8 # Do not require test.pl, this file has its own framework.
13 use feature 'unicode_strings';
24 if ( $Config{usecrosscompile} ) {
25 print "1..0 # Not all files are available during cross-compilation\n";
28 if ($^O eq 'dec_osf') {
29 print "1..0 # $^O cannot handle this test\n";
32 if ( $ENV{'PERL_BUILD_PACKAGING'} ) {
33 print "1..0 # This distro may have modified some files in cpan/. Skipping validation. \n";
36 require '../regen/regen_lib.pl';
45 podcheck.t - Look for possible problems in the Perl pods
50 ./perl -I../lib porting/podcheck.t [--show_all] [--cpan] [--deltas]
51 [--counts] [--pedantic] [FILE ...]
53 ./perl -I../lib porting/podcheck.t --add_link MODULE ...
55 ./perl -I../lib porting/podcheck.t --regen
59 podcheck.t is an extension of Pod::Checker. It looks for pod errors and
60 potential errors in the files given as arguments, or if none specified, in all
61 pods in the distribution workspace, except certain known special ones
62 (specified below). It does additional checking beyond that done by
63 Pod::Checker, and keeps a database of known potential problems, and will
64 fail a pod only if the number of such problems differs from that given in the
67 The additional checks it always makes are:
71 =item Cross-pod link checking
73 Pod::Checker verifies that links to an internal target in a pod are not
74 broken. podcheck.t extends that (when called without FILE arguments) to
75 external links. It does this by gathering up all the possible targets in the
76 workspace, and cross-checking them. It also checks that a non-broken link
77 points to just one target. (The destination pod could have two targets with
80 The way that the C<LE<lt>E<gt>> pod command works (for links outside the pod)
81 is to actually create a link to C<search.cpan.org> with an embedded query for
82 the desired pod or man page. That means that links outside the distribution
83 are valid. podcheck.t doesn't verify the validity of such links, but instead
84 keeps a database of those known to be valid. This means that if a link to a
85 target not on the list is created, the target needs to be added to the data
86 base. This is accomplished via the L<--add_link|/--add_link MODULE ...>
87 option to podcheck.t, described below.
89 =item An internal link that isn't so specified
91 If a link is broken, but there is an existing internal target of the same
92 name, it is likely that the internal target was meant, and the C<"/"> is
93 missing from the C<LE<lt>E<gt>> pod command.
95 =item Missing or duplicate NAME or missing NAME short description
97 A pod can't be linked to unless it has a unique name.
98 And a NAME should have a dash and short description after it.
100 =item Occurrences of the Unicode replacement character
102 L<Pod::Simple> replaces bytes that aren't valid according to the document's
103 encoding (declared or auto-detected) with C<\N{REPLACEMENT CHARACTER}>.
107 If the C<PERL_POD_PEDANTIC> environment variable is set or the C<--pedantic>
108 command line argument is provided then a few more checks are made.
109 The pedantic checks are:
113 =item Verbatim paragraphs that wrap in an 80 (including 1 spare) column window
115 It's annoying to have lines wrap when displaying pod documentation in a
116 terminal window. This checks that all verbatim lines fit in a standard 80
117 column window, even when using a pager that reserves a column for its own use.
118 (Thus the check is for a net of 79 columns.)
119 For those lines that don't fit, it tells you how much needs to be cut in
122 Often, the easiest thing to do to gain space for these is to lower the indent
125 =item Items that perhaps should be links
127 There are mentions of apparent files in the pods that perhaps should be links
128 instead, using C<LE<lt>...E<gt>>
130 =item Items that perhaps should be C<FE<lt>...E<gt>>
132 What look like path names enclosed in C<CE<lt>...E<gt>> should perhaps have
133 C<FE<lt>...E<gt>> mark-up instead.
137 A number of issues raised by podcheck.t and by the base Pod::Checker are not
138 really problems, but merely potential problems, that is, false positives.
139 After inspecting them and
140 deciding that they aren't real problems, it is possible to shut up this program
141 about them, unlike base Pod::Checker. For a valid link to an outside module
142 or man page, call podcheck.t with the C<--add_link> option to add it to the
143 the database of known links; for other causes, call podcheck.t with the C<--regen>
144 option to regenerate the entire database. This tells it that all existing
145 issues are to not be mentioned again.
147 C<--regen> isn't fool-proof. The database merely keeps track of the number of these
148 potential problems of each type for each pod. If a new problem of a given
149 type is introduced into the pod, podcheck.t will spit out all of them. You
150 then have to figure out which is the new one, and should it be changed or not.
151 But doing it this way insulates the database from having to keep track of line
152 numbers of problems, which may change, or the exact wording of each problem
153 which might also change without affecting whether it is a problem or not.
155 Also, if the count of potential problems of a given type for a pod decreases,
156 the database must be regenerated so that it knows the new number. The program
157 gives instructions when this happens.
159 Some pods will have varying numbers of problems of a given type. This can
160 be handled by manually editing the database file (see L</FILES>), and setting
161 the number of those problems for that pod to a negative number. This will
162 cause the corresponding error to always be suppressed no matter how many there
165 Another problem is that there is currently no check that modules listed as
166 valid in the database
167 actually are. Thus any errors introduced there will remain there.
169 =head2 Specially handled pods
175 This pod is generated by pasting bits from other pods. Errors in those bits
176 will show up as errors here, as well as for those other pods. Therefore
177 errors here are suppressed, and the pod is checked only to verify that nodes
178 within it actually exist that are externally linked to.
182 The current perldelta pod is initialized from a template that contains
183 placeholder text. Some of this text is in the form of links that don't really
184 exist. Any such links that are listed in C<@perldelta_ignore_links> will not
185 generate messages. It is presumed that these links will be cleaned up when
186 the perldelta is cleaned up for release since they should be marked with
189 =item Porting/perldelta_template.pod
191 This is not a pod, but a template for C<perldelta>. Any errors introduced
192 here will show up when C<perldelta> is created from it.
194 =item cpan-upstream pods
196 See the L</--cpan> option documentation
200 See the L</--deltas> option documentation
208 =item --add_link MODULE ...
210 Use this option to teach podcheck.t that the C<MODULE>s or man pages actually
211 exist, and to silence any messages that links to them are broken.
213 podcheck.t checks that links within the Perl core distribution are valid, but
214 it doesn't check links to man pages or external modules. When it finds
215 a broken link, it checks its database of external modules and man pages,
216 and only if not found there does it raise a message. This option just adds
217 the list of modules and man page references that follow it on the command line
223 ./perl -I../lib porting/podcheck.t --add_link Unicode::Casing
225 causes the external module "Unicode::Casing" to be added to the database, so
226 C<LE<lt>Unicode::CasingE<gt>> will be considered valid.
230 Regenerate the database used by podcheck.t to include all the existing
231 potential problems. Future runs of the program will not then flag any of
232 these. Setting this option also sets C<--pedantic>.
236 Normally, all pods in the cpan directory are skipped, except to make sure that
237 any blead-upstream links to such pods are valid.
238 This option will cause cpan upstream pods to be fully checked.
242 Normally, all old perldelta pods are skipped, except to make sure that
243 any links to such pods are valid. This is because they are considered
244 stable, and perhaps trying to fix them will cause changes that will
245 misrepresent Perl's history. But, this option will cause them to be fully
250 Normally, if the number of potential problems of a given type found for a
251 pod matches the expected value in the database, they will not be displayed.
252 This option forces the database to be ignored during the run, so all potential
253 problems are displayed and will fail their respective pod test. Specifying
254 any particular FILES to operate on automatically selects this option.
258 Instead of testing, this just dumps the counts of the occurrences of the
259 various types of potential problems in the database.
263 There are three potential problems that are not checked for by default.
264 This options enables them. The environment variable C<PERL_POD_PEDANTIC>
265 can be set to 1 to enable this option also.
266 This option is set when C<--regen> is used.
272 The database is stored in F<t/porting/known_pod_issues.dat>
280 # VMS builds have a '.com' appended to utility and script names, and it adds a
281 # trailing dot for any other file name that doesn't have a dot in it. The db
282 # is stored without those things. This regex allows for these special file
283 # names to be dealt with. It needs to be interpolated into a larger regex
284 # that furnishes the closing boundary.
285 my $vms_re = qr/ \. (?: com )? /x;
287 # Some filenames in the MANIFEST match $vms_re, and so must not be handled the
288 # same way that that the special vms ones are. This hash lists those.
289 my %special_vms_files;
291 # This is to get this to work across multiple file systems, including those
292 # that are not case sensitive. The db is stored in lower case, Un*x style,
293 # and all file name comparisons are done that way.
294 sub canonicalize($) {
296 my ($volume, $directories, $file)
297 = File::Spec->splitpath(File::Spec->canonpath($input));
298 # Assumes $volume is constant for everything in this directory structure
299 $directories = "" if ! $directories;
300 $file = "" if ! $file;
301 $file = lc join '/', File::Spec->splitdir($directories), $file;
302 $file =~ s! / /+ !/!gx; # Multiple slashes => single slash
304 # The db is stored without the special suffixes that are there in VMS, so
305 # strip them off to get the comparable name. But some files on all
306 # platforms have these suffixes, so this shouldn't happen for them, as any
307 # of their db entries will have the suffixes in them. The hash has been
308 # populated with these files.
310 && $file =~ / ( $vms_re ) $ /x
311 && ! exists $special_vms_files{$file})
313 $file =~ s/ $1 $ //x;
318 #####################################################
319 # HOW IT WORKS (in general)
321 # If not called with specific files to check, the directory structure is
322 # examined for files that have pods in them. Files that might not have to be
323 # fully parsed (e.g. in cpan) are parsed enough at this time to find their
324 # pod's NAME, and to get a checksum.
326 # Those kinds of files are sorted last, but otherwise the pods are parsed with
327 # the package coded here, My::Pod::Checker, which is an extension to
328 # Pod::Checker that adds some tests and suppresses others that aren't
329 # appropriate. The latter module has no provision for capturing diagnostics,
330 # so a package, Tie_Array_to_FH, is used to force them to be placed into an
331 # array instead of printed.
333 # Parsing the files builds up a list of links. The files are gone through
334 # again, doing cross-link checking and outputting all saved-up problems with
337 # Sorting the files last that potentially don't need to be fully parsed allows
338 # us to not parse them unless there is a link to an internal anchor in them
339 # from something that we have already parsed. Keeping checksums allows us to
340 # not parse copies of other pods.
342 #####################################################
344 # 1 => Exclude low priority messages that aren't likely to be problems, and
345 # has many false positives; higher numbers give more messages.
346 my $Warnings_Level = 200;
348 # perldelta during construction may have place holder links. N.B. This
349 # variable is referred to by name in release_managers_guide.pod
350 our @perldelta_ignore_links = ( "XXX", "perl5YYYdelta", "perldiag/message" );
352 # To see if two pods with the same NAME are actually copies of the same pod,
353 # which is not an error, it uses a checksum to save work.
354 my $digest_type = "SHA-1";
356 my $original_dir = File::Spec->rel2abs(File::Spec->curdir);
357 my $data_dir = File::Spec->catdir($original_dir, 'porting');
358 my $known_issues = File::Spec->catfile($data_dir, 'known_pod_issues.dat');
359 my $MANIFEST = File::Spec->catfile(File::Spec->updir($original_dir), 'MANIFEST');
362 my $MAX_LINE_LENGTH = 79; # 79 columns
363 my $INDENT = 7; # default nroff indent
365 # Our warning messages. Better not have [('"] in them, as those are used as
366 # delimiters for variable parts of the messages by poderror.
367 my $broken_link = "Apparent broken link";
368 my $broken_internal_link = "Apparent internal link is missing its forward slash";
369 my $multiple_targets = "There is more than one target";
370 my $duplicate_name = "Pod NAME already used";
371 my $no_name = "There is no NAME";
372 my $missing_name_description = "The NAME should have a dash and short description after it";
373 my $replacement_character = "Unicode replacement character found";
374 # the pedantic warnings messages
375 my $line_length = "Verbatim line length including indents exceeds $MAX_LINE_LENGTH by";
376 my $C_not_linked = "? Should you be using L<...> instead of";
377 my $C_with_slash = "? Should you be using F<...> or maybe L<...> instead of";
379 # objects, tests, etc can't be pods, so don't look for them. Also skip
380 # files output by the patch program. Could also ignore most of .gitignore
381 # files, but not all, so don't.
383 my $obj_ext = $Config{'obj_ext'}; $obj_ext =~ tr/.//d; # dot will be added back
384 my $lib_ext = $Config{'lib_ext'}; $lib_ext =~ tr/.//d;
385 my $lib_so = $Config{'so'}; $lib_so =~ tr/.//d;
386 my $dl_ext = $Config{'dlext'}; $dl_ext =~ tr/.//d;
388 # Not really pods, but can look like them.
389 my %excluded_files = (
390 canonicalize("lib/unicore/mktables") => 1,
391 canonicalize("Porting/make-rmg-checklist") => 1,
392 canonicalize("Porting/perldelta_template.pod") => 1,
393 canonicalize("regen/feature.pl") => 1,
394 canonicalize("regen/warnings.pl") => 1,
395 canonicalize("autodoc.pl") => 1,
396 canonicalize("configpm") => 1,
397 canonicalize("miniperl") => 1,
398 canonicalize("perl") => 1,
399 canonicalize('cpan/Pod-Perldoc/corpus/no-head.pod') => 1,
400 canonicalize('cpan/Pod-Perldoc/corpus/perlfunc.pod') => 1,
401 canonicalize('cpan/Pod-Perldoc/corpus/utf8.pod') => 1,
402 canonicalize("lib/unicore/mktables") => 1,
405 # This list should not include anything for which case sensitivity is
406 # important, as it won't work on VMS, and won't show up until tested on VMS.
407 # All or almost all such files should be listed in the MANIFEST, so that can
408 # be examined for them, and each such file explicitly excluded, as is done for
409 # .PL files in the loop just below this. For files not catchable this way,
410 # is_pod_file() can be used to exclude these at a finer grained level.
411 my $non_pods = qr/ (?: \.
412 (?: [achot] | zip | gz | bz2 | jar | tar | tgz
413 | orig | rej | patch # Patch program output
414 | sw[op] | \#.* # Editor droppings
415 | old # buildtoc output
416 | xs # pod should be in the .pm file
417 | al # autosplit files
418 | bs # bootstrap files
419 | (?i:sh) # shell scripts, hints, templates
420 | lst # assorted listing files
421 | bat # Windows,Netware,OS2 batch files
422 | cmd # Windows,Netware,OS2 command files
423 | lis # VMS compiler listings
424 | map # VMS linker maps
425 | opt # VMS linker options files
426 | mms # MM(K|S) description files
427 | ts # timestamp files generated during build
428 | $obj_ext # object files
429 | exe # $Config{'exe_ext'} might be empty string
430 | $lib_ext # object libraries
431 | $lib_so # shared libraries
432 | $dl_ext # dynamic libraries
433 | gif # GIF images (example files from CGI.pm)
434 | eg # examples from libnet
438 ) | ~$ | \ \(Autosaved\)\.txt$ # Other editor droppings
439 | ^cxx\$demangler_db\.$ # VMS name mangler database
440 | ^typemap\.?$ # typemap files
441 | ^(?i:Makefile\.PL)$
444 # Matches something that looks like a file name, but is enclosed in C<...>
445 my $C_path_re = qr{ ^
446 # exclude various things that have slashes
447 # in them but aren't paths
449 (?: (?: s | qr | m | tr | y ) / ) # regexes
450 | \d+/\d+ \b # probable fractions
458 /? # Optional initial slash
459 \w+ # First component of path, doesn't begin with
461 (?: / [-\w]+ )+ # Subsequent path components
462 (?: \. \w+ )? # Optional trailing dot and suffix
463 >* # Any enclosed L< F< have matching closing >
467 # '.PL' files should be excluded, as they aren't final pods, but often contain
468 # material used in generating pods, and so can look like a pod. We can't use
469 # the regexp above because case sensisitivity is important for these, as some
470 # '.pl' files should be examined for pods. Instead look through the MANIFEST
471 # for .PL files and get their full path names, so we can exclude each such
472 # file explicitly. This works because other porting tests prohibit having two
473 # files with the same names except for case.
474 open my $manifest_fh, '<:bytes', $MANIFEST or die "Can't open $MANIFEST";
475 while (<$manifest_fh>) {
477 # While we have MANIFEST open, on VMS platforms, look for files that match
478 # the magic VMS file names that have to be handled specially. Add these
479 # to the list of them.
480 if ($^O eq 'VMS' && / ^ ( [^\t]* $vms_re ) \t /x) {
481 $special_vms_files{$1} = 1;
483 if (/ ^ ( [^\t]* \. PL ) \t /x) {
484 $excluded_files{canonicalize($1)} = 1;
487 close $manifest_fh, or die "Can't close $MANIFEST";
490 # Pod::Checker messages to suppress
491 my @suppressed_messages = (
492 # We catch independently the ones that are real problems.
493 qr/multiple occurrences \(\d+\) of link target/,
495 "unescaped <>", # Not every '<' or '>' need be escaped
496 qr/No items in =over/, # i.e., a blockquote, which we consider legal
500 # Returns bool as to if input message is one that is to be suppressed
504 return grep { $message =~ /^$_/i } @suppressed_messages;
507 { # Closure to contain a simple subset of test.pl. This is to get rid of the
508 # unnecessary 'failed at' messages that would otherwise be output pointing
509 # to a particular line in this file.
511 my $current_test = 0;
516 $planned = $plan{tests} + 1; # +1 for final test that files haven't
518 print "1..$planned\n";
529 print "not " unless $success;
530 print "ok $current_test - $message\n";
536 my $n = @_ ? shift : 1;
539 print "ok $current_test # skip $why\n";
541 no warnings 'exiting';
546 my ($andle, $message) = @_;
550 print $andle $message =~ s/^/# /mgr;
555 sub note { unshift @_, \*STDOUT; goto &_note }
557 sub diag { unshift @_, \*STDERR; goto &_note }
560 if ($planned && $planned != $current_test) {
562 "# Looks like you planned $planned tests but ran $current_test.\n";
567 # List of known potential problems by pod and type.
570 # Pods given by the keys contain an interior node that is referred to from
572 my %has_referred_to_node;
580 my $do_upstream_cpan = 0; # Assume that are to skip anything in /cpan
581 my $do_deltas = 0; # And stable perldeltas
583 while (@ARGV && substr($ARGV[0], 0, 1) eq '-') {
584 my $arg = shift @ARGV;
586 $arg =~ s/^--/-/; # Treat '--' the same as a single '-'
587 if ($arg eq '-regen') {
591 elsif ($arg eq '-add_link') {
594 elsif ($arg eq '-cpan') {
595 $do_upstream_cpan = 1;
597 elsif ($arg eq '-deltas') {
600 elsif ($arg eq '-show_all') {
603 elsif ($arg eq '-counts') {
606 elsif ($arg eq '-pedantic') {
611 Unknown option '$arg'
613 Usage: $0 [ --regen | --cpan | --show_all | FILE ... | --add_link MODULE ... ]\n"
614 --add_link -> Add the MODULE and man page references to the database
615 --regen -> Regenerate the data file for $0
616 --cpan -> Include files in the cpan subdirectory.
617 --deltas -> Include stable perldeltas
618 --show_all -> Show all known potential problems
619 --counts -> Don't test, but give summary counts of the currently
621 --pedantic -> Check for overly long lines in verbatim blocks
626 $pedantic = 1 if exists $ENV{PERL_POD_PEDANTIC} and $ENV{PERL_POD_PEDANTIC};
629 my $cpan_or_deltas = $do_upstream_cpan || $do_deltas;
630 if (($regen + $show_all + $show_counts + $add_link + $cpan_or_deltas ) > 1) {
631 croak "--regen, --show_all, --counts, and --add_link are mutually exclusive\n and none can be run with --cpan nor --deltas";
634 my $has_input_files = @files;
638 if (! $has_input_files) {
639 croak "--add_link requires at least one module or man page reference";
642 elsif ($has_input_files) {
643 if ($regen || $show_counts || $do_upstream_cpan || $do_deltas) {
644 croak "--regen, --counts, --deltas, and --cpan can't be used since using specific files";
646 foreach my $file (@files) {
647 croak "Can't read file '$file'" if ! -r $file;
651 our %problems; # potential problems found in this run
653 package My::Pod::Checker { # Extend Pod::Checker
654 use parent 'Pod::Checker';
656 # Uses inside out hash to protect from typos
657 # For new fields, remember to add to destructor DESTROY()
658 my %CFL_text; # The text comprising the current C<>, F<>, or L<>
659 my %C_text; # If defined, are in a C<> section, and includes
660 # the accumulated text from that
661 my %current_indent; # Current line's indent
662 my %filename; # The pod is store in this file
663 my %in_CFL; # count of stacked C<>, F<>, L<> directives
664 my %indents; # Stack of indents from =over's in effect for
666 my %in_for; # true if in a =for or =begin
667 my %in_NAME; # true if within NAME section
668 my %in_begin; # true if within =begin section
669 my %in_X; # true if in a X<>
670 my %linkable_item; # Bool: if the latest =item is linkable. It isn't
671 # for bullet and number lists
672 my %linkable_nodes; # Pod::Checker adds all =items to its node list,
673 # but not all =items are linkable to
674 my %running_CFL_text; # The current text that is being accumulated until
675 # an end_FOO is found, and this includes any C<>,
676 # F<>, or L<> directives.
677 my %running_simple_text; # The currentt text that is being accumulated
678 # until an end_FOO is found, and all directives
679 # have been expanded into plain text
680 my %command_count; # Number of commands seen
681 my %seen_pod_cmd; # true if have =pod earlier
682 my %skip; # is SKIP set for this pod
683 my %start_line; # the first input line number in the the thing
684 # currently being worked on
687 my $addr = Scalar::Util::refaddr $_[0];
688 delete $CFL_text{$addr};
689 delete $C_text{$addr};
690 delete $command_count{$addr};
691 delete $current_indent{$addr};
692 delete $filename{$addr};
693 delete $in_begin{$addr};
694 delete $in_CFL{$addr};
695 delete $indents{$addr};
696 delete $in_for{$addr};
697 delete $in_NAME{$addr};
699 delete $linkable_item{$addr};
700 delete $linkable_nodes{$addr};
701 delete $running_CFL_text{$addr};
702 delete $running_simple_text{$addr};
703 delete $seen_pod_cmd{$addr};
705 delete $start_line{$addr};
711 my $filename = shift;
713 my $self = $class->SUPER::new(-quiet => 1,
714 -warnings => $Warnings_Level);
715 my $addr = Scalar::Util::refaddr $self;
716 $command_count{$addr} = 0;
717 $current_indent{$addr} = 0;
718 $filename{$addr} = $filename;
719 $in_begin{$addr} = 0;
723 $linkable_item{$addr} = 0;
724 $seen_pod_cmd{$addr} = 0;
728 # re's for messages that Pod::Checker outputs
729 my $location = qr/ \b (?:in|at|on|near) \s+ /xi;
730 my $optional_location = qr/ (?: $location )? /xi;
731 my $line_reference = qr/ [('"]? $optional_location \b line \s+
732 (?: \d+ | EOF | \Q???\E | - )
735 sub poderror { # Called to register a potential problem
737 # This adds an extra field to the parent hash, 'parameter'. It is
738 # used to extract the variable parts of a message leaving just the
739 # constant skeleton. This in turn allows the message to be
740 # categorized better, so that it shows up as a single type in our
741 # database, with the specifics of each occurrence not being stored with
747 my $addr = Scalar::Util::refaddr $self;
748 return if $skip{$addr};
750 # Input can be a string or hash. If a string, parse it to separate
751 # out the line number and convert to a hash for easier further
754 if (ref $opts ne 'HASH') {
755 $message = join "", $opts, @_;
757 if ($message =~ s/\s*($line_reference)//) {
758 ($line_number = $1) =~ s/\s*$optional_location//;
761 $line_number = '???';
763 $opts = { -msg => $message, -line => $line_number };
765 $message = $opts->{'-msg'};
769 $message =~ s/^\d+\s+//;
770 return if main::suppressed($message);
772 $self->SUPER::poderror($opts, @_);
774 $opts->{parameter} = "" unless $opts->{parameter};
776 # The variable parts of the message tend to be enclosed in '...',
777 # "....", or (...). Extract them and put them in an extra field,
778 # 'parameter'. This is trickier because the matching delimiter to a
779 # '(' is its mirror, and not itself. Text::Balanced could be used
781 while ($message =~ m/ \s* $optional_location ( [('"] )/xg) {
784 $delimiter = ')' if $delimiter eq '(';
786 # If there is no ending delimiter, don't consider it to be a
787 # variable part. Most likely it is a contraction like "Don't"
788 last unless $message =~ m/\G .+? \Q$delimiter/xg;
790 my $length = $+[0] - $start;
792 # Get the part up through the closing delimiter
793 my $special = substr($message, $start, $length);
794 $special =~ s/^\s+//; # No leading whitespace
796 # And add that variable part to the parameter, while removing it
797 # from the message. This isn't a foolproof way of finding the
798 # variable part. For example '(s)' can occur in e.g.,
800 if ($special ne '(s)') {
801 substr($message, $start, $length) = "";
802 pos $message = $start;
803 $opts->{-msg} = $message;
804 $opts->{parameter} .= " " if $opts->{parameter};
805 $opts->{parameter} .= $special;
809 # Extract any additional line number given. This is often the
810 # beginning location of something whereas the main line number gives
812 if ($message =~ /( $line_reference )/xi) {
814 while ($message =~ s/\s*\Q$line_ref//) {
815 $opts->{-msg} = $message;
816 $opts->{parameter} .= " " if $opts->{parameter};
817 $opts->{parameter} .= $line_ref;
821 Carp::carp("Couldn't extract line number from '$message'") if $message =~ /line \d+/;
822 push @{$problems{$filename{$addr}}{$message}}, $opts;
823 #push @{$problems{$self->get_filename}{$message}}, $opts;
826 # In the next subroutines, we keep track of the text of the current
827 # innermost thing, like F<fooC<bar>baz>. The things we care about raising
828 # messages about in this program all come from a single sequence of
829 # characters uninterrupted by other pod commands. Therefore we don't have
830 # to worry about recursion, and we can just set the string we care about
831 # to empty on entrance to each command.
834 # This is called by the parent class to deal with any straight text.
835 # We mostly just append this to the running current value which will
836 # be dealt with upon the end of the current construct, like a
837 # paragraph. But certain things don't contribute to checking the pod
838 # and are ignored. We also have set flags to indicate this text is
839 # going towards constructing certain constructs, and handle those
843 my $addr = Scalar::Util::refaddr $self;
845 my $return = $self->SUPER::handle_text(@_);
847 if ($in_X{$addr} || $in_for{$addr}) { # ignore
851 my $text = join "\n", @_;
852 $running_simple_text{$addr} .= $text;
854 # Keep separate tabs on C<>, F<>, and L<> directives, and one
855 # especially for C<> ones.
856 if ($in_CFL{$addr}) {
857 $CFL_text{$addr} .= $text;
858 $C_text{$addr} .= $text if defined $C_text{$addr};
861 # This variable is updated instead in the corresponding C, F, or L
863 $running_CFL_text{$addr} .= $text;
866 # do this line-by-line so we can get the right line number
867 my @lines = split /^/, $running_simple_text{$addr};
868 for my $i (0..$#lines) {
869 if ($lines[$i] =~ m/\N{REPLACEMENT CHARACTER}/) {
870 $self->poderror({ -line => $start_line{$addr} + $i,
871 -msg => $replacement_character,
872 parameter => "possibly invalid ". $self->encoding . " input at character " . pos $lines[$i],
879 # The start_FOO routines check that somehow a C<> construct hasn't escaped
880 # without being checked, and initialize things, and call the parent
881 # class's equivalent routine.
883 # The end_FOO routines close things off, and check the text that has been
884 # accumulated for FOO, then call the parent's corresponding routine.
888 check_see_but_not_link($self);
890 my $addr = Scalar::Util::refaddr $self;
891 $start_line{$addr} = $_[0]->{start_line};
892 $running_CFL_text{$addr} = "";
893 $running_simple_text{$addr} = "";
894 return $self->SUPER::start_Para(@_);
897 sub start_item_text {
899 check_see_but_not_link($self);
901 my $addr = Scalar::Util::refaddr $self;
902 $start_line{$addr} = $_[0]->{start_line};
903 $running_CFL_text{$addr} = "";
904 $running_simple_text{$addr} = "";
906 # This is the only =item that is linkable
907 $linkable_item{$addr} = 1;
909 return $self->SUPER::start_item_text(@_);
912 sub start_item_number {
914 check_see_but_not_link($self);
916 my $addr = Scalar::Util::refaddr $self;
917 $start_line{$addr} = $_[0]->{start_line};
918 $running_CFL_text{$addr} = "";
919 $running_simple_text{$addr} = "";
921 return $self->SUPER::start_item_number(@_);
924 sub start_item_bullet {
926 check_see_but_not_link($self);
928 my $addr = Scalar::Util::refaddr $self;
929 $start_line{$addr} = $_[0]->{start_line};
930 $running_CFL_text{$addr} = "";
931 $running_simple_text{$addr} = "";
933 return $self->SUPER::start_item_bullet(@_);
936 sub end_item { # No difference in =item types endings
938 check_see_but_not_link($self);
939 return $self->SUPER::end_item(@_);
944 check_see_but_not_link($self);
946 my $addr = Scalar::Util::refaddr $self;
947 $start_line{$addr} = $_[0]->{start_line};
948 $running_CFL_text{$addr} = "";
949 $running_simple_text{$addr} = "";
951 # Save this indent on a stack, and keep track of total indent
952 my $indent = $_[0]{'indent'};
953 push @{$indents{$addr}}, $indent;
954 $current_indent{$addr} += $indent;
956 return $self->SUPER::start_over(@_);
959 sub end_over_bullet { shift->end_over(@_) }
960 sub end_over_number { shift->end_over(@_) }
961 sub end_over_text { shift->end_over(@_) }
962 sub end_over_block { shift->end_over(@_) }
963 sub end_over_empty { shift->end_over(@_) }
966 check_see_but_not_link($self);
968 my $addr = Scalar::Util::refaddr $self;
971 if (@{$indents{$addr}}) {
972 $current_indent{$addr} -= pop @{$indents{$addr}};
975 # =back without corresponding =over, but should have
977 $current_indent{$addr} = 0;
981 sub check_see_but_not_link {
983 # Looks through accumulated text for current element that includes the
984 # C<>, F<>, and L<> directives for ones that look like they are
985 # C<link> instead of L<link>.
988 my $addr = Scalar::Util::refaddr $self;
990 return unless defined $running_CFL_text{$addr};
992 while ($running_CFL_text{$addr} =~ m{
993 ( (?: \w+ \s+ )* ) # The phrase before, if any
997 ( [^<]*? ) # The not < excludes nested C<L<...
1000 ( \s+ (?: under | in ) \s+ L< )?
1003 my $prefix = $1 // "";
1004 my $construct = $2; # The whole thing, like C<...>
1007 my $trailing = $5; # After the whole thing ending in "L<"
1009 # If the full phrase is something like, "you might see C<", or
1010 # similar, it really isn't a reference to a link. The ones I saw
1011 # all had the word "you" in them; and the "you" wasn't the
1012 # beginning of a sentence.
1013 if ($prefix !~ / \b you \b /x) {
1015 # Now, find what the module or man page name within the
1016 # construct would be if it actually has L<> syntax. If it
1017 # doesn't have that syntax, will set the module to the entire
1019 if (! defined $trailing # not referring to something in another
1021 && $interior !~ /$non_pods/
1023 # There can't be spaces (I think) in module names or man
1025 && $interior !~ / \s /x
1027 # F<> that end in eg \.pl are almost certainly ok, as are
1028 # those that look like a path with multiple "/" chars
1031 && $interior !~ /\.\w+$/
1032 && $interior !~ /\/.+\//)
1035 # TODO: move the checking of $pedantic higher up
1036 $self->poderror({ -line => $start_line{$addr},
1037 -msg => $C_not_linked,
1038 parameter => $construct
1044 undef $running_CFL_text{$addr};
1049 check_see_but_not_link($self);
1051 my $addr = Scalar::Util::refaddr $self;
1052 if ($in_NAME{$addr}) {
1053 if ($running_simple_text{$addr} =~ /^\s*(\S+?)\s*$/) {
1054 $self->poderror({ -line => $start_line{$addr},
1055 -msg => $missing_name_description,
1058 $in_NAME{$addr} = 0;
1060 $self->SUPER::end_Para(@_);
1065 check_see_but_not_link($self);
1067 my $addr = Scalar::Util::refaddr $self;
1068 $start_line{$addr} = $_[0]->{start_line};
1069 $running_CFL_text{$addr} = "";
1070 $running_simple_text{$addr} = "";
1072 return $self->SUPER::start_head1(@_);
1075 sub end_head1 { # This is called at the end of the =head line.
1077 check_see_but_not_link($self);
1079 my $addr = Scalar::Util::refaddr $self;
1081 $in_NAME{$addr} = 1 if $running_simple_text{$addr} eq 'NAME';
1082 return $self->SUPER::end_head(@_);
1085 sub start_Verbatim {
1087 check_see_but_not_link($self);
1089 my $addr = Scalar::Util::refaddr $self;
1090 $running_simple_text{$addr} = "";
1091 $start_line{$addr} = $_[0]->{start_line};
1092 return $self->SUPER::start_Verbatim(@_);
1097 my $addr = Scalar::Util::refaddr $self;
1099 # Pick up the name if it looks like one, since the parent class
1100 # doesn't handle verbatim NAMEs
1102 && $running_simple_text{$addr} =~ /^\s*(\S+?)\s*[,-]/)
1107 my $indent = $self->get_current_indent;
1109 # Look at each line to verify it is short enough
1110 my @lines = split /^/, $running_simple_text{$addr};
1111 for my $i (0 .. @lines - 1) {
1112 $lines[$i] =~ s/\s+$//;
1113 my $exceeds = length(Text::Tabs::expand($lines[$i]))
1114 + $indent - $MAX_LINE_LENGTH;
1115 next unless $exceeds > 0;
1117 $self->poderror({ -line => $start_line{$addr} + $i,
1118 -msg => $line_length,
1119 parameter => "+$exceeds (including " . ($indent - $INDENT) . " from =over's)",
1123 undef $running_simple_text{$addr};
1125 # Parent class didn't bother to define this
1126 #return $self->SUPER::SUPER::end_Verbatim(@_);
1131 my $addr = Scalar::Util::refaddr $self;
1133 $C_text{$addr} = "";
1135 # If not in a stacked set of C<>, F<> and L<>, initialize the text for
1137 $CFL_text{$addr} = "" if ! $in_CFL{$addr};
1140 return $self->SUPER::start_C(@_);
1145 my $addr = Scalar::Util::refaddr $self;
1147 $CFL_text{$addr} = "" if ! $in_CFL{$addr};
1149 return $self->SUPER::start_F(@_);
1154 my $addr = Scalar::Util::refaddr $self;
1156 $CFL_text{$addr} = "" if ! $in_CFL{$addr};
1158 return $self->SUPER::start_L(@_);
1163 my $addr = Scalar::Util::refaddr $self;
1165 # Warn if looks like a file or link enclosed instead by this C<>
1166 if ($C_text{$addr} =~ qr/^ $C_path_re $/x) {
1167 # Here it does look like it could be be a file path or a link.
1168 # But some varieties of regex patterns could also fit with what we
1169 # have so far. Weed those out as best we can. '/foo/' is almost
1170 # certainly meant to be a pattern, as is '/foo/g'.
1172 if ($C_text{$addr} !~ qr| ^ / [^/]* / ( [msixpodualngcr]* ) $ |x) {
1177 # Here, it looks like a pattern potentially followed by some
1178 # modifiers. To make doubly sure, don't count as patterns
1179 # those constructs which have more occurrences (generally 1)
1180 # of a modifier than is legal.
1182 map { $counts{$_}++ } split "", $1;
1183 foreach my $modifier (keys %counts) {
1184 if ($counts{$modifier} > (($modifier eq 'a')
1192 $is_pattern = 1 unless defined $is_pattern;
1195 unless ($is_pattern) {
1196 $self->poderror({ -line => $start_line{$addr},
1197 -msg => $C_with_slash,
1198 parameter => "C<$C_text{$addr}>"
1202 undef $C_text{$addr};
1204 # Add the current text to the running total. This was not done in
1205 # handle_text(), because it just sees the plain text of the innermost
1206 # stacked directive. We want to keep all the directive names
1207 # enclosing the text. Otherwise the fact that C<L<foobar>> is to a
1208 # link would be lost, as the L<> would be gone.
1209 $CFL_text{$addr} = "C<$CFL_text{$addr}>";
1211 # Add this text to the the whole running total only if popping this
1212 # directive off the stack leaves it empty. As long as something is on
1213 # the stack, it gets added to $CFL_text (just above). It is only
1214 # entirely constructed when the stack is empty.
1216 $running_CFL_text{$addr} .= $CFL_text{$addr} if ! $in_CFL{$addr};
1218 return $self->SUPER::end_C(@_);
1223 my $addr = Scalar::Util::refaddr $self;
1225 $CFL_text{$addr} = "F<$CFL_text{$addr}>";
1227 $running_CFL_text{$addr} .= $CFL_text{$addr} if ! $in_CFL{$addr};
1228 return $self->SUPER::end_F(@_);
1233 my $addr = Scalar::Util::refaddr $self;
1235 $CFL_text{$addr} = "L<$CFL_text{$addr}>";
1237 $running_CFL_text{$addr} .= $CFL_text{$addr} if ! $in_CFL{$addr};
1238 return $self->SUPER::end_L(@_);
1243 my $addr = Scalar::Util::refaddr $self;
1246 return $self->SUPER::start_X(@_);
1251 my $addr = Scalar::Util::refaddr $self;
1254 return $self->SUPER::end_X(@_);
1259 my $addr = Scalar::Util::refaddr $self;
1262 return $self->SUPER::start_for(@_);
1267 my $addr = Scalar::Util::refaddr $self;
1270 return $self->SUPER::end_for(@_);
1274 my ($self, $link) = @_;
1276 if ($link && $link->type eq 'pod') {
1277 my $page = $link->page;
1278 my $node = $link->node;
1280 # If the hyperlink is to an interior node of another page, save it
1281 # so that we can see if we need to parse normally skipped files.
1282 $has_referred_to_node{$page} = 1 if $node;
1284 # Ignore certain placeholder links in perldelta. Check if the
1285 # link is page-level, and also check if to a node within the page
1286 if ( $self->name && $self->name eq "perldelta"
1287 && (( grep { $page eq $_ } @perldelta_ignore_links)
1289 && (grep { "$page/$node" eq $_ } @perldelta_ignore_links)
1295 return $self->SUPER::hyperlink($link);
1302 $text =~ s/\s+$//s; # strip trailing whitespace
1303 $text =~ s/\s+/ /gs; # collapse whitespace
1304 my $addr = Scalar::Util::refaddr $self;
1305 push(@{$linkable_nodes{$addr}}, $text) if
1306 ! $current_indent{$addr}
1307 || $linkable_item{$addr};
1309 return $self->SUPER::node($_[0]);
1312 sub get_current_indent {
1313 return $INDENT + $current_indent{Scalar::Util::refaddr $_[0]};
1317 return $filename{Scalar::Util::refaddr $_[0]};
1320 sub linkable_nodes {
1321 my $linkables = $linkable_nodes{Scalar::Util::refaddr $_[0]};
1322 return undef unless $linkables;
1327 return $skip{Scalar::Util::refaddr $_[0]} // 0;
1332 $skip{Scalar::Util::refaddr $self} = shift;
1334 # If skipping, no need to keep the problems for it
1335 delete $problems{$self->get_filename};
1339 sub parse_from_file {
1340 # This overrides the super class method so that if an open fails on a
1341 # transitory file, it doesn't croak. It returns 1 if it did find the
1342 # file, 0 if it didn't
1345 my $filename = shift;
1346 # ignores 2nd param, which is output file. Always uses undef
1348 if (open my $in_fh, '<:bytes', $filename) {
1349 $self->SUPER::parse_from_file($in_fh, undef);
1354 # If couldn't open file, perhaps it was transitory, and hence not an error
1355 return 0 unless -e $filename;
1357 die "Can't open '$filename': $!\n";
1361 package Tie_Array_to_FH { # So printing actually goes to an array
1367 my $array_ref = shift;
1369 my $self = bless \do{ my $anonymous_scalar }, $class;
1370 $array{Scalar::Util::refaddr $self} = $array_ref;
1377 push @{$array{Scalar::Util::refaddr $self}}, @_;
1383 my %filename_to_checker; # Map a filename to its pod checker object
1384 my %id_to_checker; # Map a checksum to its pod checker object
1385 my %nodes; # key is filename, values are nodes in that file.
1386 my %nodes_first_word; # same, but value is first word of each node
1387 my %valid_modules; # List of modules known to exist outside us.
1388 my %digests; # checksums of files, whose names are the keys
1389 my %filename_to_pod; # Map a filename to its pod NAME
1390 my %files_with_unknown_issues;
1391 my %files_with_fixes;
1394 open $data_fh, '<:bytes', $known_issues or die "Can't open $known_issues";
1396 my %counts; # For --counts param, count of each issue type
1397 my %suppressed_files; # Files with at least one issue type to suppress
1399 # This file is the data file for $0.
1400 # There are three types of lines.
1401 # Comment lines are white-space only or begin with a '#', like this one. Any
1402 # changes you make to the comment lines will be lost when the file is
1404 # Lines without tab characters are simply NAMES of pods that the program knows
1405 # will have links to them and the program does not check if those links are
1407 # All other lines should have three fields, each separated by a tab. The
1408 # first field is the name of a pod; the second field is an error message
1409 # generated by this program; and the third field is a count of how many
1410 # known instances of that message there are in the pod. -1 means that the
1411 # program can expect any number of this type of message.
1414 my @existing_issues;
1417 while (<$data_fh>) { # Read the database
1419 next if /^\s*(?:#|$)/; # Skip comment and empty lines
1422 if ($add_link) { # The issues are saved and later output unchanged
1423 push @existing_issues, $_;
1427 # Keep track of counts of each issue type for each file
1428 my ($filename, $message, $count) = split /\t/;
1429 $known_problems{$filename}{$message} = $count;
1432 if ($count < 0) { # -1 means to suppress this issue type
1433 $suppressed_files{$filename} = $filename;
1436 $counts{$message} += $count;
1440 else { # Lines without a tab are modules known to be valid
1441 $valid_modules{$_} = 1
1447 $copy_fh = open_new($known_issues);
1449 # Check for basic sanity, and add each command line argument
1450 foreach my $module (@files) {
1451 die "\"$module\" does not look like a module or man page"
1452 # Must look like (A or A::B or A::B::C ..., or foo(3C)
1453 if $module !~ /^ (?: \w+ (?: :: \w+ )* | \w+ \( \d \w* \) ) $/x;
1454 $valid_modules{$module} = 1
1456 my_safer_print($copy_fh, $HEADER);
1457 foreach (sort { lc $a cmp lc $b } keys %valid_modules) {
1458 my_safer_print($copy_fh, $_, "\n");
1461 # The rest of the db file is output unchanged.
1462 my_safer_print($copy_fh, join "\n", @existing_issues, "");
1464 close_and_rename($copy_fh);
1470 foreach my $message (sort keys %counts) {
1471 $total += $counts{$message};
1472 note(Text::Tabs::expand("$counts{$message}\t$message"));
1474 note("-----\n" . Text::Tabs::expand("$total\tknown potential issues"));
1475 if (%suppressed_files) {
1476 note("\nFiles that have all messages of at least one type suppressed:");
1477 note(join ",", keys %suppressed_files);
1482 # re to match files that are to be parsed only if there is an internal link
1483 # to them. It does not include cpan, as whether those are parsed depends
1484 # on a switch. Currently, only perltoc and the stable perldelta.pod's
1485 # are included. The latter all have characters between 'perl' and
1486 # 'delta'. (Actually the currently developed one matches as well, but
1487 # is a duplicate of perldelta.pod, so can be skipped, so fine for it to
1489 my $only_for_interior_links_re = qr/ ^ pod\/perltoc.pod $
1491 unless ($do_deltas) {
1492 $only_for_interior_links_re = qr/$only_for_interior_links_re |
1493 \b perl \d+ delta \. pod \b
1500 sub output_thanks ($$$$) { # Called when an issue has been fixed
1501 my $filename = shift;
1502 my $original_count = shift;
1503 my $current_count = shift;
1504 my $message = shift;
1506 $files_with_fixes{$filename} = 1;
1508 my $fixed_count = $original_count - $current_count;
1509 my $a_problem = ($fixed_count == 1) ? "a problem" : "multiple problems";
1510 my $another_problem = ($fixed_count == 1) ? "another problem" : "another set of problems";
1514 There were $original_count occurrences (now $current_count) in this pod of type
1519 There are no longer any problems found in this pod!
1526 Thanks for fixing $a_problem!
1528 Now you must teach $0 that this was fixed.
1533 Thanks for fixing $another_problem.
1542 sub my_safer_print { # print, with error checking for outputting to db
1543 my ($fh, @lines) = @_;
1545 if (! print $fh @lines) {
1546 my $save_error = $!;
1548 die "Write failure: $save_error";
1552 sub extract_pod { # Extracts just the pod from a file; returns undef if file
1554 my $filename = shift;
1559 # Arrange for the output of Pod::Parser to be collected in an array we can
1560 # look at instead of being printed
1561 tie *ALREADY_FH, 'Tie_Array_to_FH', \@pod;
1562 if (open my $in_fh, '<:bytes', $filename) {
1563 my $parser = Pod::Parser->new();
1564 $parser->parse_from_filehandle($in_fh, *ALREADY_FH);
1567 return join "", @pod
1570 # The file should already have been opened once to get here, so if that
1571 # fails, something is wrong. It's possible that a transitory file
1572 # containing a pod would get here, so if the file no longer exists just
1574 return unless -e $filename;
1575 die "Can't open '$filename': $!\n";
1578 my $digest = Digest->new($digest_type);
1580 # This is used as a callback from File::Find::find(), which always constructs
1581 # pathnames using Unix separators
1583 # If $_ is a pod file, add it to the lists and do other prep work.
1586 # Don't look at files in directories that are for tests, nor those
1587 # beginning with a dot
1588 if (m!/t\z! || m!/\.!) {
1589 $File::Find::prune = 1;
1594 return unless -r && -s; # Can't check it if can't read it; no need to
1596 return unless -f || -l; # Weird file types won't be pods
1598 my ($leaf) = m!([^/]+)\z!;
1599 if (m!/\.! # No hidden Unix files
1600 || $leaf =~ $non_pods) {
1601 note("Not considering $_") if DEBUG;
1605 my $filename = $File::Find::name;
1607 # $filename is relative, like './path'. Strip that initial part away.
1608 $filename =~ s!^\./!! or die 'Unexpected pathname "$filename"';
1610 return if $excluded_files{canonicalize($filename)};
1615 if (! open $candidate, '<:bytes', $_) {
1617 # If a transitory file was found earlier, the open could fail
1618 # legitimately and we just skip the file; also skip it if it is a
1619 # broken symbolic link, as it is probably just a build problem;
1620 # certainly not a file that we would want to check the pod of.
1621 # Otherwise fail it here and no reason to process it further.
1622 # (But the test count will be off too)
1623 ok(0, "Can't open '$filename': $!")
1624 if -r $filename && ! -l $filename;
1630 # If the file is a .pm or .pod, having any initial '=' on a line is
1631 # grounds for testing it. Otherwise, require a head1 NAME line to
1632 # consider it as a potential pod
1633 if ($filename =~ /\.(?:pm|pod)/) {
1634 return unless $contents =~ /^=/m;
1636 return unless $contents =~ /^=head1 +NAME/m;
1639 # Here, we know that the file is a pod. Add it to the list of files
1640 # to check and create a checker object for it.
1642 push @files, $filename;
1643 my $checker = My::Pod::Checker->new($filename);
1644 $filename_to_checker{$filename} = $checker;
1646 # In order to detect duplicate pods and only analyze them once, we
1647 # compute checksums for the file, so don't have to do an exact
1648 # compare. Note that if the pod is just part of the file, the
1649 # checksums can differ for the same pod. That special case is handled
1650 # later, since if the checksums of the whole file are the same, that
1651 # case won't even come up. We don't need the checksums for files that
1652 # we parse only if there is a link to its interior, but we do need its
1653 # NAME, which is also retrieved in the code below.
1655 if ($filename =~ / (?: ^(cpan|lib|ext|dist)\/ )
1656 | $only_for_interior_links_re
1658 $digest->add($contents);
1659 $digests{$filename} = $digest->digest;
1661 # lib files aren't analyzed if they are duplicates of files copied
1662 # there from some other directory. But to determine this, we need
1663 # to know their NAMEs. We might as well find the NAME now while
1664 # the file is open. Similarly, cpan files aren't analyzed unless
1665 # we're analyzing all of them, or this particular file is linked
1666 # to by a file we are analyzing, and thus we will want to verify
1667 # that the target exists in it. We need to know at least the NAME
1668 # to see if it's worth analyzing, or so we can determine if a lib
1669 # file is a copy of a cpan one.
1670 if ($filename =~ m{ (?: ^ (?: cpan | lib ) / )
1671 | $only_for_interior_links_re
1673 if ($contents =~ /^=head1 +NAME.*/mg) {
1674 # The NAME is the first non-spaces on the line up to a
1675 # comma, dash or end of line. Otherwise, it's invalid and
1676 # this pod doesn't have a legal name that we're smart
1677 # enough to find currently. But the parser will later
1678 # find it if it thinks there is a legal name, and set the
1680 if ($contents =~ /\G # continue from the line after =head1
1681 \s* # ignore any empty lines
1683 # ignore =for paragraphs followed by empty
1685 (?: ^ =for .*? \n (?: [^\s]*? \n )* \s* )*
1687 ^ \s* ( \S+?) \s* (?: [,-] | $ )/mx) {
1689 $checker->name($name);
1690 $id_to_checker{$name} = $checker
1691 if $filename =~ m{^cpan/};
1694 elsif ($filename =~ m{^cpan/}) {
1695 $id_to_checker{$digests{$filename}} = $checker;
1701 } # End of is_pod_file()
1703 # Start of real code that isn't processing the command line (except the
1704 # db is read in above, as is processing of the --add_link option).
1705 # Here, @files contains list of files on the command line. If have any of
1706 # these, unconditionally test them, and show all the errors, even the known
1707 # ones, and, since not testing other pods, don't do cross-pod link tests.
1708 # (Could add extra code to do cross-pod tests for the ones in the list.)
1710 if ($has_input_files) {
1711 undef %known_problems;
1712 $do_upstream_cpan = $do_deltas = 1; # In case one of the inputs is one
1715 else { # No input files -- go find all the possibilities.
1717 $copy_fh = open_new($known_issues);
1718 note("Regenerating $known_issues, please be patient...");
1719 print $copy_fh $HEADER;
1722 # Move to the directory above us, but have to adjust @INC to account for
1724 s{^\.\./lib$}{lib} for @INC;
1725 chdir File::Spec->updir;
1727 # And look in this directory and all its subdirectories
1728 find( {wanted => \&is_pod_file, no_chdir => 1}, '.');
1730 # Add ourselves to the test
1731 push @files, "t/porting/podcheck.t";
1734 # Now we know how many tests there will be.
1735 plan (tests => scalar @files) if ! $regen;
1738 # Sort file names so we get consistent results, and to put cpan last,
1739 # preceded by the ones that we don't generally parse. This is because both
1740 # these classes are generally parsed only if there is a link to the interior
1741 # of them, and we have to parse all others first to guarantee that they don't
1742 # have such a link. 'lib' files come just before these, as some of these are
1743 # duplicates of others. We already have figured this out when gathering the
1744 # data as a special case for all such files, but this, while unnecessary,
1745 # puts the derived file last in the output. 'readme' files come before those,
1746 # as those also could be duplicates of others, which are considered the
1747 # primary ones. These currently aren't figured out when gathering data, so
1749 @files = sort { if ($a =~ /^cpan/) {
1750 return 1 if $b !~ /^cpan/;
1751 return lc $a cmp lc $b;
1753 elsif ($b =~ /^cpan/) {
1756 elsif ($a =~ /$only_for_interior_links_re/) {
1757 return 1 if $b !~ /$only_for_interior_links_re/;
1758 return lc $a cmp lc $b;
1760 elsif ($b =~ /$only_for_interior_links_re/) {
1763 elsif ($a =~ /^lib/) {
1764 return 1 if $b !~ /^lib/;
1765 return lc $a cmp lc $b;
1767 elsif ($b =~ /^lib/) {
1769 } elsif ($a =~ /\breadme\b/i) {
1770 return 1 if $b !~ /\breadme\b/i;
1771 return lc $a cmp lc $b;
1773 elsif ($b =~ /\breadme\b/i) {
1777 return lc $a cmp lc $b;
1782 # Now go through all the files and parse them
1784 foreach my $filename (@files) {
1786 note("parsing $filename") if DEBUG;
1788 # We may have already figured out some things in the process of generating
1789 # the file list. If so, we have a $checker object already. But if not,
1791 my $checker = $filename_to_checker{$filename};
1793 $checker = My::Pod::Checker->new($filename);
1794 $filename_to_checker{$filename} = $checker;
1797 # We have set the name in the checker object if there is a possibility
1798 # that no further parsing is necessary, but otherwise do the parsing now.
1799 if (! $checker->name) {
1800 if (! $checker->parse_from_file($filename, undef)) {
1801 $checker->set_skip("$filename is transitory");
1808 if ($checker->num_errors() < 0) { # Returns negative if not a pod
1809 $checker->set_skip("$filename is not a pod");
1813 # Here, is a pod. See if it is one that has already been tested,
1814 # or should be tested under another directory. Use either its NAME
1815 # if it has one, or a checksum if not.
1816 my $name = $checker->name;
1823 my $digest = Digest->new($digest_type);
1824 my $contents = extract_pod($filename);
1826 # If the return is undef, it means that $filename was a transitory
1828 next FILE unless defined $contents;
1829 $digest->add($contents);
1830 $id = $digest->digest;
1833 # If there is a match for this pod with something that we've already
1834 # processed, don't process it, and output why.
1836 if (defined ($prior_checker = $id_to_checker{$id})
1837 && $prior_checker != $checker) # Could have defined the checker
1838 # earlier without pursuing it
1841 # If the pods are identical, then it's just a copy, and isn't an
1842 # error. First use the checksums we have already computed to see
1843 # if the entire files are identical, which means that the pods are
1845 my $prior_filename = $prior_checker->get_filename;
1847 || ($digests{$prior_filename}
1848 && $digests{$filename}
1849 && $digests{$prior_filename} eq $digests{$filename}));
1851 # If they differ, it could be that the files differ for some
1852 # reason, but the pods they contain are identical. Extract the
1853 # pods and do the comparisons on just those.
1854 if (! $same && $name) {
1855 my $contents = extract_pod($filename);
1857 # If return is <undef>, it means that $filename no longer
1858 # exists. This means it was a transitory file, and should not
1860 next FILE unless defined $contents;
1862 my $prior_contents = extract_pod($prior_filename);
1864 # If return is <undef>, it means that $prior_filename no
1865 # longer exists. This means it was a transitory file, and
1866 # should not have been tested, but we already did process it.
1867 # What we should do now is to back-out its records, and
1868 # process $filename in its stead. But backing out is not so
1869 # simple, and so I'm (khw) skipping that unless and until
1870 # experience shows that it is needed. We do go process
1871 # $filename, and there are potential false positive conflicts
1872 # with the transitory $prior_contents, and rerunning the test
1873 # should cause it to succeed.
1874 goto process_this_pod unless defined $prior_contents;
1876 $same = $prior_contents eq $contents;
1879 use File::Basename 'basename';
1881 $checker->set_skip("The pod of $filename is a duplicate of "
1882 . "the pod for $prior_filename");
1883 } elsif ($prior_filename =~ /\breadme\b/i) {
1884 $checker->set_skip("$prior_filename is a README apparently for $filename");
1885 } elsif ($filename =~ /\breadme\b/i) {
1886 $checker->set_skip("$filename is a README apparently for $prior_filename");
1887 } elsif (! $do_upstream_cpan
1888 && $filename =~ /^cpan/
1889 && $prior_filename =~ /^cpan/)
1891 $checker->set_skip("CPAN is upstream for $filename");
1892 } elsif ( $filename =~ /^utils/ or $prior_filename =~ /^utils/ ) {
1893 $checker->set_skip("$filename copy is in utils/");
1894 } elsif ($prior_filename =~ /^(?:cpan|ext|dist)/
1895 && $filename !~ /^(?:cpan|ext|dist)/
1896 && basename($prior_filename) eq basename($filename))
1898 $checker->set_skip("$filename: Need to run make?");
1899 } else { # Here have two pods with identical names that differ
1900 $prior_checker->poderror(
1901 { -msg => $duplicate_name,
1903 parameter => "'$filename' also has NAME '$name'"
1906 { -msg => $duplicate_name,
1908 parameter => "'$prior_filename' also has NAME '$name'"
1911 # Changing the names helps later.
1912 $prior_checker->name("$name version arbitrarily numbered 1");
1913 $checker->name("$name version arbitrarily numbered 2");
1916 # In any event, don't process this pod that has the same name as
1924 $id_to_checker{$id} = $checker;
1926 my $parsed_for_links = ", but parsed for its interior links";
1927 if ((! $do_upstream_cpan && $filename =~ /^cpan/)
1928 || $filename =~ $only_for_interior_links_re)
1930 if ($filename =~ /^cpan/) {
1931 $checker->set_skip("CPAN is upstream for $filename");
1933 elsif ($filename =~ /perl\d+delta/) {
1935 $checker->set_skip("$filename is a stable perldelta");
1938 elsif ($filename =~ /perltoc/) {
1939 $checker->set_skip("$filename dependent on component pods");
1942 croak("Unexpected file '$filename' encountered that has parsing for interior-linking only");
1945 if ($name && $has_referred_to_node{$name}) {
1946 $checker->set_skip($checker->get_skip() . $parsed_for_links);
1950 # Need a name in order to process it, because not meaningful
1951 # otherwise, and also can't test links to this without a name.
1952 if (!defined $name) {
1953 $checker->poderror( { -msg => $no_name,
1959 # For skipped files, just get its NAME
1961 if (($skip = $checker->get_skip()) && $skip !~ /$parsed_for_links/)
1963 $checker->node($name) if $name;
1966 if (! $checker->parse_from_file($filename, undef)) {
1967 $checker->set_skip("$filename is transitory");
1972 # Go through everything in the file that could be an anchor that
1973 # could be a link target. Count how many there are of the same name.
1974 foreach my $node ($checker->linkable_nodes) {
1975 next FILE if ! $node; # Can be empty is like '=item *'
1976 $nodes{$name}{$node}++;
1978 # Experiments have shown that cpan search can figure out the
1979 # target of a link even if the exact wording is incorrect, as long
1980 # as the first word is. This happens frequently in perlfunc.pod,
1981 # where the link will be just to the function, but the target
1982 # entry also includes parameters to the function.
1983 my $first_word = $node;
1984 if ($first_word =~ s/^(\S+)\s+\S.*/$1/) {
1985 $nodes_first_word{$name}{$first_word} = $node;
1988 $filename_to_pod{$filename} = $name;
1992 # Here, all files have been parsed, and all links and link targets are stored.
1993 # Now go through the files again and see which don't have matches.
1994 if (! $has_input_files) {
1995 foreach my $filename (@files) {
1996 next if $filename_to_checker{$filename}->get_skip;
1998 my $checker = $filename_to_checker{$filename};
1999 foreach my $link ($checker->hyperlinks()) {
2000 my $linked_to_page = $link->page;
2001 next unless $linked_to_page; # intra-file checks are handled by std
2003 # Currently, we assume all external links are valid
2004 next if $link->type eq 'url';
2006 # Initialize the potential message.
2007 my %problem = ( -msg => $broken_link,
2008 -line => $link->line,
2009 parameter => "to \"$linked_to_page\"",
2012 # See if we have found the linked-to_file in our parse
2013 if (exists $nodes{$linked_to_page}) {
2014 my $node = $link->node;
2016 # If link is only to the page-level, already have it
2019 # If link is to a node that exists in the file, is ok
2020 if ($nodes{$linked_to_page}{$node}) {
2022 # But if the page has multiple targets with the same name,
2023 # it's ambiguous which one this should be to.
2024 if ($nodes{$linked_to_page}{$node} > 1) {
2025 $problem{-msg} = $multiple_targets;
2026 $problem{parameter} = "in $linked_to_page that $node could be pointing to";
2027 $checker->poderror(\%problem);
2029 } elsif (! $nodes_first_word{$linked_to_page}{$node}) {
2031 # Here the link target was not found, either exactly or to
2032 # the first word. Is an error.
2033 $problem{parameter} =~ s,"$,/$node",;
2034 $checker->poderror(\%problem);
2037 } # Linked-to-file not in parse; maybe is in exception list
2038 elsif (! exists $valid_modules{$link->page}) {
2040 # Here, is a link to a target that we can't find. Check if
2041 # there is an internal link on the page with the target name.
2042 # If so, it could be that they just forgot the initial '/'
2043 # But perldelta is handled specially: only do this if the
2044 # broken link isn't one of the known bad ones (that are
2045 # placemarkers and should be removed for the final)
2046 my $NAME = $filename_to_pod{$filename};
2047 if (! defined $NAME) {
2048 $checker->poderror(\%problem);
2051 if ($nodes{$NAME}{$linked_to_page}) {
2052 $problem{-msg} = $broken_internal_link;
2054 $checker->poderror(\%problem);
2061 # If regenerating the data file, start with the modules for which we don't
2062 # check targets. If you change the sort order, you need to run --regen before
2063 # committing so that future commits that do run regen don't show irrelevant
2066 foreach (sort { lc $a cmp lc $b } keys %valid_modules) {
2067 my_safer_print($copy_fh, $_, "\n");
2071 # Now ready to output the messages.
2072 foreach my $filename (@files) {
2073 my $canonical = canonicalize($filename);
2075 my $skip = $filename_to_checker{$filename}->get_skip // "";
2078 foreach my $message ( sort keys %{$problems{$filename}}) {
2081 # Preserve a negative setting.
2082 if ($known_problems{$canonical}{$message}
2083 && $known_problems{$canonical}{$message} < 0)
2085 $count = $known_problems{$canonical}{$message};
2088 $count = @{$problems{$filename}{$message}};
2090 my_safer_print($copy_fh, $canonical . "\t$message\t$count\n");
2095 skip($skip, 1) if $skip;
2097 my $thankful_diagnostics = 0;
2100 my $total_known = 0;
2101 foreach my $message ( sort keys %{$problems{$filename}}) {
2102 $known_problems{$canonical}{$message} = 0
2103 if ! $known_problems{$canonical}{$message};
2104 my $diagnostic = "";
2105 my $problem_count = scalar @{$problems{$filename}{$message}};
2106 $total_known += $problem_count;
2107 next if $known_problems{$canonical}{$message} < 0;
2108 if ($problem_count > $known_problems{$canonical}{$message}) {
2110 # Here we are about to output all the messages for this type,
2111 # subtract back this number we previously added in.
2112 $total_known -= $problem_count;
2114 $diagnostic .= $indent . qq{"$message"};
2115 if ($problem_count > 2) {
2116 $diagnostic .= " ($problem_count occurrences,"
2117 . " expected $known_problems{$canonical}{$message})";
2119 foreach my $problem (@{$problems{$filename}{$message}}) {
2120 $diagnostic .= " " if $problem_count == 1;
2121 $diagnostic .= "\n$indent$indent";
2122 $diagnostic .= "$problem->{parameter}" if $problem->{parameter};
2123 $diagnostic .= " near line $problem->{-line} of "
2125 $diagnostic .= " $problem->{comment}" if $problem->{comment};
2127 $diagnostic .= "\n";
2128 $files_with_unknown_issues{$filename} = 1;
2129 } elsif ($problem_count < $known_problems{$canonical}{$message}) {
2130 $diagnostic = output_thanks($filename, $known_problems{$canonical}{$message}, $problem_count, $message);
2131 $thankful_diagnostics++;
2133 push @diagnostics, $diagnostic if $diagnostic;
2136 # The above loop has output messages where there are current potential
2137 # issues. But it misses where there were some that have been entirely
2138 # fixed. For those, we need to look through the old issues
2139 foreach my $message ( sort keys %{$known_problems{$canonical}}) {
2140 next if $problems{$filename}{$message};
2141 next if ! $known_problems{$canonical}{$message};
2142 next if $known_problems{$canonical}{$message} < 0; # Preserve negs
2144 next if !$pedantic and $message =~
2145 /^(?:\Q$line_length\E|\Q$C_not_linked\E|\Q$C_with_slash\E)/;
2147 my $diagnostic = output_thanks($filename, $known_problems{$canonical}{$message}, 0, $message);
2148 push @diagnostics, $diagnostic if $diagnostic;
2149 $thankful_diagnostics++ if $diagnostic;
2152 my $output = "POD of $filename";
2153 $output .= ", excluding $total_known not shown known potential problems"
2155 if (@diagnostics && @diagnostics == $thankful_diagnostics) {
2156 # Output fixed issues as passing to-do tests, so they do not
2157 # cause failures, but t/harness still flags them.
2158 $output .= " # TODO"
2160 ok(@diagnostics == $thankful_diagnostics, $output);
2162 diag(join "", @diagnostics,
2163 "See end of this test output for your options on silencing this");
2166 delete $known_problems{$canonical};
2171 && ! ok (keys %known_problems == 0, "The known problems database ($data_dir/known_pod_issues.dat) includes no references to non-existent files"))
2173 note("The following files were not found: "
2174 . join ", ", keys %known_problems);
2175 note("They will automatically be removed from the db the next time");
2176 note(" cd t; ./perl -I../lib porting/podcheck.t --regen");
2181 run this test script by hand, using the following formula (on
2182 Un*x-like machines):
2184 ./perl -I../lib porting/podcheck.t --regen
2187 if (%files_with_unknown_issues) {
2188 my $were_count_files = scalar keys %files_with_unknown_issues;
2189 $were_count_files = ($were_count_files == 1)
2190 ? "was $were_count_files file"
2191 : "were $were_count_files files";
2192 my $message = <<EOF;
2194 HOW TO GET ${\__FILE__} TO PASS
2196 There $were_count_files that had new potential problems identified.
2197 Some of them may be real, and some of them may be false positives because
2198 this program isn't as smart as it likes to think it is. You can teach this
2199 program to ignore the issues it has identified, and hence pass, by doing the
2202 1) If a problem is about a link to an unknown module or man page that
2203 you know exists, re-run the command something like:
2204 ./perl -I../lib porting/podcheck.t --add_link MODULE man_page ...
2205 (MODULEs should look like Foo::Bar, and man_pages should look like
2206 bar(3c); don't do this for a module or man page that you aren't sure
2207 about; instead treat as another type of issue and follow the
2208 instructions below.)
2210 2) For other issues, decide if each should be fixed now or not. Fix the
2211 ones you decided to, and rerun this test to verify that the fixes
2214 3) If there remain false positive or problems that you don't plan to fix right
2217 That should cause all current potential problems to be accepted by
2218 the program, so that the next time it runs, they won't be flagged.
2220 if (%files_with_fixes) {
2221 $message .= " This step will also take care of the files that have fixes in them\n";
2225 For a few files, such as perltoc, certain issues will always be
2226 expected, and more of the same will be added over time. For those,
2227 before you do the regen, you can edit
2229 and find the entry for the module's file and specific error message,
2230 and change the count of known potential problems to -1.
2234 } elsif (%files_with_fixes) {
2236 To teach this test script that the potential problems have been fixed,
2243 chdir $original_dir || die "Can't change directories to $original_dir";
2244 close_and_rename($copy_fh);