Filter::Simple with immediately following "no"

[p5pRT]

Migrated from rt.perl.org#107726 (status was 'resolved')

Searchable as RT107726$

[p5pRT]

From user42@zip.com.au

With recent debian perl 5.14.2, if a Filter​::Simple module is turned off
with a "no" immediately after being turned on, like

  use MyFilter;
  no MyFilter;

then the remaining script after the "no" is ignored. The attached
foo.el script and MyFilter.pm module show the problem. Running

  perl foo.el

I expected it to print "this part is not filtered", but it doesn't. If
you put a blank line after "use MyFilter" it does.

I struck this with Smart​::Comments when I'd deleted everything between
use and no, but left those lines intending perhaps to add back some code
there shortly ...

Nosing around Filter​::Simple it looks like when the terminator is found
on the very first call of the filter_add() handler, the handler returns
0, which the perl parser perhaps takes to mean EOF, and therefore stops
reading.

I expect it ought to return with "no MyFilter" or suitably mangled
terminator, to have that run as code.

Dunno if an empty "use/no" block ought to be shown to the FILTER{} etc
code of the subclass. I suppose currently a "use MyFilter" immediately
followed by EOF has not called with an empty, so maybe no empties should
be shown to it, by default at least.

[p5pRT]

From user42@zip.com.au

foo.pl

[p5pRT]

From user42@zip.com.au

MyFilter.pm

[p5pRT]

From user42@zip.com.au

---

---

Flags​:
  category=library
  severity=medium
  module=Filter​::Simple

---

Site configuration information for perl 5.14.2​:

Configured by Debian Project at Mon Nov 28 21​:49​:25 UTC 2011.

Summary of my perl5 (revision 5 version 14 subversion 2) configuration​:
 
  Platform​:
  osname=linux, osvers=2.6.32-5-686, archname=i486-linux-gnu-thread-multi-64int
  uname='linux callisto 2.6.32-5-686 #1 smp mon oct 3 04​:15​:24 utc 2011 i686 gnulinux '
  config_args='-Dusethreads -Duselargefiles -Dccflags=-DDEBIAN -Dcccdlflags=-fPIC -Darchname=i486-linux-gnu -Dprefix=/usr -Dprivlib=/usr/share/perl/5.14 -Darchlib=/usr/lib/perl/5.14 -Dvendorprefix=/usr -Dvendorlib=/usr/share/perl5 -Dvendorarch=/usr/lib/perl5 -Dsiteprefix=/usr/local -Dsitelib=/usr/local/share/perl/5.14.2 -Dsitearch=/usr/local/lib/perl/5.14.2 -Dman1dir=/usr/share/man/man1 -Dman3dir=/usr/share/man/man3 -Dsiteman1dir=/usr/local/man/man1 -Dsiteman3dir=/usr/local/man/man3 -Duse64bitint -Dman1ext=1 -Dman3ext=3perl -Dpager=/usr/bin/sensible-pager -Uafs -Ud_csh -Ud_ualarm -Uusesfio -Uusenm -Ui_libutil -DDEBUGGING=-g -Doptimize=-O2 -Duseshrplib -Dlibperl=libperl.so.5.14.2 -des'
  hint=recommended, useposix=true, d_sigaction=define
  useithreads=define, usemultiplicity=define
  useperlio=define, d_sfio=undef, uselargefiles=define, usesocks=undef
  use64bitint=define, use64bitall=undef, uselongdouble=undef
  usemymalloc=n, bincompat5005=undef
  Compiler​:
  cc='cc', ccflags ='-D_REENTRANT -D_GNU_SOURCE -DDEBIAN -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64',
  optimize='-O2 -g',
  cppflags='-D_REENTRANT -D_GNU_SOURCE -DDEBIAN -fno-strict-aliasing -pipe -fstack-protector -I/usr/local/include'
  ccversion='', gccversion='4.6.2', gccosandvers=''
  intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=12345678
  d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=12
  ivtype='long long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
  alignbytes=4, prototype=define
  Linker and Libraries​:
  ld='cc', ldflags =' -fstack-protector -L/usr/local/lib'
  libpth=/usr/local/lib /lib/i386-linux-gnu /lib/../lib /usr/lib/i386-linux-gnu /usr/lib/../lib /lib /usr/lib /usr/lib64
  libs=-lgdbm -lgdbm_compat -ldb -ldl -lm -lpthread -lc -lcrypt
  perllibs=-ldl -lm -lpthread -lc -lcrypt
  libc=, so=so, useshrplib=true, libperl=libperl.so.5.14.2
  gnulibc_version='2.13'
  Dynamic Linking​:
  dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
  cccdlflags='-fPIC', lddlflags='-shared -O2 -g -L/usr/local/lib -fstack-protector'

Locally applied patches​:

[p5pRT]

From @rjbs

I have attempted to reproduce your problem in blead with the script you provided, but was
unable to.

  ~/tmp/FS$ cat foo.pl
  use lib '.';
  use MyFilter;
  no MyFilter;
  print "this part is not filtered\n";
  ~/tmp/FS$ cat MyFilter.pm
  package MyFilter;
  use Filter​::Simple;
  FILTER {
  print "FILTER code runs\n";
  }
  1;
  ~/tmp/FS$ perl foo.pl
  FILTER code runs
  this part is not filtered

I thought maybe the problem had been fixed since your report, but with 5.14.2, I still get the
same result.

  ~/tmp/FS$ perlbrew use 14.2
  ~/tmp/FS$ perl -v |head -2

  This is perl 5, version 14, subversion 2 (v5.14.2) built for darwin-2level
  ~/tmp/FS$ perl foo.pl
  FILTER code runs
  this part is not filtered

I am tentatively rejecting this bug. If you can provide further steps to reproduce it, please reply
and we can re-open it.

--
rjbs

[p5pRT]

The RT System itself - Status changed from 'new' to 'open'

[p5pRT]

@rjbs - Status changed from 'open' to 'rejected'

[p5pRT]

From user42@zip.com.au

"Ricardo SIGNES via RT" <perlbug-followup@​perl.org> writes​:

~/tmp/FS$ cat MyFilter.pm
package MyFilter;
use Filter​::Simple;
FILTER {
print "FILTER code runs\n";
}
1;

The MyFilter.pm must end as I posted without "1;". In that case it
still fails for me, with recent debian i386 5.14.2 and current filter
1.49 and Filter​::Simple 0.88. Such an end without "1;" is per the
examples in demo/*.pm.

--
If a grizzly bear is coming at you past advice has been "stand still".
Recently this has been revised to "run if you want to". Running is the
reason you have legs and during your last few seconds of life you may as
well put them to their intended use.

[p5pRT]

@rjbs - Status changed from 'rejected' to 'open'

[p5pRT]

From @rjbs

Whoops! Yes, verified.

So, the question I had was, then, "what is the magic true value?" It had the be the result of
calling FILTER. I hadn't included a semicolon after the subroutine passed to FILTER, so 1 became
its argument, making "1" the terminator for the filter, rather than the "no MyFilter" it expected.

I think the little state machine that gathers lines has an off-by-one, although I've really only
given it a cursory glance. I suggest this patch​:

Inline Patch

diff --git a/dist/Filter-Simple/lib/Filter/Simple.pm b/dist/Filter-Simple/lib/Filter/Simple.pm
index d1da0b2..f0ad0f3 100644
--- a/dist/Filter-Simple/lib/Filter/Simple.pm
+++ b/dist/Filter-Simple/lib/Filter/Simple.pm
@@ -198,6 +198,7 @@ sub gen_filter_import {
                     if ($terminator{terminator} &&
                         m/$terminator{terminator}/) {
                         $lastline = $_;
+                        $count++;
                         last;
                     }
                     $data .= $_;

Other eyes are encouraged.

--
rjbs

[p5pRT]

From user42@zip.com.au

"Ricardo SIGNES via RT" <perlbug-followup@​perl.org> writes​:

making "1" the terminator for the filter,

Ah, something subtle there.

I think the little state machine

It's actually a slurp through to the terminator, no? The docs could
make that clearer, since the bare filter stuff is normally line by line.

+ $count++;

Yes, looks likely. It actually only needs a flag rather than a count
anyway I suppose.

[p5pRT]

From @rjbs

* Kevin Ryde <user42@​zip.com.au> [2013-09-12T17​:57​:07]

Yes, looks likely. It actually only needs a flag rather than a count
anyway I suppose.

Could you write a test for the bug? Then I'll apply your test and my patch and
we'll high five. :)

--
rjbs

[p5pRT]

From user42@zip.com.au

"Ricardo Signes via RT" <perlbug-followup@​perl.org> writes​:

Could you write a test for the bug?

Yes.

I wouldn't mind also if the docs began by summarizing the good things of
Filter​::Simple rather than describing how hard is Filter​::Util​::Call
:-).

[p5pRT]

From @rjbs

* Kevin Ryde <user42@​zip.com.au> [2013-09-16T05​:01​:50]

"Ricardo Signes via RT" <perlbug-followup@​perl.org> writes​:

Could you write a test for the bug?

Yes.

I wouldn't mind also if the docs began by summarizing the good things of
Filter​::Simple rather than describing how hard is Filter​::Util​::Call
:-).

Such a patch would also be welcome! The days are clearly long gone when
anybody needs to be convinved to use Filter​::Simple rather than
filter​::Util​::Call. :)

--
rjbs

[p5pRT]

@dcollinsn - Status changed from 'open' to 'stalled'