6 # Do not require test.pl, this file has its own framework.
11 use feature 'unicode_strings';
22 if ( $Config{usecrosscompile} ) {
23 print "1..0 # Not all files are available during cross-compilation\n";
26 if ($^O eq 'dec_osf') {
27 print "1..0 # $^O cannot handle this test\n";
30 require '../regen/regen_lib.pl';
39 podcheck.t - Look for possible problems in the Perl pods
44 ./perl -I../lib porting/podcheck.t [--show_all] [--cpan] [--deltas]
45 [--counts] [--pedantic] [FILE ...]
47 ./perl -I../lib porting/podcheck.t --add_link MODULE ...
49 ./perl -I../lib porting/podcheck.t --regen
53 podcheck.t is an extension of Pod::Checker. It looks for pod errors and
54 potential errors in the files given as arguments, or if none specified, in all
55 pods in the distribution workspace, except certain known special ones
56 (specified below). It does additional checking beyond that done by
57 Pod::Checker, and keeps a database of known potential problems, and will
58 fail a pod only if the number of such problems differs from that given in the
59 database. It also suppresses the C<(section) deprecated> message from
60 Pod::Checker, since specifying the man page section number is quite proper to do.
62 The additional checks it always makes are:
66 =item Cross-pod link checking
68 Pod::Checker verifies that links to an internal target in a pod are not
69 broken. podcheck.t extends that (when called without FILE arguments) to
70 external links. It does this by gathering up all the possible targets in the
71 workspace, and cross-checking them. It also checks that a non-broken link
72 points to just one target. (The destination pod could have two targets with
75 The way that the C<LE<lt>E<gt>> pod command works (for links outside the pod)
76 is to actually create a link to C<search.cpan.org> with an embedded query for
77 the desired pod or man page. That means that links outside the distribution
78 are valid. podcheck.t doesn't verify the validity of such links, but instead
79 keeps a database of those known to be valid. This means that if a link to a
80 target not on the list is created, the target needs to be added to the data
81 base. This is accomplished via the L<--add_link|/--add_link MODULE ...>
82 option to podcheck.t, described below.
84 =item An internal link that isn't so specified
86 If a link is broken, but there is an existing internal target of the same
87 name, it is likely that the internal target was meant, and the C<"/"> is
88 missing from the C<LE<lt>E<gt>> pod command.
90 =item Missing or duplicate NAME or missing NAME short description
92 A pod can't be linked to unless it has a unique name.
93 And a NAME should have a dash and short description after it.
95 =item =encoding statement issues
97 This indicates if an C<=encoding> statement should be present, or moved to the
102 If the C<PERL_POD_PEDANTIC> environment variable is set or the C<--pedantic>
103 command line argument is provided then a few more checks are made.
104 The pedantic checks are:
108 =item Verbatim paragraphs that wrap in an 80 (including 1 spare) column window
110 It's annoying to have lines wrap when displaying pod documentation in a
111 terminal window. This checks that all verbatim lines fit in a standard 80
112 column window, even when using a pager that reserves a column for its own use.
113 (Thus the check is for a net of 79 columns.)
114 For those lines that don't fit, it tells you how much needs to be cut in
117 Often, the easiest thing to do to gain space for these is to lower the indent
120 =item Items that perhaps should be links
122 There are mentions of apparent files in the pods that perhaps should be links
123 instead, using C<LE<lt>...E<gt>>
125 =item Items that perhaps should be C<FE<lt>...E<gt>>
127 What look like path names enclosed in C<CE<lt>...E<gt>> should perhaps have
128 C<FE<lt>...E<gt>> mark-up instead.
132 A number of issues raised by podcheck.t and by the base Pod::Checker are not
133 really problems, but merely potential problems, that is, false positives.
134 After inspecting them and
135 deciding that they aren't real problems, it is possible to shut up this program
136 about them, unlike base Pod::Checker. For a valid link to an outside module
137 or man page, call podcheck.t with the C<--add_link> option to add it to the
138 the database of known links; for other causes, call podcheck.t with the C<--regen>
139 option to regenerate the entire database. This tells it that all existing
140 issues are to not be mentioned again.
142 C<--regen> isn't fool-proof. The database merely keeps track of the number of these
143 potential problems of each type for each pod. If a new problem of a given
144 type is introduced into the pod, podcheck.t will spit out all of them. You
145 then have to figure out which is the new one, and should it be changed or not.
146 But doing it this way insulates the database from having to keep track of line
147 numbers of problems, which may change, or the exact wording of each problem
148 which might also change without affecting whether it is a problem or not.
150 Also, if the count of potential problems of a given type for a pod decreases,
151 the database must be regenerated so that it knows the new number. The program
152 gives instructions when this happens.
154 Some pods will have varying numbers of problems of a given type. This can
155 be handled by manually editing the database file (see L</FILES>), and setting
156 the number of those problems for that pod to a negative number. This will
157 cause the corresponding error to always be suppressed no matter how many there
160 Another problem is that there is currently no check that modules listed as
161 valid in the database
162 actually are. Thus any errors introduced there will remain there.
164 =head2 Specially handled pods
170 This pod is generated by pasting bits from other pods. Errors in those bits
171 will show up as errors here, as well as for those other pods. Therefore
172 errors here are suppressed, and the pod is checked only to verify that nodes
173 within it actually exist that are externally linked to.
177 The current perldelta pod is initialized from a template that contains
178 placeholder text. Some of this text is in the form of links that don't really
179 exist. Any such links that are listed in C<@perldelta_ignore_links> will not
180 generate messages. It is presumed that these links will be cleaned up when
181 the perldelta is cleaned up for release since they should be marked with
184 =item Porting/perldelta_template.pod
186 This is not a pod, but a template for C<perldelta>. Any errors introduced
187 here will show up when C<perldelta> is created from it.
189 =item cpan-upstream pods
191 See the L</--cpan> option documentation
195 See the L</--deltas> option documentation
203 =item --add_link MODULE ...
205 Use this option to teach podcheck.t that the C<MODULE>s or man pages actually
206 exist, and to silence any messages that links to them are broken.
208 podcheck.t checks that links within the Perl core distribution are valid, but
209 it doesn't check links to man pages or external modules. When it finds
210 a broken link, it checks its database of external modules and man pages,
211 and only if not found there does it raise a message. This option just adds
212 the list of modules and man page references that follow it on the command line
218 ./perl -I../lib porting/podcheck.t --add_link Unicode::Casing
220 causes the external module "Unicode::Casing" to be added to the database, so
221 C<LE<lt>Unicode::CasingE<gt>> will be considered valid.
225 Regenerate the database used by podcheck.t to include all the existing
226 potential problems. Future runs of the program will not then flag any of
227 these. Setting this option also sets C<--pedantic>.
231 Normally, all pods in the cpan directory are skipped, except to make sure that
232 any blead-upstream links to such pods are valid.
233 This option will cause cpan upstream pods to be fully checked.
237 Normally, all old perldelta pods are skipped, except to make sure that
238 any links to such pods are valid. This is because they are considered
239 stable, and perhaps trying to fix them will cause changes that will
240 misrepresent Perl's history. But, this option will cause them to be fully
245 Normally, if the number of potential problems of a given type found for a
246 pod matches the expected value in the database, they will not be displayed.
247 This option forces the database to be ignored during the run, so all potential
248 problems are displayed and will fail their respective pod test. Specifying
249 any particular FILES to operate on automatically selects this option.
253 Instead of testing, this just dumps the counts of the occurrences of the
254 various types of potential problems in the database.
258 There are three potential problems that are not checked for by default.
259 This options enables them. The environment variable C<PERL_POD_PEDANTIC>
260 can be set to 1 to enable this option also.
261 This option is set when C<--regen> is used.
267 The database is stored in F<t/porting/known_pod_issues.dat>
275 # VMS builds have a '.com' appended to utility and script names, and it adds a
276 # trailing dot for any other file name that doesn't have a dot in it. The db
277 # is stored without those things. This regex allows for these special file
278 # names to be dealt with. It needs to be interpolated into a larger regex
279 # that furnishes the closing boundary.
280 my $vms_re = qr/ \. (?: com )? /x;
282 # Some filenames in the MANIFEST match $vms_re, and so must not be handled the
283 # same way that that the special vms ones are. This hash lists those.
284 my %special_vms_files;
286 # This is to get this to work across multiple file systems, including those
287 # that are not case sensitive. The db is stored in lower case, Un*x style,
288 # and all file name comparisons are done that way.
289 sub canonicalize($) {
291 my ($volume, $directories, $file)
292 = File::Spec->splitpath(File::Spec->canonpath($input));
293 # Assumes $volume is constant for everything in this directory structure
294 $directories = "" if ! $directories;
295 $file = "" if ! $file;
296 $file = lc join '/', File::Spec->splitdir($directories), $file;
297 $file =~ s! / /+ !/!gx; # Multiple slashes => single slash
299 # The db is stored without the special suffixes that are there in VMS, so
300 # strip them off to get the comparable name. But some files on all
301 # platforms have these suffixes, so this shouldn't happen for them, as any
302 # of their db entries will have the suffixes in them. The hash has been
303 # populated with these files.
305 && $file =~ / ( $vms_re ) $ /x
306 && ! exists $special_vms_files{$file})
308 $file =~ s/ $1 $ //x;
313 #####################################################
314 # HOW IT WORKS (in general)
316 # If not called with specific files to check, the directory structure is
317 # examined for files that have pods in them. Files that might not have to be
318 # fully parsed (e.g. in cpan) are parsed enough at this time to find their
319 # pod's NAME, and to get a checksum.
321 # Those kinds of files are sorted last, but otherwise the pods are parsed with
322 # the package coded here, My::Pod::Checker, which is an extension to
323 # Pod::Checker that adds some tests and suppresses others that aren't
324 # appropriate. The latter module has no provision for capturing diagnostics,
325 # so a package, Tie_Array_to_FH, is used to force them to be placed into an
326 # array instead of printed.
328 # Parsing the files builds up a list of links. The files are gone through
329 # again, doing cross-link checking and outputting all saved-up problems with
332 # Sorting the files last that potentially don't need to be fully parsed allows
333 # us to not parse them unless there is a link to an internal anchor in them
334 # from something that we have already parsed. Keeping checksums allows us to
335 # not parse copies of other pods.
337 #####################################################
339 # 1 => Exclude low priority messages that aren't likely to be problems, and
340 # has many false positives; higher numbers give more messages.
341 my $Warnings_Level = 200;
343 # perldelta during construction may have place holder links. N.B. This
344 # variable is referred to by name in release_managers_guide.pod
345 our @perldelta_ignore_links = ( "XXX", "perl5YYYdelta", "perldiag/message" );
347 # To see if two pods with the same NAME are actually copies of the same pod,
348 # which is not an error, it uses a checksum to save work.
349 my $digest_type = "SHA-1";
351 my $original_dir = File::Spec->rel2abs(File::Spec->curdir);
352 my $data_dir = File::Spec->catdir($original_dir, 'porting');
353 my $known_issues = File::Spec->catfile($data_dir, 'known_pod_issues.dat');
354 my $MANIFEST = File::Spec->catfile(File::Spec->updir($original_dir), 'MANIFEST');
357 my $MAX_LINE_LENGTH = 79; # 79 columns
358 my $INDENT = 7; # default nroff indent
360 # Our warning messages. Better not have [('"] in them, as those are used as
361 # delimiters for variable parts of the messages by poderror.
362 my $broken_link = "Apparent broken link";
363 my $broken_internal_link = "Apparent internal link is missing its forward slash";
364 my $multiple_targets = "There is more than one target";
365 my $duplicate_name = "Pod NAME already used";
366 my $need_encoding = "Should have =encoding statement because have non-ASCII";
367 my $encoding_first = "=encoding must be first command (if present)";
368 my $no_name = "There is no NAME";
369 my $missing_name_description = "The NAME should have a dash and short description after it";
370 # the pedantic warnings messages
371 my $line_length = "Verbatim line length including indents exceeds $MAX_LINE_LENGTH by";
372 my $C_not_linked = "? Should you be using L<...> instead of";
373 my $C_with_slash = "? Should you be using F<...> or maybe L<...> instead of";
375 # objects, tests, etc can't be pods, so don't look for them. Also skip
376 # files output by the patch program. Could also ignore most of .gitignore
377 # files, but not all, so don't.
379 my $obj_ext = $Config{'obj_ext'}; $obj_ext =~ tr/.//d; # dot will be added back
380 my $lib_ext = $Config{'lib_ext'}; $lib_ext =~ tr/.//d;
381 my $lib_so = $Config{'so'}; $lib_so =~ tr/.//d;
382 my $dl_ext = $Config{'dlext'}; $dl_ext =~ tr/.//d;
384 # Not really pods, but can look like them.
385 my %excluded_files = (
386 canonicalize("lib/unicore/mktables") => 1,
387 canonicalize("Porting/make-rmg-checklist") => 1,
388 # this one is a POD, but unfinished, so skip
390 canonicalize("Porting/perl5200delta.pod") => 1,
391 canonicalize("Porting/perldelta_template.pod") => 1,
392 canonicalize("regen/feature.pl") => 1,
393 canonicalize("regen/warnings.pl") => 1,
394 canonicalize("autodoc.pl") => 1,
395 canonicalize("configpm") => 1,
396 canonicalize("miniperl") => 1,
397 canonicalize("perl") => 1,
398 canonicalize('cpan/Pod-Perldoc/corpus/no-head.pod') => 1,
399 canonicalize('cpan/Pod-Perldoc/corpus/perlfunc.pod') => 1,
400 canonicalize('cpan/Pod-Perldoc/corpus/utf8.pod') => 1,
401 canonicalize("lib/unicore/mktables") => 1,
404 # This list should not include anything for which case sensitivity is
405 # important, as it won't work on VMS, and won't show up until tested on VMS.
406 # All or almost all such files should be listed in the MANIFEST, so that can
407 # be examined for them, and each such file explicitly excluded, as is done for
408 # .PL files in the loop just below this. For files not catchable this way,
409 # is_pod_file() can be used to exclude these at a finer grained level.
410 my $non_pods = qr/ (?: \.
411 (?: [achot] | zip | gz | bz2 | jar | tar | tgz
412 | orig | rej | patch # Patch program output
413 | sw[op] | \#.* # Editor droppings
414 | old # buildtoc output
415 | xs # pod should be in the .pm file
416 | al # autosplit files
417 | bs # bootstrap files
418 | (?i:sh) # shell scripts, hints, templates
419 | lst # assorted listing files
420 | bat # Windows,Netware,OS2 batch files
421 | cmd # Windows,Netware,OS2 command files
422 | lis # VMS compiler listings
423 | map # VMS linker maps
424 | opt # VMS linker options files
425 | mms # MM(K|S) description files
426 | ts # timestamp files generated during build
427 | $obj_ext # object files
428 | exe # $Config{'exe_ext'} might be empty string
429 | $lib_ext # object libraries
430 | $lib_so # shared libraries
431 | $dl_ext # dynamic libraries
432 | gif # GIF images (example files from CGI.pm)
433 | eg # examples from libnet
436 ) | ~$ | \ \(Autosaved\)\.txt$ # Other editor droppings
437 | ^cxx\$demangler_db\.$ # VMS name mangler database
438 | ^typemap\.?$ # typemap files
439 | ^(?i:Makefile\.PL)$
442 # '.PL' files should be excluded, as they aren't final pods, but often contain
443 # material used in generating pods, and so can look like a pod. We can't use
444 # the regexp above because case sensisitivity is important for these, as some
445 # '.pl' files should be examined for pods. Instead look through the MANIFEST
446 # for .PL files and get their full path names, so we can exclude each such
447 # file explicitly. This works because other porting tests prohibit having two
448 # files with the same names except for case.
449 open my $manifest_fh, '<:bytes', $MANIFEST or die "Can't open $MANIFEST";
450 while (<$manifest_fh>) {
452 # While we have MANIFEST open, on VMS platforms, look for files that match
453 # the magic VMS file names that have to be handled specially. Add these
454 # to the list of them.
455 if ($^O eq 'VMS' && / ^ ( [^\t]* $vms_re ) \t /x) {
456 $special_vms_files{$1} = 1;
458 if (/ ^ ( [^\t]* \. PL ) \t /x) {
459 $excluded_files{canonicalize($1)} = 1;
462 close $manifest_fh, or die "Can't close $MANIFEST";
465 # Pod::Checker messages to suppress
466 my @suppressed_messages = (
467 "(section) in", # Checker is wrong to flag this
468 "multiple occurrence of link target", # We catch independently the ones
469 # that are real problems.
471 "Entity number out of range", # Checker outputs this for anything above
472 # 255, but in fact all Unicode is valid
473 "No items in =over", # ie a blockquote
477 # Returns bool as to if input message is one that is to be suppressed
480 return grep { $message =~ /^\Q$_/i } @suppressed_messages;
483 { # Closure to contain a simple subset of test.pl. This is to get rid of the
484 # unnecessary 'failed at' messages that would otherwise be output pointing
485 # to a particular line in this file.
487 my $current_test = 0;
492 $planned = $plan{tests} + 1; # +1 for final test that files haven't
494 print "1..$planned\n";
505 print "not " unless $success;
506 print "ok $current_test - $message\n";
512 my $n = @_ ? shift : 1;
515 print "ok $current_test # skip $why\n";
517 no warnings 'exiting';
526 print $message =~ s/^/# /mgr;
532 if ($planned && $planned != $current_test) {
534 "# Looks like you planned $planned tests but ran $current_test.\n";
539 # List of known potential problems by pod and type.
542 # Pods given by the keys contain an interior node that is referred to from
544 my %has_referred_to_node;
552 my $do_upstream_cpan = 0; # Assume that are to skip anything in /cpan
553 my $do_deltas = 0; # And stable perldeltas
555 while (@ARGV && substr($ARGV[0], 0, 1) eq '-') {
556 my $arg = shift @ARGV;
558 $arg =~ s/^--/-/; # Treat '--' the same as a single '-'
559 if ($arg eq '-regen') {
563 elsif ($arg eq '-add_link') {
566 elsif ($arg eq '-cpan') {
567 $do_upstream_cpan = 1;
569 elsif ($arg eq '-deltas') {
572 elsif ($arg eq '-show_all') {
575 elsif ($arg eq '-counts') {
578 elsif ($arg eq '-pedantic') {
583 Unknown option '$arg'
585 Usage: $0 [ --regen | --cpan | --show_all | FILE ... | --add_link MODULE ... ]\n"
586 --add_link -> Add the MODULE and man page references to the database
587 --regen -> Regenerate the data file for $0
588 --cpan -> Include files in the cpan subdirectory.
589 --deltas -> Include stable perldeltas
590 --show_all -> Show all known potential problems
591 --counts -> Don't test, but give summary counts of the currently
593 --pedantic -> Check for overly long lines in verbatim blocks
598 $pedantic = 1 if exists $ENV{PERL_POD_PEDANTIC} and $ENV{PERL_POD_PEDANTIC};
601 my $cpan_or_deltas = $do_upstream_cpan || $do_deltas;
602 if (($regen + $show_all + $show_counts + $add_link + $cpan_or_deltas ) > 1) {
603 croak "--regen, --show_all, --counts, and --add_link are mutually exclusive\n and none can be run with --cpan nor --deltas";
606 my $has_input_files = @files;
610 if (! $has_input_files) {
611 croak "--add_link requires at least one module or man page reference";
614 elsif ($has_input_files) {
615 if ($regen || $show_counts || $do_upstream_cpan || $do_deltas) {
616 croak "--regen, --counts, --deltas, and --cpan can't be used since using specific files";
618 foreach my $file (@files) {
619 croak "Can't read file '$file'" if ! -r $file;
623 our %problems; # potential problems found in this run
625 package My::Pod::Checker { # Extend Pod::Checker
626 use parent 'Pod::Checker';
628 # Uses inside out hash to protect from typos
629 # For new fields, remember to add to destructor DESTROY()
630 my %indents; # Stack of indents from =over's in effect for
632 my %current_indent; # Current line's indent
633 my %filename; # The pod is store in this file
634 my %skip; # is SKIP set for this pod
635 my %in_NAME; # true if within NAME section
636 my %in_begin; # true if within =begin section
637 my %linkable_item; # Bool: if the latest =item is linkable. It isn't
638 # for bullet and number lists
639 my %linkable_nodes; # Pod::Checker adds all =items to its node list,
640 # but not all =items are linkable to
641 my %seen_encoding_cmd; # true if have =encoding earlier
642 my %command_count; # Number of commands seen
643 my %seen_pod_cmd; # true if have =pod earlier
644 my %warned_encoding; # true if already have warned about =encoding
648 my $addr = Scalar::Util::refaddr $_[0];
649 delete $command_count{$addr};
650 delete $current_indent{$addr};
651 delete $filename{$addr};
652 delete $in_begin{$addr};
653 delete $indents{$addr};
654 delete $in_NAME{$addr};
655 delete $linkable_item{$addr};
656 delete $linkable_nodes{$addr};
657 delete $seen_encoding_cmd{$addr};
658 delete $seen_pod_cmd{$addr};
660 delete $warned_encoding{$addr};
666 my $filename = shift;
668 my $self = $class->SUPER::new(-quiet => 1,
669 -warnings => $Warnings_Level);
670 my $addr = Scalar::Util::refaddr $self;
671 $command_count{$addr} = 0;
672 $current_indent{$addr} = 0;
673 $filename{$addr} = $filename;
674 $in_begin{$addr} = 0;
676 $linkable_item{$addr} = 0;
677 $seen_encoding_cmd{$addr} = 0;
678 $seen_pod_cmd{$addr} = 0;
679 $warned_encoding{$addr} = 0;
683 # re's for messages that Pod::Checker outputs
684 my $location = qr/ \b (?:in|at|on|near) \s+ /xi;
685 my $optional_location = qr/ (?: $location )? /xi;
686 my $line_reference = qr/ [('"]? $optional_location \b line \s+
687 (?: \d+ | EOF | \Q???\E | - )
690 sub poderror { # Called to register a potential problem
692 # This adds an extra field to the parent hash, 'parameter'. It is
693 # used to extract the variable parts of a message leaving just the
694 # constant skeleton. This in turn allows the message to be
695 # categorized better, so that it shows up as a single type in our
696 # database, with the specifics of each occurrence not being stored with
702 my $addr = Scalar::Util::refaddr $self;
703 return if $skip{$addr};
705 # Input can be a string or hash. If a string, parse it to separate
706 # out the line number and convert to a hash for easier further
709 if (ref $opts ne 'HASH') {
710 $message = join "", $opts, @_;
712 if ($message =~ s/\s*($line_reference)//) {
713 ($line_number = $1) =~ s/\s*$optional_location//;
716 $line_number = '???';
718 $opts = { -msg => $message, -line => $line_number };
720 $message = $opts->{'-msg'};
724 $message =~ s/^\d+\s+//;
725 return if main::suppressed($message);
727 $self->SUPER::poderror($opts, @_);
729 $opts->{parameter} = "" unless $opts->{parameter};
731 # The variable parts of the message tend to be enclosed in '...',
732 # "....", or (...). Extract them and put them in an extra field,
733 # 'parameter'. This is trickier because the matching delimiter to a
734 # '(' is its mirror, and not itself. Text::Balanced could be used
736 while ($message =~ m/ \s* $optional_location ( [('"] )/xg) {
739 $delimiter = ')' if $delimiter eq '(';
741 # If there is no ending delimiter, don't consider it to be a
742 # variable part. Most likely it is a contraction like "Don't"
743 last unless $message =~ m/\G .+? \Q$delimiter/xg;
745 my $length = $+[0] - $start;
747 # Get the part up through the closing delimiter
748 my $special = substr($message, $start, $length);
749 $special =~ s/^\s+//; # No leading whitespace
751 # And add that variable part to the parameter, while removing it
752 # from the message. This isn't a foolproof way of finding the
753 # variable part. For example '(s)' can occur in e.g.,
755 if ($special ne '(s)') {
756 substr($message, $start, $length) = "";
757 pos $message = $start;
758 $opts->{-msg} = $message;
759 $opts->{parameter} .= " " if $opts->{parameter};
760 $opts->{parameter} .= $special;
764 # Extract any additional line number given. This is often the
765 # beginning location of something whereas the main line number gives
767 if ($message =~ /( $line_reference )/xi) {
769 while ($message =~ s/\s*\Q$line_ref//) {
770 $opts->{-msg} = $message;
771 $opts->{parameter} .= " " if $opts->{parameter};
772 $opts->{parameter} .= $line_ref;
776 Carp::carp("Couldn't extract line number from '$message'") if $message =~ /line \d+/;
777 push @{$problems{$filename{$addr}}{$message}}, $opts;
778 #push @{$problems{$self->get_filename}{$message}}, $opts;
781 sub check_encoding { # Does it need an =encoding statement?
782 my ($self, $paragraph, $line_num, $pod_para) = @_;
784 # Do nothing if there is an =encoding in the file, or if the line
785 # doesn't require an =encoding, or have already warned.
786 my $addr = Scalar::Util::refaddr $self;
787 return if $seen_encoding_cmd{$addr}
788 || $warned_encoding{$addr}
789 || $paragraph !~ /\P{ASCII}/;
791 $warned_encoding{$addr} = 1;
792 my ($file, $line) = $pod_para->file_line;
793 $self->poderror({ -line => $line, -file => $file,
794 -msg => $need_encoding
800 my ($self, $paragraph, $line_num, $pod_para) = @_;
801 $self->check_encoding($paragraph, $line_num, $pod_para);
803 $self->SUPER::verbatim($paragraph, $line_num, $pod_para);
805 my $addr = Scalar::Util::refaddr $self;
807 # Pick up the name, since the parent class doesn't in verbatim
808 # NAMEs; so treat as non-verbatim. The parent class only allows one
809 # paragraph in a NAME section, so if there is an extra blank line, it
810 # will trigger a message, but such a blank line is harmless, so skip
812 if ($in_NAME{$addr} && $paragraph =~ /\S/) {
813 $self->textblock($paragraph, $line_num, $pod_para);
816 my @lines = split /^/, $paragraph;
817 for my $i (0 .. @lines - 1) {
818 if ( my $encoding = $seen_encoding_cmd{$addr} ) {
820 $lines[$i] = Encode::decode($encoding, $lines[$i]);
822 $lines[$i] =~ s/\s+$//;
823 my $indent = $self->get_current_indent;
825 if ($pedantic) { # TODO: this check should be moved higher
826 # to avoid more unnecessary work
827 my $exceeds = length(Text::Tabs::expand($lines[$i]))
828 + $indent - $MAX_LINE_LENGTH;
829 next unless $exceeds > 0;
830 my ($file, $line) = $pod_para->file_line;
831 $self->poderror({ -line => $line + $i, -file => $file,
832 -msg => $line_length,
833 parameter => "+$exceeds (including " . ($indent - $INDENT) . " from =over's)",
840 my ($self, $paragraph, $line_num, $pod_para) = @_;
841 $self->check_encoding($paragraph, $line_num, $pod_para);
843 $self->SUPER::textblock($paragraph, $line_num, $pod_para);
845 my ($file, $line) = $pod_para->file_line;
846 my $addr = Scalar::Util::refaddr $self;
847 if ($in_NAME{$addr}) {
849 my $text = $self->interpolate($paragraph, $line_num);
850 if ($text =~ /^\s*(\S+?)\s*$/) {
852 $self->poderror({ -line => $line, -file => $file,
853 -msg => $missing_name_description,
858 $paragraph = join " ", split /^/, $paragraph;
860 # Matches something that looks like a file name, but is enclosed in
862 my $C_path_re = qr{ \b ( C<
863 # exclude various things that have slashes
864 # in them but aren't paths
866 (?: (?: s | qr | m) / ) # regexes
867 | \d+/\d+> # probable fractions
872 | - # File names don't begin with "-"
874 [-\w]+ (?: / [-\w]+ )+ (?: \. \w+ )? > )
877 # If looks like a reference to other documentation by containing the
878 # word 'See' and then a likely pod directive, warn.
879 while ($paragraph =~ m{
880 ( (?: \w+ \s+ )* ) # The phrase before, if any
884 ( [^<]*? ) # The not < excludes nested C<L<...
887 ( \s+ (?: under | in ) \s+ L< )?
889 my $prefix = $1 // "";
890 my $construct = $2; # The whole thing, like C<...>
893 my $trailing = $5; # After the whole thing ending in "L<"
895 # If the full phrase is something like, "you might see C<", or
896 # similar, it really isn't a reference to a link. The ones I saw
897 # all had the word "you" in them; and the "you" wasn't the
898 # beginning of a sentence.
899 if ($prefix !~ / \b you \b /x) {
901 # Now, find what the module or man page name within the
902 # construct would be if it actually has L<> syntax. If it
903 # doesn't have that syntax, will set the module to the entire
906 (?: [^|]+ \| )? # Optional arbitrary text ending
908 ( .+? ) # module, etc. name
909 (?: \/ .+ )? # target within module
913 if (! defined $trailing # not referring to something in another
915 && $interior !~ /$non_pods/
917 # C<> that look like files have their own message below, so
919 && $construct !~ /$C_path_re/g
921 # There can't be spaces (I think) in module names or man
923 && $module !~ / \s /x
925 # F<> that end in eg \.pl are almost certainly ok, as are
926 # those that look like a path with multiple "/" chars
929 && $interior !~ /\.\w+$/
930 && $interior !~ /\/.+\//)
933 # TODO: move the checking of $pedantic higher up
934 $self->poderror({ -line => $line, -file => $file,
935 -msg => $C_not_linked,
936 parameter => $construct
941 while ($paragraph =~ m/$C_path_re/g) {
943 # TODO: move the checking of $pedantic higher up
944 $self->poderror({ -line => $line, -file => $file,
945 -msg => $C_with_slash,
946 parameter => $construct
953 my ($self, $cmd, $paragraph, $line_num, $pod_para) = @_;
954 my $addr = Scalar::Util::refaddr $self;
956 $seen_pod_cmd{$addr}++;
958 elsif ($cmd eq "encoding") {
959 my ($file, $line) = $pod_para->file_line;
960 $seen_encoding_cmd{$addr} = $paragraph; # for later decoding
961 if ($command_count{$addr} != 1 && $seen_pod_cmd{$addr}) {
962 $self->poderror({ -line => $line, -file => $file,
963 -msg => $encoding_first
967 $self->check_encoding($paragraph, $line_num, $pod_para);
969 # Pod::Check treats all =items as linkable, but the bullet and
970 # numbered lists really aren't. So keep our own list. This has to be
971 # processed before SUPER is called so that the list is started before
972 # the rest of it gets parsed.
973 if ($cmd eq 'item') { # Not linkable if item begins with * or a digit
974 $linkable_item{$addr} = ($paragraph !~ / ^ \s*
976 | \d+ \.? (?: \$ | \s+ )
982 $self->SUPER::command($cmd, $paragraph, $line_num, $pod_para);
984 $command_count{$addr}++;
986 $in_NAME{$addr} = 0; # Will change to 1 below if necessary
987 $in_begin{$addr} = 0; # ibid
988 if ($cmd eq 'over') {
989 my $text = $self->interpolate($paragraph, $line_num);
990 my $indent = 4; # default
991 $indent = $1 if $text && $text =~ /^\s*(\d+)\s*$/;
992 push @{$indents{$addr}}, $indent;
993 $current_indent{$addr} += $indent;
995 elsif ($cmd eq 'back') {
996 if (@{$indents{$addr}}) {
997 $current_indent{$addr} -= pop @{$indents{$addr}};
1000 # =back without corresponding =over, but should have
1002 $current_indent{$addr} = 0;
1005 elsif ($cmd =~ /^head/) {
1006 if (! $in_begin{$addr}) {
1008 # If a particular formatter, then this command doesn't really
1010 $current_indent{$addr} = 0;
1011 undef @{$indents{$addr}};
1014 my $text = $self->interpolate($paragraph, $line_num);
1015 $in_NAME{$addr} = 1 if $cmd eq 'head1'
1016 && $text && $text =~ /^NAME\b/;
1018 elsif ($cmd eq 'begin') {
1019 $in_begin{$addr} = 1;
1029 if ($_[0] && ($page = $_[0][1]{'-page'})) {
1030 my $node = $_[0][1]{'-node'};
1032 # If the hyperlink is to an interior node of another page, save it
1033 # so that we can see if we need to parse normally skipped files.
1034 $has_referred_to_node{$page} = 1 if $node;
1036 # Ignore certain placeholder links in perldelta. Check if the
1037 # link is page-level, and also check if to a node within the page
1038 if ($self->name && $self->name eq "perldelta"
1039 && ((grep { $page eq $_ } @perldelta_ignore_links)
1041 && (grep { "$page/$node" eq $_ } @perldelta_ignore_links)
1046 return $self->SUPER::hyperlink($_[0]);
1053 $text =~ s/\s+$//s; # strip trailing whitespace
1054 $text =~ s/\s+/ /gs; # collapse whitespace
1055 my $addr = Scalar::Util::refaddr $self;
1056 push(@{$linkable_nodes{$addr}}, $text) if
1057 ! $current_indent{$addr}
1058 || $linkable_item{$addr};
1060 return $self->SUPER::node($_[0]);
1063 sub get_current_indent {
1064 return $INDENT + $current_indent{Scalar::Util::refaddr $_[0]};
1068 return $filename{Scalar::Util::refaddr $_[0]};
1071 sub linkable_nodes {
1072 my $linkables = $linkable_nodes{Scalar::Util::refaddr $_[0]};
1073 return undef unless $linkables;
1078 return $skip{Scalar::Util::refaddr $_[0]} // 0;
1083 $skip{Scalar::Util::refaddr $self} = shift;
1085 # If skipping, no need to keep the problems for it
1086 delete $problems{$self->get_filename};
1090 sub parse_from_file {
1091 # This overrides the super class method so that if an open fails on a
1092 # transitory file, it doesn't croak. It returns 1 if it did find the
1093 # file, 0 if it didn't
1096 my $filename = shift;
1097 # ignores 2nd param, which is output file. Always uses undef
1099 if (open my $in_fh, '<:bytes', $filename) {
1100 $self->SUPER::parse_from_filehandle($in_fh, undef);
1105 # If couldn't open file, perhaps it was transitory, and hence not an error
1106 return 0 unless -e $filename;
1108 die "Can't open '$filename': $!\n";
1112 package Tie_Array_to_FH { # So printing actually goes to an array
1118 my $array_ref = shift;
1120 my $self = bless \do{ my $anonymous_scalar }, $class;
1121 $array{Scalar::Util::refaddr $self} = $array_ref;
1128 push @{$array{Scalar::Util::refaddr $self}}, @_;
1134 my %filename_to_checker; # Map a filename to it's pod checker object
1135 my %id_to_checker; # Map a checksum to it's pod checker object
1136 my %nodes; # key is filename, values are nodes in that file.
1137 my %nodes_first_word; # same, but value is first word of each node
1138 my %valid_modules; # List of modules known to exist outside us.
1139 my %digests; # checksums of files, whose names are the keys
1140 my %filename_to_pod; # Map a filename to its pod NAME
1141 my %files_with_unknown_issues;
1142 my %files_with_fixes;
1145 open $data_fh, '<:bytes', $known_issues or die "Can't open $known_issues";
1147 my %counts; # For --counts param, count of each issue type
1148 my %suppressed_files; # Files with at least one issue type to suppress
1150 # This file is the data file for $0.
1151 # There are three types of lines.
1152 # Comment lines are white-space only or begin with a '#', like this one. Any
1153 # changes you make to the comment lines will be lost when the file is
1155 # Lines without tab characters are simply NAMES of pods that the program knows
1156 # will have links to them and the program does not check if those links are
1158 # All other lines should have three fields, each separated by a tab. The
1159 # first field is the name of a pod; the second field is an error message
1160 # generated by this program; and the third field is a count of how many
1161 # known instances of that message there are in the pod. -1 means that the
1162 # program can expect any number of this type of message.
1165 my @existing_issues;
1168 while (<$data_fh>) { # Read the database
1170 next if /^\s*(?:#|$)/; # Skip comment and empty lines
1173 if ($add_link) { # The issues are saved and later output unchanged
1174 push @existing_issues, $_;
1178 # Keep track of counts of each issue type for each file
1179 my ($filename, $message, $count) = split /\t/;
1180 $known_problems{$filename}{$message} = $count;
1183 if ($count < 0) { # -1 means to suppress this issue type
1184 $suppressed_files{$filename} = $filename;
1187 $counts{$message} += $count;
1191 else { # Lines without a tab are modules known to be valid
1192 $valid_modules{$_} = 1
1198 $copy_fh = open_new($known_issues);
1200 # Check for basic sanity, and add each command line argument
1201 foreach my $module (@files) {
1202 die "\"$module\" does not look like a module or man page"
1203 # Must look like (A or A::B or A::B::C ..., or foo(3C)
1204 if $module !~ /^ (?: \w+ (?: :: \w+ )* | \w+ \( \d \w* \) ) $/x;
1205 $valid_modules{$module} = 1
1207 my_safer_print($copy_fh, $HEADER);
1208 foreach (sort { lc $a cmp lc $b } keys %valid_modules) {
1209 my_safer_print($copy_fh, $_, "\n");
1212 # The rest of the db file is output unchanged.
1213 my_safer_print($copy_fh, join "\n", @existing_issues, "");
1215 close_and_rename($copy_fh);
1221 foreach my $message (sort keys %counts) {
1222 $total += $counts{$message};
1223 note(Text::Tabs::expand("$counts{$message}\t$message"));
1225 note("-----\n" . Text::Tabs::expand("$total\tknown potential issues"));
1226 if (%suppressed_files) {
1227 note("\nFiles that have all messages of at least one type suppressed:");
1228 note(join ",", keys %suppressed_files);
1233 # re to match files that are to be parsed only if there is an internal link
1234 # to them. It does not include cpan, as whether those are parsed depends
1235 # on a switch. Currently, only perltoc and the stable perldelta.pod's
1236 # are included. The latter all have characters between 'perl' and
1237 # 'delta'. (Actually the currently developed one matches as well, but
1238 # is a duplicate of perldelta.pod, so can be skipped, so fine for it to
1240 my $only_for_interior_links_re = qr/ ^ pod\/perltoc.pod $
1242 unless ($do_deltas) {
1243 $only_for_interior_links_re = qr/$only_for_interior_links_re |
1244 \b perl \d+ delta \. pod \b
1251 sub output_thanks ($$$$) { # Called when an issue has been fixed
1252 my $filename = shift;
1253 my $original_count = shift;
1254 my $current_count = shift;
1255 my $message = shift;
1257 $files_with_fixes{$filename} = 1;
1259 my $fixed_count = $original_count - $current_count;
1260 my $a_problem = ($fixed_count == 1) ? "a problem" : "multiple problems";
1261 my $another_problem = ($fixed_count == 1) ? "another problem" : "another set of problems";
1265 There were $original_count occurrences (now $current_count) in this pod of type
1270 There are no longer any problems found in this pod!
1277 Thanks for fixing $a_problem!
1279 Now you must teach $0 that this was fixed.
1284 Thanks for fixing $another_problem.
1293 sub my_safer_print { # print, with error checking for outputting to db
1294 my ($fh, @lines) = @_;
1296 if (! print $fh @lines) {
1297 my $save_error = $!;
1299 die "Write failure: $save_error";
1303 sub extract_pod { # Extracts just the pod from a file; returns undef if file
1305 my $filename = shift;
1309 # Arrange for the output of Pod::Parser to be collected in an array we can
1310 # look at instead of being printed
1311 tie *ALREADY_FH, 'Tie_Array_to_FH', \@pod;
1312 if (open my $in_fh, '<:bytes', $filename) {
1313 my $parser = Pod::Parser->new();
1314 $parser->parse_from_filehandle($in_fh, *ALREADY_FH);
1317 return join "", @pod
1320 # The file should already have been opened once to get here, so if that
1321 # fails, something is wrong. It's possible that a transitory file
1322 # containing a pod would get here, so if the file no longer exists just
1324 return unless -e $filename;
1325 die "Can't open '$filename': $!\n";
1328 my $digest = Digest->new($digest_type);
1330 # This is used as a callback from File::Find::find(), which always constructs
1331 # pathnames using Unix separators
1333 # If $_ is a pod file, add it to the lists and do other prep work.
1336 # Don't look at files in directories that are for tests, nor those
1337 # beginning with a dot
1338 if (m!/t\z! || m!/\.!) {
1339 $File::Find::prune = 1;
1344 return unless -r && -s; # Can't check it if can't read it; no need to
1346 return unless -f || -l; # Weird file types won't be pods
1348 my ($leaf) = m!([^/]+)\z!;
1349 if (m!/\.! # No hidden Unix files
1350 || $leaf =~ $non_pods) {
1351 note("Not considering $_") if DEBUG;
1355 my $filename = $File::Find::name;
1357 # $filename is relative, like './path'. Strip that initial part away.
1358 $filename =~ s!^\./!! or die 'Unexpected pathname "$filename"';
1360 return if $excluded_files{canonicalize($filename)};
1365 if (! open $candidate, '<:bytes', $_) {
1367 # If a transitory file was found earlier, the open could fail
1368 # legitimately and we just skip the file; also skip it if it is a
1369 # broken symbolic link, as it is probably just a build problem;
1370 # certainly not a file that we would want to check the pod of.
1371 # Otherwise fail it here and no reason to process it further.
1372 # (But the test count will be off too)
1373 ok(0, "Can't open '$filename': $!")
1374 if -r $filename && ! -l $filename;
1380 # If the file is a .pm or .pod, having any initial '=' on a line is
1381 # grounds for testing it. Otherwise, require a head1 NAME line to
1382 # consider it as a potential pod
1383 if ($filename =~ /\.(?:pm|pod)/) {
1384 return unless $contents =~ /^=/m;
1386 return unless $contents =~ /^=head1 +NAME/m;
1389 # Here, we know that the file is a pod. Add it to the list of files
1390 # to check and create a checker object for it.
1392 push @files, $filename;
1393 my $checker = My::Pod::Checker->new($filename);
1394 $filename_to_checker{$filename} = $checker;
1396 # In order to detect duplicate pods and only analyze them once, we
1397 # compute checksums for the file, so don't have to do an exact
1398 # compare. Note that if the pod is just part of the file, the
1399 # checksums can differ for the same pod. That special case is handled
1400 # later, since if the checksums of the whole file are the same, that
1401 # case won't even come up. We don't need the checksums for files that
1402 # we parse only if there is a link to its interior, but we do need its
1403 # NAME, which is also retrieved in the code below.
1405 if ($filename =~ / (?: ^(cpan|lib|ext|dist)\/ )
1406 | $only_for_interior_links_re
1408 $digest->add($contents);
1409 $digests{$filename} = $digest->digest;
1411 # lib files aren't analyzed if they are duplicates of files copied
1412 # there from some other directory. But to determine this, we need
1413 # to know their NAMEs. We might as well find the NAME now while
1414 # the file is open. Similarly, cpan files aren't analyzed unless
1415 # we're analyzing all of them, or this particular file is linked
1416 # to by a file we are analyzing, and thus we will want to verify
1417 # that the target exists in it. We need to know at least the NAME
1418 # to see if it's worth analyzing, or so we can determine if a lib
1419 # file is a copy of a cpan one.
1420 if ($filename =~ m{ (?: ^ (?: cpan | lib ) / )
1421 | $only_for_interior_links_re
1423 if ($contents =~ /^=head1 +NAME.*/mg) {
1424 # The NAME is the first non-spaces on the line up to a
1425 # comma, dash or end of line. Otherwise, it's invalid and
1426 # this pod doesn't have a legal name that we're smart
1427 # enough to find currently. But the parser will later
1428 # find it if it thinks there is a legal name, and set the
1430 if ($contents =~ /\G # continue from the line after =head1
1431 \s* # ignore any empty lines
1432 ^ \s* ( \S+?) \s* (?: [,-] | $ )/mx) {
1434 $checker->name($name);
1435 $id_to_checker{$name} = $checker
1436 if $filename =~ m{^cpan/};
1439 elsif ($filename =~ m{^cpan/}) {
1440 $id_to_checker{$digests{$filename}} = $checker;
1446 } # End of is_pod_file()
1448 # Start of real code that isn't processing the command line (except the
1449 # db is read in above, as is processing of the --add_link option).
1450 # Here, @files contains list of files on the command line. If have any of
1451 # these, unconditionally test them, and show all the errors, even the known
1452 # ones, and, since not testing other pods, don't do cross-pod link tests.
1453 # (Could add extra code to do cross-pod tests for the ones in the list.)
1455 if ($has_input_files) {
1456 undef %known_problems;
1457 $do_upstream_cpan = $do_deltas = 1; # In case one of the inputs is one
1460 else { # No input files -- go find all the possibilities.
1462 $copy_fh = open_new($known_issues);
1463 note("Regenerating $known_issues, please be patient...");
1464 print $copy_fh $HEADER;
1467 # Move to the directory above us, but have to adjust @INC to account for
1469 s{^\.\./lib$}{lib} for @INC;
1470 chdir File::Spec->updir;
1472 # And look in this directory and all its subdirectories
1473 find( {wanted => \&is_pod_file, no_chdir => 1}, '.');
1475 # Add ourselves to the test
1476 push @files, "t/porting/podcheck.t";
1479 # Now we know how many tests there will be.
1480 plan (tests => scalar @files) if ! $regen;
1483 # Sort file names so we get consistent results, and to put cpan last,
1484 # preceded by the ones that we don't generally parse. This is because both
1485 # these classes are generally parsed only if there is a link to the interior
1486 # of them, and we have to parse all others first to guarantee that they don't
1487 # have such a link. 'lib' files come just before these, as some of these are
1488 # duplicates of others. We already have figured this out when gathering the
1489 # data as a special case for all such files, but this, while unnecessary,
1490 # puts the derived file last in the output. 'readme' files come before those,
1491 # as those also could be duplicates of others, which are considered the
1492 # primary ones. These currently aren't figured out when gathering data, so
1494 @files = sort { if ($a =~ /^cpan/) {
1495 return 1 if $b !~ /^cpan/;
1496 return lc $a cmp lc $b;
1498 elsif ($b =~ /^cpan/) {
1501 elsif ($a =~ /$only_for_interior_links_re/) {
1502 return 1 if $b !~ /$only_for_interior_links_re/;
1503 return lc $a cmp lc $b;
1505 elsif ($b =~ /$only_for_interior_links_re/) {
1508 elsif ($a =~ /^lib/) {
1509 return 1 if $b !~ /^lib/;
1510 return lc $a cmp lc $b;
1512 elsif ($b =~ /^lib/) {
1514 } elsif ($a =~ /\breadme\b/i) {
1515 return 1 if $b !~ /\breadme\b/i;
1516 return lc $a cmp lc $b;
1518 elsif ($b =~ /\breadme\b/i) {
1522 return lc $a cmp lc $b;
1527 # Now go through all the files and parse them
1529 foreach my $filename (@files) {
1531 note("parsing $filename") if DEBUG;
1533 # We may have already figured out some things in the process of generating
1534 # the file list. If so, we have a $checker object already. But if not,
1536 my $checker = $filename_to_checker{$filename};
1538 $checker = My::Pod::Checker->new($filename);
1539 $filename_to_checker{$filename} = $checker;
1542 # We have set the name in the checker object if there is a possibility
1543 # that no further parsing is necessary, but otherwise do the parsing now.
1544 if (! $checker->name) {
1545 if (! $checker->parse_from_file($filename, undef)) {
1546 $checker->set_skip("$filename is transitory");
1553 if ($checker->num_errors() < 0) { # Returns negative if not a pod
1554 $checker->set_skip("$filename is not a pod");
1558 # Here, is a pod. See if it is one that has already been tested,
1559 # or should be tested under another directory. Use either its NAME
1560 # if it has one, or a checksum if not.
1561 my $name = $checker->name;
1568 my $digest = Digest->new($digest_type);
1569 my $contents = extract_pod($filename);
1571 # If the return is undef, it means that $filename was a transitory
1573 next FILE unless defined $contents;
1574 $digest->add($contents);
1575 $id = $digest->digest;
1578 # If there is a match for this pod with something that we've already
1579 # processed, don't process it, and output why.
1581 if (defined ($prior_checker = $id_to_checker{$id})
1582 && $prior_checker != $checker) # Could have defined the checker
1583 # earlier without pursuing it
1586 # If the pods are identical, then it's just a copy, and isn't an
1587 # error. First use the checksums we have already computed to see
1588 # if the entire files are identical, which means that the pods are
1590 my $prior_filename = $prior_checker->get_filename;
1592 || ($digests{$prior_filename}
1593 && $digests{$filename}
1594 && $digests{$prior_filename} eq $digests{$filename}));
1596 # If they differ, it could be that the files differ for some
1597 # reason, but the pods they contain are identical. Extract the
1598 # pods and do the comparisons on just those.
1599 if (! $same && $name) {
1600 my $contents = extract_pod($filename);
1602 # If return is <undef>, it means that $filename no longer
1603 # exists. This means it was a transitory file, and should not
1605 next FILE unless defined $contents;
1607 my $prior_contents = extract_pod($prior_filename);
1609 # If return is <undef>, it means that $prior_filename no
1610 # longer exists. This means it was a transitory file, and
1611 # should not have been tested, but we already did process it.
1612 # What we should do now is to back-out its records, and
1613 # process $filename in its stead. But backing out is not so
1614 # simple, and so I'm (khw) skipping that unless and until
1615 # experience shows that it is needed. We do go process
1616 # $filename, and there are potential false positive conflicts
1617 # with the transitory $prior_contents, and rerunning the test
1618 # should cause it to succeed.
1619 goto process_this_pod unless defined $prior_contents;
1621 $same = $prior_contents eq $contents;
1624 use File::Basename 'basename';
1626 $checker->set_skip("The pod of $filename is a duplicate of "
1627 . "the pod for $prior_filename");
1628 } elsif ($prior_filename =~ /\breadme\b/i) {
1629 $checker->set_skip("$prior_filename is a README apparently for $filename");
1630 } elsif ($filename =~ /\breadme\b/i) {
1631 $checker->set_skip("$filename is a README apparently for $prior_filename");
1632 } elsif (! $do_upstream_cpan
1633 && $filename =~ /^cpan/
1634 && $prior_filename =~ /^cpan/)
1636 $checker->set_skip("CPAN is upstream for $filename");
1637 } elsif ( $filename =~ /^utils/ or $prior_filename =~ /^utils/ ) {
1638 $checker->set_skip("$filename copy is in utils/");
1639 } elsif ($prior_filename =~ /^(?:cpan|ext|dist)/
1640 && $filename !~ /^(?:cpan|ext|dist)/
1641 && basename($prior_filename) eq basename($filename))
1643 $checker->set_skip("$filename: Need to run make?");
1644 } else { # Here have two pods with identical names that differ
1645 $prior_checker->poderror(
1646 { -msg => $duplicate_name,
1648 parameter => "'$filename' also has NAME '$name'"
1651 { -msg => $duplicate_name,
1653 parameter => "'$prior_filename' also has NAME '$name'"
1656 # Changing the names helps later.
1657 $prior_checker->name("$name version arbitrarily numbered 1");
1658 $checker->name("$name version arbitrarily numbered 2");
1661 # In any event, don't process this pod that has the same name as
1669 $id_to_checker{$id} = $checker;
1671 my $parsed_for_links = ", but parsed for its interior links";
1672 if ((! $do_upstream_cpan && $filename =~ /^cpan/)
1673 || $filename =~ $only_for_interior_links_re)
1675 if ($filename =~ /^cpan/) {
1676 $checker->set_skip("CPAN is upstream for $filename");
1678 elsif ($filename =~ /perl\d+delta/) {
1680 $checker->set_skip("$filename is a stable perldelta");
1683 elsif ($filename =~ /perltoc/) {
1684 $checker->set_skip("$filename dependent on component pods");
1687 croak("Unexpected file '$filename' encountered that has parsing for interior-linking only");
1690 if ($name && $has_referred_to_node{$name}) {
1691 $checker->set_skip($checker->get_skip() . $parsed_for_links);
1695 # Need a name in order to process it, because not meaningful
1696 # otherwise, and also can't test links to this without a name.
1697 if (!defined $name) {
1698 $checker->poderror( { -msg => $no_name,
1704 # For skipped files, just get its NAME
1706 if (($skip = $checker->get_skip()) && $skip !~ /$parsed_for_links/)
1708 $checker->node($name) if $name;
1711 if (! $checker->parse_from_file($filename, undef)) {
1712 $checker->set_skip("$filename is transitory");
1717 # Go through everything in the file that could be an anchor that
1718 # could be a link target. Count how many there are of the same name.
1719 foreach my $node ($checker->linkable_nodes) {
1720 next FILE if ! $node; # Can be empty is like '=item *'
1721 if (exists $nodes{$name}{$node}) {
1722 $nodes{$name}{$node}++;
1725 $nodes{$name}{$node} = 1;
1728 # Experiments have shown that cpan search can figure out the
1729 # target of a link even if the exact wording is incorrect, as long
1730 # as the first word is. This happens frequently in perlfunc.pod,
1731 # where the link will be just to the function, but the target
1732 # entry also includes parameters to the function.
1733 my $first_word = $node;
1734 if ($first_word =~ s/^(\S+)\s+\S.*/$1/) {
1735 $nodes_first_word{$name}{$first_word} = $node;
1738 $filename_to_pod{$filename} = $name;
1742 # Here, all files have been parsed, and all links and link targets are stored.
1743 # Now go through the files again and see which don't have matches.
1744 if (! $has_input_files) {
1745 foreach my $filename (@files) {
1746 next if $filename_to_checker{$filename}->get_skip;
1747 my $checker = $filename_to_checker{$filename};
1748 foreach my $link ($checker->hyperlink) {
1749 my $linked_to_page = $link->[1]->page;
1750 next unless $linked_to_page; # intra-file checks are handled by std
1753 # Initialize the potential message.
1754 my %problem = ( -msg => $broken_link,
1755 -line => $link->[0],
1756 parameter => "to \"$linked_to_page\"",
1759 # See if we have found the linked-to_file in our parse
1760 if (exists $nodes{$linked_to_page}) {
1761 my $node = $link->[1]->node;
1763 # If link is only to the page-level, already have it
1766 # Transform pod language to what we are expecting
1767 $node =~ s,E<sol>,/,g;
1768 $node =~ s/E<verbar>/|/g;
1770 # If link is to a node that exists in the file, is ok
1771 if ($nodes{$linked_to_page}{$node}) {
1773 # But if the page has multiple targets with the same name,
1774 # it's ambiguous which one this should be to.
1775 if ($nodes{$linked_to_page}{$node} > 1) {
1776 $problem{-msg} = $multiple_targets;
1777 $problem{parameter} = "in $linked_to_page that $node could be pointing to";
1778 $checker->poderror(\%problem);
1780 } elsif (! $nodes_first_word{$linked_to_page}{$node}) {
1782 # Here the link target was not found, either exactly or to
1783 # the first word. Is an error.
1784 $problem{parameter} =~ s,"$,/$node",;
1785 $checker->poderror(\%problem);
1788 } # Linked-to-file not in parse; maybe is in exception list
1789 elsif (! exists $valid_modules{$link->[1]->page}) {
1791 # Here, is a link to a target that we can't find. Check if
1792 # there is an internal link on the page with the target name.
1793 # If so, it could be that they just forgot the initial '/'
1794 # But perldelta is handled specially: only do this if the
1795 # broken link isn't one of the known bad ones (that are
1796 # placemarkers and should be removed for the final)
1797 my $NAME = $filename_to_pod{$filename};
1798 if (! defined $NAME) {
1799 $checker->poderror(\%problem);
1802 if ($nodes{$NAME}{$linked_to_page}) {
1803 $problem{-msg} = $broken_internal_link;
1805 $checker->poderror(\%problem);
1812 # If regenerating the data file, start with the modules for which we don't
1813 # check targets. If you change the sort order, you need to run --regen before
1814 # committing so that future commits that do run regen don't show irrelevant
1817 foreach (sort { lc $a cmp lc $b } keys %valid_modules) {
1818 my_safer_print($copy_fh, $_, "\n");
1822 # Now ready to output the messages.
1823 foreach my $filename (@files) {
1824 my $canonical = canonicalize($filename);
1826 my $skip = $filename_to_checker{$filename}->get_skip // "";
1829 foreach my $message ( sort keys %{$problems{$filename}}) {
1832 # Preserve a negative setting.
1833 if ($known_problems{$canonical}{$message}
1834 && $known_problems{$canonical}{$message} < 0)
1836 $count = $known_problems{$canonical}{$message};
1839 $count = @{$problems{$filename}{$message}};
1841 my_safer_print($copy_fh, $canonical . "\t$message\t$count\n");
1846 skip($skip, 1) if $skip;
1848 my $thankful_diagnostics = 0;
1851 my $total_known = 0;
1852 foreach my $message ( sort keys %{$problems{$filename}}) {
1853 $known_problems{$canonical}{$message} = 0
1854 if ! $known_problems{$canonical}{$message};
1855 my $diagnostic = "";
1856 my $problem_count = scalar @{$problems{$filename}{$message}};
1857 $total_known += $problem_count;
1858 next if $known_problems{$canonical}{$message} < 0;
1859 if ($problem_count > $known_problems{$canonical}{$message}) {
1861 # Here we are about to output all the messages for this type,
1862 # subtract back this number we previously added in.
1863 $total_known -= $problem_count;
1865 $diagnostic .= $indent . qq{"$message"};
1866 if ($problem_count > 2) {
1867 $diagnostic .= " ($problem_count occurrences,"
1868 . " expected $known_problems{$canonical}{$message})";
1870 foreach my $problem (@{$problems{$filename}{$message}}) {
1871 $diagnostic .= " " if $problem_count == 1;
1872 $diagnostic .= "\n$indent$indent";
1873 $diagnostic .= "$problem->{parameter}" if $problem->{parameter};
1874 $diagnostic .= " near line $problem->{-line}";
1875 $diagnostic .= " $problem->{comment}" if $problem->{comment};
1877 $diagnostic .= "\n";
1878 $files_with_unknown_issues{$filename} = 1;
1879 } elsif ($problem_count < $known_problems{$canonical}{$message}) {
1880 $diagnostic = output_thanks($filename, $known_problems{$canonical}{$message}, $problem_count, $message);
1881 $thankful_diagnostics++;
1883 push @diagnostics, $diagnostic if $diagnostic;
1886 # The above loop has output messages where there are current potential
1887 # issues. But it misses where there were some that have been entirely
1888 # fixed. For those, we need to look through the old issues
1889 foreach my $message ( sort keys %{$known_problems{$canonical}}) {
1890 next if $problems{$filename}{$message};
1891 next if ! $known_problems{$canonical}{$message};
1892 next if $known_problems{$canonical}{$message} < 0; # Preserve negs
1894 next if !$pedantic and $message =~
1895 /^(?:\Q$line_length\E|\Q$C_not_linked\E|\Q$C_with_slash\E)/;
1897 my $diagnostic = output_thanks($filename, $known_problems{$canonical}{$message}, 0, $message);
1898 push @diagnostics, $diagnostic if $diagnostic;
1899 $thankful_diagnostics++ if $diagnostic;
1902 my $output = "POD of $filename";
1903 $output .= ", excluding $total_known not shown known potential problems"
1905 if (@diagnostics && @diagnostics == $thankful_diagnostics) {
1906 # Output fixed issues as passing to-do tests, so they do not
1907 # cause failures, but t/harness still flags them.
1908 $output .= " # TODO"
1910 ok(@diagnostics == $thankful_diagnostics, $output);
1912 note(join "", @diagnostics,
1913 "See end of this test output for your options on silencing this");
1916 delete $known_problems{$canonical};
1921 && ! ok (keys %known_problems == 0, "The known problems database includes no references to non-existent files"))
1923 note("The following files were not found: "
1924 . join ", ", keys %known_problems);
1925 note("They will automatically be removed from the db the next time");
1926 note(" cd t; ./perl -I../lib porting/podcheck.t --regen");
1931 run this test script by hand, using the following formula (on
1932 Un*x-like machines):
1934 ./perl -I../lib porting/podcheck.t --regen
1937 if (%files_with_unknown_issues) {
1938 my $were_count_files = scalar keys %files_with_unknown_issues;
1939 $were_count_files = ($were_count_files == 1)
1940 ? "was $were_count_files file"
1941 : "were $were_count_files files";
1942 my $message = <<EOF;
1944 HOW TO GET THIS .t TO PASS
1946 There $were_count_files that had new potential problems identified.
1947 Some of them may be real, and some of them may be false positives because
1948 this program isn't as smart as it likes to think it is. You can teach this
1949 program to ignore the issues it has identified, and hence pass, by doing the
1952 1) If a problem is about a link to an unknown module or man page that
1953 you know exists, re-run the command something like:
1954 ./perl -I../lib porting/podcheck.t --add_link MODULE man_page ...
1955 (MODULEs should look like Foo::Bar, and man_pages should look like
1956 bar(3c); don't do this for a module or man page that you aren't sure
1957 about; instead treat as another type of issue and follow the
1958 instructions below.)
1960 2) For other issues, decide if each should be fixed now or not. Fix the
1961 ones you decided to, and rerun this test to verify that the fixes
1964 3) If there remain false positive or problems that you don't plan to fix right
1967 That should cause all current potential problems to be accepted by
1968 the program, so that the next time it runs, they won't be flagged.
1970 if (%files_with_fixes) {
1971 $message .= " This step will also take care of the files that have fixes in them\n";
1975 For a few files, such as perltoc, certain issues will always be
1976 expected, and more of the same will be added over time. For those,
1977 before you do the regen, you can edit
1979 and find the entry for the module's file and specific error message,
1980 and change the count of known potential problems to -1.
1984 } elsif (%files_with_fixes) {
1986 To teach this test script that the potential problems have been fixed,
1993 chdir $original_dir || die "Can't change directories to $original_dir";
1994 close_and_rename($copy_fh);