+(x) axa (0,2)(0,1)(1,2)
+
+E (a*){2}(x) x (0,1)(0,0)(0,1)
+E (a*){2}(x) ax (0,2)(1,1)(1,2)
+E (a*){2}(x) axa (0,2)(1,1)(1,2)
diff --git a/test/regex/repetition.dat b/test/regex/repetition.dat
new file mode 100644
index 000000000..b54a2c6f4
--- /dev/null
+++ b/test/regex/repetition.dat
@@ -0,0 +1,79 @@
+NOTE implicit vs. explicit repetitions : 2002-08-01
+#
+# Glenn Fowler <gsf@research.att.com>
+# conforming matches (column 4) must match one of the following BREs
+# NOMATCH
+# (0,.)\((\(.\),\(.\))(?,?)(\2,\3)\)*
+# (0,.)\((\(.\),\(.\))(\2,\3)(?,?)\)*
+# i.e., each 3-tuple has two identical elements and one (?,?)
+#
+
+E ((..)|(.)) NULL NOMATCH
+E ((..)|(.))((..)|(.)) NULL NOMATCH
+E ((..)|(.))((..)|(.))((..)|(.)) NULL NOMATCH
+
+E ((..)|(.)){1} NULL NOMATCH
+E ((..)|(.)){2} NULL NOMATCH
+E ((..)|(.)){3} NULL NOMATCH
+
+E ((..)|(.))* NULL (0,0)
+
+E ((..)|(.)) a (0,1)(0,1)(?,?)(0,1)
+E ((..)|(.))((..)|(.)) a NOMATCH
+E ((..)|(.))((..)|(.))((..)|(.)) a NOMATCH
+
+E ((..)|(.)){1} a (0,1)(0,1)(?,?)(0,1)
+E ((..)|(.)){2} a NOMATCH
+E ((..)|(.)){3} a NOMATCH
+
+E ((..)|(.))* a (0,1)(0,1)(?,?)(0,1)
+
+E ((..)|(.)) aa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.))((..)|(.)) aa (0,2)(0,1)(?,?)(0,1)(1,2)(?,?)(1,2)
+E ((..)|(.))((..)|(.))((..)|(.)) aa NOMATCH
+
+E ((..)|(.)){1} aa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.)){2} aa (0,2)(1,2)(?,?)(1,2)
+E ((..)|(.)){3} aa NOMATCH
+
+E ((..)|(.))* aa (0,2)(0,2)(0,2)(?,?)
+
+E ((..)|(.)) aaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.))((..)|(.)) aaa (0,3)(0,2)(0,2)(?,?)(2,3)(?,?)(2,3)
+E ((..)|(.))((..)|(.))((..)|(.)) aaa (0,3)(0,1)(?,?)(0,1)(1,2)(?,?)(1,2)(2,3)(?,?)(2,3)
+
+E ((..)|(.)){1} aaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.)){2} aaa (0,3)(2,3)(?,?)(2,3)
+E ((..)|(.)){3} aaa (0,3)(2,3)(?,?)(2,3)
+
+E ((..)|(.))* aaa (0,3)(2,3)(?,?)(2,3)
+
+E ((..)|(.)) aaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.))((..)|(.)) aaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)
+E ((..)|(.))((..)|(.))((..)|(.)) aaaa (0,4)(0,2)(0,2)(?,?)(2,3)(?,?)(2,3)(3,4)(?,?)(3,4)
+
+E ((..)|(.)){1} aaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.)){2} aaaa (0,4)(2,4)(2,4)(?,?)
+E ((..)|(.)){3} aaaa (0,4)(3,4)(?,?)(3,4)
+
+E ((..)|(.))* aaaa (0,4)(2,4)(2,4)(?,?)
+
+E ((..)|(.)) aaaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.))((..)|(.)) aaaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)
+E ((..)|(.))((..)|(.))((..)|(.)) aaaaa (0,5)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)(4,5)(?,?)(4,5)
+
+E ((..)|(.)){1} aaaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.)){2} aaaaa (0,4)(2,4)(2,4)(?,?)
+E ((..)|(.)){3} aaaaa (0,5)(4,5)(?,?)(4,5)
+
+E ((..)|(.))* aaaaa (0,5)(4,5)(?,?)(4,5)
+
+E ((..)|(.)) aaaaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.))((..)|(.)) aaaaaa (0,4)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)
+E ((..)|(.))((..)|(.))((..)|(.)) aaaaaa (0,6)(0,2)(0,2)(?,?)(2,4)(2,4)(?,?)(4,6)(4,6)(?,?)
+
+E ((..)|(.)){1} aaaaaa (0,2)(0,2)(0,2)(?,?)
+E ((..)|(.)){2} aaaaaa (0,4)(2,4)(2,4)(?,?)
+E ((..)|(.)){3} aaaaaa (0,6)(4,6)(4,6)(?,?)
+
+E ((..)|(.))* aaaaaa (0,6)(4,6)(4,6)(?,?)
diff --git a/test/regex/rightassoc.dat b/test/regex/rightassoc.dat
new file mode 100644
index 000000000..ed7f28e62
--- /dev/null
+++ b/test/regex/rightassoc.dat
@@ -0,0 +1,16 @@
+NOTE left-assoc:pass-none right-assoc:pass-all : 2002-04-29
+
+E (a|ab)(c|bcd)(d*) abcd (0,4)(0,2)(2,3)(3,4)
+E (a|ab)(bcd|c)(d*) abcd (0,4)(0,2)(2,3)(3,4)
+E (ab|a)(c|bcd)(d*) abcd (0,4)(0,2)(2,3)(3,4)
+E (ab|a)(bcd|c)(d*) abcd (0,4)(0,2)(2,3)(3,4)
+
+E (a*)(b|abc)(c*) abc (0,3)(0,1)(1,2)(2,3)
+E (a*)(abc|b)(c*) abc (0,3)(0,1)(1,2)(2,3)
+E (a*)(b|abc)(c*) abc (0,3)(0,1)(1,2)(2,3)
+E (a*)(abc|b)(c*) abc (0,3)(0,1)(1,2)(2,3)
+
+E (a|ab)(c|bcd)(d|.*) abcd (0,4)(0,2)(2,3)(3,4)
+E (a|ab)(bcd|c)(d|.*) abcd (0,4)(0,2)(2,3)(3,4)
+E (ab|a)(c|bcd)(d|.*) abcd (0,4)(0,2)(2,3)(3,4)
+E (ab|a)(bcd|c)(d|.*) abcd (0,4)(0,2)(2,3)(3,4)
diff --git a/test/regex/testregex.c b/test/regex/testregex.c
new file mode 100644
index 000000000..7b86ab79e
--- /dev/null
+++ b/test/regex/testregex.c
@@ -0,0 +1,2121 @@
+#pragma prototyped noticed
+
+/*
+ * regex(3) test harness
+ *
+ * build: cc -o testregex testregex.c
+ * help: testregex --man
+ * note: REG_* features are detected by #ifdef; if REG_* are enums
+ * then supply #define REG_foo REG_foo for each enum REG_foo
+ *
+ * Glenn Fowler <gsf@research.att.com>
+ * AT&T Labs Research
+ *
+ * PLEASE: publish your tests so everyone can benefit
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of THIS SOFTWARE FILE (the "Software"), to deal in the Software
+ * without restriction, including without limitation the rights to use,
+ * copy, modify, merge, publish, distribute, and/or sell copies of the
+ * Software, and to permit persons to whom the Software is furnished to do
+ * so, subject to the following disclaimer:
+ *
+ * THIS SOFTWARE IS PROVIDED BY AT&T ``AS IS'' AND ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL AT&T BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+static const char id[] = "\n@(#)$Id: testregex (AT&T Research) 2005-05-20 $\0\n";
+
+#if _PACKAGE_ast
+#include <ast.h>
+#else
+#include <sys/types.h>
+#endif
+
+#include <stdio.h>
+#include <regex.h>
+#include <ctype.h>
+#include <setjmp.h>
+#include <signal.h>
+#include <string.h>
+#include <unistd.h>
+
+#ifdef __STDC__
+#include <stdlib.h>
+#include <locale.h>
+#endif
+
+#if !_PACKAGE_ast
+#undef REG_DISCIPLINE
+#endif
+
+#ifndef REG_DELIMITED
+#undef _REG_subcomp
+#endif
+
+#define TEST_ARE 0x00000001
+#define TEST_BRE 0x00000002
+#define TEST_ERE 0x00000004
+#define TEST_KRE 0x00000008
+#define TEST_LRE 0x00000010
+#define TEST_SRE 0x00000020
+
+#define TEST_EXPAND 0x00000040
+#define TEST_LENIENT 0x00000080
+
+#define TEST_QUERY 0x00000100
+#define TEST_SUB 0x00000200
+#define TEST_UNSPECIFIED 0x00000400
+#define TEST_VERIFY 0x00000800
+#define TEST_AND 0x00001000
+#define TEST_OR 0x00002000
+
+#define TEST_DELIMIT 0x00010000
+#define TEST_OK 0x00020000
+#define TEST_SAME 0x00040000
+
+#define TEST_ACTUAL 0x00100000
+#define TEST_BASELINE 0x00200000
+#define TEST_FAIL 0x00400000
+#define TEST_PASS 0x00800000
+#define TEST_SUMMARY 0x01000000
+
+#define TEST_IGNORE_ERROR 0x02000000
+#define TEST_IGNORE_OVER 0x04000000
+#define TEST_IGNORE_POSITION 0x08000000
+
+#define TEST_CATCH 0x10000000
+#define TEST_VERBOSE 0x20000000
+
+#define TEST_GLOBAL (TEST_ACTUAL|TEST_AND|TEST_BASELINE|TEST_CATCH|TEST_FAIL|TEST_IGNORE_ERROR|TEST_IGNORE_OVER|TEST_IGNORE_POSITION|TEST_OR|TEST_PASS|TEST_SUMMARY|TEST_VERBOSE)
+
+#ifdef REG_DISCIPLINE
+
+
+#include <stk.h>
+
+typedef struct Disc_s
+{
+ regdisc_t disc;
+ int ordinal;
+ Sfio_t* sp;
+} Disc_t;
+
+static void*
+compf(const regex_t* re, const char* xstr, size_t xlen, regdisc_t* disc)
+{
+ Disc_t* dp = (Disc_t*)disc;
+
+ return (void*)++dp->ordinal;
+}
+
+static int
+execf(const regex_t* re, void* data, const char* xstr, size_t xlen, const char* sstr, size_t slen, char** snxt, regdisc_t* disc)
+{
+ Disc_t* dp = (Disc_t*)disc;
+
+ sfprintf(dp->sp, "{%-.*s}(%d:%d)", xlen, xstr, (int)data, slen);
+ return atoi(xstr);
+}
+
+static void*
+resizef(void* handle, void* data, size_t size)
+{
+ if (!size)
+ return 0;
+ return stkalloc((Sfio_t*)handle, size);
+}
+
+#endif
+
+#ifndef NiL
+#ifdef __STDC__
+#define NiL 0
+#else
+#define NiL (char*)0
+#endif
+#endif
+
+#define H(x) do{if(html)fprintf(stderr,x);}while(0)
+#define T(x) fprintf(stderr,x)
+
+static void
+help(int html)
+{
+H("<!DOCTYPE HTML PUBLIC \"-//IETF//DTD HTML//EN\">\n");
+H("<HTML>\n");
+H("<HEAD>\n");
+H("<TITLE>testregex man document</TITLE>\n");
+H("</HEAD>\n");
+H("<BODY bgcolor=white>\n");
+H("<PRE>\n");
+T("NAME\n");
+T(" testregex - regex(3) test harness\n");
+T("\n");
+T("SYNOPSIS\n");
+T(" testregex [ options ]\n");
+T("\n");
+T("DESCRIPTION\n");
+T(" testregex reads regex(3) test specifications, one per line, from the\n");
+T(" standard input and writes one output line for each failed test. A\n");
+T(" summary line is written after all tests are done. Each successful\n");
+T(" test is run again with REG_NOSUB. Unsupported features are noted\n");
+T(" before the first test, and tests requiring these features are\n");
+T(" silently ignored.\n");
+T("\n");
+T("OPTIONS\n");
+T(" -c catch signals and non-terminating calls\n");
+T(" -e ignore error return mismatches\n");
+T(" -h list help on standard error\n");
+T(" -n do not repeat successful tests with regnexec()\n");
+T(" -o ignore match[] overrun errors\n");
+T(" -p ignore negative position mismatches\n");
+T(" -s use stack instead of malloc\n");
+T(" -x do not repeat successful tests with REG_NOSUB\n");
+T(" -v list each test line\n");
+T(" -A list failed test lines with actual answers\n");
+T(" -B list all test lines with actual answers\n");
+T(" -F list failed test lines\n");
+T(" -P list passed test lines\n");
+T(" -S output one summary line\n");
+T("\n");
+T("INPUT FORMAT\n");
+T(" Input lines may be blank, a comment beginning with #, or a test\n");
+T(" specification. A specification is five fields separated by one\n");
+T(" or more tabs. NULL denotes the empty string and NIL denotes the\n");
+T(" 0 pointer.\n");
+T("\n");
+T(" Field 1: the regex(3) flags to apply, one character per REG_feature\n");
+T(" flag. The test is skipped if REG_feature is not supported by the\n");
+T(" implementation. If the first character is not [BEASKL] then the\n");
+T(" specification is a global control line. One or more of [BEASKL] may be\n");
+T(" specified; the test will be repeated for each mode.\n");
+T("\n");
+T(" B basic BRE (grep, ed, sed)\n");
+T(" E REG_EXTENDED ERE (egrep)\n");
+T(" A REG_AUGMENTED ARE (egrep with negation)\n");
+T(" S REG_SHELL SRE (sh glob)\n");
+T(" K REG_SHELL|REG_AUGMENTED KRE (ksh glob)\n");
+T(" L REG_LITERAL LRE (fgrep)\n");
+T("\n");
+T(" a REG_LEFT|REG_RIGHT implicit ^...$\n");
+T(" b REG_NOTBOL lhs does not match ^\n");
+T(" c REG_COMMENT ignore space and #...\\n\n");
+T(" d REG_SHELL_DOT explicit leading . match\n");
+T(" e REG_NOTEOL rhs does not match $\n");
+T(" f REG_MULTIPLE multiple \\n separated patterns\n");
+T(" g FNM_LEADING_DIR testfnmatch only -- match until /\n");
+T(" h REG_MULTIREF multiple digit backref\n");
+T(" i REG_ICASE ignore case\n");
+T(" j REG_SPAN . matches \\n\n");
+T(" k REG_ESCAPE \\ to ecape [...] delimiter\n");
+T(" l REG_LEFT implicit ^...\n");
+T(" m REG_MINIMAL minimal match\n");
+T(" n REG_NEWLINE explicit \\n match\n");
+T(" o REG_ENCLOSED (|&) magic inside [@|&](...)\n");
+T(" p REG_SHELL_PATH explicit / match\n");
+T(" q REG_DELIMITED delimited pattern\n");
+T(" r REG_RIGHT implicit ...$\n");
+T(" s REG_SHELL_ESCAPED \\ not special\n");
+T(" t REG_MUSTDELIM all delimiters must be specified\n");
+T(" u standard unspecified behavior -- errors not counted\n");
+T(" w REG_NOSUB no subexpression match array\n");
+T(" x REG_LENIENT let some errors slide\n");
+T(" y REG_LEFT regexec() implicit ^...\n");
+T(" z REG_NULL NULL subexpressions ok\n");
+T(" $ expand C \\c escapes in fields 2 and 3\n");
+T(" / field 2 is a regsubcomp() expression\n");
+T("\n");
+T(" Field 1 control lines:\n");
+T("\n");
+T(" C set LC_COLLATE and LC_CTYPE to locale in field 2\n");
+T("\n");
+T(" ?test ... output field 5 if passed and != EXPECTED, silent otherwise\n");
+T(" &test ... output field 5 if current and previous passed\n");
+T(" |test ... output field 5 if current passed and previous failed\n");
+T(" ; ... output field 2 if previous failed\n");
+T(" {test ... skip if failed until }\n");
+T(" } end of skip\n");
+T("\n");
+T(" : comment comment copied as output NOTE\n");
+T(" :comment:test :comment: ignored\n");
+T(" N[OTE] comment comment copied as output NOTE\n");
+T(" T[EST] comment comment\n");
+T("\n");
+T(" number use number for nmatch (20 by default)\n");
+T("\n");
+T(" Field 2: the regular expression pattern; SAME uses the pattern from\n");
+T(" the previous specification.\n");
+T("\n");
+T(" Field 3: the string to match.\n");
+T("\n");
+T(" Field 4: the test outcome. This is either one of the posix error\n");
+T(" codes (with REG_ omitted) or the match array, a list of (m,n)\n");
+T(" entries with m and n being first and last+1 positions in the\n");
+T(" field 3 string, or NULL if REG_NOSUB is in effect and success\n");
+T(" is expected. BADPAT is acceptable in place of any regcomp(3)\n");
+T(" error code. The match[] array is initialized to (-2,-2) before\n");
+T(" each test. All array elements from 0 to nmatch-1 must be specified\n");
+T(" in the outcome. Unspecified endpoints (offset -1) are denoted by ?.\n");
+T(" Unset endpoints (offset -2) are denoted by X. {x}(o:n) denotes a\n");
+T(" matched (?{...}) expression, where x is the text enclosed by {...},\n");
+T(" o is the expression ordinal counting from 1, and n is the length of\n");
+T(" the unmatched portion of the subject string. If x starts with a\n");
+T(" number then that is the return value of re_execf(), otherwise 0 is\n");
+T(" returned.\n");
+T("\n");
+T(" Field 5: optional comment appended to the report.\n");
+T("\n");
+T("CAVEAT\n");
+T(" If a regex implementation misbehaves with memory then all bets are off.\n");
+T("\n");
+T("CONTRIBUTORS\n");
+T(" Glenn Fowler gsf@research.att.com (ksh strmatch, regex extensions)\n");
+T(" David Korn dgk@research.att.com (ksh glob matcher)\n");
+T(" Doug McIlroy mcilroy@dartmouth.edu (ast regex/testre in C++)\n");
+T(" Tom Lord lord@regexps.com (rx tests)\n");
+T(" Henry Spencer henry@zoo.toronto.edu (original public regex)\n");
+T(" Andrew Hume andrew@research.att.com (gre tests)\n");
+T(" John Maddock John_Maddock@compuserve.com (regex++ tests)\n");
+T(" Philip Hazel ph10@cam.ac.uk (pcre tests)\n");
+T(" Ville Laurikari vl@iki.fi (libtre tests)\n");
+H("</PRE>\n");
+H("</BODY>\n");
+H("</HTML>\n");
+}
+
+#ifndef elementsof
+#define elementsof(x) (sizeof(x)/sizeof(x[0]))
+#endif
+
+#ifndef streq
+#define streq(a,b) (*(a)==*(b)&&!strcmp(a,b))
+#endif
+
+#define HUNG 2
+#define NOTEST (~0)
+
+#ifndef REG_TEST_DEFAULT
+#define REG_TEST_DEFAULT 0
+#endif
+
+#ifndef REG_EXEC_DEFAULT
+#define REG_EXEC_DEFAULT 0
+#endif
+
+static const char* unsupported[] =
+{
+ "BASIC",
+#ifndef REG_EXTENDED
+ "EXTENDED",
+#endif
+#ifndef REG_AUGMENTED
+ "AUGMENTED",
+#endif
+#ifndef REG_SHELL
+ "SHELL",
+#endif
+
+#ifndef REG_COMMENT
+ "COMMENT",
+#endif
+#ifndef REG_DELIMITED
+ "DELIMITED",
+#endif
+#ifndef REG_DISCIPLINE
+ "DISCIPLINE",
+#endif
+#ifndef REG_ESCAPE
+ "ESCAPE",
+#endif
+#ifndef REG_ICASE
+ "ICASE",
+#endif
+#ifndef REG_LEFT
+ "LEFT",
+#endif
+#ifndef REG_LENIENT
+ "LENIENT",
+#endif
+#ifndef REG_LITERAL
+ "LITERAL",
+#endif
+#ifndef REG_MINIMAL
+ "MINIMAL",
+#endif
+#ifndef REG_MULTIPLE
+ "MULTIPLE",
+#endif
+#ifndef REG_MULTIREF
+ "MULTIREF",
+#endif
+#ifndef REG_MUSTDELIM
+ "MUSTDELIM",
+#endif
+#ifndef REG_NEWLINE
+ "NEWLINE",
+#endif
+#ifndef REG_NOTBOL
+ "NOTBOL",
+#endif
+#ifndef REG_NOTEOL
+ "NOTEOL",
+#endif
+#ifndef REG_NULL
+ "NULL",
+#endif
+#ifndef REG_RIGHT
+ "RIGHT",
+#endif
+#ifndef REG_SHELL_DOT
+ "SHELL_DOT",
+#endif
+#ifndef REG_SHELL_ESCAPED
+ "SHELL_ESCAPED",
+#endif
+#ifndef REG_SHELL_GROUP
+ "SHELL_GROUP",
+#endif
+#ifndef REG_SHELL_PATH
+ "SHELL_PATH",
+#endif
+#ifndef REG_SPAN
+ "SPAN",
+#endif
+#if REG_NOSUB & REG_TEST_DEFAULT
+ "SUBMATCH",
+#endif
+#if !_REG_nexec
+ "regnexec",
+#endif
+#if !_REG_subcomp
+ "regsubcomp",
+#endif
+ 0
+};
+
+#ifndef REG_COMMENT
+#define REG_COMMENT NOTEST
+#endif
+#ifndef REG_DELIMITED
+#define REG_DELIMITED NOTEST
+#endif
+#ifndef REG_ESCAPE
+#define REG_ESCAPE NOTEST
+#endif
+#ifndef REG_ICASE
+#define REG_ICASE NOTEST
+#endif
+#ifndef REG_LEFT
+#define REG_LEFT NOTEST
+#endif
+#ifndef REG_LENIENT
+#define REG_LENIENT 0
+#endif
+#ifndef REG_MINIMAL
+#define REG_MINIMAL NOTEST
+#endif
+#ifndef REG_MULTIPLE
+#define REG_MULTIPLE NOTEST
+#endif
+#ifndef REG_MULTIREF
+#define REG_MULTIREF NOTEST
+#endif
+#ifndef REG_MUSTDELIM
+#define REG_MUSTDELIM NOTEST
+#endif
+#ifndef REG_NEWLINE
+#define REG_NEWLINE NOTEST
+#endif
+#ifndef REG_NOTBOL
+#define REG_NOTBOL NOTEST
+#endif
+#ifndef REG_NOTEOL
+#define REG_NOTEOL NOTEST
+#endif
+#ifndef REG_NULL
+#define REG_NULL NOTEST
+#endif
+#ifndef REG_RIGHT
+#define REG_RIGHT NOTEST
+#endif
+#ifndef REG_SHELL_DOT
+#define REG_SHELL_DOT NOTEST
+#endif
+#ifndef REG_SHELL_ESCAPED
+#define REG_SHELL_ESCAPED NOTEST
+#endif
+#ifndef REG_SHELL_GROUP
+#define REG_SHELL_GROUP NOTEST
+#endif
+#ifndef REG_SHELL_PATH
+#define REG_SHELL_PATH NOTEST
+#endif
+#ifndef REG_SPAN
+#define REG_SPAN NOTEST
+#endif
+
+#define REG_UNKNOWN (-1)
+
+#ifndef REG_ENEWLINE
+#define REG_ENEWLINE (REG_UNKNOWN-1)
+#endif
+#ifndef REG_ENULL
+#ifndef REG_EMPTY
+#define REG_ENULL (REG_UNKNOWN-2)
+#else
+#define REG_ENULL REG_EMPTY
+#endif
+#endif
+#ifndef REG_ECOUNT
+#define REG_ECOUNT (REG_UNKNOWN-3)
+#endif
+#ifndef REG_BADESC
+#define REG_BADESC (REG_UNKNOWN-4)
+#endif
+#ifndef REG_EMEM
+#define REG_EMEM (REG_UNKNOWN-5)
+#endif
+#ifndef REG_EHUNG
+#define REG_EHUNG (REG_UNKNOWN-6)
+#endif
+#ifndef REG_EBUS
+#define REG_EBUS (REG_UNKNOWN-7)
+#endif
+#ifndef REG_EFAULT
+#define REG_EFAULT (REG_UNKNOWN-8)
+#endif
+#ifndef REG_EFLAGS
+#define REG_EFLAGS (REG_UNKNOWN-9)
+#endif
+#ifndef REG_EDELIM
+#define REG_EDELIM (REG_UNKNOWN-9)
+#endif
+
+static const struct { int code; char* name; } codes[] =
+{
+ REG_UNKNOWN, "UNKNOWN",
+ REG_NOMATCH, "NOMATCH",
+ REG_BADPAT, "BADPAT",
+ REG_ECOLLATE, "ECOLLATE",
+ REG_ECTYPE, "ECTYPE",
+ REG_EESCAPE, "EESCAPE",
+ REG_ESUBREG, "ESUBREG",
+ REG_EBRACK, "EBRACK",
+ REG_EPAREN, "EPAREN",
+ REG_EBRACE, "EBRACE",
+ REG_BADBR, "BADBR",
+ REG_ERANGE, "ERANGE",
+ REG_ESPACE, "ESPACE",
+ REG_BADRPT, "BADRPT",
+ REG_ENEWLINE, "ENEWLINE",
+ REG_ENULL, "ENULL",
+ REG_ECOUNT, "ECOUNT",
+ REG_BADESC, "BADESC",
+ REG_EMEM, "EMEM",
+ REG_EHUNG, "EHUNG",
+ REG_EBUS, "EBUS",
+ REG_EFAULT, "EFAULT",
+ REG_EFLAGS, "EFLAGS",
+ REG_EDELIM, "EDELIM",
+};
+
+static struct
+{
+ regmatch_t NOMATCH;
+ int errors;
+ int extracted;
+ int ignored;
+ int lineno;
+ int passed;
+ int signals;
+ int unspecified;
+ int verify;
+ int warnings;
+ char* file;
+ char* stack;
+ char* which;
+ jmp_buf gotcha;
+#ifdef REG_DISCIPLINE
+ Disc_t disc;
+#endif
+} state;
+
+static void
+quote(char* s, int len, unsigned long test)
+{
+ unsigned char* u = (unsigned char*)s;
+ unsigned char* e;
+ int c;
+
+ if (!u)
+ printf("NIL");
+ else if (!*u && len <= 1)
+ printf("NULL");
+ else if (test & TEST_EXPAND)
+ {
+ if (len < 0)
+ len = strlen((char*)u);
+ e = u + len;
+ if (test & TEST_DELIMIT)
+ printf("\"");
+ while (u < e)
+ switch (c = *u++)
+ {
+ case '\\':
+ printf("\\\\");
+ break;
+ case '"':
+ if (test & TEST_DELIMIT)
+ printf("\\\"");
+ else
+ printf("\"");
+ break;
+ case '\a':
+ printf("\\a");
+ break;
+ case '\b':
+ printf("\\b");
+ break;
+ case 033:
+ printf("\\e");
+ break;
+ case '\f':
+ printf("\\f");
+ break;
+ case '\n':
+ printf("\\n");
+ break;
+ case '\r':
+ printf("\\r");
+ break;
+ case '\t':
+ printf("\\t");
+ break;
+ case '\v':
+ printf("\\v");
+ break;
+ default:
+ if (!iscntrl(c) && isprint(c))
+ putchar(c);
+ else
+ printf("\\x%02x", c);
+ break;
+ }
+ if (test & TEST_DELIMIT)
+ printf("\"");
+ }
+ else
+ printf("%s", s);
+}
+
+static void
+report(char* comment, char* fun, char* re, char* s, int len, char* msg, int flags, unsigned long test)
+{
+ if (state.file)
+ printf("%s:", state.file);
+ printf("%d:", state.lineno);
+ if (re)
+ {
+ printf(" ");
+ quote(re, -1, test|TEST_DELIMIT);
+ if (s)
+ {
+ printf(" versus ");
+ quote(s, len, test|TEST_DELIMIT);
+ }
+ }
+ if (test & TEST_UNSPECIFIED)
+ {
+ state.unspecified++;
+ printf(" unspecified behavior");
+ }
+ else
+ state.errors++;
+ if (state.which)
+ printf(" %s", state.which);
+ if (flags & REG_NOSUB)
+ printf(" NOSUB");
+ if (fun)
+ printf(" %s", fun);
+ if (comment[strlen(comment)-1] == '\n')
+ printf(" %s", comment);
+ else
+ {
+ printf(" %s: ", comment);
+ if (msg)
+ printf("%s: ", msg);
+ }
+}
+
+static void
+error(regex_t* preg, int code)
+{
+ char* msg;
+ char buf[256];
+
+ switch (code)
+ {
+ case REG_EBUS:
+ msg = "bus error";
+ break;
+ case REG_EFAULT:
+ msg = "memory fault";
+ break;
+ case REG_EHUNG:
+ msg = "did not terminate";
+ break;
+ default:
+ regerror(code, preg, msg = buf, sizeof buf);
+ break;
+ }
+ printf("%s\n", msg);
+}
+
+static void
+bad(char* comment, char* re, char* s, int len, unsigned long test)
+{
+ printf("bad test case ");
+ report(comment, NiL, re, s, len, NiL, 0, test);
+ exit(1);
+}
+
+static int
+escape(char* s)
+{
+ char* b;
+ char* t;
+ char* q;
+ char* e;
+ int c;
+
+ for (b = t = s; *t = *s; s++, t++)
+ if (*s == '\\')
+ switch (*++s)
+ {
+ case '\\':
+ break;
+ case 'a':
+ *t = '\a';
+ break;
+ case 'b':
+ *t = '\b';
+ break;
+ case 'c':
+ if (*t = *++s)
+ *t &= 037;
+ else
+ s--;
+ break;
+ case 'e':
+ case 'E':
+ *t = 033;
+ break;
+ case 'f':
+ *t = '\f';
+ break;
+ case 'n':
+ *t = '\n';
+ break;
+ case 'r':
+ *t = '\r';
+ break;
+ case 's':
+ *t = ' ';
+ break;
+ case 't':
+ *t = '\t';
+ break;
+ case 'v':
+ *t = '\v';
+ break;
+ case 'u':
+ case 'x':
+ c = 0;
+ q = c == 'u' ? (s + 5) : (char*)0;
+ e = s + 1;
+ while (!e || !q || s < q)
+ {
+ switch (*++s)
+ {
+ case 'a': case 'b': case 'c': case 'd': case 'e': case 'f':
+ c = (c << 4) + *s - 'a' + 10;
+ continue;
+ case 'A': case 'B': case 'C': case 'D': case 'E': case 'F':
+ c = (c << 4) + *s - 'A' + 10;
+ continue;
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
+ c = (c << 4) + *s - '0';
+ continue;
+ case '{':
+ case '[':
+ if (s != e)
+ {
+ s--;
+ break;
+ }
+ e = 0;
+ continue;
+ case '}':
+ case ']':
+ if (e)
+ s--;
+ break;
+ default:
+ s--;
+ break;
+ }
+ break;
+ }
+ *t = c;
+ break;
+ case '0': case '1': case '2': case '3':
+ case '4': case '5': case '6': case '7':
+ c = *s - '0';
+ q = s + 2;
+ while (s < q)
+ {
+ switch (*++s)
+ {
+ case '0': case '1': case '2': case '3':
+ case '4': case '5': case '6': case '7':
+ c = (c << 3) + *s - '0';
+ break;
+ default:
+ q = --s;
+ break;
+ }
+ }
+ *t = c;
+ break;
+ default:
+ *(s + 1) = 0;
+ bad("invalid C \\ escape\n", s - 1, NiL, 0, 0);
+ }
+ return t - b;
+}
+
+static void
+matchoffprint(int off)
+{
+ switch (off)
+ {
+ case -2:
+ printf("X");
+ break;
+ case -1:
+ printf("?");
+ break;
+ default:
+ printf("%d", off);
+ break;
+ }
+}
+
+static void
+matchprint(regmatch_t* match, int nmatch, int nsub, char* ans, unsigned long test)
+{
+ int i;
+
+ for (; nmatch > nsub + 1; nmatch--)
+ if ((match[nmatch-1].rm_so != -1 || match[nmatch-1].rm_eo != -1) && (!(test & TEST_IGNORE_POSITION) || match[nmatch-1].rm_so >= 0 && match[nmatch-1].rm_eo >= 0))
+ break;
+ for (i = 0; i < nmatch; i++)
+ {
+ printf("(");
+ matchoffprint(match[i].rm_so);
+ printf(",");
+ matchoffprint(match[i].rm_eo);
+ printf(")");
+ }
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE)))
+ {
+ if (ans)
+ printf(" expected: %s", ans);
+ printf("\n");
+ }
+}
+
+static int
+matchcheck(regmatch_t* match, int nmatch, int nsub, char* ans, char* re, char* s, int len, int flags, unsigned long test)
+{
+ char* p;
+ int i;
+ int m;
+ int n;
+
+ if (streq(ans, "OK"))
+ return test & (TEST_BASELINE|TEST_PASS|TEST_VERIFY);
+ for (i = 0, p = ans; i < nmatch && *p; i++)
+ {
+ if (*p == '{')
+ {
+#ifdef REG_DISCIPLINE
+ char* x;
+
+ x = sfstruse(state.disc.sp);
+ if (strcmp(p, x))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ return 0;
+ report("callout failed", NiL, re, s, len, NiL, flags, test);
+ quote(p, -1, test);
+ printf(" expected, ");
+ quote(x, -1, test);
+ printf(" returned\n");
+ }
+#endif
+ break;
+ }
+ if (*p++ != '(')
+ bad("improper answer\n", re, s, -1, test);
+ if (*p == '?')
+ {
+ m = -1;
+ p++;
+ }
+ else
+ m = strtol(p, &p, 10);
+ if (*p++ != ',')
+ bad("improper answer\n", re, s, -1, test);
+ if (*p == '?')
+ {
+ n = -1;
+ p++;
+ }
+ else
+ n = strtol(p, &p, 10);
+ if (*p++ != ')')
+ bad("improper answer\n", re, s, -1, test);
+ if (m!=match[i].rm_so || n!=match[i].rm_eo)
+ {
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY)))
+ {
+ report("failed: match was", NiL, re, s, len, NiL, flags, test);
+ matchprint(match, nmatch, nsub, ans, test);
+ }
+ return 0;
+ }
+ }
+ for (; i < nmatch; i++)
+ {
+ if (match[i].rm_so!=-1 || match[i].rm_eo!=-1)
+ {
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_VERIFY)))
+ {
+ if ((test & TEST_IGNORE_POSITION) && (match[i].rm_so<0 || match[i].rm_eo<0))
+ {
+ state.ignored++;
+ return 0;
+ }
+ if (!(test & TEST_SUMMARY))
+ {
+ report("failed: match was", NiL, re, s, len, NiL, flags, test);
+ matchprint(match, nmatch, nsub, ans, test);
+ }
+ }
+ return 0;
+ }
+ }
+ if (!(test & TEST_IGNORE_OVER) && match[nmatch].rm_so != state.NOMATCH.rm_so)
+ {
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY)))
+ {
+ report("failed: overran match array", NiL, re, s, len, NiL, flags, test);
+ matchprint(match, nmatch + 1, nsub, NiL, test);
+ }
+ return 0;
+ }
+ return 1;
+}
+
+static void
+sigunblock(int s)
+{
+#ifdef SIG_SETMASK
+ int op;
+ sigset_t mask;
+
+ sigemptyset(&mask);
+ if (s)
+ {
+ sigaddset(&mask, s);
+ op = SIG_UNBLOCK;
+ }
+ else op = SIG_SETMASK;
+ sigprocmask(op, &mask, NiL);
+#else
+#ifdef sigmask
+ sigsetmask(s ? (sigsetmask(0L) & ~sigmask(s)) : 0L);
+#endif
+#endif
+}
+
+static void
+gotcha(int sig)
+{
+ int ret;
+
+ signal(sig, gotcha);
+ alarm(0);
+ state.signals++;
+ switch (sig)
+ {
+ case SIGALRM:
+ ret = REG_EHUNG;
+ break;
+ case SIGBUS:
+ ret = REG_EBUS;
+ break;
+ default:
+ ret = REG_EFAULT;
+ break;
+ }
+ sigunblock(sig);
+ longjmp(state.gotcha, ret);
+}
+
+static char*
+getline(FILE* fp)
+{
+ static char buf[32 * 1024];
+
+ register char* s = buf;
+ register char* e = &buf[sizeof(buf)];
+ register char* b;
+
+ for (;;)
+ {
+ if (!(b = fgets(s, e - s, fp)))
+ return 0;
+ state.lineno++;
+ s += strlen(s);
+ if (s == b || *--s != '\n' || s == b || *(s - 1) != '\\')
+ {
+ *s = 0;
+ break;
+ }
+ s--;
+ }
+ return buf;
+}
+
+static unsigned long
+note(unsigned long level, char* msg, unsigned long skip, unsigned long test)
+{
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_SUMMARY)) && !skip)
+ {
+ printf("NOTE\t");
+ if (msg)
+ printf("%s: ", msg);
+ printf("skipping lines %d", state.lineno);
+ }
+ return skip | level;
+}
+
+#define TABS(n) &ts[7-((n)&7)]
+
+static char ts[] = "\t\t\t\t\t\t\t";
+
+static unsigned long
+extract(int* tabs, char* spec, char* re, char* s, char* ans, char* msg, char* accept, regmatch_t* match, int nmatch, int nsub, unsigned long skip, unsigned long level, unsigned long test)
+{
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_OK|TEST_PASS|TEST_SUMMARY))
+ {
+ state.extracted = 1;
+ if (test & TEST_OK)
+ {
+ state.passed++;
+ if ((test & TEST_VERIFY) && !(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_SUMMARY)))
+ {
+ if (msg && strcmp(msg, "EXPECTED"))
+ printf("NOTE\t%s\n", msg);
+ return skip;
+ }
+ test &= ~(TEST_PASS|TEST_QUERY);
+ }
+ if (test & (TEST_QUERY|TEST_VERIFY))
+ {
+ if (test & TEST_BASELINE)
+ test &= ~(TEST_BASELINE|TEST_PASS);
+ else
+ test |= TEST_PASS;
+ skip |= level;
+ }
+ if (!(test & TEST_OK))
+ {
+ if (test & TEST_UNSPECIFIED)
+ state.unspecified++;
+ else
+ state.errors++;
+ }
+ if (test & (TEST_PASS|TEST_SUMMARY))
+ return skip;
+ test &= ~TEST_DELIMIT;
+ printf("%s%s", spec, TABS(*tabs++));
+ if ((test & (TEST_BASELINE|TEST_SAME)) == (TEST_BASELINE|TEST_SAME))
+ printf("SAME");
+ else
+ quote(re, -1, test);
+ printf("%s", TABS(*tabs++));
+ quote(s, -1, test);
+ printf("%s", TABS(*tabs++));
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE)) || !accept && !match)
+ printf("%s", ans);
+ else if (accept)
+ printf("%s", accept);
+ else
+ matchprint(match, nmatch, nsub, NiL, test);
+ if (msg)
+ printf("%s%s", TABS(*tabs++), msg);
+ putchar('\n');
+ }
+ else if (test & TEST_QUERY)
+ skip = note(level, msg, skip, test);
+ else if (test & TEST_VERIFY)
+ state.extracted = 1;
+ return skip;
+}
+
+static int
+catchfree(regex_t* preg, int flags, int* tabs, char* spec, char* re, char* s, char* ans, char* msg, char* accept, regmatch_t* match, int nmatch, int nsub, unsigned long skip, unsigned long level, unsigned long test)
+{
+ int eret;
+
+ if (!(test & TEST_CATCH))
+ {
+ regfree(preg);
+ eret = 0;
+ }
+ else if (!(eret = setjmp(state.gotcha)))
+ {
+ alarm(HUNG);
+ regfree(preg);
+ alarm(0);
+ }
+ else if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ extract(tabs, spec, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test);
+ else
+ {
+ report("failed", "regfree", re, NiL, -1, msg, flags, test);
+ error(preg, eret);
+ }
+ return eret;
+}
+
+int
+main(int argc, char** argv)
+{
+ int flags;
+ int cflags;
+ int eflags;
+ int nmatch;
+ int nexec;
+ int nstr;
+ int cret;
+ int eret;
+ int nsub;
+ int i;
+ int j;
+ int expected;
+ int got;
+ int locale;
+ int subunitlen;
+ int testno;
+ unsigned long level;
+ unsigned long skip;
+ char* p;
+ char* line;
+ char* spec;
+ char* re;
+ char* s;
+ char* ans;
+ char* msg;
+ char* fun;
+ char* ppat;
+ char* subunit;
+ char* version;
+ char* field[6];
+ char* delim[6];
+ FILE* fp;
+ int tabs[6];
+ char unit[64];
+ regmatch_t match[100];
+ regex_t preg;
+
+ static char pat[32 * 1024];
+
+ int nonosub = REG_NOSUB == 0;
+ int nonexec = 0;
+
+ unsigned long test = 0;
+
+ static char* filter[] = { "-", 0 };
+
+ state.NOMATCH.rm_so = state.NOMATCH.rm_eo = -2;
+ p = unit;
+ version = (char*)id + 10;
+ while (p < &unit[sizeof(unit)-1] && (*p = *version++) && !isspace(*p))
+ p++;
+ *p = 0;
+ while ((p = *++argv) && *p == '-')
+ for (;;)
+ {
+ switch (*++p)
+ {
+ case 0:
+ break;
+ case 'c':
+ test |= TEST_CATCH;
+ continue;
+ case 'e':
+ test |= TEST_IGNORE_ERROR;
+ continue;
+ case 'h':
+ case '?':
+ help(0);
+ return 2;
+ case '-':
+ help(p[1] == 'h');
+ return 2;
+ case 'n':
+ nonexec = 1;
+ continue;
+ case 'o':
+ test |= TEST_IGNORE_OVER;
+ continue;
+ case 'p':
+ test |= TEST_IGNORE_POSITION;
+ continue;
+ case 's':
+#ifdef REG_DISCIPLINE
+ if (!(state.stack = stkalloc(stkstd, 0)))
+ fprintf(stderr, "%s: out of space [stack]", unit);
+ state.disc.disc.re_resizef = resizef;
+ state.disc.disc.re_resizehandle = (void*)stkstd;
+#endif
+ continue;
+ case 'x':
+ nonosub = 1;
+ continue;
+ case 'v':
+ test |= TEST_VERBOSE;
+ continue;
+ case 'A':
+ test |= TEST_ACTUAL;
+ continue;
+ case 'B':
+ test |= TEST_BASELINE;
+ continue;
+ case 'F':
+ test |= TEST_FAIL;
+ continue;
+ case 'P':
+ test |= TEST_PASS;
+ continue;
+ case 'S':
+ test |= TEST_SUMMARY;
+ continue;
+ default:
+ fprintf(stderr, "%s: %c: invalid option\n", unit, *p);
+ return 2;
+ }
+ break;
+ }
+ if (!*argv)
+ argv = filter;
+ locale = 0;
+ while (state.file = *argv++)
+ {
+ if (streq(state.file, "-") || streq(state.file, "/dev/stdin") || streq(state.file, "/dev/fd/0"))
+ {
+ state.file = 0;
+ fp = stdin;
+ }
+ else if (!(fp = fopen(state.file, "r")))
+ {
+ fprintf(stderr, "%s: %s: cannot read\n", unit, state.file);
+ return 2;
+ }
+ testno = state.errors = state.ignored = state.lineno = state.passed =
+ state.signals = state.unspecified = state.warnings = 0;
+ skip = 0;
+ level = 1;
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_SUMMARY)))
+ {
+ printf("TEST\t%s ", unit);
+ if (s = state.file)
+ {
+ subunit = p = 0;
+ for (;;)
+ {
+ switch (*s++)
+ {
+ case 0:
+ break;
+ case '/':
+ subunit = s;
+ continue;
+ case '.':
+ p = s - 1;
+ continue;
+ default:
+ continue;
+ }
+ break;
+ }
+ if (!subunit)
+ subunit = state.file;
+ if (p < subunit)
+ p = s - 1;
+ subunitlen = p - subunit;
+ printf("%-.*s ", subunitlen, subunit);
+ }
+ else
+ subunit = 0;
+ for (s = version; *s && (*s != ' ' || *(s + 1) != '$'); s++)
+ putchar(*s);
+ if (test & TEST_CATCH)
+ printf(", catch");
+ if (test & TEST_IGNORE_ERROR)
+ printf(", ignore error code mismatches");
+ if (test & TEST_IGNORE_POSITION)
+ printf(", ignore negative position mismatches");
+#ifdef REG_DISCIPLINE
+ if (state.stack)
+ printf(", stack");
+#endif
+ if (test & TEST_VERBOSE)
+ printf(", verbose");
+ printf("\n");
+#ifdef REG_VERSIONID
+ if (regerror(REG_VERSIONID, NiL, pat, sizeof(pat)) > 0)
+ s = pat;
+ else
+#endif
+#ifdef REG_TEST_VERSION
+ s = REG_TEST_VERSION;
+#else
+ s = "regex";
+#endif
+ printf("NOTE\t%s\n", s);
+ if (elementsof(unsupported) > 1)
+ {
+#if (REG_TEST_DEFAULT & (REG_AUGMENTED|REG_EXTENDED|REG_SHELL)) || !defined(REG_EXTENDED)
+ i = 0;
+#else
+ i = REG_EXTENDED != 0;
+#endif
+ for (got = 0; i < elementsof(unsupported) - 1; i++)
+ {
+ if (!got)
+ {
+ got = 1;
+ printf("NOTE\tunsupported: %s", unsupported[i]);
+ }
+ else
+ printf(",%s", unsupported[i]);
+ }
+ if (got)
+ printf("\n");
+ }
+ }
+#ifdef REG_DISCIPLINE
+ state.disc.disc.re_version = REG_VERSION;
+ state.disc.disc.re_compf = compf;
+ state.disc.disc.re_execf = execf;
+ if (!(state.disc.sp = sfstropen()))
+ bad("out of space [discipline string stream]\n", NiL, NiL, 0, 0);
+ preg.re_disc = &state.disc.disc;
+#endif
+ if (test & TEST_CATCH)
+ {
+ signal(SIGALRM, gotcha);
+ signal(SIGBUS, gotcha);
+ signal(SIGSEGV, gotcha);
+ }
+ while (p = getline(fp))
+ {
+
+ /* parse: */
+
+ line = p;
+ if (*p == ':' && !isspace(*(p + 1)))
+ {
+ while (*++p && *p != ':');
+ if (!*p++)
+ {
+ if (test & TEST_BASELINE)
+ printf("%s\n", line);
+ continue;
+ }
+ }
+ while (isspace(*p))
+ p++;
+ if (*p == 0 || *p == '#' || *p == 'T')
+ {
+ if (test & TEST_BASELINE)
+ printf("%s\n", line);
+ continue;
+ }
+ if (*p == ':' || *p == 'N')
+ {
+ if (test & TEST_BASELINE)
+ printf("%s\n", line);
+ else if (!(test & (TEST_ACTUAL|TEST_FAIL|TEST_PASS|TEST_SUMMARY)))
+ {
+ while (*++p && !isspace(*p));
+ while (isspace(*p))
+ p++;
+ printf("NOTE %s\n", p);
+ }
+ continue;
+ }
+ j = 0;
+ i = 0;
+ field[i++] = p;
+ for (;;)
+ {
+ switch (*p++)
+ {
+ case 0:
+ p--;
+ j = 0;
+ goto checkfield;
+ case '\t':
+ *(delim[i] = p - 1) = 0;
+ j = 1;
+ checkfield:
+ s = field[i - 1];
+ if (streq(s, "NIL"))
+ field[i - 1] = 0;
+ else if (streq(s, "NULL"))
+ *s = 0;
+ while (*p == '\t')
+ {
+ p++;
+ j++;
+ }
+ tabs[i - 1] = j;
+ if (!*p)
+ break;
+ if (i >= elementsof(field))
+ bad("too many fields\n", NiL, NiL, 0, 0);
+ field[i++] = p;
+ /*FALLTHROUGH*/
+ default:
+ continue;
+ }
+ break;
+ }
+ if (!(spec = field[0]))
+ bad("NIL spec\n", NiL, NiL, 0, 0);
+
+ /* interpret: */
+
+ cflags = REG_TEST_DEFAULT;
+ eflags = REG_EXEC_DEFAULT;
+ test &= TEST_GLOBAL;
+ state.extracted = 0;
+ nmatch = 20;
+ nsub = -1;
+ for (p = spec; *p; p++)
+ {
+ if (isdigit(*p))
+ {
+ nmatch = strtol(p, &p, 10);
+ if (nmatch >= elementsof(match))
+ bad("nmatch must be < 100\n", NiL, NiL, 0, 0);
+ p--;
+ continue;
+ }
+ switch (*p)
+ {
+ case 'A':
+ test |= TEST_ARE;
+ continue;
+ case 'B':
+ test |= TEST_BRE;
+ continue;
+ case 'C':
+ if (!(test & TEST_QUERY) && !(skip & level))
+ bad("locale must be nested\n", NiL, NiL, 0, 0);
+ test &= ~TEST_QUERY;
+ if (locale)
+ bad("locale nesting not supported\n", NiL, NiL, 0, 0);
+ if (i != 2)
+ bad("locale field expected\n", NiL, NiL, 0, 0);
+ if (!(skip & level))
+ {
+#if defined(LC_COLLATE) && defined(LC_CTYPE)
+ s = field[1];
+ if (!s || streq(s, "POSIX"))
+ s = "C";
+ if (!(ans = setlocale(LC_COLLATE, s)) || streq(ans, "C") || streq(ans, "POSIX") || !(ans = setlocale(LC_CTYPE, s)) || streq(ans, "C") || streq(ans, "POSIX"))
+ skip = note(level, s, skip, test);
+ else
+ {
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_SUMMARY)))
+ printf("NOTE \"%s\" locale\n", s);
+ locale = level;
+ }
+#else
+ skip = note(level, skip, test, "locales not supported");
+#endif
+ }
+ cflags = NOTEST;
+ continue;
+ case 'E':
+ test |= TEST_ERE;
+ continue;
+ case 'K':
+ test |= TEST_KRE;
+ continue;
+ case 'L':
+ test |= TEST_LRE;
+ continue;
+ case 'S':
+ test |= TEST_SRE;
+ continue;
+
+ case 'a':
+ cflags |= REG_LEFT|REG_RIGHT;
+ continue;
+ case 'b':
+ eflags |= REG_NOTBOL;
+ continue;
+ case 'c':
+ cflags |= REG_COMMENT;
+ continue;
+ case 'd':
+ cflags |= REG_SHELL_DOT;
+ continue;
+ case 'e':
+ eflags |= REG_NOTEOL;
+ continue;
+ case 'f':
+ cflags |= REG_MULTIPLE;
+ continue;
+ case 'g':
+ cflags |= NOTEST;
+ continue;
+ case 'h':
+ cflags |= REG_MULTIREF;
+ continue;
+ case 'i':
+ cflags |= REG_ICASE;
+ continue;
+ case 'j':
+ cflags |= REG_SPAN;
+ continue;
+ case 'k':
+ cflags |= REG_ESCAPE;
+ continue;
+ case 'l':
+ cflags |= REG_LEFT;
+ continue;
+ case 'm':
+ cflags |= REG_MINIMAL;
+ continue;
+ case 'n':
+ cflags |= REG_NEWLINE;
+ continue;
+ case 'o':
+ cflags |= REG_SHELL_GROUP;
+ continue;
+ case 'p':
+ cflags |= REG_SHELL_PATH;
+ continue;
+ case 'q':
+ cflags |= REG_DELIMITED;
+ continue;
+ case 'r':
+ cflags |= REG_RIGHT;
+ continue;
+ case 's':
+ cflags |= REG_SHELL_ESCAPED;
+ continue;
+ case 't':
+ cflags |= REG_MUSTDELIM;
+ continue;
+ case 'u':
+ test |= TEST_UNSPECIFIED;
+ continue;
+ case 'w':
+ cflags |= REG_NOSUB;
+ continue;
+ case 'x':
+ if (REG_LENIENT)
+ cflags |= REG_LENIENT;
+ else
+ test |= TEST_LENIENT;
+ continue;
+ case 'y':
+ eflags |= REG_LEFT;
+ continue;
+ case 'z':
+ cflags |= REG_NULL;
+ continue;
+
+ case '$':
+ test |= TEST_EXPAND;
+ continue;
+
+ case '/':
+ test |= TEST_SUB;
+ continue;
+
+ case '?':
+ test |= TEST_VERIFY;
+ test &= ~(TEST_AND|TEST_OR);
+ state.verify = state.passed;
+ continue;
+ case '&':
+ test |= TEST_VERIFY|TEST_AND;
+ test &= ~TEST_OR;
+ continue;
+ case '|':
+ test |= TEST_VERIFY|TEST_OR;
+ test &= ~TEST_AND;
+ continue;
+ case ';':
+ test |= TEST_OR;
+ test &= ~TEST_AND;
+ continue;
+
+ case '{':
+ level <<= 1;
+ if (skip & (level >> 1))
+ {
+ skip |= level;
+ cflags = NOTEST;
+ }
+ else
+ {
+ skip &= ~level;
+ test |= TEST_QUERY;
+ }
+ continue;
+ case '}':
+ if (level == 1)
+ bad("invalid {...} nesting\n", NiL, NiL, 0, 0);
+ if ((skip & level) && !(skip & (level>>1)))
+ {
+ if (!(test & (TEST_BASELINE|TEST_SUMMARY)))
+ {
+ if (test & (TEST_ACTUAL|TEST_FAIL))
+ printf("}\n");
+ else if (!(test & TEST_PASS))
+ printf("-%d\n", state.lineno);
+ }
+ }
+#if defined(LC_COLLATE) && defined(LC_CTYPE)
+ else if (locale & level)
+ {
+ locale = 0;
+ if (!(skip & level))
+ {
+ s = "C";
+ setlocale(LC_COLLATE, s);
+ setlocale(LC_CTYPE, s);
+ if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_SUMMARY)))
+ printf("NOTE \"%s\" locale\n", s);
+ else if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_PASS))
+ printf("}\n");
+ }
+ else if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL))
+ printf("}\n");
+ }
+#endif
+ level >>= 1;
+ cflags = NOTEST;
+ continue;
+
+ default:
+ bad("bad spec\n", spec, NiL, 0, test);
+ break;
+
+ }
+ break;
+ }
+ if ((cflags|eflags) == NOTEST || (skip & level) && (test & TEST_BASELINE))
+ {
+ if (test & TEST_BASELINE)
+ {
+ while (i > 1)
+ *delim[--i] = '\t';
+ printf("%s\n", line);
+ }
+ continue;
+ }
+ if (test & TEST_OR)
+ {
+ if (!(test & TEST_VERIFY))
+ {
+ test &= ~TEST_OR;
+ if (state.passed == state.verify && i > 1)
+ printf("NOTE\t%s\n", field[1]);
+ continue;
+ }
+ else if (state.passed > state.verify)
+ continue;
+ }
+ else if (test & TEST_AND)
+ {
+ if (state.passed == state.verify)
+ continue;
+ state.passed = state.verify;
+ }
+ if (i < 4)
+ bad("too few fields\n", NiL, NiL, 0, test);
+ while (i < elementsof(field))
+ field[i++] = 0;
+ if (re = field[1])
+ {
+ if (streq(re, "SAME"))
+ {
+ re = ppat;
+ test |= TEST_SAME;
+ }
+ else
+ {
+ if (test & TEST_EXPAND)
+ escape(re);
+ strcpy(ppat = pat, re);
+ }
+ }
+ else
+ ppat = 0;
+ nstr = -1;
+ if ((s = field[2]) && (test & TEST_EXPAND))
+ {
+ nstr = escape(s);
+#if _REG_nexec
+ if (nstr != strlen(s))
+ nexec = nstr;
+#endif
+ }
+ if (!(ans = field[3]))
+ bad("NIL answer\n", NiL, NiL, 0, test);
+ msg = field[4];
+ fflush(stdout);
+ if (test & TEST_SUB)
+#if _REG_subcomp
+ cflags |= REG_DELIMITED;
+#else
+ continue;
+#endif
+
+ compile:
+
+ if (state.extracted || (skip & level))
+ continue;
+#if !(REG_TEST_DEFAULT & (REG_AUGMENTED|REG_EXTENDED|REG_SHELL))
+#ifdef REG_EXTENDED
+ if (REG_EXTENDED != 0 && (test & TEST_BRE))
+#else
+ if (test & TEST_BRE)
+#endif
+ {
+ test &= ~TEST_BRE;
+ flags = cflags;
+ state.which = "BRE";
+ }
+ else
+#endif
+#ifdef REG_EXTENDED
+ if (test & TEST_ERE)
+ {
+ test &= ~TEST_ERE;
+ flags = cflags | REG_EXTENDED;
+ state.which = "ERE";
+ }
+ else
+#endif
+#ifdef REG_AUGMENTED
+ if (test & TEST_ARE)
+ {
+ test &= ~TEST_ARE;
+ flags = cflags | REG_AUGMENTED;
+ state.which = "ARE";
+ }
+ else
+#endif
+#ifdef REG_LITERAL
+ if (test & TEST_LRE)
+ {
+ test &= ~TEST_LRE;
+ flags = cflags | REG_LITERAL;
+ state.which = "LRE";
+ }
+ else
+#endif
+#ifdef REG_SHELL
+ if (test & TEST_SRE)
+ {
+ test &= ~TEST_SRE;
+ flags = cflags | REG_SHELL;
+ state.which = "SRE";
+ }
+ else
+#ifdef REG_AUGMENTED
+ if (test & TEST_KRE)
+ {
+ test &= ~TEST_KRE;
+ flags = cflags | REG_SHELL | REG_AUGMENTED;
+ state.which = "KRE";
+ }
+ else
+#endif
+#endif
+ {
+ if (test & (TEST_BASELINE|TEST_PASS|TEST_VERIFY))
+ extract(tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test|TEST_OK);
+ continue;
+ }
+ if ((test & (TEST_QUERY|TEST_VERBOSE|TEST_VERIFY)) == TEST_VERBOSE)
+ {
+ printf("test %-3d %s ", state.lineno, state.which);
+ quote(re, -1, test|TEST_DELIMIT);
+ printf(" ");
+ quote(s, nstr, test|TEST_DELIMIT);
+ printf("\n");
+ }
+
+ nosub:
+ fun = "regcomp";
+#if _REG_nexec
+ if (nstr >= 0 && nstr != strlen(s))
+ nexec = nstr;
+
+ else
+#endif
+ nexec = -1;
+ if (state.extracted || (skip & level))
+ continue;
+ if (!(test & TEST_QUERY))
+ testno++;
+#ifdef REG_DISCIPLINE
+ if (state.stack)
+ stkset(stkstd, state.stack, 0);
+ flags |= REG_DISCIPLINE;
+ state.disc.ordinal = 0;
+ sfstrseek(state.disc.sp, 0, SEEK_SET);
+#endif
+ if (!(test & TEST_CATCH))
+ cret = regcomp(&preg, re, flags);
+ else if (!(cret = setjmp(state.gotcha)))
+ {
+ alarm(HUNG);
+ cret = regcomp(&preg, re, flags);
+ alarm(0);
+ }
+#if _REG_subcomp
+ if (!cret && (test & TEST_SUB))
+ {
+ fun = "regsubcomp";
+ p = re + preg.re_npat;
+ if (!(test & TEST_CATCH))
+ cret = regsubcomp(&preg, p, NiL, 0, 0);
+ else if (!(cret = setjmp(state.gotcha)))
+ {
+ alarm(HUNG);
+ cret = regsubcomp(&preg, p, NiL, 0, 0);
+ alarm(0);
+ }
+ if (!cret && *(p += preg.re_npat) && !(preg.re_sub->re_flags & REG_SUB_LAST))
+ {
+ if (catchfree(&preg, flags, tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test))
+ continue;
+ cret = REG_EFLAGS;
+ }
+ }
+#endif
+ if (!cret)
+ {
+ if (!(flags & REG_NOSUB) && nsub < 0 && *ans == '(')
+ {
+ for (p = ans; *p; p++)
+ if (*p == '(')
+ nsub++;
+ else if (*p == '{')
+ nsub--;
+ if (nsub >= 0)
+ {
+ if (test & TEST_IGNORE_OVER)
+ {
+ if (nmatch > nsub)
+ nmatch = nsub + 1;
+ }
+ else if (nsub != preg.re_nsub)
+ {
+ if (nsub > preg.re_nsub)
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, "OK", NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("re_nsub incorrect", fun, re, NiL, -1, msg, flags, test);
+ printf("at least %d expected, %d returned\n", nsub, preg.re_nsub);
+ state.errors++;
+ }
+ }
+ else
+ nsub = preg.re_nsub;
+ }
+ }
+ }
+ if (!(test & TEST_SUB) && *ans && *ans != '(' && !streq(ans, "OK") && !streq(ans, "NOMATCH"))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, "OK", NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else if (!(test & TEST_LENIENT))
+ {
+ report("failed", fun, re, NiL, -1, msg, flags, test);
+ printf("%s expected, OK returned\n", ans);
+ }
+ catchfree(&preg, flags, tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test);
+ continue;
+ }
+ }
+ else
+ {
+ if (test & TEST_LENIENT)
+ /* we'll let it go this time */;
+ else if (!*ans || ans[0]=='(' || cret == REG_BADPAT && streq(ans, "NOMATCH"))
+ {
+ got = 0;
+ for (i = 1; i < elementsof(codes); i++)
+ if (cret==codes[i].code)
+ got = i;
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, codes[got].name, NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("failed", fun, re, NiL, -1, msg, flags, test);
+ printf("%s returned: ", codes[got].name);
+ error(&preg, cret);
+ }
+ }
+ else
+ {
+ expected = got = 0;
+ for (i = 1; i < elementsof(codes); i++)
+ {
+ if (streq(ans, codes[i].name))
+ expected = i;
+ if (cret==codes[i].code)
+ got = i;
+ }
+ if (!expected)
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, codes[got].name, NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("failed: invalid error code", NiL, re, NiL, -1, msg, flags, test);
+ printf("%s expected, %s returned\n", ans, codes[got].name);
+ }
+ }
+ else if (cret != codes[expected].code && cret != REG_BADPAT)
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, codes[got].name, NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else if (test & TEST_IGNORE_ERROR)
+ state.ignored++;
+ else
+ {
+ report("should fail and did", fun, re, NiL, -1, msg, flags, test);
+ printf("%s expected, %s returned: ", ans, codes[got].name);
+ state.errors--;
+ state.warnings++;
+ error(&preg, cret);
+ }
+ }
+ }
+ goto compile;
+ }
+
+#if _REG_nexec
+ execute:
+ if (nexec >= 0)
+ fun = "regnexec";
+ else
+#endif
+ fun = "regexec";
+
+ for (i = 0; i < elementsof(match); i++)
+ match[i] = state.NOMATCH;
+
+#if _REG_nexec
+ if (nexec >= 0)
+ {
+ eret = regnexec(&preg, s, nexec, nmatch, match, eflags);
+ s[nexec] = 0;
+ }
+ else
+#endif
+ {
+ if (!(test & TEST_CATCH))
+ eret = regexec(&preg, s, nmatch, match, eflags);
+ else if (!(eret = setjmp(state.gotcha)))
+ {
+ alarm(HUNG);
+ eret = regexec(&preg, s, nmatch, match, eflags);
+ alarm(0);
+ }
+ }
+#if _REG_subcomp
+ if ((test & TEST_SUB) && !eret)
+ {
+ fun = "regsubexec";
+ if (!(test & TEST_CATCH))
+ eret = regsubexec(&preg, s, nmatch, match);
+ else if (!(eret = setjmp(state.gotcha)))
+ {
+ alarm(HUNG);
+ eret = regsubexec(&preg, s, nmatch, match);
+ alarm(0);
+ }
+ }
+#endif
+ if (flags & REG_NOSUB)
+ {
+ if (eret)
+ {
+ if (eret != REG_NOMATCH || !streq(ans, "NOMATCH"))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, "NOMATCH", NiL, 0, 0, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("REG_NOSUB failed", fun, re, s, nstr, msg, flags, test);
+ error(&preg, eret);
+ }
+ }
+ }
+ else if (streq(ans, "NOMATCH"))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, NiL, match, nmatch, nsub, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("should fail and didn't", fun, re, s, nstr, msg, flags, test);
+ error(&preg, eret);
+ }
+ }
+ }
+ else if (eret)
+ {
+ if (eret != REG_NOMATCH || !streq(ans, "NOMATCH"))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, "NOMATCH", NiL, 0, nsub, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("failed", fun, re, s, nstr, msg, flags, test);
+ if (eret != REG_NOMATCH)
+ error(&preg, eret);
+ else if (*ans)
+ printf("expected: %s\n", ans);
+ else
+ printf("\n");
+ }
+ }
+ }
+ else if (streq(ans, "NOMATCH"))
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, NiL, match, nmatch, nsub, skip, level, test|TEST_DELIMIT);
+ else
+ {
+ report("should fail and didn't", fun, re, s, nstr, msg, flags, test);
+ matchprint(match, nmatch, nsub, NiL, test);
+ }
+ }
+#if _REG_subcomp
+ else if (test & TEST_SUB)
+ {
+ p = preg.re_sub->re_buf;
+ if (strcmp(p, ans))
+ {
+ report("failed", fun, re, s, nstr, msg, flags, test);
+ quote(ans, -1, test|TEST_DELIMIT);
+ printf(" expected, ");
+ quote(p, -1, test|TEST_DELIMIT);
+ printf(" returned\n");
+ }
+ }
+#endif
+ else if (!*ans)
+ {
+ if (match[0].rm_so != state.NOMATCH.rm_so)
+ {
+ if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test);
+ else
+ {
+ report("failed: no match but match array assigned", NiL, re, s, nstr, msg, flags, test);
+ matchprint(match, nmatch, nsub, NiL, test);
+ }
+ }
+ }
+ else if (matchcheck(match, nmatch, nsub, ans, re, s, nstr, flags, test))
+ {
+#if _REG_nexec
+ if (nexec < 0 && !nonexec)
+ {
+ nexec = nstr >= 0 ? nstr : strlen(s);
+ s[nexec] = '\n';
+ testno++;
+ goto execute;
+ }
+#endif
+ if (!(test & (TEST_SUB|TEST_VERIFY)) && !nonosub)
+ {
+ if (catchfree(&preg, flags, tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test))
+ continue;
+ flags |= REG_NOSUB;
+ goto nosub;
+ }
+ if (test & (TEST_BASELINE|TEST_PASS|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, NiL, match, nmatch, nsub, skip, level, test|TEST_OK);
+ }
+ else if (test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS|TEST_QUERY|TEST_SUMMARY|TEST_VERIFY))
+ skip = extract(tabs, line, re, s, ans, msg, NiL, match, nmatch, nsub, skip, level, test|TEST_DELIMIT);
+ if (catchfree(&preg, flags, tabs, line, re, s, ans, msg, NiL, NiL, 0, 0, skip, level, test))
+ continue;
+ goto compile;
+ }
+ if (test & TEST_SUMMARY)
+ printf("tests=%-4d errors=%-4d warnings=%-2d ignored=%-2d unspecified=%-2d signals=%d\n", testno, state.errors, state.warnings, state.ignored, state.unspecified, state.signals);
+ else if (!(test & (TEST_ACTUAL|TEST_BASELINE|TEST_FAIL|TEST_PASS)))
+ {
+ printf("TEST\t%s", unit);
+ if (subunit)
+ printf(" %-.*s", subunitlen, subunit);
+ printf(", %d test%s", testno, testno == 1 ? "" : "s");
+ if (state.ignored)
+ printf(", %d ignored mismatche%s", state.ignored, state.ignored == 1 ? "" : "s");
+ if (state.warnings)
+ printf(", %d warning%s", state.warnings, state.warnings == 1 ? "" : "s");
+ if (state.unspecified)
+ printf(", %d unspecified difference%s", state.unspecified, state.unspecified == 1 ? "" : "s");
+ if (state.signals)
+ printf(", %d signal%s", state.signals, state.signals == 1 ? "" : "s");
+ printf(", %d error%s\n", state.errors, state.errors == 1 ? "" : "s");
+ }
+ if (fp != stdin)
+ fclose(fp);
+ }
+ return 0;
+}
diff --git a/test/regex/testregexc.c b/test/regex/testregexc.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexc.c
@@ -0,0 +1 @@
+#include "testregex.c"
diff --git a/test/regex/testregexf.c b/test/regex/testregexf.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexf.c
@@ -0,0 +1 @@
+#include "testregex.c"
diff --git a/test/regex/testregexl.c b/test/regex/testregexl.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexl.c
@@ -0,0 +1 @@
+#include "testregex.c"
diff --git a/test/regex/testregexn.c b/test/regex/testregexn.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexn.c
@@ -0,0 +1 @@
+#include "testregex.c"
diff --git a/test/regex/testregexp.c b/test/regex/testregexp.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexp.c
@@ -0,0 +1 @@
+#include "testregex.c"
diff --git a/test/regex/testregexr.c b/test/regex/testregexr.c
new file mode 100644
index 000000000..46b3ee848
--- /dev/null
+++ b/test/regex/testregexr.c
@@ -0,0 +1 @@
+#include "testregex.c"