| View previous topic :: View next topic |
| Author |
Message |
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
 Posted: Thu Jan 25, 2018 11:07 am Post subject: duplicate bans remover |
 |
|
is there a way to have a monitor for set bans to check for duplicate bans within same range and have the less restrictive ones removed
for example :
| Code: |
+b *!*@lets.try.this
+b *!*@*.try.this
+b *!*@*.this
|
to have it leave only the most restrictive no matter what order they have been added so as well if added in this order in reverse to still have leave the
most restrictive :
| Code: |
+b *!*@*.this
+b *!*@*.try.this
+b *!*@lets.try.this
|
Pixelz provides this proc as start im not sure how to proper code additional code to make it useable
| Code: | proc findRedundantBans {chan} {
set redundant ""
foreach ban1 [chanbans $chan] {
set ban1 [lindex $ban1 0]
foreach ban2 [chanbans $chan] {
set ban2 [lindex $ban2 0]
if {$ban1 eq $ban2} {
continue
} elseif {[matchaddr $ban1 $ban2]} {
lappend redundant $ban2
}
}
}
return [lsort -unique $redundant]
} |
|
|
| Back to top |
|
 |
caesar
Mint Rubber
Joined: 14 Oct 2001
Posts: 3741
Location: Mint Factory
|
| Posted: Sat Jan 27, 2018 4:54 am Post subject: |
|
|
I've seen that you posted a couple of times about this matter and since didn't get any answer was thinking about having a look at this.
On what network are you trying to add those bans? I'm asking cos for example on Undernet the IRC server will remove the other bans if there's a more restrictive one on it's own. I pushed a ban via the bot from DCC chat:
| Code: |
<cez> .tcl pushmode #channel +b "*!*@something.here.com"
<bot> Tcl:
<cez> .tcl pushmode #channel +b "*!*@*.here.com"
<bot> Tcl:
|
and the result I see on the #channel channel is:
| Code: |
* bot sets mode: -b+b *!*@something.here.com *!*@*.here.com
|
I even tried the same thing with my mIRC client and the result is the same:
| Code: |
* cez sets mode: +b *!*@something.here.com
* cez sets mode: -b+b *!*@something.here.com *!*@*.here.com
|
so this means the server dose the heavy lifting and leaves only the most restrictive ban on.
_________________
Once the game is over, the king and the pawn go back in the same box. |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Sat Jan 27, 2018 8:37 am Post subject: |
|
|
thanx for the reply caesar,
its mainly meant for dalnet and some other nets wich dont have that in place to track duplicate bans |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Sun Jan 28, 2018 11:41 am Post subject: |
|
|
| as networks like dalnet and others are hammered by all kind of floods ending up with a full ban list the smart banlist managing should clean it up and take the load off a bit leaving buffer for extra bans |
|
| Back to top |
|
|
SpiKe^^
Owner
Joined: 12 May 2006
Posts: 792
Location: Tennessee, USA
|
| Posted: Sun Jan 28, 2018 4:13 pm Post subject: duplicate bans remover |
|
|
Test the function of proc findRedundantBans with this very short script.
| Code: |
bind pub o|o .cleanbans checkChanBans
proc checkChanBans {nk uh hn ch tx} {
foreach ban [findRedundantBans $ch] {
pushmode $ch -b $ban
}
}
proc findRedundantBans {ch} { ;# thanks Pixelz #
set redundant ""
foreach ban1 [chanbans $ch] {
set ban1 [lindex $ban1 0]
foreach ban2 [chanbans $ch] {
set ban2 [lindex $ban2 0]
if {$ban1 eq $ban2} { continue }
if {[matchaddr $ban1 $ban2]} {
lappend redundant $ban2
}
}
}
return [lsort -unique $redundant]
}
|
_________________
SpiKe^^
Get BogusTrivia 2.06.4.7 at www.mytclscripts.com
or visit the New Tcl Acrhive at www.tclarchive.org
. |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Sun Jan 28, 2018 6:30 pm Post subject: |
|
|
tnx spike i tested it it seems to work well
is there a way to have it work on ban settings rather than manually cleanup |
|
| Back to top |
|
|
SpiKe^^
Owner
Joined: 12 May 2006
Posts: 792
Location: Tennessee, USA
|
| Posted: Sun Jan 28, 2018 9:42 pm Post subject: duplicate bans remover |
|
|
Try this.
| Code: |
bind mode - "#% +b" checkRedundant
proc checkRedundant {nk uh hn ch mod ban} {
foreach ban2 [chanbans $ch] { set ban2 [lindex $ban2 0]
if {$ban eq $ban2} { continue }
if {[matchaddr $ban $ban2]} { pushmode $ch -b $ban2
} elseif {[matchaddr $ban2 $ban]} { pushmode $ch -b $ban }
}
}
bind pub o|o .cleanbans checkChanBans
proc checkChanBans {nk uh hn ch tx} {
foreach ban [findRedundantBans $ch] {
pushmode $ch -b $ban
}
}
proc findRedundantBans {ch} { ;# thanks Pixelz #
set redundant ""
foreach ban1 [chanbans $ch] {
set ban1 [lindex $ban1 0]
foreach ban2 [chanbans $ch] {
set ban2 [lindex $ban2 0]
if {$ban1 eq $ban2} { continue }
if {[matchaddr $ban1 $ban2]} {
lappend redundant $ban2
}
}
}
return [lsort -unique $redundant]
}
|
_________________
SpiKe^^
Get BogusTrivia 2.06.4.7 at www.mytclscripts.com
or visit the New Tcl Acrhive at www.tclarchive.org
. |
|
| Back to top |
|
|
caesar
Mint Rubber
Joined: 14 Oct 2001
Posts: 3741
Location: Mint Factory
|
| Posted: Mon Jan 29, 2018 2:23 am Post subject: |
|
|
Ah, matchaddr! Don't know why I din't think about this. Probably because got a lot of stuff on my mind at work.. 
Anyway, Pixelz came with a good idea to use that lsort -unique BUT it's used at the end of the proc when should be used at the start on the banlist the bot fetches because this way the most restrictive bans are sorted out first.
| Code: |
% puts $banlist
*!*@*.here.com *!*@something.here.com *!*@foo.some.com *!*@*.foo.com *!foo@blah.com
% lsort -unique $banlist
*!*@*.foo.com *!*@*.here.com *!*@foo.some.com *!*@something.here.com *!foo@blah.com
|
this way the most restrictive ones are kept and actually the "redundant" ones are removed.
I know SpiKe at times can't be bothered to correct the code a user came with, but there's no point into looping and asking for the bans twice when can define a variable containing the banlist and loop tough it as many times you want later.
| Code: |
proc findRedundantBans {chan} {
if {![botonchan $chan]} return
set chanbans [chanbans $chan]
foreach {ban bywho age} [join $chanbans] {
lappend banlist $ban
}
set banlist [lsort -unique $banlist]
set len [llength $banlist]
while {$len > 0} {
set b1 [lindex [split $banlist] 0]
set banlist [lreplace $banlist 0 0]
foreach b2 $banlist {
if {[matchaddr $b1 $b2]} {
if {[info exists redundant]} {
if {[expr [lsearch -nocase [join $redundant] $b2]] > -1} continue
}
lappend redundant $b2
}
}
incr len -1
}
if {[info exists redundant]} {
return $redundant
}
}
|
Tested briefly so you need to do more tests and let me know if get any errors.
I would do a few changes to other stuff as well and turn them into:
| Code: |
bind mode - "#% +b" checkRedundant
bind pub o|o .cleanbans checkChanBans
proc checkRedundant {nk uh hn ch mod ban} {
purgeRedundant $ch
}
proc checkChanBans {nk uh hn ch tx} {
purgeRedundant $ch
}
proc purgeRedundant {ch} {
if {![botisop $ch]} return
set redundant [findRedundantBans $ch]
set len [llength $redundant]
set max 6
while {$len > 0} {
if {$len > $max} {
set mode [string repeat "b" $max]
set bans [join [lrange $redundant 0 [expr {$max - 1}]]]
set redundant [lrange $redundant $max end]
incr len -$max
} else {
set mode [string repeat "b" $len]
set bans [join $redundant]
set len 0
}
puthelp "MODE $ch -$mode $bans"
}
}
|
_________________
Once the game is over, the king and the pawn go back in the same box. |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Mon Jan 29, 2018 8:39 am Post subject: |
|
|
both work only spikes unsets bans in stacked unlike ceasars
ceasar code only stacks mode settings when less restrictive is set last by chanop while |
|
| Back to top |
|
|
caesar
Mint Rubber
Joined: 14 Oct 2001
Posts: 3741
Location: Mint Factory
|
| Posted: Mon Jan 29, 2018 10:29 am Post subject: |
|
|
Replace
| Code: |
puthelp "MODE $ch -$mode $bans"
|
with: | Code: |
pushmode $ch -$mode $bans
|
but I can't test if multiple bans will removed at the same time OR will get some weird behavior, at least I got some on Undernet.
Anyway, there's a key difference in the way I loop vs. the way Spike's code loops. And no, is not about me using while and a foreach and him using two foreach loops cos it's basically the same thing.
The difference lays in:
| Code: |
set b1 [lindex [split $banlist] 0]
set banlist [lreplace $banlist 0 0]
|
Let me explain. Let's say the channel banlist is a list of numbers: 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10 (in my example will use from 1 to 5). Now, if we look at the code:
| Code: |
foreach ban1 [chanbans $ch] {
set ban1 [lindex $ban1 0]
foreach ban2 [chanbans $ch] {
set ban2 [lindex $ban2 0]
if {$ban1 eq $ban2} { continue }
if {[matchaddr $ban1 $ban2]} {
lappend redundant $ban2
}
}
}
|
And simplify it to see what's going on:
| Code: |
foreach a [split $bans] {
foreach b [split $bans] {
puts "$a vs $b"
}
}
|
the result will be:
| Code: |
% foreach a [split $bans] { foreach b [split $bans] { puts "$a vs $b" } }
1 vs 1
1 vs 2
1 vs 3
1 vs 4
1 vs 5
2 vs 1
2 vs 2
2 vs 3
2 vs 4
2 vs 5
3 vs 1
3 vs 2
3 vs 3
3 vs 4
3 vs 5
4 vs 1
4 vs 2
4 vs 3
4 vs 4
4 vs 5
5 vs 1
5 vs 2
5 vs 3
5 vs 4
5 vs 5
|
meaning 25 attempts to match one ban mask with another. With just a single lines inside the first loop that will basically remove the first element from the list every time the loop executes:
| Code: |
foreach a [split $bans] {
set bans [lreplace $bans 0 0]
foreach b [split $bans] {
puts "$a vs $b"
}
}
|
the result will be:
| Code: |
% foreach a [split $bans] { set bans [lreplace $bans 0 0]; foreach b [split $bans] { puts "$a vs $b" } }
1 vs 2
1 vs 3
1 vs 4
1 vs 5
2 vs 3
2 vs 4
2 vs 5
3 vs 4
3 vs 5
4 vs 5
|
meaning 10 attempts to match one ban mask with another. That's 15 matches that have been already done, meaning a 60% decrease in time used for matching bans.
If where to compare 30 bans then this means 900 attempts vs. 435 attempts ONLY by adding that lreplace line. 
| Code: |
% set bans [list 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
% set c 0
0
% foreach a [split $bans] { foreach b [split $bans] { incr c } }; puts $c
900
|
vs.
| Code: |
% set bans [list 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
% set c 0
0
% foreach a [split $bans] { set bans [lreplace $bans 0 0]; foreach b [split $bans] { incr c } }; puts $c
435
|
This means an 51.66% less matching. If where to compare 50 vs. same 50 then there would be 2500 matches vs. 1225 with just a single lreplace line. That's INSANE!
I just noticed that I forgot to take that if {$ban1 eq $ban2} { continue } line into account, but this doesn't make much of a difference, basically with 10 bans instead of 100 will do only 90 matches, still double of the other method that has the lreplace trick in. 
I think we can push this a notch further by adding another lreplace, but need to do some testing first.
_________________
Once the game is over, the king and the pawn go back in the same box. |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Mon Jan 29, 2018 11:29 am Post subject: |
|
|
| hm ok yes i think u already mentioned it since what i also discovered after testing is the more bans match then it removes most but leaves like 2 or 3 |
|
| Back to top |
|
|
caesar
Mint Rubber
Joined: 14 Oct 2001
Posts: 3741
Location: Mint Factory
|
| Posted: Mon Jan 29, 2018 1:48 pm Post subject: |
|
|
Something is not working or what are you saying? Mind posting an example so I can come on Dalnet and test stuff on my own?
_________________
Once the game is over, the king and the pawn go back in the same box. |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Mon Jan 29, 2018 3:25 pm Post subject: |
|
|
it doesnt remove all matching duplicates
| Quote: | sets mode: +b *!*@*.IP
sets mode: +bbbbbbbbbbbbbbb *!*@RifSytes-zcd.299.27.20.IP *!*@RifSytes-599.460.33.69.IP *!*@RifSytes-3g5.144.20.79.IP *!*@RifSytes-066.185.88.80.IP *!*@RifSytes-ll9.546.74.68.IP *!*@RifSytes-6u3.339.13.92.IP *!*@RifSytes-kqs.407.57.83.IP *!*@RifSytes-b7v.883.54.19.IP *!*@RifSytes-56d.862.69.27.IP *!*@RifSytes-ox5.511.75.48.IP *!*@RifSytes-6sw.885.18.89.IP *!*@RifSytes-5i6.288.53.70.IP *!*@RifSytes-v32.936.41.94.IP *!*@RifSytes-54f.459.78.41.IP *!*@RifSytes-bwp.836.85.33.IP
sets mode: +bbbbbbbbbbbbbbb *!*@RifSytes-m36.196.43.47.IP *!*@RifSytes-915.770.49.94.IP *!*@RifSytes-5qy.659.08.92.IP *!*@RifSytes-3ut.148.34.53.IP *!*@RifSytes-3o9.200.73.87.IP *!*@RifSytes-pm7.923.75.08.IP *!*@RifSytes-996.695.55.62.IP *!*@RifSytes-gk0.421.24.66.IP *!*@RifSytes-071.544.81.59.IP *!*@RifSytes-sv8.105.76.27.IP *!*@RifSytes-5ob.997.12.63.IP *!*@RifSytes-nh3.819.91.12.IP *!*@RifSytes-50n.997.16.02.IP *!*@RifSytes-4q7.694.07.31.IP *!*@RifSytes-o8a.187.92.31.IP
sets mode: -bbbbbb *!*@RifSytes-zcd.299.27.20.IP *!*@RifSytes-599.460.33.69.IP *!*@RifSytes-3g5.144.20.79.IP *!*@RifSytes-066.185.88.80.IP *!*@RifSytes-ll9.546.74.68.IP *!*@RifSytes-6u3.339.13.92.IP
sets mode: -bbbbbb *!*@RifSytes-kqs.407.57.83.IP *!*@RifSytes-b7v.883.54.19.IP *!*@RifSytes-56d.862.69.27.IP *!*@RifSytes-ox5.511.75.48.IP *!*@RifSytes-6sw.885.18.89.IP *!*@RifSytes-5i6.288.53.70.IP
sets mode: -bb *!*@RifSytes-v32.936.41.94.IP *!*@RifSytes-54f.459.78.41.IP
sets mode: -bbbbbb *!*@RifSytes-m36.196.43.47.IP *!*@RifSytes-915.770.49.94.IP *!*@RifSytes-5qy.659.08.92.IP *!*@RifSytes-3ut.148.34.53.IP *!*@RifSytes-3o9.200.73.87.IP *!*@RifSytes-pm7.923.75.08.IP
sets mode: -bbbbbb *!*@RifSytes-996.695.55.62.IP *!*@RifSytes-gk0.421.24.66.IP *!*@RifSytes-071.544.81.59.IP *!*@RifSytes-sv8.105.76.27.IP *!*@RifSytes-5ob.997.12.63.IP *!*@RifSytes-nh3.819.91.12.IP
sets mode: -bb *!*@RifSytes-50n.997.16.02.IP *!*@RifSytes-4q7.694.07.31.IP |
|
|
| Back to top |
|
|
SpiKe^^
Owner
Joined: 12 May 2006
Posts: 792
Location: Tennessee, USA
|
| Posted: Mon Jan 29, 2018 6:32 pm Post subject: |
|
|
In that test above, the script appears to have left 2 bans that should have been removed.
This seems like an extreme example, no one would have 30 similar bans set on a channel.
Which of these 2 scripts are you running??
_________________
SpiKe^^
Get BogusTrivia 2.06.4.7 at www.mytclscripts.com
or visit the New Tcl Acrhive at www.tclarchive.org
.
Last edited by SpiKe^^ on Tue Jan 30, 2018 9:32 am; edited 1 time in total |
|
| Back to top |
|
|
simo
Owner
Joined: 22 Mar 2015
Posts: 941
|
| Posted: Tue Jan 30, 2018 6:00 am Post subject: |
|
|
Its an extreme example Indeed with 30 duplicates, i also tested with like 15 duplicates with same result
also i thought of the example of *!*@* set by accident will result in entire ban list removal and should prob be exempted from triggering
I used your version in this test spike |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
