path: root/CVSROOT/log_accum.pl

#!/usr/bin/perl -w
#
# $FreeBSD$
#
# Perl filter to handle the log messages from the checkin of files in
# a directory.  This script will group the lists of files by log
# message, and mail a single consolidated log message at the end of
# the commit.
#
# This file assumes a pre-commit checking program that leaves the
# names of the first and last commit directories in a temporary file.
#
# Originally by David Hampton <hampton@cisco.com>
#
# Extensively hacked for FreeBSD by Peter Wemm <peter@netplex.com.au>,
#  with parts stolen from Greg Woods <woods@most.wierd.com> version.
#
# Extensively cleaned up and re-worked to use an external configuration
# file by Josef Karthauser <joe@tao.org.uk>.

require 5.003;      # might work with older perl5

use strict;
use lib $ENV{CVSROOT};
use CVSROOT::cfg;


############################################################
#
# Constants
#
############################################################
my $STATE_NONE    = 0;
my $STATE_CHANGED = 1;
my $STATE_ADDED   = 2;
my $STATE_REMOVED = 3;
my $STATE_LOG     = 4;

my $BASE_FN       = "$cfg::TMPDIR/$cfg::FILE_PREFIX";
my $LAST_FILE     = $cfg::LAST_FILE;
my $CHANGED_FILE  = "$BASE_FN.changed";
my $ADDED_FILE    = "$BASE_FN.added";
my $REMOVED_FILE  = "$BASE_FN.removed";
my $LOG_FILE      = "$BASE_FN.log";
my $SUMMARY_FILE  = "$BASE_FN.summary";
my $MAIL_FILE     = "$BASE_FN.mail";
my $SUBJ_FILE     = "$BASE_FN.subj";
my $TAGS_FILE     = "$BASE_FN.tags";

my $CVSROOT       = $ENV{'CVSROOT'} || "/home/ncvs";

############################################################
#
# Subroutines
#
############################################################

# Remove the temporary files.
sub cleanup_tmpfiles {
    my @files;      # The list of temporary files.

    # Don't clean up afterwards if in debug mode.
    return if $cfg::DEBUG;

    opendir DIR, $cfg::TMPDIR or die "Cannot open directory: $cfg::TMPDIR!";
    push @files, grep /^$cfg::FILE_PREFIX\..*$/, readdir(DIR);
    closedir DIR;

    foreach (@files) {
        unlink "$cfg::TMPDIR/$_";
    }
}


# Append a line to a named file.
sub append_line {
    my $filename = shift;   # File to append to.
    my $line = shift;   # Line to append.

    open FILE, ">>$filename" or
        die "Cannot open for append file $filename.\n";
    print FILE "$line\n";
    close FILE;
}


# Read the first line from a named file.
sub read_line {
    my $filename = shift;   # The file to read the line from.

    my $line;       # The line read from the file.

    open FILE, "<$filename" or die "Cannot read file $filename!";
    $line = <FILE>;
    close FILE;
    chomp $line;

    return $line;
}


# Return the contents of a file as a list of strings,
# with trailing line feeds removed.
# Return an empty list of the file couldn't be opened for some reason.
sub read_logfile {
    my $filename = shift;   # The file to read from.

    my @text = ();      # The contents of the file.

    if (open FILE, "<$filename") {
        while (<FILE>) {
            chomp;
            push @text, $_;
        }
        close FILE;
    }

    return @text;
}


# Write a list to a file.
sub write_logfile {
    my $filename = shift;   # File to write to.
    my @lines = @_;     # Contents to write to file.

    open FILE, ">$filename" or
        die "Cannot open for write log file $filename.";
    print FILE join("\n", @lines), "\n";
    close FILE;
}


sub format_names {
    my $dir = shift;
    my @files = @_;

    my $indent = length($dir);
    $indent = 20 if $indent < 20;

    my $format = "    %-" . sprintf("%d", $indent) . "s ";

    my @lines = (sprintf($format, $dir));

    if ($cfg::DEBUG) {
        print STDERR "format_names(): dir = ", $dir;
        #print STDERR "; tag = ", $tag;
        print STDERR "; files = ", join(":", @files), ".\n";
    }

    foreach my $file (@files) {
        if (length($lines[$#lines]) + length($file) > 66) {
            $lines[++$#lines] = sprintf($format, "", "");
        }
        $lines[$#lines] .= $file . " ";
    }

    return @lines;
}


sub format_lists {
    my $header = shift;
    my @lines = @_;

    print STDERR "format_lists(): ", join(":", @lines), "\n" if $cfg::DEBUG;

    my $lastdir = '';
    my $lastsep = '';
    my $tag = '';
    my @files = ();
    my @text = ();
    foreach my $line (@lines) {
        if ($line =~ /.*\/$/) {
            push @text, &format_names($lastdir, @files) if $lastdir;
            @files = ();

            $lastdir = $line;
            $lastdir =~ s,/$,,;

            $tag = "";  # next thing is a tag
        } elsif (!$tag) {
            $tag = $line;
            next if $header . $tag eq $lastsep;

            $lastsep = $header . $tag;
            if ($tag eq 'HEAD') {
                push @text, "  $header files:";
            } else {
                push @text, sprintf("  %-22s (Branch: %s)",
                    "$header files:", $tag);
            }
        } else {
            push @files, $line;
        }
    }
    push @text, &format_names($lastdir, sort @files);

    return @text;
}


sub append_names_to_file {
    my $filename = shift;
    my $dir = shift;
    my $tag = shift;
    my @files = @_;

    return unless @files;

    open FILE, ">>$filename" or die "Cannot append to file $filename.";

    print FILE $dir, "\n";
    print FILE $tag, "\n";
    print FILE join("\n", @files), "\n";
    close FILE;
}


#
# Use cvs status to obtain the current revision number of a given file.
#
sub get_revision_number {
    my $file = shift;

    my $rcsfile = "";
    my $revision = "";

    open(RCS, "-|") || exec 'cvs', '-Qn', 'status', $file;
    while (<RCS>) {
        if (/^[ \t]*Repository revision/) {
            chomp;
            my @revline = split;
            $revision = $revline[2];
            $revline[3] =~ m|^$CVSROOT/+(.*),v$|;
            $rcsfile = $1;
            last;
        }
    }
    close RCS;

    $rcsfile =~ s|/Attic/|/|;   # Remove 'Attic/' if present.
    return($revision, $rcsfile);
}

#
# Count the number of lines in a given revision of a file.
#
sub count_lines_in_revision {
    my $file = shift;   # File in repository.
    my $rev = shift;    # Revision number.

    my $lines = 0;
    open(RCS, "-|") || exec 'cvs', '-Qn', 'update', '-p', "-r$rev", $file;
    while (<RCS>) {
        ++$lines;
    }
    close RCS;

    return $lines;
}


#
# Summarise details of the file modifications.
#
sub change_summary_changed {
    my $outfile = shift;        # File name of output file.
    my @filenames = @_;     # List of files to check.

    foreach my $file (@filenames) {
        next unless $file;

        my $delta = "";
        my ($rev, $rcsfile) = get_revision_number($file);

        if ($rev and $rcsfile) {
            open(RCS, "-|") || exec 'cvs', '-Qn', 'log', "-r$rev", $file;
            while (<RCS>) {
                if (/^date:.*lines:\s(.*)$/) {
                    $delta = $1;
                    last;
                }
            }
            close RCS;
        }

        &append_line($outfile, "$rev,$delta,,$rcsfile");
    }
}

#
# Summarise details of added files.
#
sub change_summary_added {
    my $outfile = shift;        # File name of output file.
    my @filenames = @_;     # List of files to check.

    foreach my $file (@filenames) {
        next unless $file;

        my $delta = "";
        my ($rev, $rcsfile) = get_revision_number($file);

        if ($rev and $rcsfile) {
            my $lines = count_lines_in_revision($file, $rev);
            $delta = "+$lines -0";
        }

        &append_line($outfile, "$rev,$delta,new,$rcsfile");
    }
}

#
# Summarise details of removed files.
#
sub change_summary_removed {
    my $outfile = shift;        # File name of output file.
    my @filenames = @_;     # List of files to check.

    foreach my $file (@filenames) {
        next unless $file;

        my $delta = "";
        my ($rev, $rcsfile) = get_revision_number($file);

        if ($rev and $rcsfile) {
            $rev =~ /(?:(.*)\.)?([^\.]+)\.([^\.]+)$/;
            my ($base, $r1, $r2) = ($1, $2, $3);

            my $prevrev = "";
            if ($r2 == 1) {
                $prevrev = $base;
            } else {
                $prevrev = "$base." if $base;
                $prevrev .= "$r1." . ($r2 - 1);
            }

            my $lines = count_lines_in_revision($file, $prevrev);
            $delta = "+0 -$lines";
        }

        &append_line($outfile, "$rev,$delta,dead,$rcsfile");
    }
}

sub build_header {
    delete $ENV{'TZ'};

    my $datestr = `/bin/date +"%Y/%m/%d %H:%M:%S %Z"`;
    chomp $datestr;

    my $login = $ENV{'USER'} || getlogin ||
        (getpwuid($<))[0] || sprintf("uid#%d",$<);
    my $header = sprintf("%-8s    %s", $login, $datestr);

    my @text;
    push @text, $header;
    push @text, "";
    push @text, "  $cfg::MAILBANNER", "" if $cfg::MAILBANNER;

    return @text;
}

# !!! Mailing-list and commitlog history file mappings here !!!
# This needs pulling out as a configuration block somewhere so
# that others can easily change it.
sub mlist_map {
    my $dir = shift;    # Directory name
   
    return 'cvs-CVSROOT'    if $dir =~ /^CVSROOT\//;
    return 'cvs-ports'  if $dir =~ /^ports\//;
    return 'cvs-www'    if $dir =~ /^www\//;
    return 'cvs-doc'    if $dir =~ /^doc\//;
    return 'cvs-distrib'    if $dir =~ /^distrib\//;

    return 'cvs-other'  unless $dir =~ /^src\//;

    $dir =~ s,^src/,,;

    return 'cvs-bin'    if $dir =~ /^bin\//;
    return 'cvs-contrib'    if $dir =~ /^contrib\//;
    return 'cvs-eBones' if $dir =~ /^eBones\//;
    return 'cvs-etc'    if $dir =~ /^etc\//;
    return 'cvs-games'  if $dir =~ /^games\//;
    return 'cvs-gnu'    if $dir =~ /^gnu\//;
    return 'cvs-include'    if $dir =~ /^include\//;
    return 'cvs-kerberosIV' if $dir =~ /^kerberosIV\//;
    return 'cvs-lib'    if $dir =~ /^lib\//;
    return 'cvs-libexec'    if $dir =~ /^libexec\//;
    return 'cvs-lkm'    if $dir =~ /^lkm\//;
    return 'cvs-release'    if $dir =~ /^release\//;
    return 'cvs-sbin'   if $dir =~ /^sbin\//;
    return 'cvs-share'  if $dir =~ /^share\//;
    return 'cvs-sys'    if $dir =~ /^sys\//;
    return 'cvs-tools'  if $dir =~ /^tools\//;
    return 'cvs-usrbin' if $dir =~ /^usr\.bin\//;
    return 'cvs-usrsbin'    if $dir =~ /^usr\.sbin\//;

    return 'cvs-user';
}    

sub do_changes_file {
    my @text = @_;

    my %unique = ();
    my @mailaddrs = &read_logfile($MAIL_FILE);
    foreach my $category (@mailaddrs) {
        next if ($unique{$category});
        $unique{$category} = 1;
        if ($category =~ /^cvs-/) {
            # convert mailing list name back to category
            chomp $category;
            $category =~ s/^cvs-//;

            my $changes = "$CVSROOT/CVSROOT/commitlogs/$category";
            open CHANGES, ">>$changes"
                or die "Cannot open $changes.\n";
            print CHANGES map { "$_\n" } @text;
            print CHANGES "\n\n\n";
            close CHANGES;
        }
    }
}

sub mail_notification {
    my @text = @_;

    print "Mailing the commit message...\n";

# This is turned off since the To: lines go overboard.
# Also it has bit-rotted since, and can't just be switched on again.
# - but keep it for the time being in case we do something like cvs-stable
#   my @mailaddrs = &read_logfile($MAIL_FILE);
#   print(MAIL 'To: cvs-committers' . $dom . ", cvs-all" . $dom);
#   foreach $line (@mailaddrs) {
#       next if ($unique{$line});
#       $unique{$line} = 1;
#       next if /^cvs-/;
#       print(MAIL ", " . $line . $dom);
#   }
#   print(MAIL "\n");

    my @email = ();

    my $to = $cfg::MAILADDRS;
    push @email, "To: $to" if $cfg::ADD_TO_LINE;

    my $subject = 'Subject: cvs commit:';
    my @subj = &read_logfile($SUBJ_FILE);
    my $subjlines = 0;
    my $subjwords = 0;  # minimum of two "words" per line
    LINE: foreach my $line (@subj) {
        foreach my $word (split(/ /, $line)) {
            if ($subjwords > 2 &&
                length($subject . " " . $word) > 75) {
                if ($subjlines > 2) {
                    $subject .= " ...";
                }
                push @email, $subject;
                if ($subjlines > 2) {
                    $subject = "";
                    last LINE;
                }

                # rfc822 continuation line
                $subject = "        ";
                $subjwords = 0;
                $subjlines++;
            }
            $subject .= " " . $word;
            $subjwords++;
        }
    }
    push @email, $subject if $subject;

    # If required add a header to the mail msg showing
    # which branches were modified during the commit.
    if ($cfg::MAIL_BRANCH_HDR) {
        my %tags = map { $_ => 1 } &read_logfile($TAGS_FILE);
        push @email, $cfg::MAIL_BRANCH_HDR . ": " .
            join(",", sort keys %tags);
    }

    push @email, "";
    push @email, @text;

    # Transform the email message?
    if (defined($cfg::MAIL_TRANSFORM) && $cfg::MAIL_TRANSFORM) {
        die 'log_accum.pl: $cfg::MAIL_TRANSFORM isn\'t a sub!'
            unless ref($cfg::MAIL_TRANSFORM) eq "CODE";

        if ($cfg::DEBUG) {
            print "Email transform.\n";
            print map { "Before: $_\n" } @email;
        }

        @email = &$cfg::MAIL_TRANSFORM(@email);

        print map { "After: $_\n" } @email if $cfg::DEBUG;
    }

    # Send the email.
    open MAIL, "| $cfg::MAILCMD $to"
        or die "Please check $cfg::MAILCMD.";
    print MAIL map { "$_\n" } @email;
    close MAIL;
}

# Return the length of the longest value in the list.
sub longest_value {
    my @values = @_;

    my @sorted = sort { $b <=> $a } map { length $_ } @values;
    return $sorted[0];
}

sub format_summaries {
    my @filenames = @_;

    my @revs;
    my @deltas;
    my @files;
    my @statuses;

    # Parse the summary file.
    foreach my $filename (@filenames) {
        open FILE, $filename or next;
        while (<FILE>) {
            chomp;
            my ($r, $d, $s, $f) = split(/,/, $_, 4);
            push @revs, $r;
            push @deltas, $d;
            push @statuses, $s;
            push @files, $f;
        }
        close FILE;
    }    

    # Format the output, extra spaces after "Changes"
    # to match historic formatting.
    my $r_max = longest_value("Revision", @revs) + 2;
    my $d_max = longest_value("Changes  ", @deltas) + 2;

    my @text;
    my $fmt = "%-" . $r_max . "s%-" . $d_max . "s%s";
    push @text, sprintf $fmt, "Revision", "Changes", "Path";

    my @order = sort { $files[$a] cmp $files[$b] } (0 .. $#revs);
    foreach (@order) {
        my $file = $files[$_];
        my $status = $statuses[$_];
        $file .= " ($status)" if $status;
        push @text, sprintf $fmt, $revs[$_], $deltas[$_], $file;
    }

    return @text;
}

#############################################################
#
# Main Body
#
############################################################

#
# Setup environment
#
umask (002);

#
# Initialize basic variables
#
my $input_params = $ARGV[0];
my ($directory, @filenames) = split " ", $input_params;
#@files = split(' ', $input_params);

my @path = split('/', $directory);
my $dir;
if ($#path == 0) {
    $dir = ".";
} else {
    $dir = join('/', @path[1..$#path]);
}
$dir = $dir . "/";

#
# Throw some values at the developer if in debug mode
#
if ($cfg::DEBUG) {
    print "ARGV      - ", join(":", @ARGV), "\n";
    print "directory - ", $directory, "\n";
    print "filenames - ", join(":", @filenames), "\n";
    print "path      - ", join(":", @path), "\n";
    print "dir       - ", $dir, "\n";
    print "pid       - ", $cfg::PID, "\n";
}

# Was used for To: lines, still used for commitlogs naming.
&append_line($MAIL_FILE, &mlist_map("$directory/"));
&append_line($SUBJ_FILE, "$directory " . join(" ", sort @filenames));

#
# Check for a new directory first.  This will always appear as a
# single item in the argument list, and an empty log message.
#
if ($input_params =~ /New directory/) {
    my @text = &build_header();

    push @text, "  $input_params";
    &do_changes_file(@text);
    &mail_notification(@text) if $cfg::MAIL_ON_DIR_CREATION;
    &cleanup_tmpfiles();
    exit 0;
}

#
# Check for an import command.  This will always appear as a
# single item in the argument list, and a log message.
#
if ($input_params =~ /Imported sources/) {
    my @text = &build_header();

    push @text, "  $input_params";

    while (<STDIN>) {
        chomp;
        push @text, "  $_";
    }

    &do_changes_file(@text);
    &mail_notification(@text);
    &cleanup_tmpfiles();
    exit 0;
}    

#
# Iterate over the body of the message collecting information.
#
my %added_files;        # Hashes containing lists of files
my %changed_files;      # that have been changed, keyed
my %removed_files;      # by branch tag.

my @log_lines;          # The lines of the log message.

my $tag = "HEAD";       # Default branch is HEAD.
my $state = $STATE_NONE;    # Initially in no state.

while (<STDIN>) {
    s/[ \t\n]+$//;      # delete trailing space

    # parse the revision tag if it exists.
    if (/^Revision\/Branch:(.*)/)   { $tag = $1;     next; }
    if (/^[ \t]+Tag: (.*)/)     { $tag = $1;     next; }
    if (/^[ \t]+No tag$/)       { $tag = "HEAD"; next; }

    # check for a state change, guarding against similar markers
    # in the log message itself.
    unless ($state == $STATE_LOG) {
        if (/^Modified Files/)  { $state = $STATE_CHANGED; next; }
        if (/^Added Files/) { $state = $STATE_ADDED;   next; }
        if (/^Removed Files/)   { $state = $STATE_REMOVED; next; }
        if (/^Log Message/) { $state = $STATE_LOG;     next; }
    }

    # don't so anything if we're not in a state.
    next if $state == $STATE_NONE;

    # collect the log line (ignoring empty template entries)?
    if ($state == $STATE_LOG) {
        # XXX this stuff should come from the configuration file.
        unless (
            /^PR:$/i ||
            /^Reviewed by:$/i ||
            /^Submitted by:$/i ||
            /^Obtained from:$/i ||
            /^MFC after:$/i ||
            /^Approved by:$/i) {
            push @log_lines, $_;
        }
        next;
    }

    # otherwise collect information about which files changed.
    my @files = split;
    push @{ $changed_files{$tag} }, @files if $state == $STATE_CHANGED;
    push @{ $added_files{$tag} },   @files if $state == $STATE_ADDED;
    push @{ $removed_files{$tag} }, @files if $state == $STATE_REMOVED;
}
&append_line($TAGS_FILE, $tag);

#
# Strip leading and trailing blank lines from the log message.  Also
# compress multiple blank lines in the body of the message down to a
# single blank line.
# (Note, this only does the mail and changes log, not the rcs log).
#
while ($#log_lines > -1) {
    last if ($log_lines[0] ne "");
    shift(@log_lines);
}
while ($#log_lines > -1) {
    last if ($log_lines[$#log_lines] ne "");
    pop(@log_lines);
}
for (my $l = $#log_lines; $l > 0; $l--) {
    if (($log_lines[$l - 1] eq "") && ($log_lines[$l] eq "")) {
        splice(@log_lines, $l, 1);
    }
}

#
# Find the log file that matches this log message
#
my $message_index;      # The index of this log message
for ($message_index = 0; ; $message_index++) {
    last unless -e "$LOG_FILE.$message_index";

    my @text = &read_logfile("$LOG_FILE.$message_index");
    last if  $#text == -1;
    last if join(" ", @log_lines) eq join(" ", @text);
}

#
# Spit out the information gathered in this pass.
#
foreach my $tag ( keys %added_files ) {
    &append_names_to_file("$ADDED_FILE.$message_index",   $dir, $tag,
        @{ $added_files{$tag} });
}
foreach my $tag ( keys %changed_files ) {
    &append_names_to_file("$CHANGED_FILE.$message_index", $dir, $tag,
        @{ $changed_files{$tag} });
}
foreach my $tag ( keys %removed_files ) {
    &append_names_to_file("$REMOVED_FILE.$message_index", $dir, $tag,
        @{ $removed_files{$tag} });
}
&write_logfile("$LOG_FILE.$message_index", @log_lines);

#
# Save the info for the commit summary.
#
foreach my $tag ( keys %added_files ) {
    &change_summary_added("$SUMMARY_FILE.$message_index",
        @{ $added_files{$tag} });
}
foreach my $tag ( keys %changed_files ) {
    &change_summary_changed("$SUMMARY_FILE.$message_index",
        @{ $changed_files{$tag} });
}
foreach my $tag ( keys %removed_files ) {
    &change_summary_removed("$SUMMARY_FILE.$message_index",
        @{ $removed_files{$tag} });
}

#
# Check whether this is the last directory.  If not, quit.
# The last directory name was written by commit_prep.pl on
# the way in.
#
if (-e $LAST_FILE) {
    $_ = &read_line($LAST_FILE);
    my $tmpfiles = $directory;
    $tmpfiles =~ s,([^a-zA-Z0-9_/]),\\$1,g;
    if (! grep(/$tmpfiles$/, $_)) {
        print "More commits to come...\n";
        exit 0
    }
}

#
# This is it.  The commits are all finished.  Lump everything together
# into a single message, fire a copy off to the mailing list, and drop
# it on the end of the Changes file.
#

#
# Produce the final compilation of the log messages
#
for (my $i = 0; ; $i++) {
    last unless -e "$LOG_FILE.$i";

    my @log_msg = &build_header();

    my @mod_lines = &read_logfile("$CHANGED_FILE.$i");
    push @log_msg, &format_lists("Modified", @mod_lines) if @mod_lines;

    my @add_lines = &read_logfile("$ADDED_FILE.$i");
    push @log_msg, &format_lists("Added", @add_lines) if @add_lines;

    my @rem_lines = &read_logfile("$REMOVED_FILE.$i");
    push @log_msg, &format_lists("Removed", @rem_lines) if @rem_lines;

    my @msg_lines = &read_logfile("$LOG_FILE.$i");
    push @log_msg, "  Log:", (map { "  $_" } @msg_lines) if @msg_lines;


    if (-e "$SUMMARY_FILE.$i") {
        push @log_msg, "  ", map {"  $_"}
            format_summaries("$SUMMARY_FILE.$i");
    }

    #
    # Add a copy of the message in the relevant log files.
    #
    &do_changes_file(@log_msg);

    #
    # Mail out the notification.
    #
    &mail_notification(@log_msg);
}

&cleanup_tmpfiles();
exit 0;
# EOF