In perl.git, the branch blead has been updated <http://perl5.git.perl.org/perl.git/commitdiff/49143bdc2f098cbb6c3513be408d13f261a699c1?hp=acfee34976362908a3d7644d6d99606b73a3c7a6>
- Log ----------------------------------------------------------------- commit 49143bdc2f098cbb6c3513be408d13f261a699c1 Merge: acfee34 3c7fa35 Author: Steve Hay <[email protected]> Date: Thu Dec 19 17:53:01 2013 +0000 Merge branch 'perldelta' into blead Add a new mode to Porting/corelist-perldelta.pl to update the module changes in pod/perldelta.pod for the perl you are currently building. This can be run as: ./perl -Ilib Porting/corelist-perldelta.pl --mode=update pod/perldelta.pod *after* updating Module::CoreList with something like: ./perl -Ilib Porting/corelist.pl cpan There is still scope for improvement in the handling of removed modules, and in automatically cribbing information from the Changes files of those modules that have been updated, but this is a big step in the right direction. commit 3c7fa35436acaf242bfb96d4c5abe2547dc847e8 Author: Steve Hay <[email protected]> Date: Thu Dec 19 17:25:28 2013 +0000 Porting/corelist-perldelta.pl - Minor tidy-ups M Porting/corelist-perldelta.pl commit 2c5d738be772a93cfd7b67a57547f36c84b696f6 Author: Steve Hay <[email protected]> Date: Tue Dec 17 08:49:33 2013 +0000 Update various instructions about the use of Porting/corelist-perldelta.pl M Porting/how_to_write_a_perldelta.pod M Porting/perl5200delta.pod M Porting/perldelta_template.pod M Porting/release_managers_guide.pod M pod/perldelta.pod commit 12671d75a06b8564b0143f67d4b5294a71471360 Author: Steve Hay <[email protected]> Date: Tue Dec 17 08:26:17 2013 +0000 Add Abir Viqar to AUTHORS M AUTHORS commit 59d2de2549d6941aca18a448b88ef767570bd223 Author: Abir Viqar <[email protected]> Date: Sat Dec 14 14:07:28 2013 -0500 Porting/corelist-perldelta.pl - Make documentation clearer * Fix incorrect usage example for update * Document that removed modules are currently ignored * Clarify why some distributions have to be manually checked * Clarify why a distribution may not be listed in Module::CoreList M Porting/corelist-perldelta.pl commit 318148ea073588bee044d98810cbfcc8227a3067 Author: Abir Viqar <[email protected]> Date: Sat Dec 14 14:06:35 2013 -0500 Porting/corelist-perldelta.pl - Skip dummy items during update M Porting/corelist-perldelta.pl commit 9c7cb4611e3c450ad0766327af388f5dce504c02 Author: Abir Viqar <[email protected]> Date: Sat Dec 14 14:05:43 2013 -0500 Porting/corelist-perldelta.pl - Use Unix newlines in perldelta M Porting/corelist-perldelta.pl commit 5e1cf63ba35995df5a77e4c4eb5eb266ab2fc9b5 Author: Abir Viqar <[email protected]> Date: Sat Dec 14 14:03:59 2013 -0500 Porting/corelist-perldelta.pl - Default to comparing last two versions do_update_existing() previously required versions to be specified M Porting/corelist-perldelta.pl commit 3bb80250e94a706e7c07159410002bd1a5299d4f Author: Abir Viqar <[email protected]> Date: Tue Oct 22 13:31:18 2013 -0400 Porting/corelist-perldelta.pl - Make do_generate use DeltaUpdater DeltaUpdater::sections_to_pod() makes generate_section() redundant M Porting/corelist-perldelta.pl commit 5c51ce30bd69b2b9dfdd34fbdb12a1ca3dfb789e Author: Abir Viqar <[email protected]> Date: Sun Oct 6 10:58:27 2013 -0400 Porting/corelist-perldelta.pl - Add a new mode update This adds a new mode, update, which given an the path to an existing perldelta file, will add missing entries and update incorrect version information. This commit introduces a new module, DeltaUpdater, which is used for pod manipulation. M Porting/corelist-perldelta.pl commit fc353283d46e6633d67fe5ff79bc64b8aeafcc5d Author: Abir Viqar <[email protected]> Date: Thu Oct 3 17:14:34 2013 -0400 Porting/corelist-perldelta.pl - Make do_check less noisy The removed modules as returned by corelist_perldelta contains too many false positives to be useful. Also, explictly state whether the difference is from the pod or the results returned from corelist_perldelta(). M Porting/corelist-perldelta.pl commit f40724d58902d75124135a26602bf5d12c481c67 Author: Abir Viqar <[email protected]> Date: Thu Oct 3 17:01:54 2013 -0400 Porting/corelist-perldelta.pl - Improve corelist_delta corelist_delta now goes through almost all of the core distributions. The problem with the previous approach was that the keys of %Modules in Porting/Maintainers.pl do not all correspond to a valid distribution or do not correspond to a module as listed in Module::CoreList. This commit also updates the callers of the function. M Porting/corelist-perldelta.pl commit c014dfade439c530cde995c21eddff12e9179657 Author: Abir Viqar <[email protected]> Date: Thu Oct 3 16:48:34 2013 -0400 Porting/corelist-perldelta.pl - Cleanup DeltaParser * Comment generation of accessor methods * Make _parse_delta() more explicit by calling the section parsing methods explicitly * Ensure that the new_modules, updated_modules, and removed_modules attributes exist even if the perldelta file does not contain the corresponding sections M Porting/corelist-perldelta.pl commit 3668a9e5f3fcda99d7e32f1ac571123b116971cd Author: Abir Viqar <[email protected]> Date: Thu Oct 3 16:44:38 2013 -0400 Porting/corelist-perldelta.pl - Improve version parsing Catch more forms of how updated modules have been listed in past perldelta M Porting/corelist-perldelta.pl commit 5c404e2c7db2b223a9681288c75a963c279640f9 Author: Abir Viqar <[email protected]> Date: Thu Oct 3 16:40:13 2013 -0400 Porting/corelist.pl - DeltaParser: match modules in L<> and C<> Module names have been listed in both M Porting/corelist-perldelta.pl commit b88937b3b612eec078920cec65cfef193797f906 Author: Abir Viqar <[email protected]> Date: Thu Oct 3 16:36:26 2013 -0400 Porting/corelist-perldelta.pl - Remove trailing whitespace M Porting/corelist-perldelta.pl commit 67e92e89d4a2f7b9a3a401ea926f97ce445961a5 Author: Abir Viqar <[email protected]> Date: Sun Oct 6 10:52:21 2013 -0400 Porting/corelist-perldelta.pl - fix usage example lib needs to be included when generating the module changes with the currently build Perl M Porting/corelist-perldelta.pl commit 5e0e6da6f877531a1108852368d8ab0810f8662d Author: Abir Viqar <[email protected]> Date: Thu Oct 3 16:32:42 2013 -0400 Porting/corelist-perldelta.pl - Use Module::CoreList in dist/ Ensure that the latest version of Module::CoreList is used when running the script M Porting/corelist-perldelta.pl ----------------------------------------------------------------------- Summary of changes: AUTHORS | 1 + Porting/corelist-perldelta.pl | 753 ++++++++++++++++++++++++++++++----- Porting/how_to_write_a_perldelta.pod | 43 +- Porting/perl5200delta.pod | 8 +- Porting/perldelta_template.pod | 8 +- Porting/release_managers_guide.pod | 7 +- pod/perldelta.pod | 8 +- 7 files changed, 675 insertions(+), 153 deletions(-) diff --git a/AUTHORS b/AUTHORS index 689303c..428e1d2 100644 --- a/AUTHORS +++ b/AUTHORS @@ -23,6 +23,7 @@ Augustina Blair <[email protected]> Abe Timmerman <[email protected]> Abhijit Menon-Sen <[email protected]> Abigail <[email protected]> +Abir Viqar <[email protected]> Achim Bohnet <[email protected]> Adam Flott <[email protected]> Adam Kennedy <[email protected]> diff --git a/Porting/corelist-perldelta.pl b/Porting/corelist-perldelta.pl index e6fb582..fd0b2e2 100755 --- a/Porting/corelist-perldelta.pl +++ b/Porting/corelist-perldelta.pl @@ -4,14 +4,18 @@ use strict; use warnings; use lib 'Porting'; use Maintainers qw/%Modules/; +use lib 'dist/Module-CoreList/lib'; use Module::CoreList; use Getopt::Long; =head1 USAGE # generate the module changes for the Perl you are currently building - ./perl Porting/corelist-perldelta.pl - + ./perl -Ilib Porting/corelist-perldelta.pl + + # update the module changes for the Perl you are currently building + ./perl -Ilib Porting/corelist-perldelta.pl --mode=update pod/perldelta.pod + # generate a diff between the corelist sections of two perldelta* files: perl Porting/corelist-perldelta.pl --mode=check 5.017001 5.017002 <perl5172delta.pod @@ -25,9 +29,11 @@ The part to check the diff wants to be run with a Perl that has an up-to-date L<Module::CoreList>, but needs the outside L<Algorithm::Diff>. Ideally, the program will be split into two separate programs, one -to generate the text and one to show the diff between the +to generate the text and one to show the diff between the corelist sections of the last perldelta and the next perldelta. +Currently no information about Removed Modules is displayed in any of the +modes. =cut my %sections = ( @@ -44,60 +50,27 @@ my %titles = ( my $deprecated; -#--------------------------------------------------------------------------# - -sub added { - my ($mod, $old_v, $new_v) = @_; - say "=item *\n"; - say "L<$mod> $new_v has been added to the Perl core.\n"; -} - -sub updated { - my ($mod, $old_v, $new_v) = @_; - say "=item *\n"; - say "L<$mod> has been upgraded from version $old_v to $new_v.\n"; - if ( $deprecated->{$mod} ) { - say "NOTE: L<$mod> is deprecated and may be removed from a future version of Perl.\n"; - } -} - -sub removed { - my ($mod, $old_v, $new_v) = @_; - say "=item *\n"; - say "C<$mod> has been removed from the Perl core. Prior version was $old_v.\n"; -} - -sub generate_section { - my ($title, $item_sub, @mods ) = @_; - return unless @mods; - - say "=head2 $title\n"; - say "=over 4\n"; - - for my $tuple ( sort { lc($a->[0]) cmp lc($b->[0]) } @mods ) { - my ($mod,$old_v,$new_v) = @$tuple; - $old_v //= q('undef'); - $new_v //= q('undef'); - $item_sub->($mod, $old_v, $new_v); - } - - say "=back\n"; -} - -#--------------------------------------------------------------------------# - sub run { my %opt = (mode => 'generate'); GetOptions(\%opt, - 'mode|m:s', # 'generate', 'check' + 'mode|m:s', # 'generate', 'check', 'update' ); # by default, compare latest two version in CoreList; my @versions = sort keys %Module::CoreList::version; - my ($old, $new) = (shift @ARGV, shift @ARGV); - $old ||= $versions[-2]; - $new ||= $versions[-1]; + my $old = $versions[-2]; + my $new = $versions[-1]; + + # use the provided versions if present + # @ARGV >=2 means [old_version] [new_version] [path/to/file] + if ( @ARGV >= 2) { + ($old, $new) = (shift @ARGV, shift @ARGV); + die "$old is an invalid version\n" if not exists + $Module::CoreList::version{$old}; + die "$new is an invalid verison\n" if not exists + $Module::CoreList::version{$new}; + } if ( $opt{mode} eq 'generate' ) { do_generate($old => $new); @@ -105,6 +78,9 @@ sub run { elsif ( $opt{mode} eq 'check' ) { do_check(\*ARGV, $old => $new); } + elsif ( $opt{mode} eq 'update' ) { + do_update_existing(shift @ARGV, $old => $new); + } else { die "Unrecognized mode '$opt{mode}'\n"; } @@ -112,35 +88,146 @@ sub run { exit 0; } +# Given two perl versions, it returns a list describing the core distributions that have changed. +# The first three elements are hashrefs corresponding to new, updated, and removed modules +# and are of the form (mostly, see the special remarks about removed): +# 'Distribution Name' => ['Distribution Name', previous version number, current version number] +# where the version number is undef if the distribution did not exist. +# The fourth element is an arrayref of core distribution names of those distribution for which it +# is unknown whether they have changed and therefore need to be manually checked. +# +# In most cases, the distribution name in %Modules corresponds to the module that is representative +# of the distribution as listed in Module::CoreList. However, there are a few distribution names +# that do not correspond to a module. %distToModules has been created which maps the distribution +# name to a representative module. The representative module was chosen by either looking at the +# Makefile of the distribution or by seeing which module the distribution has been traditionally +# listed under in past perldeltas. +# +# There are a few distributions for which there is no single representative module (e.g. libnet). +# These distributions are returned as the last element of the list. +# +# %Modules contains a final key, _PERLLIB, which contains a list of modules that are owned by p5p. +# This list contains modules and pragmata that may also be present in Module::CoreList. +# A list of modules are in the list @unclaimedModules, which were manually listed based on whether +# they were independent modules and whether they have been listed in past perldeltas. +# The pragmata were found by doing something like: +# say for sort grep { $_ eq lc $_ and !exists $Modules{$_}} +# keys %{$Module::CoreList::version{'5.019003'}} +# and manually filtering out pragamata that were already covered. +# +# It is currently not possible to differentiate between a removed module and a removed +# distribution. Therefore, the removed hashref contains every module that has been removed, even if +# the module's corresponding distribution has not been removed. + sub corelist_delta { my ($old, $new) = @_; my $corelist = \%Module::CoreList::version; - + my %changes = Module::CoreList::changes_between( $old, $new ); $deprecated = $Module::CoreList::deprecated{$new}; - my (@new,@deprecated,@removed,@pragmas,@modules); + my $getModifyType = sub { + my $data = shift; + if ( exists $data->{left} and exists $data->{right} ) { + return 'updated'; + } + elsif ( !exists $data->{left} and exists $data->{right} ) { + return 'new'; + } + elsif ( exists $data->{left} and !exists $data->{right} ) { + return 'removed'; + } + return undef; + }; + + my @unclaimedModules = qw/AnyDBM_File B B::Concise B::Deparse Benchmark Class::Struct Config::Extensions DB DBM_Filter Devel::Peek DirHandle DynaLoader English Errno ExtUtils::Embed ExtUtils::Minip ... [675 chars truncated] + my @unclaimedPragmata = qw/_charnames arybase attributes blib bytes charnames deprecate diagnostics encoding feature fields filetest inc::latest integer less locale mro open ops overload overloadin ... [48 chars truncated] + my @unclaimed = (@unclaimedModules, @unclaimedPragmata); + + my %distToModules = ( + 'IO-Compress' => [ + { + 'name' => 'IO-Compress', + 'modification' => $getModifyType->( $changes{'IO::Compress::Base'} ), + 'data' => $changes{'IO::Compress::Base'} + } + ], + 'Locale-Codes' => [ + { + 'name' => 'Locale::Codes', + 'modification' => $getModifyType->( $changes{'Locale::Codes'} ), + 'data' => $changes{'Locale::Codes'} + } + ], + 'PathTools' => [ + { + 'name' => 'File::Spec', + 'modification' => $getModifyType->( $changes{'Cwd'} ), + 'data' => $changes{'Cwd'} + } + ], + 'Scalar-List-Utils' => [ + { + 'name' => 'List::Util', + 'modification' => $getModifyType->( $changes{'List::Util'} ), + 'data' => $changes{'List::Util'} + }, + { + 'name' => 'Scalar::Util', + 'modification' => $getModifyType->( $changes{'Scalar::Util'} ), + 'data' => $changes{'Scalar::Util'} + } + ], + 'Text-Tabs+Wrap' => [ + { + 'name' => 'Text::Tabs', + 'modification' => $getModifyType->( $changes{'Text::Tabs'} ), + 'data' => $changes{'Text::Tabs'} + }, + { + 'name' => 'Text::Wrap', + 'modification' => $getModifyType->( $changes{'Text::Wrap'} ), + 'data' => $changes{'Text::Wrap'} + } + ], + ); + + # structure is (new|removed|updated) => [ [ModuleName, previousVersion, newVersion] ] + my $deltaGrouping = {}; + + # list of distributions listed in %Modules that need to be manually checked because there is no module that represents it + my @manuallyCheck; # %Modules defines what is currently in core for my $k ( keys %Modules ) { - next unless exists $corelist->{$new}{$k}; - my $old_ver = $corelist->{$old}{$k}; - my $new_ver = $corelist->{$new}{$k}; - # in core but not in last corelist - if ( ! exists $corelist->{$old}{$k} ) { - push @new, [$k, undef, $new_ver]; + next if $k eq '_PERLLIB'; #these are taken care of by being listed in @unclaimed + next if Module::CoreList::is_core($k) and !exists $changes{$k}; #modules that have not changed + + my ( $distName, $modifyType, $data ); + + if ( exists $changes{$k} ) { + $distName = $k; + $modifyType = $getModifyType->( $changes{$k} ); + $data = $changes{$k}; } - # otherwise just pragmas or modules - else { - my $old_ver = $corelist->{$old}{$k}; - my $new_ver = $corelist->{$new}{$k}; - next unless defined $old_ver && defined $new_ver && $old_ver ne $new_ver; - my $tuple = [ $k, $old_ver, $new_ver ]; - if ( $k eq lc $k ) { - push @pragmas, $tuple; - } - else { - push @modules, $tuple; + elsif ( exists $distToModules{$k} ) { + # modification will be undef if the distribution has not changed + my @modules = grep { $_->{modification} } @{ $distToModules{$k} }; + for (@modules) { + $deltaGrouping->{ $_->{modification} }->{ $_->{name} } = [ $_->{name}, $_->{data}->{left}, $_->{data}->{right} ]; } + next; + } + else { + push @manuallyCheck, $k and next; + } + + $deltaGrouping->{$modifyType}->{$distName} = [ $distName, $data->{left}, $data->{right} ]; + } + + for my $k (@unclaimed) { + if ( exists $changes{$k} ) { + $deltaGrouping->{ $getModifyType->( $changes{$k} ) }->{$k} = + [ $k, $changes{$k}->{left}, $changes{$k}->{right} ]; } } @@ -150,33 +237,78 @@ sub corelist_delta { # important. That's the best we can do without a historical Maintainers.pl for my $k ( keys %{ $corelist->{$old} } ) { if ( ! exists $corelist->{$new}{$k} ) { - push @removed, [$k, $corelist->{$old}{$k}, undef]; + $deltaGrouping->{'removed'}->{$k} = [ $k, $corelist->{$old}{$k}, undef ]; } } - return (\@new, \@removed, \@pragmas, \@modules); + return ( + \%{ $deltaGrouping->{'new'} }, + \%{ $deltaGrouping->{'removed'} }, + \%{ $deltaGrouping->{'updated'} }, + \@manuallyCheck + ); +} + +# currently does not update the Removed Module section +sub do_update_existing { + my ( $existing, $old, $new ) = @_; + + my ( $added, $removed, $updated, $manuallyCheck ) = corelist_delta( $old => $new ); + if ($manuallyCheck) { + print "It cannot be determined whether the following distributions have changed.\n"; + print "Please check and list accordingly:\n"; + say "\t* $_" for sort @{$manuallyCheck}; + print "\n"; + } + + my $data = { + new => $added, + updated => $updated, + #removed => $removed, ignore removed for now + }; + + my $text = DeltaUpdater::transform_pod( $existing, $data ); + open my $out, '>', $existing or die "can't open perldelta file $existing: $!"; + binmode($out); + print $out $text; + close $out; + say "The New and Updated Modules and Pragamata sections in $existing have been updated"; + say "Please ensure the Removed Modules and Pragmata section is up-to-date"; } sub do_generate { my ($old, $new) = @_; - my ($added, $removed, $pragmas, $modules) = corelist_delta($old => $new); + my ($added, $removed, $updated, $manuallyCheck) = corelist_delta($old => $new); + + if ($manuallyCheck) { + print "\nXXXIt cannot be determined whether the following distributions have changed.\n"; + print "Please check and list accordingly:\n"; + say "\t$_" for @{$manuallyCheck}; + print "\n"; + } - generate_section($titles{new}, \&added, @{ $added }); - generate_section($titles{updated}, \&updated, @{ $pragmas }, @{ $modules }); - generate_section($titles{removed}, \&removed, @{ $removed }); + my $data = { + new => $added, + updated => $updated, + #removed => $removed, ignore removed for now + }; + + say DeltaUpdater::sections_to_pod($data) } sub do_check { my ($in, $old, $new) = @_; my $delta = DeltaParser->new($in); - my ($added, $removed, $pragmas, $modules) = corelist_delta($old => $new); + my ($added, $removed, $updated) = corelist_delta($old => $new); - for my $ck (['new', $delta->new_modules, $added], - ['removed', $delta->removed_modules, $removed], - ['updated', $delta->updated_modules, [@{ $modules }, @{ $pragmas }]]) { + # because of the difficulty in identifying the distribution for removed modules + # don't bother checking them + for my $ck ([ 'new', $delta->new_modules, $added ], + #[ 'removed', $delta->removed_modules, $removed ], + [ 'updated', $delta->updated_modules, $updated ] ) { my @delta = @{ $ck->[1] }; - my @corelist = sort { lc $a->[0] cmp lc $b->[0] } @{ $ck->[2] }; + my @corelist = sort { lc $a->[0] cmp lc $b->[0] } values %{ $ck->[2] }; printf $ck->[0] . ":\n"; @@ -196,9 +328,9 @@ sub do_check { $sep = "---\n"; printf "%d,%dc%d,%d\n", $diff->Get(qw( Min1 Max1 Min2 Max2 )); } - print "< $_\n" for $diff->Items(1); + print "Delta< $_\n" for $diff->Items(1); print $sep; - print "> $_\n" for $diff->Items(2); + print "Corelist> $_\n" for $diff->Items(2); } print "\n"; @@ -206,6 +338,404 @@ sub do_check { } { + + package DeltaUpdater; + use List::Util 'reduce'; + + sub get_section_name_from_heading { + my $heading = shift; + while (my ($key, $expression) = each %sections) { + if ($heading =~ $expression) { + return $titles{$key}; + } + } + die "$heading did not match any section"; + } + + sub is_desired_section_name { + for (values %sections) { + return 1 if $_[0] =~ $_; + } + return 0; + } + + # verify the module and pragmata in the section, changing the stated version if necessary + # this subroutine warns if the module name cannot be parsed or if it is not listed in + # the results returned from corelist_delta() + # + # a side-effect of calling this function is that modules present in the section are + # removed from $data, resulting in $data containing only those modules and pragmata + # that were not listed in the perldelta file. This means we can then pass $data to + # add_to_section() without worrying about filtering out duplicates + sub update_section { + my ( $section, $data, $title ) = @_; + my @items = @{ $section->{items} }; + + for my $item (@items) { + + my $content = $item->{text}; + my $module = $item->{name}; + + #skip dummy items + next if !$module and $content =~ /\s*xx*\s*/i; + + say "Could not parse module name; line is:\n\t$content" and next unless $module; + + if ( !$data->{$title}{$module} ) { + print "$module is not listed as being $title in Module::CoreList.\n"; + print "Ensure Module::CoreList has been updated and\n"; + print "check to see that the distribution is not listed under another name.\n\n"; + next; + } + + if ( $title eq 'new' ) { + my ($new) = $content =~ /(\d[^\s]+)\s+has\s+been.*$/m; + say "Could not parse new version for $module; line is:\n\t$content" and next unless $new; + if ( $data->{$title}{$module}[2] ne $new ) { + say "$module: new version differs; version in pod: $new; version in corelist: " . $data->{$title}{$module}[2]; + } + $content =~ s/\d[^\s]+(\s+has\s+been.*$)/$data->{$title}{$module}[2].$1/me; + } + + elsif ( $title eq 'updated' ) { + my ( $prev, $new ) = $content =~ /from\s+(?:version\s+)?(\d[^\s]+)\s+to\s+(?:version\s+)?(\d[^\s,]+?)(?=[\s,]|\.\s|\.$|$).*/s; + say "Could not parse old and new version for $module; line is:\n\t$content" and next + unless $prev and $new; + if ( $data->{$title}{$module}[1] ne $prev ) { + say "$module: previous version differs; version in pod: $prev; version in corelist: " . $data->{$title}{$module}[1]; + } + if ( $data->{$title}{$module}[2] ne $new ) { + say "$module: new version differs; version in pod: $new; version in corelist: " . $data->{$title}{$module}[2]; + } + $content =~ + s/(from\s+(?:version\s+)?)\d[^\s]+(\s+to\s+(?:version\s+)?)\d[^\s,]+?(?=[\s,]|\.\s|\.$|$)(.*)/$1.$data->{$title}{$module}[1].$2.$data->{$title}{$module}[2].$3/se; + } + + elsif ( $title eq 'removed' ) { + my ($prev) = $content =~ /^.*?was\s+(\d[^\s]+?)/m; + say "Could not parse old version for $module; line is:\n\t$content" and next unless $prev; + if ( $data->{$title}{$module}[1] ne $prev ) { + say "$module: previous version differs; $prev " . $data->{$title}{$module}[1]; + } + $content =~ s/(^.*?was\s+)\d[^\s]+?/$1.$data->{$title}{$module}[1]/me; + } + + delete $data->{$title}{$module}; + $item->{text} = $content; + } + return $section; + } + + # add modules and pragmata present in $data to the section + sub add_to_section { + my ( $section, $data, $title ) = @_; + + #undef is a valid version name in Module::CoreList so supress warnings about concatenating undef values + no warnings 'uninitialized'; + for ( values %{ $data->{$title} } ) { + my ( $mod, $old_v, $new_v ) = @{$_}; + my ( $item, $text ); + + $item = { name => $mod, text => "=item *\n" }; + if ( $title eq 'new' ) { + $text = "L<$mod> $new_v has been added to the Perl core.\n"; + } + + elsif ( $title eq 'updated' ) { + $text = "L<$mod> has been upgraded from version $old_v to $new_v.\n"; + if ( $deprecated->{$mod} ) { + $text .= "NOTE: L<$mod> is deprecated and may be removed from a future version of Perl.\n"; + } + } + + elsif ( $title eq 'removed' ) { + $text = "C<$mod> has been removed from the Perl core. Prior version was $old_v.\n"; + } + + $item->{text} .= "\n$text\n"; + push @{ $section->{items} }, $item; + } + return $section; + } + + sub sort_items_in_section { + my ($section) = @_; + + # if we could not parse the module name, it will be uninitalized + # in sort. This is not a problem as it will just result in these + # sections being placed near the begining of the section + no warnings 'uninitialized'; + $section->{items} = + [ sort { lc $a->{name} cmp lc $b->{name} } @{ $section->{items} } ]; + return $section; + } + + # given a hashref of the form returned by corelist_delta() + # and a hash structured as documented in transform_pod(), it returns + # a pod string representation of the sections, creating sections + # if necessary + sub sections_to_pod { + my ( $data, %sections ) = @_; + my $out = ''; + + for ( + ( + [ 'New Modules and Pragmata', 'new' ], + [ 'Updated Modules and Pragmata', 'updated' ], + [ 'Removed Modules and Pragmata', 'removed' ] + ) + ) + { + my ( $section_name, $title ) = @{$_}; + + my $section = $sections{$section_name} // { + name => $section_name, + preceding_text => "=head2 $_->[0]\n=over 4\n", + following_text => "=back\n", + items => [], + manual => 1 + }; + + $section = update_section( $section, $data, $title ); + $section = add_to_section( $section, $data, $title ); + $section = sort_items_in_section( $section ); + + next if $section->{manual} and scalar @{ $section->{items} } == 0; + + my $items = reduce { no warnings 'once'; $a . $b->{text} } + ( '', @{ $section->{items} } ); + $out .= + ( $section->{preceding_text} // '' ) + . $items + . ( $section->{following_text} // '' ); + } + return $out; + } + + # given a filename corresponding to an existing perldelta file + # and a hashref of the form returned by corelist_delta(), it + # returns a string of the resulting file after the module + # information has been added. + sub transform_pod { + my ( $existing, $data ) = @_; + + # will contain hashrefs corresponding to new, updated and removed + # modules and pragmata keyed by section name + # each section is hashref of the structure + # preceding_text => Text occurring before and including the over + # region containing the list of modules, + # items => [Arrayref of hashrefs corresponding to a module + # entry], + # an entry has the form: + # name => Module name or undef if the name could not be determined + # text => The text of the entry, including the item heading + # + # following_text => Any text not corresponding to a module + # that occurs after the first module + # + # the sections are converted to a pod string by calling sections_to_pod() + my %sections; + + # we are in the Modules_and_Pragmata's section + my $in_Modules_and_Pragmata; + + # we are the Modules_and_Pragmata's section but have not + # encountered any of the desired sections. We use this + # flag to determine whether we should append the text to $out + # or we need to delay appending until the module listings are + # processed and instead append to $append_to_out + my $in_Modules_and_Pragmata_preamble; + + my $done_processing_Modules_and_Pragmata; + + my $current_section; + + # $nested_element_level == 0 : not in an over region, treat lines as text + # $nested_element_level == 1 : presumably in the top over region that + # corresponds to the module listing. Treat + # each item as a module + # $nested_element_level > 1 : we only consider these values when we are in an item + # We treat lines as the text of the current item. + my $nested_element_level = 0; + + my $current_item; + my $need_to_parse_module_name; + + my $out = ''; + my $append_to_out = ''; + + open my $fh, '<', $existing or die "can't open perldelta file $existing: $!"; + binmode($fh); + + while (<$fh>) { + # treat the rest of the file as plain text + if ($done_processing_Modules_and_Pragmata) { + $out .= $_; + next; + } + + elsif ( !$in_Modules_and_Pragmata ) { + # entering Modules and Pragmata + if (/^=head1 Modules and Pragmata/) { + $in_Modules_and_Pragmata = 1; + $in_Modules_and_Pragmata_preamble = 1; + } + $out .= $_; + next; + } + + # leaving Modules and Pragmata + elsif (/^=head1/) { + if ($current_section) { + push @{ $current_section->{items} }, $current_item + if $current_item; + $sections{ $current_section->{name} } = $current_section; + } + $done_processing_Modules_and_Pragmata = 1; + $out .= + sections_to_pod( $data, %sections ) . $append_to_out . $_; + next; + } + + # new section in Modules and Pragmata + elsif (/^=head2 (.*?)$/) { + my $name = $1; + if ($current_section) { + push @{ $current_section->{items} }, $current_item + if $current_item; + $sections{ $current_section->{name} } = $current_section; + undef $current_section; + } + + if ( is_desired_section_name($name) ) { + undef $in_Modules_and_Pragmata_preamble; + if ( $nested_element_level > 0 ) { + die "Unexpected head2 at line no. $."; + } + my $title = get_section_name_from_heading($name); + if ( exists $sections{$title} ) { + die "$name occurred twice at line no. $."; + } + $current_section = {}; + $current_section->{name} = $title; + $current_section->{preceding_text} = $_; + $current_section->{items} = []; + $nested_element_level = 0; + next; + } + + # otherwise treat section as plain text + else { + if ($in_Modules_and_Pragmata_preamble) { + $out .= $_; + } + else { + $append_to_out .= $_; + } + next; + } + } + + elsif ($current_section) { + + # not in an over region + if ( $nested_element_level == 0 ) { + if (/^=over/) { + $nested_element_level++; + } + if ( scalar @{ $current_section->{items} } > 0 ) { + $current_section->{following_text} .= $_; + } + else { + $current_section->{preceding_text} .= $_; + } + next; + } + + if ($current_item) { + if ($need_to_parse_module_name) { + # the item may not have a parsable module name, which means that + # $current_item->{name} will never be defined. + if (/^(?:L|C)<(.+?)>/) { + $current_item->{name} = $1; + undef $need_to_parse_module_name; + } + # =item or =back signals the end of an item + # block, which we handle below + if ( !/^=(?:item|back)/ ) { + $current_item->{text} .= $_; + next; + } + } + # currently in an over region + # treat text inside region as plain text + if ( $nested_element_level > 1 ) { + if (/^=back/) { + $nested_element_level--; + } + elsif (/^=over/) { + $nested_element_level++; + } + $current_item->{text} .= $_; + next; + } + # entering over region + if (/^=over/) { + $nested_element_level++; + $current_item->{text} .= $_; + next; + } + # =item or =back signals the end of an item + # block, which we handle below + if ( !/^=(?:item|back)/ ) { + $current_item->{text} .= $_; + next; + } + } + + if (/^=item \*/) { + push @{ $current_section->{items} }, $current_item + if $current_item; + $current_item = { text => $_ }; + $need_to_parse_module_name = 1; + next; + } + + if (/^=back/) { + push @{ $current_section->{items} }, $current_item + if $current_item; + undef $current_item; + $nested_element_level--; + } + + if ( scalar @{ $current_section->{items} } == 0 ) { + $current_section->{preceding_text} .= $_; + } + else { + $current_section->{following_text} .= $_; + } + next; + } + + # text in Modules and Pragmata not in a head2 region + else { + if ($in_Modules_and_Pragmata_preamble) { + $out .= $_; + } + else { + $append_to_out .= $_; + } + next; + } + } + close $fh; + die 'Never saw Modules and Pragmata section' unless $in_Modules_and_Pragmata; + return $out; + } + +} + +{ package DeltaParser; use Pod::Simple::SimpleTree; @@ -223,6 +753,10 @@ sub do_check { return $self; } + # creates the accessor methods: + # new_modules + # updated_modules + # removed_modules for my $k (keys %sections) { no strict 'refs'; my $m = "${k}_modules"; @@ -232,29 +766,17 @@ sub do_check { sub _parse_delta { my ($self, $pod) = @_; - map { - my ($t, $s) = @{ $_ }; - - # Keep the section title if it has one: - if( $s->[0]->[0] eq 'head2' ) { - #warn "Keeping section title '$s->[0]->[2]'"; - $titles{ $t } = $s->[0]->[2] - if $s->[0]->[2]; - }; - - $self->${\"_parse_${t}_section"}($s) - } map { - my $s = $self->_look_for_section($pod => $sections{$_}) - or die "failed to parse $_ section"; - [$_, $s]; - } keys %sections; - - for my $s (keys %sections) { - my $m = "${s}_modules"; - - $self->{$m} = [sort { - lc $a->[0] cmp lc $b->[0] - } @{ $self->{$m} }]; + my $new_section = $self->_look_for_section( $pod, $sections{new} ); + my $updated_section = $self->_look_for_section( $pod, $sections{updated} ); + my $removed_section = $self->_look_for_section( $pod, $sections{removed} ); + + $self->_parse_new_section($new_section); + $self->_parse_updated_section($updated_section); + $self->_parse_removed_section($removed_section); + + for (qw/new_modules updated_modules removed_modules/) { + $self->{$_} = + [ sort { lc $a->[0] cmp lc $b->[0] } @{ $self->{$_} } ]; } return; @@ -263,6 +785,8 @@ sub do_check { sub _parse_new_section { my ($self, $section) = @_; + $self->{new_modules} = []; + return unless $section; $self->{new_modules} = $self->_parse_section($section => sub { my ($el) = @_; @@ -278,13 +802,25 @@ sub do_check { sub _parse_updated_section { my ($self, $section) = @_; + $self->{updated_modules} = []; + return unless $section; $self->{updated_modules} = $self->_parse_section($section => sub { my ($el) = @_; my ($first, $second) = @{ $el }[2, 3]; my $module = $first->[2]; + + # the regular expression matches the following: + # from VERSION_NUMBER to VERSION_NUMBER + # from VERSION_NUMBER to VERSION_NUMBER. + # from version VERSION_NUMBER to version VERSION_NUMBER. + # from VERSION_NUMBER to VERSION_NUMBER and MODULE from VERSION_NUMBER to VERSION_NUMBER + # from VERSION_NUMBER to VERSION_NUMBER, and MODULE from VERSION_NUMBER to VERSION_NUMBER + # + # some perldeltas contain more than one module listed in an entry, this only attempts to match the + # first module my ($old, $new) = $second =~ - /from\s+(?:version\s+)?(\d[^\s]+)\s+to\s+(\d[^\s]+?)\.?$/; + /from\s+(?:version\s+)?(\d[^\s]+)\s+to\s+(?:version\s+)?(\d[^\s,]+?)(?=[\s,]|\.\s|\.$|$).*/s; warn "Unable to extract old or new version of $module from perldelta" if !defined $old || !defined $new; @@ -297,6 +833,9 @@ sub do_check { sub _parse_removed_section { my ($self, $section) = @_; + + $self->{removed_modules} = []; + return unless $section; $self->{removed_modules} = $self->_parse_section($section => sub { my ($el) = @_; @@ -316,7 +855,7 @@ sub do_check { my ($el) = @_; return unless ref $el && $el->[0] =~ /^item-/ && @{ $el } > 2 && ref $el->[2]; - return unless $el->[2]->[0] eq 'C'; + return unless $el->[2]->[0] =~ /C|L/; return 1; }); @@ -349,7 +888,7 @@ sub do_check { sub { my ($el) = @_; my ($heading) = $el->[0] =~ /^head(\d)$/; - my $f = $heading && $el->[2] =~ /^$section/; + my $f = $heading && $el->[2] =~ /^$section/; $level = $heading if $f && !$level; return $f; }, diff --git a/Porting/how_to_write_a_perldelta.pod b/Porting/how_to_write_a_perldelta.pod index a7bfc4a..7b9e770 100644 --- a/Porting/how_to_write_a_perldelta.pod +++ b/Porting/how_to_write_a_perldelta.pod @@ -126,28 +126,24 @@ source tree. =item Modules and Pragmata -All changes to installed files in F<ext/> and F<lib/> go here, in a list -ordered by distribution name. Minimally it should be the module version, -but it's more useful to the end user to give a paragraph's summary of the -module's changes. In an ideal world, dual-life modules would have a -F<Changes> file that could be cribbed. +All changes to installed files in F<cpan/>, F<dist/>, F<ext/> and F<lib/> go +here, in a list ordered by distribution name. Minimally it should be the +module version, but it's more useful to the end user to give a paragraph's +summary of the module's changes. In an ideal world, dual-life modules would +have a F<Changes> file that could be cribbed. Whilst this section could be built by incrementally working through change descriptions applying to files, this is prone to error. It's better to -collate changes to F<ext/> and F<lib/> by module, and then summarise all -changes to a module as a group. This can be done by partitioning directories -within F<ext/> and F<lib/> to a number of people. - -B<FIXME> - this could be automated better +collate changes by module, and then summarise all changes to a module as a +group. If Module::CoreList has been updated, then F<Porting/corelist-perldelta.pl> -will automatically print out several sections if relevant that can be -pasted into F<perldelta>: +will automatically update two sections in F<perldelta>: * New Modules and Pragmata - * Pragmata Changes - * Updated Modules - * Removed Modules and Pragmata + * Updated Modules and Pragmata + +(Currently, it does not update the Removed Modules and Pragmata section.) Each section will have stub entries following a template like this: @@ -162,23 +158,10 @@ added by hand. A more adventurous enhancement would be to automate grabbing the changelogs for dual lived modules. For each of them, grab the relevant changes files from CPAN for the old and new versions, and if the old one is -a strict subset of the new one, splice the extra lines right into the output, -as a basis for summarising. +a strict subset of the new one, use the extra lines as a basis for summarising. (And if not, experiment with using F<git> to get the relevant part of changelog -for the particular file in core) - -These could also be enhanced further by using a Pod parser module to produce -a parse tree of F<perl${whatever}delta.pod>, and splicing in the updates -correctly without throwing existing entries away. - -If you think that's nuts, take a look at what F<pod/buildtoc> already does to -splice into existing Makefiles on various platforms: - -http://perl5.git.perl.org/perl.git/blob/blead:/pod/buildtoc#l498 - -Perl is this really powerful language for text manipulation. And fun to -play with. We need to get that message out. :-) +for the particular file in core.) =item Utility Changes diff --git a/Porting/perl5200delta.pod b/Porting/perl5200delta.pod index 864a4b6..f4350aa 100644 --- a/Porting/perl5200delta.pod +++ b/Porting/perl5200delta.pod @@ -95,11 +95,9 @@ XXX XXX All changes to installed files in F<cpan/>, F<dist/>, F<ext/> and F<lib/> go here. If Module::CoreList is updated, generate an initial draft of the -following sections using F<Porting/corelist-perldelta.pl>, which prints stub -entries to STDOUT. Results can be pasted in place of the '=head2' entries -below. A paragraph summary for important changes should then be added by hand. -In an ideal world, dual-life modules would have a F<Changes> file that could be -cribbed. +following sections using F<Porting/corelist-perldelta.pl>. A paragraph summary +for important changes should then be added by hand. In an ideal world, +dual-life modules would have a F<Changes> file that could be cribbed. [ Within each section, list entries as a =item entry ] diff --git a/Porting/perldelta_template.pod b/Porting/perldelta_template.pod index bb48abe..ca762fa 100644 --- a/Porting/perldelta_template.pod +++ b/Porting/perldelta_template.pod @@ -95,11 +95,9 @@ XXX XXX All changes to installed files in F<cpan/>, F<dist/>, F<ext/> and F<lib/> go here. If Module::CoreList is updated, generate an initial draft of the -following sections using F<Porting/corelist-perldelta.pl>, which prints stub -entries to STDOUT. Results can be pasted in place of the '=head2' entries -below. A paragraph summary for important changes should then be added by hand. -In an ideal world, dual-life modules would have a F<Changes> file that could be -cribbed. +following sections using F<Porting/corelist-perldelta.pl>. A paragraph summary +for important changes should then be added by hand. In an ideal world, +dual-life modules would have a F<Changes> file that could be cribbed. [ Within each section, list entries as a =item entry ] diff --git a/Porting/release_managers_guide.pod b/Porting/release_managers_guide.pod index 37ccf37..c1eeff1 100644 --- a/Porting/release_managers_guide.pod +++ b/Porting/release_managers_guide.pod @@ -642,7 +642,12 @@ section, which can be generated with something like: $ perl Porting/acknowledgements.pl v5.15.0..HEAD -Fill in the "Updated Modules" section now that Module::CoreList is updated. +Fill in the "New/Updated Modules" sections now that Module::CoreList is updated: + + $ ./perl -Ilib Porting/corelist-perldelta.pl --mode=update pod/perldelta.pod + +Ideally, also fill in a summary of the major changes to each module for which +an entry has been added by F<corelist-perldelta.pl>. Re-read the perldelta to try to find any embarrassing typos and thinkos; remove any C<TODO> or C<XXX> flags; update the "Known Problems" section diff --git a/pod/perldelta.pod b/pod/perldelta.pod index 108c6a7..b8f7c5a 100644 --- a/pod/perldelta.pod +++ b/pod/perldelta.pod @@ -117,11 +117,9 @@ XXX XXX All changes to installed files in F<cpan/>, F<dist/>, F<ext/> and F<lib/> go here. If Module::CoreList is updated, generate an initial draft of the -following sections using F<Porting/corelist-perldelta.pl>, which prints stub -entries to STDOUT. Results can be pasted in place of the '=head2' entries -below. A paragraph summary for important changes should then be added by hand. -In an ideal world, dual-life modules would have a F<Changes> file that could be -cribbed. +following sections using F<Porting/corelist-perldelta.pl>. A paragraph summary +for important changes should then be added by hand. In an ideal world, +dual-life modules would have a F<Changes> file that could be cribbed. [ Within each section, list entries as a =item entry ] -- Perl5 Master Repository
