#! /usr/bin/perl
# $Id: grep-changelog,v 1.2 2000/05/05 13:19:05 gerd Exp $

# Copyright (C) 1999, 2000 Free Software Foundation, Inc.
#
# This file is part of GNU Emacs.
#
# GNU Emacs is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
#
# GNU Emacs is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with GNU Emacs; see the file COPYING.  If not, write to the
# Free Software Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.


# Extract entries from ChangeLogs matching specified criteria.
# Optionally format the resulting output to a form suitable for RCS
# logs, like they are used in Emacs, for example.  In this format,
# author lines leading spaces, and file names are removed.

require 5;

# Parse command line options.

use Getopt::Long;
$result = GetOptions ("author=s" => \$author,
		      "text=s"  => \$regexp,
		      "exclude=s"  => \$exclude,
		      "from-date=s" => \$from_date,
		      "to-date=s" => \$to_date,
		      "rcs-log" => \$rcs_log,
		      "with-date" => \$with_date,
		      "version" => \$version,
		      "help"    => \$help);

# If date options are specified, check that they have the format
# YYYY-MM-DD.

$result = 0 if $from_date && $from_date !~ /^\d\d\d\d-\d\d-\d\d$/;
$result = 0 if $to_date && $to_date !~ /^\d\d\d\d-\d\d-\d\d$/;

# Print usage information and exit when necessary.

if ($result == 0 || $help) {
    print <<USAGE;
Usage: $0 [options] [CHANGELOG...]
Print entries in ChangeLogs matching various criteria.  Valid options
are

  --author=AUTHOR         match entries whose author line matches
			  regular expression AUTHOR
  --text=TEXT             match entries whose text matches regular
			  expression TEXT.
  --exclude=TEXT	  exclude entries matching TEXT.
  --from-date=YYYY-MM-DD  match entries not older than given date
  --to-date=YYYY-MM-DD    match entries not younger than given date
  --rcs-log		  format output suitable for RCS log entries.
  --with-date		  print short date line in RCS log
  --version		  print version info
  --help		  print this help

If no CHANGELOG is specified scan the files "ChangeLog" and
"ChangeLog.[9-1]" in the current directory.  Old-style dates in ChangeLogs
are not recognized.
USAGE
    exit $help ? 0 : 1;
}

# Print version info and exit if `--version' was specified.

if ($version) {
    print "0.1\n";
    exit 0;
}


# Value is non-zero if HEADER matches according to command line
# options specified, i.e. it matches $author, and its date is in
# the range $from_date <= date <= $to_date.

sub header_match_p ($) {
    my $header = shift;

    # No match if AUTHOR-regexp specified and doesn't match.
    return 0 if $author && $header !~ /$author/;

    # Check that the date of the entry matches if date options
    # `--from-date' and/or `--to-date' were specified .  Old-style
    # dates in ChangeLogs are not recognized, and never match.
    if ($from_date || $to_date) {
	if ($header =~ /^(\d\d\d\d-\d\d-\d\d)/) {
	    my $date = $1;
	    return 0 if $from_date && $date lt $from_date;
	    return 0 if $to_date && $date gt $to_date;
	} else {
	    # Don't bother recognizing old-style dates.
	    return 0;
	}
    }

    return 1;
}


# Value is non-zero if ENTRY matches the criteria specified on the
# command line, i.e. it matches $regexp, and it doesn't match
# $exclude.

sub entry_match_p ($) {
    my $entry = shift;

    if ($regexp) {
	return 1 if ($entry =~ /$regexp/
		     && (!$exclude || $entry !~ $exclude));
    } else {
	return 1 if !$exclude || $entry !~ $exclude;
    }

    return 0;
}


# Print HEADER and/or ENTRY in a format suitable for what was
# specified on the command line.  If $rcs_log is specified, author
# lines are not printed, and leading spaces and file names are removed
# from ChangeLog entries.

sub print_log ($$) {
    my ($header, $entry) = @_;

    if ($rcs_log) {
	# Remove leading whitespace from entry.
	$entry =~ s/^\s+//mg;
	# Remove file name parts.
	$entry =~ s/^\*.*\(/(/mg;
	# Remove file name parts, 2.
	$entry =~ s/^\*.*://mg;
        if ($with_date) {
	    $header =~ /(\d\d\d\d-\d\d-\d\d)/;
	    print "!changelog-date $1\n";
	}
	print $entry;
    } else {
	print $header, $entry;
    }
}

# Scan LOG for matching entries, and print them to standard output.

sub parse_changelog ($) {
    my $log = shift;
    my $entry;
    my $match;

    # Open the ChangeLog.
    open (IN, "< $log") || die "Cannot open $log: $!";

    while ($line = <IN>) {
	if ($line =~ /^\S/) {
	    # Line is an author-line.  Print previous entry if
	    # it matches.
	    print_log ($header, $entry)
		if header_match_p ($header) && entry_match_p ($entry);

	    $entry = "";
  	    $header = $line;

	    # Add empty lines below the header.
	    while (($line = <IN>) && $line =~ /^\s*$/) {
	        $header = "$header$line";
	    }
        }

	if ($line =~ /^\s*\*/) {
	    # LINE is the first line of a ChangeLog entry.  Print
	    # previous entry if it matches.
	    print_log ($header, $entry)
		if header_match_p ($header) && entry_match_p ($entry);
	    $entry = $line;
	} else {
	    # Add LINE to the current entry.
	    $entry = "$entry$line";
	}
    }

    # Print last entry if it matches.
    print_log ($header, $entry)
	if header_match_p ($header) && entry_match_p ($entry);

    close IN;
}


# Main program.  Process ChangeLogs.

if (@ARGV > 0) {
    # If files were specified on the command line, parse those files.
    while ($log = shift @ARGV) {
	parse_changelog ($log);
    }
} else {
    # Parse default files ChangeLog and ChangeLog.9...ChangeLog.1 in
    # that order.
    parse_changelog ("ChangeLog");
    for ($i = 9; $i >= 1; --$i) {
	my $log = "ChangeLog.$i";
	parse_changelog ($log) if -f $log;
    }
}


# grep-changelog ends here.