use constant _PERL_LT_5_8_4 => ($] < 5.008004) ? 1 : 0;
-our $VERSION = '1.004001'; # 1.4.1
+our $VERSION = '1.004003'; # 1.4.3
sub VERSION {
for ($_[1]) {
shift->SUPER::VERSION(@_);
}
-my $extras_load_warned;
+my $extra_load_states;
our $Smells_Like_VCS = (-e '.git' || -e '.svn'
|| (-e '../../dist.ini' && (-e '../../.git' || -e '../../.svn')));
}
};
if ($extra_tests) {
- my @failed;
- if (eval { require indirect; 1 }) {
- indirect->unimport(':fatal');
- } else {
- push @failed, 'indirect';
- }
- if (eval { require multidimensional; 1 }) {
- multidimensional->unimport;
- } else {
- push @failed, 'multidimensional';
- }
- if (eval { require bareword::filehandles; 1 }) {
- bareword::filehandles->unimport;
- } else {
- push @failed, 'bareword::filehandles';
- }
- if (@failed and not $extras_load_warned++) {
- my $failed = join ' ', @failed;
- warn <<EOE;
+ $extra_load_states ||= do {
+
+ my (%rv, @failed);
+ for my $mod (qw(indirect multidimensional bareword::filehandles)) {
+ eval "require $mod; \$rv{'$mod'} = 1;" or do {
+ push @failed, $mod;
+
+ # courtesy of the 5.8 require bug
+ $mod =~ s|::|/|g;
+ delete $INC{"$mod.pm"};
+ };
+ }
+
+ if (@failed) {
+ my $failed = join ' ', @failed;
+ print STDERR <<EOE;
strictures.pm extra testing active but couldn't load all modules. Missing were:
$failed
Extra testing is auto-enabled in checkouts only, so if you're the author
-of a strictures using module you need to run:
+of a strictures-using module you need to run:
cpan indirect multidimensional bareword::filehandles
but these modules are not required by your users.
EOE
- }
+ }
+
+ \%rv;
+ };
+
+ indirect->unimport(':fatal') if $extra_load_states->{indirect};
+ multidimensional->unimport if $extra_load_states->{multidimensional};
+ bareword::filehandles->unimport if $extra_load_states->{'bareword::filehandles'};
}
}
(caller)[1] =~ /^(?:t|xt|lib|blib)/
and when either C<.git> or C<.svn> is present in the current directory (with
-the intention of only forcing extra tests on the author side) - or when C<.git>
+the intention of only forcing extra tests on the author side) -- or when C<.git>
or C<.svn> is present two directories up along with C<dist.ini> (which would
-indicate we are in a C<dzil test> operation, via L<Dist::Zilla>) -
+indicate we are in a C<dzil test> operation, via L<Dist::Zilla>) --
or when the C<PERL_STRICTURES_EXTRA> environment variable is set, in which case
use strictures 1;
about a year now. I figured it was time to make it shorter.
Things like the importer in C<use Moose> don't help me because they turn
-warnings on but don't make them fatal - which from my point of view is
-useless because I want an exception to tell me my code isn't warnings clean.
+warnings on but don't make them fatal -- which from my point of view is
+useless because I want an exception to tell me my code isn't warnings-clean.
Any time I see a warning from my code, that indicates a mistake.
-Any time my code encounters a mistake, I want a crash - not spew to STDERR
+Any time my code encounters a mistake, I want a crash -- not spew to STDERR
and then unknown (and probably undesired) subsequent behaviour.
I also want to ensure that obvious coding mistakes, like indirect object
cost of blowing things up on another machine.
Therefore, L<strictures> turns on additional checking, but only when it thinks
-it's running in a test file in a VCS checkout - though if this causes
+it's running in a test file in a VCS checkout -- although if this causes
undesired behaviour this can be overridden by setting the
C<PERL_STRICTURES_EXTRA> environment variable.
If additional useful author side checks come to mind, I'll add them to the
-C<PERL_STRICTURES_EXTRA> code path only - this will result in a minor version increase (i.e.
+C<PERL_STRICTURES_EXTRA> code path only -- this will result in a minor version increase (e.g.
1.000000 to 1.001000 (1.1.0) or similar). Any fixes only to the mechanism of
-this code will result in a subversion increas (i.e. 1.000000 to 1.000001
+this code will result in a sub-version increase (e.g. 1.000000 to 1.000001
(1.0.1)).
If the behaviour of C<use strictures> in normal mode changes in any way, that
-will constitute a major version increase - and the code already checks
+will constitute a major version increase -- and the code already checks
when its version is tested to ensure that
use strictures 1;
=head1 EXTRA TESTING RATIONALE
Every so often, somebody complains that they're deploying via C<git pull>
-and that they don't want L<strictures> to enable itself in this case - and that
+and that they don't want L<strictures> to enable itself in this case -- and that
setting C<PERL_STRICTURES_EXTRA> to 0 isn't acceptable (additional ways to
disable extra testing would be welcome but the discussion never seems to get
that far).
extra testing on via a heuristic:
The extra testing is all stuff that only ever blows up at compile time;
-this is intentional. So the oft raised concern that it's different code being
-tested is only sort of the case - none of the modules involved affect the
+this is intentional. So the oft-raised concern that it's different code being
+tested is only sort of the case -- none of the modules involved affect the
final optree to my knowledge, so the author gets some additional compile
time crashes which he/she then fixes, and the rest of the testing is
completely valid for all environments.
-The point of the extra testing - especially C<no indirect> - is to catch
+The point of the extra testing -- especially C<no indirect> -- is to catch
mistakes that newbie users won't even realise are mistakes without
help. For example,
foo { ... };
-where foo is an & prototyped sub that you forgot to import - this is
-pernicious to track down since all -seems- fine until it gets called
+where foo is an & prototyped sub that you forgot to import -- this is
+pernicious to track down since all I<seems> fine until it gets called
and you get a crash. Worse still, you can fail to have imported it due
to a circular require, at which point you have a load order dependent
-bug which I've seen before now -only- show up in production due to tiny
+bug which I've seen before now I<only> show up in production due to tiny
differences between the production and the development environment. I wrote
L<http://shadow.cat/blog/matt-s-trout/indirect-but-still-fatal/> to explain
this particular problem before L<strictures> itself existed.
-As such, in my experience so far the L<strictures> extra testing has
--avoided- production versus development differences, not caused them.
+As such, in my experience so far L<strictures>' extra testing has
+I<avoided> production versus development differences, not caused them.
Additionally, L<strictures>' policy is very much "try and provide as much
-protection as possible for newbies - who won't think about whether there's
-an option to turn on or not" - so having only the environment variable
+protection as possible for newbies -- who won't think about whether there's
+an option to turn on or not" -- so having only the environment variable
is not sufficient to achieve that (I get to explain that you need to add
-C<use strict> at least once a week on freenode #perl - newbies sometimes
+C<use strict> at least once a week on freenode #perl -- newbies sometimes
completely skip steps because they don't understand that that step
is important).
-I make no claims that the heuristic is perfect - it's already been evolved
+I make no claims that the heuristic is perfect -- it's already been evolved
significantly over time, especially for 1.004 where we changed things to
ensure it only fires on files in your checkout (rather than L<strictures>-using
modules you happened to have installed, which was just silly). However, I
hope the above clarifies why a heuristic approach is not only necessary but
-desirable from a POV of providing new users with as much safety as possible,
+desirable from a point of view of providing new users with as much safety as possible,
and will allow any future discussion on the subject to focus on "how do we
minimise annoyance to people deploying from checkouts intentionally".