4 use warnings FATAL => 'all';
6 use constant _PERL_LT_5_8_4 => ($] < 5.008004) ? 1 : 0;
8 our $VERSION = '1.004004'; # 1.4.4
12 last unless defined && !ref && int != 1;
13 die "Major version specified as $_ - this is strictures version 1";
15 # disable this since Foo->VERSION(undef) correctly returns the version
16 # and that can happen either if our caller passes undef explicitly or
17 # because the for above autovivified $_[1] - I could make it stop but
18 # it's pointless since we don't want to blow up if the caller does
19 # something valid either.
20 no warnings 'uninitialized';
21 shift->SUPER::VERSION(@_);
24 our $extra_load_states;
26 our $Smells_Like_VCS = (-e '.git' || -e '.svn'
27 || (-e '../../dist.ini' && (-e '../../.git' || -e '../../.svn')));
31 warnings->import(FATAL => 'all');
35 while (my @caller = caller(++$depth)) {
36 if ($caller[3] =~ /::BEGIN$/) {
37 # older perls report the BEGIN in the wrong file
38 $caller_file = $depth > 1 ? (caller($depth-1))[1] : $caller[1];
39 $caller_file = $caller[1];
43 my $extra_tests = do {
44 if (exists $ENV{PERL_STRICTURES_EXTRA}) {
45 if (_PERL_LT_5_8_4 and $ENV{PERL_STRICTURES_EXTRA}) {
46 die 'PERL_STRICTURES_EXTRA checks are not available on perls older than 5.8.4: '
47 . "please unset \$ENV{PERL_STRICTURES_EXTRA}\n";
49 $ENV{PERL_STRICTURES_EXTRA};
50 } elsif (! _PERL_LT_5_8_4) {
51 !!($caller_file =~ /^(?:t|xt|lib|blib)/
56 $extra_load_states ||= do {
59 foreach my $mod (qw(indirect multidimensional bareword::filehandles)) {
60 eval "require $mod; \$rv{'$mod'} = 1;" or do {
63 # courtesy of the 5.8 require bug
64 # (we do a copy because 5.16.2 at least uses the same read-only
65 # scalars for the qw() list and it doesn't seem worth a $^V check)
67 (my $file = $mod) =~ s|::|/|g;
68 delete $INC{"${file}.pm"};
73 my $failed = join ' ', @failed;
75 strictures.pm extra testing active but couldn't load all modules. Missing were:
79 Extra testing is auto-enabled in checkouts only, so if you're the author
80 of a strictures-using module you need to run:
82 cpan indirect multidimensional bareword::filehandles
84 but these modules are not required by your users.
91 indirect->unimport(':fatal') if $extra_load_states->{indirect};
92 multidimensional->unimport if $extra_load_states->{multidimensional};
93 bareword::filehandles->unimport if $extra_load_states->{'bareword::filehandles'};
102 strictures - turn on strict and make all warnings fatal
111 use warnings FATAL => 'all';
113 except when called from a file which matches:
115 (caller)[1] =~ /^(?:t|xt|lib|blib)/
117 and when either C<.git> or C<.svn> is present in the current directory (with
118 the intention of only forcing extra tests on the author side) -- or when C<.git>
119 or C<.svn> is present two directories up along with C<dist.ini> (which would
120 indicate we are in a C<dzil test> operation, via L<Dist::Zilla>) --
121 or when the C<PERL_STRICTURES_EXTRA> environment variable is set, in which case
128 use warnings FATAL => 'all';
131 no bareword::filehandles;
133 Note that C<PERL_STRICTURES_EXTRA> may at some point add even more tests, with only a minor
134 version increase, but any changes to the effect of C<use strictures> in
135 normal mode will involve a major version bump.
137 If any of the extra testing modules are not present, L<strictures> will
138 complain loudly, once, via C<warn()>, and then shut up. But you really
139 should consider installing them, they're all great anti-footgun tools.
143 I've been writing the equivalent of this module at the top of my code for
144 about a year now. I figured it was time to make it shorter.
146 Things like the importer in C<use Moose> don't help me because they turn
147 warnings on but don't make them fatal -- which from my point of view is
148 useless because I want an exception to tell me my code isn't warnings-clean.
150 Any time I see a warning from my code, that indicates a mistake.
152 Any time my code encounters a mistake, I want a crash -- not spew to STDERR
153 and then unknown (and probably undesired) subsequent behaviour.
155 I also want to ensure that obvious coding mistakes, like indirect object
156 syntax (and not so obvious mistakes that cause things to accidentally compile
157 as such) get caught, but not at the cost of an XS dependency and not at the
158 cost of blowing things up on another machine.
160 Therefore, L<strictures> turns on additional checking, but only when it thinks
161 it's running in a test file in a VCS checkout -- although if this causes
162 undesired behaviour this can be overridden by setting the
163 C<PERL_STRICTURES_EXTRA> environment variable.
165 If additional useful author side checks come to mind, I'll add them to the
166 C<PERL_STRICTURES_EXTRA> code path only -- this will result in a minor version increase (e.g.
167 1.000000 to 1.001000 (1.1.0) or similar). Any fixes only to the mechanism of
168 this code will result in a sub-version increase (e.g. 1.000000 to 1.000001
171 If the behaviour of C<use strictures> in normal mode changes in any way, that
172 will constitute a major version increase -- and the code already checks
173 when its version is tested to ensure that
177 will continue to only introduce the current set of strictures even if 2.0 is
184 This method does the setup work described above in L</DESCRIPTION>
188 This method traps the C<< strictures->VERSION(1) >> call produced by a use line
189 with a version number on it and does the version check.
191 =head1 EXTRA TESTING RATIONALE
193 Every so often, somebody complains that they're deploying via C<git pull>
194 and that they don't want L<strictures> to enable itself in this case -- and that
195 setting C<PERL_STRICTURES_EXTRA> to 0 isn't acceptable (additional ways to
196 disable extra testing would be welcome but the discussion never seems to get
199 In order to allow us to skip a couple of stages and get straight to a
200 productive conversation, here's my current rationale for turning the
201 extra testing on via a heuristic:
203 The extra testing is all stuff that only ever blows up at compile time;
204 this is intentional. So the oft-raised concern that it's different code being
205 tested is only sort of the case -- none of the modules involved affect the
206 final optree to my knowledge, so the author gets some additional compile
207 time crashes which he/she then fixes, and the rest of the testing is
208 completely valid for all environments.
210 The point of the extra testing -- especially C<no indirect> -- is to catch
211 mistakes that newbie users won't even realise are mistakes without
216 where foo is an & prototyped sub that you forgot to import -- this is
217 pernicious to track down since all I<seems> fine until it gets called
218 and you get a crash. Worse still, you can fail to have imported it due
219 to a circular require, at which point you have a load order dependent
220 bug which I've seen before now I<only> show up in production due to tiny
221 differences between the production and the development environment. I wrote
222 L<http://shadow.cat/blog/matt-s-trout/indirect-but-still-fatal/> to explain
223 this particular problem before L<strictures> itself existed.
225 As such, in my experience so far L<strictures>' extra testing has
226 I<avoided> production versus development differences, not caused them.
228 Additionally, L<strictures>' policy is very much "try and provide as much
229 protection as possible for newbies -- who won't think about whether there's
230 an option to turn on or not" -- so having only the environment variable
231 is not sufficient to achieve that (I get to explain that you need to add
232 C<use strict> at least once a week on freenode #perl -- newbies sometimes
233 completely skip steps because they don't understand that that step
236 I make no claims that the heuristic is perfect -- it's already been evolved
237 significantly over time, especially for 1.004 where we changed things to
238 ensure it only fires on files in your checkout (rather than L<strictures>-using
239 modules you happened to have installed, which was just silly). However, I
240 hope the above clarifies why a heuristic approach is not only necessary but
241 desirable from a point of view of providing new users with as much safety as possible,
242 and will allow any future discussion on the subject to focus on "how do we
243 minimise annoyance to people deploying from checkouts intentionally".
245 =head1 COMMUNITY AND SUPPORT
249 irc.perl.org #toolchain
251 (or bug 'mst' in query on there or freenode)
253 =head2 Git repository
255 Gitweb is on http://git.shadowcat.co.uk/ and the clone URL is:
257 git clone git://git.shadowcat.co.uk/p5sagit/strictures.git
259 The web interface to the repository is at:
261 http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=p5sagit/strictures.git
265 mst - Matt S. Trout (cpan:MSTROUT) <mst@shadowcat.co.uk>
269 None required yet. Maybe this module is perfect (hahahahaha ...).
273 Copyright (c) 2010 the strictures L</AUTHOR> and L</CONTRIBUTORS>
278 This library is free software and may be distributed under the same terms