no warnings doesn't overrule use warnings #1621

p5pRT · 2000-03-30T03:38:02Z

Migrated from rt.perl.org#2859 (status was 'resolved')

Searchable as RT2859$

p5pRT · 2000-03-30T03:38:02Z

From @andk

Created by [email protected]

Thanks to a bugreport by Larry Virden I could track down various warnings that
sometimes appear during CPAN.pm's r command. These can be traced back to the
following smaller testcase. The testcase is based on code found in CGI::Base.

% perl -we '
use warnings;
{
my $eval = q[
package ExtUtils::MakeMaker::_version;
no strict;

local $VERSION;
$VERSION=undef; do {
($VERSION = $Revision) =~ s/.*(\d+\.\d+).*/$1/;
}; $VERSION
];
no warnings;
eval($eval);
}
'
Use of uninitialized value in substitution (s///) at (eval 1) line 7.
%

I would expect that this should be silent.

Perl Info


Flags:
    category=core
    severity=medium

Site configuration information for perl v5.6.0:

Configured by k at Thu Mar 23 23:27:40 CET 2000.

Summary of my perl5 (revision 5.0 version 6 subversion 0) configuration:
  Platform:
    osname=linux, osvers=2.2.15pre14, archname=i686-linux
    uname='linux ak-71.mind.de 2.2.15pre14 #1 smp wed mar 15 15:37:15 cet 2000 i686 unknown '
    config_args='-Dprefix=/usr/local/perl-5.6.0 -Doptimize=-g -Dusemymalloc -des'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=undef use5005threads=undef useithreads=undef usemultiplicity=undef
    useperlio=undef d_sfio=undef uselargefiles=define 
    use64bitint=undef use64bitall=undef uselongdouble=undef usesocks=undef
  Compiler:
    cc='cc', optimize='-g', gccversion=egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
    cppflags='-DDEBUGGING -fno-strict-aliasing -I/usr/local/include'
    ccflags ='-DDEBUGGING -fno-strict-aliasing -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64'
    stdchar='char', d_stdstdio=define, usevfork=false
    intsize=4, longsize=4, ptrsize=4, doublesize=8
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
    ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=4, usemymalloc=y, prototype=define
  Linker and Libraries:
    ld='cc', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib
    libs=-lnsl -lndbm -lgdbm -ldb -ldl -lm -lc -lposix -lcrypt
    libc=/lib/libc-2.1.2.so, so=so, useshrplib=false, libperl=libperl.a
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-rdynamic'
    cccdlflags='-fpic', lddlflags='-shared -L/usr/local/lib'

Locally applied patches:
    


@INC for perl v5.6.0:
    /usr/local/perl-5.6.0/lib/5.6.0/i686-linux
    /usr/local/perl-5.6.0/lib/5.6.0
    /usr/local/perl-5.6.0/lib/site_perl/5.6.0/i686-linux
    /usr/local/perl-5.6.0/lib/site_perl/5.6.0
    /usr/local/perl-5.6.0/lib/site_perl
    .


Environment for perl v5.6.0:
    HOME=/home/k
    LANG=en_US
    LANGUAGE (unset)
    LC_ALL=en_US
    LD_LIBRARY_PATH (unset)
    LOGDIR (unset)
    PATH=/home/k/bin:/usr/local/bin:/usr/bin:/bin:/usr/X11R6/bin:/usr/local/perl/bin:/usr/X11/bin:/sbin:/usr/sbin:/usr/local/Office51/bin
    PERL_BADLANG (unset)
    SHELL=/bin/zsh

p5pRT · 2000-04-02T12:23:29Z

From @pmqs

From: Andreas J. Koenig [mailto:k@ak-71.mind.de]

...

Thanks to a bugreport by Larry Virden I could track down various warnings
that
sometimes appear during CPAN.pm's r command. These can be traced back to
the
following smaller testcase. The testcase is based on code found in
CGI::Base.

% perl -we '
use warnings;
{
my $eval = q[
package ExtUtils::MakeMaker::_version;
no strict;
        local $VERSION;
        $VERSION=undef; do \{
            $$VERSION = $Revision$ =~ s/\.\*$\\d\+\\\.\\d\+$\.\*/$1/;
        \}; $VERSION
    \];
no warnings;
eval($eval);
}
'
Use of uninitialized value in substitution (s///) at (eval 1) line 7.
%

I would expect that this should be silent.

Good catch! I amazed that didn't get picked up before.

Here is a patch that fixes it.

Paul

begin 664 lexwarn.eval.patch.gz
M'XL("!YXYS@"`VQE>'=A<FXN979A;"YP871C:`#M7&MSFT87_FS_BA,[,R2Q
M);&+;N!IIVZBMNZT'M=VDW8ZK0>AE6`L`07D2S/Y[STL2`(+!!(@Z;UHDECL
MGK/L>?9<GET<CN'P&&XM4&U[_`R>;KA@JYZF*]A\<]N[`J+`>WU@..!9V,W`
MM::.Q@!;F.99SG-])D@5N)Z:7$;@H_%Q!+`=:^2H$W@T/#UR`Q@:8P:J"X9I
M3ST<!<>Y&,*S-86!!:;E@:X^Q(<Z!<,#KN]X8`V#6TW4>S:[DZK=JR.&`WFZ
MZO&A--6$(>.W<ZP)UWAO36R'Z<QT#1S_BCEC.'<TW;^X9-ZCY=S[MNN>9RN-
MQN/C8]U&D;IF31KOK\XO&^K4TRW';?QHZ:IY]Q%'54VW,9]%[:G^7/=4IS[Z
MQS<HP$/M6SCZK]<_G8+P)("K6]/Q`/H,*%@.Z,9(9TZ`0,)*<."LJ<='FKK,
MMSR*2K&52EVDY>7!#_3,@7__*X>IDSX*^(V'0=<5G^Q`]5086N.Q]>@&O0-C
M.(2:AK="'&NM>KLN-FS[3O,05"':6C/9XZ+G\,(<L"<%ZO.FPW?OWD4N#VZG
M#'Y6':`$Q);2I`IM`Q5%\;!6JT7E;M#6<]L!1)MT%"(K1`SDWL4_?'R)MLDI
M_M,%W@3\<W/^L??#Q>7MS9NW9V'3U4]WNF%Z+GSE?[7LVM>6?8?K/CJ+Z-Q<
MW5Z_P6YT'ML8&^8(W<B^>U0=$[^[.-8K+FH,X<TKUV::H8X_G5]?KE!Y"Y]#
MI7`2R7(X*P3SYJ/[L&(LF$UU-MWOKGN]FX]Y-+[@3PYS"%='!+RJ;0^NEVBA
MI_N#QK'*B52*/LSNQ\88=)O`GC2IHJ`G19/7L!UU-%$;ODZ#8HY8CJL$F46$
M+7>&L;;<<?"S9?*H(Q(049&H0KK1J$O0B,>?2!2IG1Y_A(JG5&[.HN\X_/BF
MXQ4F.*;=@XL`\3P8W"O(6NQ!'>,BF1;$5M!OQM4+()P\P^L^G(&F6S;_A@)?
MSG`8]$W,C1]ZU]?P^AO>O"P*T/OMJO?^-IQ+63,+_,I/[/->88H_#0\=W/B'
M#82Y!Y0S_5?P:U!%8G<!G`TF5,-$&]0Q+C#7Q#):`_1)!NWZ1D9GV%7)^M0V
M,K#E&[B)8J=:9`+_B#O/[MVA6R\_.HF?VE^KT/LEPYJ7@EN,S`%#\JBI7C0L
MB\][O@;8@]QK<9.R`S`^_=+QK^6P(QIG*\0Z%9J;'E+;6<HR@T=8D0R$G9:V
MO_^`HL7MSR4#,N>_5C)[PR=*W@;+(E69R,M>IP)VTOJ&2;]3+4#K53KA/[#2
M":EI1MAAI4L+U'RU+CE&5TQ\17ZL(!X32\`ZZU`XVW?JZQA-*S1ZG;I7`1`K
M(HF?D@T&&)0/#$S5FSIL=L(7&,NWE_[F4&K2V5E#,91\<Y*I0+CMCW*%M"3$
M]^5)H;N"H1<.XD)69";CZLS=5?Z-8I3+_*TL_88%^&0C17D/0$QRSWUR-"+6
MMQ^.RXDWP_2$'<H^)927E:<:.[>SR4IW^20K*U_:7$SC)(>8O&MDTC+!]IUD
M:S$OY$AXPMX3B<5.H10JD;1UJ*)ZEK*M*%P+JW:!79P(R'N`Y;J\0OAOY15"
M9BH5]IQ7K,HOZS&+]-2R>97=7AI)K:*;+'%IQ5+>QD%&>?BLQS(JQ&U5`L@^
M^,A^XB_U+4_/>N0?"*4]\^>]B0_]><_!)S3H`],`NB"V%2HII`5$EN7$I_Z!
MRGJ/_>76*9$[P6-_?M2##;3=GAWU%#\"CU+D^5*<@'^5Z94G\&WO^XM+^`SP
M^J]/\!6ZQ!=L_(Q_EQWM).J-)RNJT4EZ2CY)5?'O&P(03+X<4^*9.H=I?J8^
MR<L$LU%(2M<Y02C&!$M`4(PY0RXHMN(=Q?C@QN=,>X)HDN?N!=*1HK";3/2R
MGF7:OZP0ST+9%:J))GM9%2H02JM0O#>Q0O$>_MN@?H6B6*%$!?]0DEZA`I7U
M*A013PFAD0J%#93,*U0>`EEOJ'TMPAXW\9N7.:OV*:?C[+Q:[2(?=3=%ZW^X
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M87ILQ!RP'I@S'%N/_NZJ;YBJ\PSF=-+'GMA1%2J,QVR$JQ<*#8R1X8%`!3!&
MIN7$&!R7_S8VV-<@]DG&!YW!K-F6XZG^_TA//RFK,D<>S)BUI7EP)#X-7WY&
M1WGS?HY1BC\$*C.)GJUI_VKZLJ[Y20ZILR=UP#1C@A-?>&7"[G3FG%&%T$-'
M,0]-T/UA^2;HK?-)+WEE*8^>\F30*IRS%,AI`<AI<<AI99#G+E-EKT1E/K2Z
M"A<S8]F*V*M%%J\1B;Q<Y-Q_WTG:&T:.X8/?AM[H&I8YPUX!4A>#/@8C9C*'
M'P2$??&**,D*D8.*>`S?SV7[SXO!YF]X03_B8M=,FSKNXOTJ!G-1['?F8M]#
M^&*5,7M@X[HN@"32%L@MJ=-JMV0"DD0[,LK9*#=_%0HAS7:W@U+-MDREE@@B
M$<5.IQ/();WDH=4A8B!/*&[M5\B'OR)"NRV:3R%\8D>)U,FG,#\J08UV/I4%
M=2(=*B7I1/SBI0=$G"/F.?>82?[0'.8O8/HR_QEZCJ_.(\:=3A3H(N\AW7:W
/#1+.ATO\"S$WW[0@20``
`
end

p5pRT · 2000-04-02T14:29:47Z

From @andk

On Sun, 2 Apr 2000 20:22:34 +0100 , "Marquess,P,Paul,NEL38 R" <paul.marquess@bt.com> said:

Here is a patch that fixes it.

Thanks, Paul. Amazing new tests, and

All tests successful.
u=1.82 s=0.68 cu=112.94 cs=13.59 scripts=230 tests=11830

p5pRT · 2000-04-24T10:26:43Z

From @gsar

On Thu, 30 Mar 2000 13:38:08 +0200, "Andreas J. Koenig" wrote:

Thanks to a bugreport by Larry Virden I could track down various warnings that
sometimes appear during CPAN.pm's r command. These can be traced back to the
following smaller testcase. The testcase is based on code found in CGI::Base.

% perl -we '
use warnings;
{
my $eval = q[
package ExtUtils::MakeMaker::_version;
no strict;
       local $VERSION;
       $VERSION=undef; do \{
           $$VERSION = $Revision$ =~ s/\.\*$\\d\+\\\.\\d\+$\.\*/$1/;
       \}; $VERSION
   \];
no warnings;
eval($eval);
}
'
Use of uninitialized value in substitution (s///) at (eval 1) line 7.
%

I would expect that this should be silent.

Try this patch.

Sarathy
gsar@ActiveState.com

Inline Patch

-----------------------------------8<-----------------------------------
Change 5934 by gsar@auger on 2000/04/24 17:16:54

	propagate lexical warnings from surrounding scope correctly
	within string eval()

Affected files ...

... //depot/perl/pp_ctl.c#198 edit
... //depot/perl/t/pragma/warn/pp_ctl#11 edit

Differences ...

==== //depot/perl/pp_ctl.c#198 (text) ====
Index: perl/pp_ctl.c
--- perl/pp_ctl.c.~1~	Mon Apr 24 10:16:59 2000
+++ perl/pp_ctl.c	Mon Apr 24 10:16:59 2000
@@ -3261,9 +3261,11 @@
     SAVEHINTS();
     PL_hints = PL_op->op_targ;
     SAVESPTR(PL_compiling.cop_warnings);
-    if (!specialWARN(PL_compiling.cop_warnings)) {
-        PL_compiling.cop_warnings = newSVsv(PL_compiling.cop_warnings) ;
-        SAVEFREESV(PL_compiling.cop_warnings) ;
+    if (specialWARN(PL_curcop->cop_warnings))
+        PL_compiling.cop_warnings = PL_curcop->cop_warnings;
+    else {
+        PL_compiling.cop_warnings = newSVsv(PL_curcop->cop_warnings);
+        SAVEFREESV(PL_compiling.cop_warnings);
     }
 
     push_return(PL_op->op_next);

==== //depot/perl/t/pragma/warn/pp_ctl#11 (text) ====
Index: perl/t/pragma/warn/pp_ctl
--- perl/t/pragma/warn/pp_ctl.~1~	Mon Apr 24 10:16:59 2000
+++ perl/t/pragma/warn/pp_ctl	Mon Apr 24 10:16:59 2000
@@ -214,4 +214,17 @@
 { bless ['A'], 'Foo' for 1..10 }
 { bless ['B'], 'Foo' for 1..10 }
 EXPECT
-
+########
+# pp_ctl.c
+use warnings;
+eval 'print $foo';
+EXPECT
+Use of uninitialized value in print at (eval 1) line 1.
+########
+# pp_ctl.c
+use warnings;
+{
+    no warnings;
+    eval 'print $foo';
+}
+EXPECT
End of Patch.

p5pRT · 2000-04-24T10:35:12Z

From @gsar

On Mon, 24 Apr 2000 10:25:08 PDT, I wrote:

Change 5934 by gsar@auger on 2000/04/24 17:16:54

propagate lexical warnings from surrounding scope correctly
within string eval()

Wups, I missed the same fix that Paul had sent due to the different subject
line. I've added his much improved testsuite for the same.

Sarathy
gsar@ActiveState.com

p5pRT · 2000-04-26T02:22:04Z

From @pmqs

From: Gurusamy Sarathy [mailto:gsar@ActiveState.com]

On Mon, 24 Apr 2000 10:25:08 PDT, I wrote:

Change 5934 by gsar@auger on 2000/04/24 17:16:54

propagate lexical warnings from surrounding scope correctly
within string eval()

Wups, I missed the same fix that Paul had sent due to the different
subject
line. I've added his much improved testsuite for the same.

Thanks Sarathy.

Paul

p5pRT closed this as completed Nov 28, 2003

p5pRT added Severity Medium distro-Linux type-library labels Oct 18, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

no warnings doesn't overrule use warnings #1621

no warnings doesn't overrule use warnings #1621

p5pRT commented Mar 30, 2000

p5pRT commented Mar 30, 2000

Uh oh!

p5pRT commented Apr 2, 2000

Uh oh!

p5pRT commented Apr 2, 2000

Uh oh!

p5pRT commented Apr 24, 2000

Uh oh!

p5pRT commented Apr 24, 2000

Uh oh!

p5pRT commented Apr 26, 2000

Uh oh!

no warnings doesn't overrule use warnings #1621

no warnings doesn't overrule use warnings #1621

Comments

p5pRT commented Mar 30, 2000

p5pRT commented Mar 30, 2000

From @andk

Created by [email protected]

Uh oh!

p5pRT commented Apr 2, 2000

From @pmqs

Uh oh!

p5pRT commented Apr 2, 2000

From @andk

Uh oh!

p5pRT commented Apr 24, 2000

From @gsar

Uh oh!

p5pRT commented Apr 24, 2000

From @gsar

Uh oh!

p5pRT commented Apr 26, 2000

From @pmqs

Uh oh!