You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@spamassassin.apache.org by da...@chaosreigns.com on 2011/12/01 18:58:13 UTC
Adding a blacklist via sa-update - would you mind?
There is some question among spamassassin developers* on whether or not
it is acceptable to increase the network load of spamassassin, by one
DNS query per email, for existing releases (version 3.3.x), by adding
one DNS blacklist to the rule set via sa-update.
This Mailspike blacklist has proved useful and reliable in (ruleqa) testing
for the last two years.
Responses that would be particularly useful:
1) I run spamassassin on a very high volume of email, and the addition
of 1 DNS query per email would cause me more problems than the
increased accuracy would be worth. And I don't already have all
network or DNS tests disabled.
2) I run spamassassin on a very high volume of email, and the addition
of 1 DNS query per email would not bother me.
If it does get added, it will be following a post to this list. If you
want to ensure you do not use this blacklist if it does get added, you can
preemptively disable it with:
score RCVD_IN_MSPIKE_ZBI 0
score RCVD_IN_MSPIKE_L5 0
score RCVD_IN_MSPIKE_L4 0
score RCVD_IN_MSPIKE_L3 0
score RCVD_IN_MSPIKE_L2 0
score RCVD_IN_MSPIKE_H2 0
score RCVD_IN_MSPIKE_H3 0
score RCVD_IN_MSPIKE_H4 0
score RCVD_IN_MSPIKE_H5 0
score RCVD_IN_MSPIKE_BL 0
score RCVD_IN_MSPIKE_WL 0
Latest RuleQA results:
http://ruleqa.spamassassin.org/?daterev=20111126&rule=%2Fmspike
The rule I'm most interested in is RCVD_IN_MSPIKE_BL, hitting 74.6% of
spam, and 0.0067% of ham (9 out of 135,096). Which places it as the
6th highest ranked rule, which is very impressive. Although these rules
are not using "reuse", which causes some bias.
MSECS SPAM% HAM% S/O RANK SCORE NAME WHO/AGE
0 82.8451 0.0044 1.000 1.00 0.00 RCVD_IN_XBL
0 76.6509 0.0007 1.000 1.00 0.00 URIBL_AB_SURBL
0 82.8057 0.0044 1.000 1.00 0.00 RAZOR2_CF_RANGE_E8_51_100
0 82.8057 0.0096 1.000 0.99 0.00 RAZOR2_CF_RANGE_51_100
0 73.0507 0.0037 1.000 0.99 0.00 DIGEST_MULTIPLE
0 74.6530 0.0067 1.000 0.99 0.00 T_RCVD_IN_MSPIKE_BL
Mailspike has assured us that they will not cause false positives due to
detecting high hit rates, and that they have "a DNSBL mirror network,
which is on standby and ready to kick in, in case we start getting too
much traffic."
So worst case, for some unforeseen reason Mailspike's network
catastrophically fails, and you get no DNS responses, resulting in no
impact on your spamassassin scores. And we remove it via sa-update 1 day
later.
One of the problems with this question is that the answers we can get on
this mailing list really don't matter. If you're reading this email, you
can disable the rule before it's added, and it won't affect you. So I can
only hope that you can provide some idea of how people not reading this
list or the dev list would feel. Are there people processing so much email
that they could be negatively impacted by 1 additional DNS query per email
who are not actively reading either of these lists?
We would prefer to only add this blacklist to future releases, version
3.4.x+. But we currently do not have a way to maintain two separate sets
of rule updates. So our options are:
1) Never add new blacklists. Not actually an option.
2) Add new blacklists only to 3.4.x+ releases, only calculate optimal
scores for 3.4.x+, and just exclude the new rules from 3.3.x releases
resulting in suboptimal 3.3.x scores. This seems to be the prevailing
option. Theoretically possible this could cause significant increase in
missed spam, actual likelihood is unknown.
3) Add new blacklists uniformly to all 3.3.x+ releases.
4) Develop a way to maintain two separate rule sets. May, realistically,
be impossible.
5) Add new blacklists to 3.4.x rules, and stop providing 3.3.x updates. We
can't expect people to just cut over suddenly like that.
Bug discussing adding this DNSBL:
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6400
If you're wondering "what about releases older than 3.3.x?", the answer is:
We stopped expecting releases that old to be useful a while ago. Please
upgrade.
* This might sound like implication that I'm a spamassassin developer. I'm
not. I do not have commit access.
--
"We will be dead soon. Is this how we want to live?"
http://www.ChaosReigns.com
Re: Adding a blacklist via sa-update - would you mind?
Posted by "Rolf E. Sonneveld" <R....@sonnection.nl>.
On 12/5/11 10:32 PM, David F. Skoll wrote:
> On Mon, 5 Dec 2011 16:27:23 -0500
> darxus@chaosreigns.com wrote:
>
>> I asked in the bug how people felt about posting to the users list
>> about it. I got no objections.
>> The complete lack of responses from anybody but you in the last 4 days
>> seems to be a loud and clear indication that nobody cares, and we
>> should go ahead with adding this useful blacklist.
> Ah... I didn't object because I don't use sa-update. Nevertheless, if
> I did use sa-update, I think I'd find the addition of a new DNSBL
> slightly disconcerting.
>
> I'm not very familiar with sa-update, but perhaps it should have
> seperate channels similar to Debian's "stable" and "testing" releases.
> A new DNSBL (IMO) would definitely be a "testing" release. End-users
> could then subscribe to whichever channel suited their risk profiles.
+1
BTW, I do use sa-update for one of my installations.
/rolf
Re: Adding a blacklist via sa-update - would you mind?
Posted by "Kevin A. McGrail" <KM...@PCCC.com>.
>> The masscheck portion of preparing the rules for publishing.
> Is that hard to automate? If it has to be run once for
> updates.spamassassin.org, is it difficult to run it twice?
> [I really know very little about the back-end SA infrastructure.]
>
> I guess you'd want to make sure that testing.updates included
> everything that update does.
Well, once I get the current updates working, I'll let you know. ;-)
But I think it's svn credential related because I used svn to commit and
I think I removed a stored credentials and that's why updates are broken
at the moment.
Right now, I'm pretty much for least path of resistance to get stuff out
the door.
HOWEVER, here's a brilliant idea. We create an "ultrasafe" rule channel
or something like that. We put scores that disable the new RBL in that
and they can pull sa-update default + the ultrasafe channels and
combined they disable all new RBLs. Then masscheck remains as-is, etc.
Regards,
KAM
Re: Adding a blacklist via sa-update - would you mind?
Posted by "David F. Skoll" <df...@roaringpenguin.com>.
On Tue, 06 Dec 2011 15:20:15 -0500
"Kevin A. McGrail" <KM...@PCCC.com> wrote:
> The masscheck portion of preparing the rules for publishing.
Is that hard to automate? If it has to be run once for
updates.spamassassin.org, is it difficult to run it twice?
[I really know very little about the back-end SA infrastructure.]
I guess you'd want to make sure that testing.updates included
everything that update does.
Regards,
David.
Re: Adding a blacklist via sa-update - would you mind?
Posted by "Kevin A. McGrail" <KM...@PCCC.com>.
On 12/6/2011 3:11 PM, David F. Skoll wrote:
> On Tue, 06 Dec 2011 14:45:49 -0500
> "Kevin A. McGrail"<KM...@PCCC.com> wrote:
>
>>>> I'm not very familiar with sa-update, but perhaps it should have
>>>> seperate channels similar to Debian's "stable" and "testing"
>>>> releases. A new DNSBL (IMO) would definitely be a "testing"
>>>> release. End-users could then subscribe to whichever channel
>>>> suited their risk profiles.
>> -1 unless someone is going to volunteer to set it up.
> Is there much to set up? sa-update already has a --channel argument.
> It seems to me all you need to do is put new DNSBLs in a testing
> channel. People who want them could include
> "--channel testing.updates.spamassassin.org" or whatever in their
> sa-update invocation.
>
> Am I missing something important?
The masscheck portion of preparing the rules for publishing.
Re: Adding a blacklist via sa-update - would you mind?
Posted by "David F. Skoll" <df...@roaringpenguin.com>.
On Tue, 06 Dec 2011 22:01:44 +0100
Karsten Bräckelmann <gu...@rudersport.de> wrote:
> On Tue, 2011-12-06 at 15:11 -0500, David F. Skoll wrote:
> > Is there much to set up? sa-update already has a --channel
> > argument. It seems to me all you need to do is put new DNSBLs in a
> > testing channel.
> New, let alone worthwhile, DNSBLs don't just pop up like that.
That's not really the point... I was trying to make the point that
having a separate channel for somewhat-experimental rules or rules
that might have a large performance impact might be a good idea. A
new RBL is (IMO) simply one example of seomthing not quite safe enough
for the main sa-update channel.
Regards,
David.
Re: Adding a blacklist via sa-update - would you mind?
Posted by Karsten Bräckelmann <gu...@rudersport.de>.
On Tue, 2011-12-06 at 15:11 -0500, David F. Skoll wrote:
> Is there much to set up? sa-update already has a --channel argument.
> It seems to me all you need to do is put new DNSBLs in a testing
> channel.
New, let alone worthwhile, DNSBLs don't just pop up like that.
If anyone wants to include a new DNSBL that shows good results, it's
trivial to do. Just get the rules from SVN. Copy-n-paste available, as
well as anonymous SVN access.
--
char *t="\10pse\0r\0dtu\0.@ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4";
main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1:
(c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}
Re: Adding a blacklist via sa-update - would you mind?
Posted by "David F. Skoll" <df...@roaringpenguin.com>.
On Tue, 06 Dec 2011 14:45:49 -0500
"Kevin A. McGrail" <KM...@PCCC.com> wrote:
> >> I'm not very familiar with sa-update, but perhaps it should have
> >> seperate channels similar to Debian's "stable" and "testing"
> >> releases. A new DNSBL (IMO) would definitely be a "testing"
> >> release. End-users could then subscribe to whichever channel
> >> suited their risk profiles.
> -1 unless someone is going to volunteer to set it up.
Is there much to set up? sa-update already has a --channel argument.
It seems to me all you need to do is put new DNSBLs in a testing
channel. People who want them could include
"--channel testing.updates.spamassassin.org" or whatever in their
sa-update invocation.
Am I missing something important?
Regards,
David.
Re: Adding a blacklist via sa-update - would you mind?
Posted by "Kevin A. McGrail" <KM...@PCCC.com>.
On 12/6/2011 2:27 PM, Dave Pooser wrote:
>> I'm not very familiar with sa-update, but perhaps it should have
>> seperate channels similar to Debian's "stable" and "testing" releases.
>> A new DNSBL (IMO) would definitely be a "testing" release. End-users
>> could then subscribe to whichever channel suited their risk profiles.
> +1
-1 unless someone is going to volunteer to set it up.
Re: Adding a blacklist via sa-update - would you mind?
Posted by Dave Pooser <da...@pooserville.com>.
>I'm not very familiar with sa-update, but perhaps it should have
>seperate channels similar to Debian's "stable" and "testing" releases.
>A new DNSBL (IMO) would definitely be a "testing" release. End-users
>could then subscribe to whichever channel suited their risk profiles.
+1
--
Dave Pooser
Cat-Herder-in-Chief, Pooserville.com
Re: Adding a blacklist via sa-update - would you mind?
Posted by "David F. Skoll" <df...@roaringpenguin.com>.
On Mon, 5 Dec 2011 16:27:23 -0500
darxus@chaosreigns.com wrote:
> I asked in the bug how people felt about posting to the users list
> about it. I got no objections.
> The complete lack of responses from anybody but you in the last 4 days
> seems to be a loud and clear indication that nobody cares, and we
> should go ahead with adding this useful blacklist.
Ah... I didn't object because I don't use sa-update. Nevertheless, if
I did use sa-update, I think I'd find the addition of a new DNSBL
slightly disconcerting.
I'm not very familiar with sa-update, but perhaps it should have
seperate channels similar to Debian's "stable" and "testing" releases.
A new DNSBL (IMO) would definitely be a "testing" release. End-users
could then subscribe to whichever channel suited their risk profiles.
Regards,
David.
RE: Adding a blacklist via sa-update - would you mind?
Posted by Brian Bebeau <BB...@trustwave.com>.
> On Thu, 2011-12-01 at 12:58 -0500, darxus@chaosreigns.com wrote:
>> Since you keep stressing the one, 1, single DNS query per message, I
>> cannot help but nit-pick -- please do have a look at the rules you're
>> talking about. Or maybe, just re-read your own comment 1 on bug 6400.
>> It's not a single query.
> Yup, sorry. I made the mistake of assuming the mailspike instructions would included everything, but turns out it's just the blacklist:
> http://mailspike.org/usage.html
> Only part I care about is the blacklist.
I would object to automatically including this. We process over a million emails a day.
The usage page is not too clear if they would mind this load for just the blacklist, but
I suspect not. In order to turn it off, I'd have to edit local.cf, make a new package,
and get IT to schedule a deployment for it to a large number of scanners. If this
went through, we'd need a good amount of notice to put that in place.
--
Brian Bebeau
Security Researcher
Spiderlabs Research
Trustwave
bbebeau@trustwave.com
This transmission may contain information that is privileged, confidential, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format.
Re: Adding a blacklist via sa-update - would you mind?
Posted by Karsten Bräckelmann <gu...@rudersport.de>.
On Mon, 2011-12-05 at 16:27 -0500, darxus@chaosreigns.com wrote:
> On 12/02, Karsten Bräckelmann wrote:
> > How very subversive...
>
> I asked in the bug how people felt about posting to the users list about
> it. I got no objections.
You got no answer. Subtle difference.
> The complete lack of responses from anybody but you in the last 4 days
> seems to be a loud and clear indication that nobody cares, and we should go
> ahead with adding this useful blacklist.
You do realize that your target audience for that question is exactly
the group hardly posting here on weekends, right?
> > > We would prefer to only add this blacklist to future releases, version
> > > 3.4.x+. But we currently do not have a way to maintain two separate sets
> > > of rule updates. So our options are:
> >
> > Discussing exactly this on the dev@ list would have been quite some
> > idea, don't ya think?
>
> I thought that question had already been sufficiently implied.
Implied would be correct. Though not specifically asked, let alone
discussed by the devs. Not everyone follows closely all bugzilla mail.
(And the Mailspike specific feature request is not the place to discuss
this much broader topic either.)
--
char *t="\10pse\0r\0dtu\0.@ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4";
main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1:
(c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}
Re: several process gocr
Posted by Walter Hurry <wa...@lavabit.com>.
On Fri, 09 Dec 2011 00:41:49 +0100, LEVEAU Stanislas wrote:
> <html>
> <head>
> <meta content="text/html; charset=ISO-8859-1"
> http-equiv="Content-Type">
> </head>
> <body bgcolor="#FFFFFF" text="#000099">
> Hi,<span id="result_box" class="" lang="en"><br>
> <br>
> <span class="hps">I use the</span> <span class="hps">plugin</span>
> <span class="hps">text OCR</span> <span class="hps">with</span>
> <span
> class="hps">spamassassin</span> <span class="hps">and</span>
> <span class="hps">amavisd-new<br>
> <br>
> </span> <span class="hps">Doc:</span> <span class="hps"><a
> class="moz-txt-link-freetext"
> href="http://www.maiamailguard.com/maia/wiki/OCR">http://
www.maiamailguard.com/maia/wiki/OCR</a><br>
> <br>
> </span><br>
> <span class="hps">And it happens</span> <span class="hps">that
> there are several</span> <span class="hps">process</span> <span
> class="hps">gocr</span><span>-consuming</span> a lot of <span
> class="hps">cpu</span><span> and block my server</span><br>
> <br>
> <span class="hps">Do you have</span> <span class="hps">an idea
> or</span>
> <span class="hps">already</span> <span class="hps">had the</span>
> <span class="hps">problem?</span><br>
> <br>
> <span class="hps">What is the best</span> <span
> class="hps">solution</span>
> <span class="hps">to scan messages</span> <span class="hps">with
> pictures</span><span>?</span> <span
> class="hps">Fuzzyocr</span><span class="">?</span></span>
> <br>
> <br>
> <br>
> OS : Redhat AS5.0<br>
> <br>
> Rpms installation<br>
> <br>
> perl-Mail-SpamAssassin-3.3.1-1<br>
> amavisd-new-2.6.6-1.el5.rf<br>
> <span id="result_box" class="" lang="en"><span
> class="hps">gocr-0.49-1.el5.rf<br>
> </span></span><br>
> <br>
> Regards<br>
> Stan<br>
> </body>
> </html>
Though one wonders why.
several process gocr
Posted by LEVEAU Stanislas <st...@ac-caen.fr>.
Hi,
I use the plugin text OCR with spamassassin and amavisd-new
Doc: http://www.maiamailguard.com/maia/wiki/OCR
And it happens that there are several process gocr-consuming a lot of
cpuand block my server
Do you have an idea or already had the problem?
What is the best solution to scan messages with pictures? Fuzzyocr?
OS : Redhat AS5.0
Rpms installation
perl-Mail-SpamAssassin-3.3.1-1
amavisd-new-2.6.6-1.el5.rf
gocr-0.49-1.el5.rf
Regards
Stan
RE: Adding a blacklist via sa-update - would you mind?
Posted by R - elists <li...@abbacomm.net>.
bottom line is
yes, we would mind
mainly about the enabled by default part
there are too many scenarios & reasons to argue for and against it at this
moment in time
- rh
Re: Adding a blacklist via sa-update - would you mind?
Posted by da...@chaosreigns.com.
On 12/02, Karsten Bräckelmann wrote:
> How very subversive...
I asked in the bug how people felt about posting to the users list about
it. I got no objections.
The complete lack of responses from anybody but you in the last 4 days
seems to be a loud and clear indication that nobody cares, and we should go
ahead with adding this useful blacklist.
> On Thu, 2011-12-01 at 12:58 -0500, darxus@chaosreigns.com wrote:
> Since you keep stressing the one, 1, single DNS query per message, I
> cannot help but nit-pick -- please do have a look at the rules you're
> talking about. Or maybe, just re-read your own comment 1 on bug 6400.
> It's not a single query.
Yup, sorry. I made the mistake of assuming the mailspike instructions
would included everything, but turns out it's just the blacklist:
http://mailspike.org/usage.html
Only part I care about is the blacklist.
> > If it does get added, it will be following a post to this list. If you
> > want to ensure you do not use this blacklist if it does get added, you can
> > preemptively disable it with:
>
> You forgot to disable and meta-out the sub-rules, most notably including
> the actual check_rbl() eval rules doing the queries. You just disabled
> the scores but did not eliminate the DNS queries.
>
> I have pointed out such omission in the past, a couple times.
That does sound familiar. Seems odd that that's necessary. So it should
look like this?
score __RCVD_IN_MSPIKE_B 0
score __RCVD_IN_MSPIKE_L 0
score __RCVD_IN_MSPIKE_Z 0
score RCVD_IN_MSPIKE_L5 0
score RCVD_IN_MSPIKE_L4 0
score RCVD_IN_MSPIKE_L3 0
score RCVD_IN_MSPIKE_L2 0
score RCVD_IN_MSPIKE_H5 0
score RCVD_IN_MSPIKE_H4 0
score RCVD_IN_MSPIKE_H3 0
score RCVD_IN_MSPIKE_H2 0
meta __RCVD_IN_MSPIKE_LOW 0
meta RCVD_IN_MSPIKE_ZBI 0
meta RCVD_IN_MSPIKE_BL 0
meta RCVD_IN_MSPIKE_WL 0
> > We would prefer to only add this blacklist to future releases, version
> > 3.4.x+. But we currently do not have a way to maintain two separate sets
> > of rule updates. So our options are:
>
> Discussing exactly this on the dev@ list would have been quite some
> idea, don't ya think?
I thought that question had already been sufficiently implied.
> > 4) Develop a way to maintain two separate rule sets. May, realistically,
> > be impossible.
>
> I wonder how hard running the re-scoring twice could be -- one time with
> the relevant rules snipered from the mass-check logs. Well, for the time
> we keep updating 3.3.x update channels in addition to the then-current
> 3.4.x, which is something not even discussed.
That might be great. When do you think that's likely to happen?
> > 5) Add new blacklists to 3.4.x rules, and stop providing 3.3.x updates. We
> > can't expect people to just cut over suddenly like that.
>
> True on the latter. However, there have been no plans (or discussion)
> yet about generating frequent rule updates for the current stable, *and*
> previous release. Stopping the updates does NOT suddenly invalidate the
> previous version. People today still happily running 3.2.x can tell you.
>
> Ironically, you're the one who pushed for 3.2.x EOL. Now your aim seems
> to be not only to maintain two stable versions, but also to keep
> generating rule updates long-term.
I wanted 3.2.x to be supported *or* for the lack of support to be clearly
stated. It seemed pretty clear it wasn't going to get any support. Mostly
because nobody seemed to know how to handle score generation for 3.2.
The spamassassin downloads page still looks like 3.2.5 is supported.
"A major release is supported for no less than 6 months after the release
of the next major release." - http://wiki.apache.org/spamassassin/ReleaseGoals
Yes, I think for at *least* 6 months after a release, rule updates should
continue to be provided for the previous release. I think 6 months is too
short.
--
"We will be dead soon. Is this how we want to live?"
http://www.ChaosReigns.com
Re: Adding a blacklist via sa-update - would you mind?
Posted by Karsten Bräckelmann <gu...@rudersport.de>.
How very subversive...
On Thu, 2011-12-01 at 12:58 -0500, darxus@chaosreigns.com wrote:
> There is some question among spamassassin developers* on whether or not
> it is acceptable to increase the network load of spamassassin, by one
> DNS query per email, for existing releases (version 3.3.x), by adding
> one DNS blacklist to the rule set via sa-update.
Since you keep stressing the one, 1, single DNS query per message, I
cannot help but nit-pick -- please do have a look at the rules you're
talking about. Or maybe, just re-read your own comment 1 on bug 6400.
It's not a single query.
> If it does get added, it will be following a post to this list. If you
> want to ensure you do not use this blacklist if it does get added, you can
> preemptively disable it with:
You forgot to disable and meta-out the sub-rules, most notably including
the actual check_rbl() eval rules doing the queries. You just disabled
the scores but did not eliminate the DNS queries.
I have pointed out such omission in the past, a couple times.
> We would prefer to only add this blacklist to future releases, version
> 3.4.x+. But we currently do not have a way to maintain two separate sets
> of rule updates. So our options are:
Discussing exactly this on the dev@ list would have been quite some
idea, don't ya think?
> 4) Develop a way to maintain two separate rule sets. May, realistically,
> be impossible.
I wonder how hard running the re-scoring twice could be -- one time with
the relevant rules snipered from the mass-check logs. Well, for the time
we keep updating 3.3.x update channels in addition to the then-current
3.4.x, which is something not even discussed.
> 5) Add new blacklists to 3.4.x rules, and stop providing 3.3.x updates. We
> can't expect people to just cut over suddenly like that.
True on the latter. However, there have been no plans (or discussion)
yet about generating frequent rule updates for the current stable, *and*
previous release. Stopping the updates does NOT suddenly invalidate the
previous version. People today still happily running 3.2.x can tell you.
Ironically, you're the one who pushed for 3.2.x EOL. Now your aim seems
to be not only to maintain two stable versions, but also to keep
generating rule updates long-term.
> Bug discussing adding this DNSBL:
> https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6400
Tracking the bug's status, not so much discussion. Heavy-load discussion
should better be done on the dev@ list, not in bugzilla.
> * This might sound like implication that I'm a spamassassin developer. I'm
> not. I do not have commit access.
--
char *t="\10pse\0r\0dtu\0.@ghno\x4e\xc8\x79\xf4\xab\x51\x8a\x10\xf4\xf4\xc4";
main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1:
(c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}}