hoi :)
I just tested my little script that can send changesets per mail.
okok, it still had a bug when I first tested it but that should be
fixed now.
If anyone is interested (perhaps for Documentation/BK-usage), here it
is:
#!/usr/bin/perl -w
# after sending an announcement (created by Documentation/BK-usage/bk-make-sum)
# just pipe your mail through this script.
# It will create one new mail per Changeset, properly threaded.
# Copyright ? 2005 Martin Waitz <[email protected]>
use strict;
my $from;
my $to;
my $cc;
my $references;
# all local repositories are in ~/src/.
# you have to adjust this function if you keep them elsewhere.
sub local_repository($) {
my $repo;
$repo= shift;
$repo =~ s,.*/,"$ENV{HOME}/src/",e;
return $repo;
}
# this checks if we are allowed to send mails with this sender
# please modify the regexp to check for your adress!
sub check_from($) {
my $from = shift;
exit 1 unless $from =~ /insert-your-email-here/; #FIXME
}
# send one changeset.
# Parameters: the cset number, description prefix and the actual description.
sub send_cset($$$$) {
my ($cset, $serial, $desc, $longdesc) = @_;
open (MAIL, "| /usr/sbin/sendmail -t") or die "fork sendmail: $!";
print MAIL "From: $from\n";
print MAIL "To: $to\n";
print MAIL "Cc: $cc\n" if $cc;
print MAIL "References: $references\n" if $references;
print MAIL "Subject: [PATCH $serial] $desc\n";
print MAIL "\n";
print MAIL "$desc\n";
print MAIL "$longdesc\n";
print MAIL "\n";
print MAIL `bk export -tpatch -du -r $cset`;
close (MAIL) or die "could not send mail: error code $?";
}
# Parse header
while (<>) {
chomp;
last if /^$/;
if (/^From:\s+(.+)$/i) {
$from = $1;
} elsif (/^To:\s+(.+)$/i) {
$to = $1;
} elsif (/^Cc:\s+(.+)$/i) {
$cc = $1;
} elsif (/^Message-Id:\s+(.+)$/i) {
$references = $1;
}
}
my @cset;
my @desc;
my @longdesc;
# Parse body
my $cset;
while (<>) {
chomp;
if (/^\s+bk pull (\S+)$/) {
chdir local_repository($1) ||
die "could not find local repository for $1\n";
} elsif (/^<\S+> \(\S+ ([\d.]+)\)$/) {
# start of new ChangeSet
$cset = $1;
} elsif (/^ (.*)$/ && $cset) {
# first line of a description
unshift @cset, $cset;
unshift @desc, $1;
unshift @longdesc, "";
$cset = "";
} elsif (/^ (.*)$/) {
# next lines of description
$longdesc[0] .= "$1\n";
}
}
check_from($from);
my $i;
my $count;
$count = scalar @cset;
foreach $i (1 .. $count) {
print "$desc[$i-1]\n";
send_cset($cset[$i-1], "$i/$count", $desc[$i-1], $longdesc[$i-1]);
# let sendmail process the mail...
sleep 3;
}
exit(0);
--
Martin Waitz
Martin Waitz wrote:
> hoi :)
>
> I just tested my little script that can send changesets per mail.
> okok, it still had a bug when I first tested it but that should be
> fixed now.
>
> If anyone is interested (perhaps for Documentation/BK-usage), here it
> is:
Putting this in Documentation/BK-usage would be fine.
> #!/usr/bin/perl -w
>
> # after sending an announcement (created by Documentation/BK-usage/bk-make-sum)
> # just pipe your mail through this script.
> # It will create one new mail per Changeset, properly threaded.
>
> # Copyright ? 2005 Martin Waitz <[email protected]>
>
> use strict;
>
> my $from;
> my $to;
> my $cc;
> my $references;
>
> # all local repositories are in ~/src/.
> # you have to adjust this function if you keep them elsewhere.
> sub local_repository($) {
> my $repo;
>
> $repo= shift;
>
> $repo =~ s,.*/,"$ENV{HOME}/src/",e;
> return $repo;
> }
>
> # this checks if we are allowed to send mails with this sender
> # please modify the regexp to check for your adress!
> sub check_from($) {
> my $from = shift;
>
> exit 1 unless $from =~ /insert-your-email-here/; #FIXME
> }
Move 'insert email here' into a default variable, a variable that can be
overridden by an environment variable.
> # send one changeset.
> # Parameters: the cset number, description prefix and the actual description.
> sub send_cset($$$$) {
> my ($cset, $serial, $desc, $longdesc) = @_;
>
> open (MAIL, "| /usr/sbin/sendmail -t") or die "fork sendmail: $!";
> print MAIL "From: $from\n";
> print MAIL "To: $to\n";
> print MAIL "Cc: $cc\n" if $cc;
> print MAIL "References: $references\n" if $references;
> print MAIL "Subject: [PATCH $serial] $desc\n";
> print MAIL "\n";
> print MAIL "$desc\n";
> print MAIL "$longdesc\n";
> print MAIL "\n";
> print MAIL `bk export -tpatch -du -r $cset`;
> close (MAIL) or die "could not send mail: error code $?";
I would suggest '-hdu' to avoid the patch header, but some may disagree.
> # Parse header
> while (<>) {
> chomp;
> last if /^$/;
>
> if (/^From:\s+(.+)$/i) {
> $from = $1;
> } elsif (/^To:\s+(.+)$/i) {
> $to = $1;
> } elsif (/^Cc:\s+(.+)$/i) {
> $cc = $1;
> } elsif (/^Message-Id:\s+(.+)$/i) {
> $references = $1;
> }
note that this misses multi-line headers. multi-line headers are those
where the second, and succeeding lines begin with whitespace.
On Thu, Mar 03, 2005 at 11:59:50AM +0100, Martin Waitz wrote:
> hoi :)
>
> I just tested my little script that can send changesets per mail.
> okok, it still had a bug when I first tested it but that should be
> fixed now.
Greg has a similar script - could you take a look and tell
which is the better (and why). We only want one in the kernel.
It is attached.
Sam
send_lots_of_emails.pl:
#!/usr/bin/perl -w
# horrible hack of a script to send off a large number of email messages, one after
# each other, all chained together. This is useful for large numbers of patches.
#
# Use at your own risk!!!!
#
# greg kroah-hartman Jan 8, 2002
# <[email protected]>a
#
# updated to give a valid subject and CC the owner of the patch - Jan 2005
#
# change this to your email address.
$from = "SOMEONE <someone\@somewhere.com>";
# modify these options each time you run the script
$to = '[email protected]';
$initial_reply_to = '<[email protected]>';
$initial_subject = "[PATCH] XXX fixes for 2.6.11-rc3";
@files = (
"rev-1.2041.patch",
"rev-1.2042.patch",
"rev-1.2043.patch",
"rev-1.2044.patch",
"rev-1.2045.patch",
"rev-1.2046.patch",
);
use Mail::Sendmail;
# we make a "fake" message id by taking the current number
# of seconds since the beginning of Unix time and tacking on
# a random number to the end, in case we are called quicker than
# 1 second since the last time we were called.
sub make_message_id
{
my $date = `date "+\%s"`;
chomp($date);
my $pseudo_rand = int (rand(4200));
$message_id = "<$date$pseudo_rand\@kroah.com>";
print "new message id = $message_id\n";
}
$cc = "";
sub send_message
{
%mail = ( To => $to,
From => $from,
CC => $cc,
Subject => $subject,
Message => $message,
'Reply-to' => "Greg K-H <greg\@kroah.com>",
'In-Reply-To' => $reply_to,
'Message-ID' => $message_id,
'X-Mailer' => "gregkh_patchbomb",
);
$mail{smtp} = 'localhost';
sendmail(%mail) or die $Mail::Sendmail::error;
print "OK. Log says:\n", $Mail::Sendmail::log;
print "\n\n"
}
$reply_to = $initial_reply_to;
make_message_id();
$subject = $initial_subject;
foreach $t (@files) {
$F = $t;
open F or die "can't open file $t";
# first line is the CC: list
$cc = <F>;
print "cc: $cc";
# second line is the Subject:
$subject = <F>;
print "subject: $subject";
undef $/;
$message = <F>; # slurp the whole file in
close F;
$/ = "\n";
send_message();
# set up for the next message
$reply_to = $message_id;
make_message_id();
# $subject = "Re: ".$initial_subject;
}
Sam Ravnborg <[email protected]> wrote:
> sub make_message_id
> {
> my $date = `date "+\%s"`;
> chomp($date);
> my $pseudo_rand = int (rand(4200));
> $message_id = "<$date$pseudo_rand\@kroah.com>";
> print "new message id = $message_id\n";
> }
Why not
use Email::MessageID;
use Net::Domain qw(hostfqdn);
sub make_message_id()
{
return Email::MessageID->new(host => hostfqdn());
}