#!/usr/bin/perl
+
+BEGIN {
+ @INC = '..' if -f '../TestInit.pm';
+}
+use TestInit qw(T); # T is chdir to the top level
+
use warnings;
use strict;
-chdir 't';
-require './test.pl';
-
+require 't/test.pl';
plan('no_plan');
-$|=1;
-
# --make-exceptions-list outputs the list of strings that don't have
# perldiag.pod entries to STDERR without TAP formatting, so they can
# easily be put in the __DATA__ section of this file. This was done
# Just add the documentation instead.
my $make_exceptions_list = ($ARGV[0]||'') eq '--make-exceptions-list';
-chdir '..' or die "Can't chdir ..: $!";
-BEGIN { defined $ENV{PERL_UNICODE} and push @INC, "lib"; }
+require 'regen/embed_lib.pl';
+# Look for functions that look like they could be diagnostic ones.
my @functions;
-
-open my $func_fh, "<", "embed.fnc" or die "Can't open embed.fnc: $!";
-
-# Look for functions in embed.fnc that look like they could be diagnostic ones.
-while (<$func_fh>) {
- chomp;
- s/^\s+//;
- while (s/\s*\\$//) { # Grab up all continuation lines, these end in \
- my $next = <$func_fh>;
- $next =~ s/^\s+//;
- chomp $next;
- $_ .= $next;
- }
- next if /^:/; # Lines beginning with colon are comments.
- next unless /\|/; # Lines without a vertical bar are something we can't deal
- # with
- my @fields = split /\s*\|\s*/;
- next unless $fields[2] =~ /warn|err|(\b|_)die|croak/i;
- push @functions, $fields[2];
-
+foreach (@{(setup_embed())[0]}) {
+ next if @$_ < 2;
+ next unless $_->[2] =~ /warn|err|(\b|_)die|croak/i;
# The flag p means that this function may have a 'Perl_' prefix
# The flag s means that this function may have a 'S_' prefix
- push @functions, "Perl_$fields[2]", if $fields[0] =~ /p/;
- push @functions, "S_$fields[2]", if $fields[0] =~ /s/;
-}
-
-close $func_fh;
+ push @functions, $_->[2];
+ push @functions, 'Perl_' . $_->[2] if $_->[0] =~ /p/;
+ push @functions, 'S_' . $_->[2] if $_->[0] =~ /s/;
+};
+my $regcomp_re = "(?<routine>(?:ckWARN(?:\\d+)?reg\\w*|vWARN\\d+))";
my $function_re = join '|', @functions;
-my $source_msg_re = "(?<routine>\\bDIE\\b|$function_re)";
+my $regcomp_fail_re = '\b(?:(?:Simple_)?v)?FAIL[2-4]?\b';
+my $source_msg_re =
+ "(?<routine>\\bDIE\\b|$function_re|$regcomp_fail_re)";
my $text_re = '"(?<text>(?:\\\\"|[^"]|"\s*[A-Z_]+\s*")*)"';
my $source_msg_call_re = qr/$source_msg_re(?:_nocontext)? \s*
\(aTHX_ \s*
(?:packWARN\d*\((?<category>.*?)\),)? \s*
$text_re /x;
my $bad_version_re = qr{BADVERSION\([^"]*$text_re};
+ $regcomp_fail_re = qr/$regcomp_fail_re\([^"]*$text_re/;
+my $regcomp_call_re = qr/$regcomp_re.*?$text_re/;
my %entries;
# Get the ignores that are compiled into this file
+my $reading_categorical_exceptions;
while (<DATA>) {
chomp;
- $entries{$_}{todo}=1;
+ $entries{$_}{todo} = 1;
+ $reading_categorical_exceptions and $entries{$_}{cattodo}=1;
+ /__CATEGORIES__/ and ++$reading_categorical_exceptions;
}
my $pod = "pod/perldiag.pod";
open my $diagfh, "<", $pod
or die "Can't open $pod: $!";
-my $category_re = qr/ [a-z0-9_]+?/; # Note: requires an initial space
+my $category_re = qr/ [a-z0-9_:]+?/; # Note: requires an initial space
my $severity_re = qr/ . (?: \| . )* /x; # A severity is a single char, but can
# be of the form 'S|P|W'
+my @same_descr;
while (<$diagfh>) {
if (m/^=item (.*)/) {
$cur_entry = $1;
- if (exists $entries{$cur_entry}) {
+ # Allow multi-line headers
+ while (<$diagfh>) {
+ if (/^\s*$/) {
+ last;
+ }
+
+ $cur_entry .= $_;
+ }
+
+ $cur_entry =~ s/\n/ /gs; # Fix multi-line headers if they have \n's
+ $cur_entry =~ s/\s+\z//;
+
+ if (exists $entries{$cur_entry} && $entries{$cur_entry}{todo}
+ && !$entries{$cur_entry}{cattodo}) {
TODO: {
local $::TODO = "Remove the TODO entry \"$cur_entry\" from DATA as it is already in $pod near line $.";
ok($cur_entry);
# overwrites one in DATA.
$entries{$cur_entry}{todo} = 0;
$entries{$cur_entry}{line_number} = $.;
- next;
}
next if ! defined $cur_entry;
if (/^ \( ( $severity_re )
# Can have multiple categories separated by commas
- (?: ( $category_re ) (?: , $category_re)* )? \) /x)
+ ( $category_re (?: , $category_re)* )? \) /x)
{
$entries{$cur_entry}{severity} = $1;
- $entries{$cur_entry}{category} = $2;
+ $entries{$cur_entry}{category} =
+ $2 && join ", ", sort split " ", $2 =~ y/,//dr;
+
+ # Record it also for other messages sharing the same description
+ @$_{qw<severity category>} =
+ @{$entries{$cur_entry}}{qw<severity category>}
+ for @same_descr;
}
elsif (! $entries{$cur_entry}{first_line} && $_ =~ /\S/) {
# that can later examine it to determine if that is ok or not
$entries{$cur_entry}{first_line} = $_;
}
+ if (/\S/) {
+ @same_descr = ();
+ }
+ else {
+ push @same_descr, $entries{$cur_entry};
+ }
}
}
NVgf => 'f',
HEKf256=>'s',
HEKf => 's',
+ UTF8f=> 's',
SVf256=>'s',
SVf32=> 's',
SVf => 's');
join '|', sort { length $b cmp length $a } keys %specialformats;
my $specialformats_re = qr/%$format_modifiers"\s*($specialformats)(\s*")?/;
-# Recursively descend looking for source files.
-my @todo = sort <*>;
-while (@todo) {
- my $todo = shift @todo;
- next if $todo ~~ ['t', 'lib', 'ext', 'dist', 'cpan'];
- # opmini.c is just a copy of op.c, so there's no need to check again.
- next if $todo eq 'opmini.c';
- if (-d $todo) {
- unshift @todo, sort glob "$todo/*";
- } elsif ($todo =~ m/\.[ch]$/) {
- check_file($todo);
- }
+open my $fh, '<', 'MANIFEST' or die "Can't open MANIFEST: $!";
+while (my $file = <$fh>) {
+ chomp $file;
+ $file =~ s/\s+.*//;
+ next unless $file =~ /\.(?:c|cpp|h|xs|y)\z/ or $file =~ /^perly\./;
+ # OS/2 extensions have never been migrated to ext/, hence the special case:
+ next if $file =~ m!\A(?:ext|dist|cpan|lib|t|os2/OS2)/!
+ && $file !~ m!\Aext/DynaLoader/!;
+ check_file($file);
}
+close $fh or die $!;
# Standardize messages with variants into the form that appears
# in perldiag.pod -- useful for things without a diag_listed_as annotation
$sub = $_;
}
next if $sub =~ m/^XS/;
- if (m</\* diag_listed_as: (.*) \*/>) {
+ if (m</\*\s*diag_listed_as: (.*?)\s*\*/>) {
$listed_as = $1;
$listed_as_line = $.+1;
}
my $multiline = 0;
# Loop to accumulate the message text all on one line.
- if (m/$source_msg_re(?:_nocontext)?\s*\(/) {
+ if (m/(?:$source_msg_re(?:_nocontext)?|$regcomp_re)\s*\(/) {
while (not m/\);$/) {
my $nextline = <$codefh>;
# Means we fell off the end of the file. Not terribly surprising;
$nextline =~ s/^\s+//;
$_ =~ s/\\$//;
# Note that we only want to do this where *both* are true.
- if ($_ =~ m/"$/ and $nextline =~ m/^"/) {
- $_ =~ s/"$//;
+ if ($_ =~ m/"\s*$/ and $nextline =~ m/^"/) {
+ $_ =~ s/"\s*$//;
$nextline =~ s/^"//;
}
$_ .= $nextline;
# The %"foo" thing needs to happen *before* this regex.
# diag($_);
# DIE is just return Perl_die
- my ($name, $category);
+ my ($name, $category, $routine);
if (/$source_msg_call_re/) {
- ($name, $category) = ($+{'text'}, $+{'category'});
+ ($name, $category, $routine) = ($+{'text'}, $+{'category'}, $+{'routine'});
+ # Sometimes the regexp will pick up too much for the category
+ # e.g., WARN_UNINITIALIZED), PL_warn_uninit_sv ... up to the next )
+ $category && $category =~ s/\).*//s;
}
elsif (/$bad_version_re/) {
($name, $category) = ($+{'text'}, undef);
}
+ elsif (/$regcomp_fail_re/) {
+ # FAIL("foo") -> "foo in regex m/%s/"
+ # vFAIL("foo") -> "foo in regex; marked by <-- HERE in m/%s/"
+ ($name, $category) = ($+{'text'}, undef);
+ $name .=
+ " in regex" . ("; marked by <-- HERE in" x /vFAIL/) . " m/%s/";
+ }
+ elsif (/$regcomp_call_re/) {
+ # vWARN/ckWARNreg("foo") -> "foo in regex; marked by <-- HERE in m/%s/
+ ($name, $category, $routine) = ($+{'text'}, undef, $+{'routine'});
+ $name .= " in regex; marked by <-- HERE in m/%s/";
+ $category = 'WARN_REGEXP';
+ if ($routine =~ /dep/) {
+ $category .= ',WARN_DEPRECATED';
+ }
+ }
else {
next;
}
- my $severity = {croak => [qw/P F/],
- die => [qw/P F/],
- warn => [qw/W D S/],
- }->{$+{'routine'}||'die'};
- my @categories;
+ # Try to guess what the severity should be. In the case of
+ # Perl_ck_warner and other _ck_ functions, we can tell whether it is
+ # a severe/default warning or no by the _d suffix. In the case of
+ # other warn functions we cannot tell, because Perl_warner may be pre-
+ # ceded by if(ckWARN) or if(ckWARN_d).
+ my $severity = !$routine ? '[PFX]'
+ : $routine =~ /warn.*_d\z/ ? '[DS]'
+ : $routine =~ /ck_warn/ ? 'W'
+ : $routine =~ /warner/ ? '[WDS]'
+ : $routine =~ /warn/ ? 'S'
+ : $routine =~ /ckWARN.*dep/ ? 'D'
+ : $routine =~ /ckWARN\d*reg/ ? 'W'
+ : $routine =~ /vWARN\d/ ? '[WDS]'
+ : '[PFX]';
+ my $categories;
if (defined $category) {
- @categories = map {s/^WARN_//; lc $_} split /\s*[|,]\s*/, $category;
+ $category =~ s/__/::/g;
+ $categories =
+ join ", ",
+ sort map {s/^WARN_//; lc $_} split /\s*[|,]\s*/, $category;
}
if ($listed_as and $listed_as_line == $. - $multiline) {
$name = $listed_as;
# inside an #if 0 block.
next if $name eq 'SKIPME';
- check_message(standardize($name),$codefn);
+ next if $name=~/\[TESTING\]/; # ignore these as they are works in progress
+
+ check_message(standardize($name),$codefn,$severity,$categories);
}
}
sub check_message {
- my($name,$codefn,$partial) = @_;
+ my($name,$codefn,$severity,$categories,$partial) = @_;
my $key = $name =~ y/\n/ /r;
my $ret;
- if (exists $entries{$key}) {
+ # Try to reduce printf() formats to simplest forms
+ # Really this should be matching %s, etc like diagnostics.pm does
+
+ # Kill flags
+ $key =~ s/%[#0\-+]/%/g;
+
+ # Kill width
+ $key =~ s/\%(\d+|\*)/%/g;
+
+ # Kill precision
+ $key =~ s/\%\.(\d+|\*)/%/g;
+
+ if (exists $entries{$key} and
+ # todo + cattodo means it is not found and it is not in the
+ # regular todo list, either
+ !$entries{$key}{todo} || !$entries{$key}{cattodo}) {
$ret = 1;
if ( $entries{$key}{seen}++ ) {
# no need to repeat entries we've tested
- } elsif ($entries{$name}{todo}) {
+ } elsif ($entries{$key}{todo}) {
TODO: {
no warnings 'once';
local $::TODO = 'in DATA';
# There is no listing, but it is in the list of exceptions. TODO FAIL.
- fail($name);
+ fail($key);
diag(
" Message '$name'\n from $codefn line $. is not listed in $pod\n".
" (but it wasn't documented in 5.10 either, so marking it TODO)."
} else {
# We found an actual valid entry in perldiag.pod for this error.
pass($key);
+
+ return $ret
+ if $entries{$key}{cattodo};
+
+ # Now check the category and severity
+
+ # Cache our severity qr thingies
+ use feature 'state';
+ state %qrs;
+ my $qr = $qrs{$severity} ||= qr/$severity/;
+
+ like($entries{$key}{severity}, $qr,
+ $severity =~ /\[/
+ ? "severity is one of $severity for $key"
+ : "severity is $severity for $key");
+
+ is($entries{$key}{category}, $categories,
+ ($categories ? "categories are [$categories]" : "no category")
+ . " for $key");
}
- # Later, should start checking that the severity is correct, too.
} elsif ($partial) {
# noop
} else {
my $ok;
if ($name =~ /\n/) {
$ok = 1;
- check_message($_,$codefn,1) or $ok = 0, last for split /\n/, $name;
+ check_message($_,$codefn,$severity,$categories,1) or $ok = 0, last
+ for split /\n/, $name;
}
if ($ok) {
# noop
# don't have to go from "meh" to perfect all at once.
#
# PLEASE DO NOT ADD TO THIS LIST. Instead, write an entry in
-# pod/perldiag.pod for your new (warning|error).
+# pod/perldiag.pod for your new (warning|error). Nevertheless,
+# listing exceptions here when this script is not smart enough
+# to recognize the messages is not so bad, as long as there are
+# entries in perldiag.
+
+# Entries after __CATEGORIES__ are those that are in perldiag but fail the
+# severity/category test.
# Also FIXME this test, as the first entry in TODO *is* covered by the
# description: Malformed UTF-8 character (%s)
__DATA__
Malformed UTF-8 character (unexpected non-continuation byte 0x%x, immediately after start byte 0x%x)
-'%c' allowed only after types %s in %s
-bad top format reference
Cannot apply "%s" in non-PerlIO perl
-Can't %s big-endian %ss on this
-Can't call mro_isa_changed_in() on anonymous symbol table
-Can't call mro_method_changed_in() on anonymous symbol table
-Can't coerce readonly %s to string
-Can't coerce readonly %s to string in %s
Can't find string terminator %c%s%c anywhere before EOF
Can't fix broken locale name "%s"
Can't get short module name from a handle
clear %s
Code missing after '/' in pack
Code missing after '/' in unpack
-Corrupted regexp opcode %d > %d
'%c' outside of string in pack
Debug leaking scalars child failed%s with errno %d: %s
'/' does not take a repeat count in %s
-Don't know how to get file name
-Don't know how to handle magic of type \%o
-Dp not implemented on this platform
Error reading "%s": %s
execl not implemented!
Filehandle STD%s reopened as %s only for input
filter_del can only delete in reverse order (currently)
YOU HAVEN'T DISABLED SET-ID SCRIPTS IN THE KERNEL YET! FIX YOUR KERNEL, PUT A C WRAPPER AROUND THIS SCRIPT, OR USE -u AND UNDUMP!
-fork() not implemented!
free %s
-Free to wrong pool %p not %p
get %s %p %p %p
-gethostent not implemented!
getpwnam returned invalid UIC %o for user "%s"
glob failed (can't start child: %s)
glob failed (child exited with status %d%s)
Illegal character %sin prototype for %s : %s
Illegal hexadecimal digit '%c' ignored
Illegal octal digit '%c' ignored
-Infinite recursion in regex
-Integer overflow in srand
-internal %<num>p might conflict with future printf extensions
Invalid argument to sv_cat_decode
Invalid range "%c-%c" in transliteration operator
Invalid separator character %c%c%c in PerlIO layer specification %s
Invalid type '%c' in unpack
Invalid type ',' in %s
ioctlsocket not implemented!
-'j' not supported on this platform
-'J' not supported on this platform
killpg not implemented!
-length() used on %s (did you mean "scalar(%s)"?)
-length() used on %hash (did you mean "scalar(keys %hash)"?)
-length() used on @array (did you mean "scalar(@array)"?)
List form of pipe open not implemented
Malformed integer in [] in %s
Malformed UTF-8 character (fatal)
Missing (suid) fd script name
More than one argument to open
More than one argument to open(,':%s')
-mprotect for %p %u failed with %d
-mprotect RW for %p %u failed with %d
+\N{} in character class restricted to one character in regex; marked by <-- HERE in m/%s/
No %s allowed while running setgid
No %s allowed with (suid) fdscript
-No such class field "%s"
Not an XSUB reference
Operator or semicolon missing before %c%s
-Pattern subroutine nesting without pos change exceeded limit in regex
-Perl %s required--this is only %s, stopped
+PerlApp::TextQuery: no arguments, please
+POSIX syntax [%c %c] is reserved for future extensions in regex; marked by <-- HERE in m/%s/
ptr wrong %p != %p fl=%x nl=%p e=%p for %d
Recompile perl with -DDEBUGGING to use -D switch (did you mean -d ?)
+Regexp modifier "%c" may appear a maximum of twice in regex; marked by <-- HERE in m/%s/
+Regexp modifier "%c" may not appear twice in regex; marked by <-- HERE in m/%s/
+Regexp modifiers "%c" and "%c" are mutually exclusive in regex; marked by <-- HERE in m/%s/
+Regexp *+ operand could be empty in regex; marked by <-- HERE in m/%s/
Repeated format line will never terminate (~~ and @#)
Reversed %c= operator
%s(%f) failed
%sCompilation failed in require
+Sequence (?%c...) not implemented in regex; marked by <-- HERE in m/%s/
+Sequence (%s...) not recognized in regex; marked by <-- HERE in m/%s/
+Sequence %s... not terminated in regex; marked by <-- HERE in m/%s/
+Sequence (?%c... not terminated in regex; marked by <-- HERE in m/%s/
+Sequence (?(%c... not terminated in regex; marked by <-- HERE in m/%s/
+Sequence (?R) not terminated in regex m/%s/
set %s %p %p %p
%s free() ignored (RMAGIC, PERL_CORE)
%s has too many errors.
System V IPC is not implemented on this machine
-T and -B not implemented on filehandles
Terminating on signal SIG%s(%d)
-The crypt() function is unimplemented due to excessive paranoia.
The crypt() function is not implemented on NetWare
The flock() function is not implemented on NetWare
The rewinddir() function is not implemented on NetWare
Unrecognized character %s; marked by <-- HERE after %s<-- HERE near column %d
Unstable directory path, current directory changed unexpectedly
Unterminated compressed integer in unpack
-Usage: CODE(0x%x)(%s)
-Usage: %s(%s)
Usage: %s::%s(%s)
Usage: File::Copy::rmscopy(from,to[,date_flag])
Usage: VMS::Filespec::candelete(spec)
Wrong syntax (suid) fd script name "%s"
'X' outside of string in %s
'X' outside of string in unpack
+Zero length \N{} in regex; marked by <-- HERE in m/%s/
+
+__CATEGORIES__
+Code point 0x%X is not Unicode, all \p{} matches fail; all \P{} matches succeed
+Code point 0x%X is not Unicode, may not be portable
+Illegal character \%o (carriage return)
+Missing argument in %s
+Unicode non-character U+%X is illegal for open interchange
+Operation "%s" returns its argument for non-Unicode code point 0x%X
+Operation "%s" returns its argument for UTF-16 surrogate U+%X
+Unicode surrogate U+%X is illegal in UTF-8
+UTF-16 surrogate U+%X
+False [] range "%s" in regex; marked by <-- HERE in m/%s/