javabdd-devel — General developer's list.

[Javabdd-devel] compiling BuDDy+JavaBDD on a 64bits architecture

[Fausto S. <fau...@un...>]

Hi to everybody!

I'm trying to compile BuDDy+JavaBDD on an Intel Xeon QuadCore 64bits box
running linux and Java 1.6. I modified the makefile in order to specify
the 64bits requirement: -fPIC instead of -m32. I has to put -fPIC
everywhere, or otherwise it does compile for a missing symbol.
Compilation works almost fine now, with the exception that I get some
warnings about the translation between 32 bits integers and pointers.
The library libbuddy.so is generated, as well as the net/ hierarchy. I
can run the NQueens test with BuDDy. However, the RubiksCube test breaks
from time to time, with a segmentation fault at the setPair() method. It
seems that everything works, except for the BDDPairing class.

I attach the Makefile that I'm using.

What am I doing wrong?

Thanks,
  Fausto

Re: [Javabdd-devel] Ranges

[Ricardo O. <n3...@ho...>]

Hi,

One more issue. I'm beginning to find this very strange.

The following code:

BDDDomain domain = bddFactory.extDomain(65535);

BDD rule = domain.varRange(0, 4000);
rule.allsat();

gives the following result:

Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 60
	at net.sf.javabdd.BuDDyFactory$BuDDyBDD.allsat0(Native Method)
	at net.sf.javabdd.BuDDyFactory$BuDDyBDD.allsat(BuDDyFactory.java:889)

Is this normal? Am I doing something wrong?

Thanks,
Ricardo.

________________________________



From: n3...@ho...
To: jw...@al...
Date: Wed, 30 Jul 2008 03:39:46 +0000
CC: jav...@li...
Subject: Re: [Javabdd-devel] Ranges




Hi,

I tried a few more things, but couldn't figure out why it could be doing this. I didn't check your code, though. I wrote a small wrapper to handle ranges. It only determines the correct BDD for a range, by removing the extra part that the varRange() method adds at the beginning (basic set difference). It seems to be enough for now, but I'm still not very sure because sometimes varRange() seems to work correctly..
Hope you have more time than me to check this out.

Ricardo.






________________________________



Date: Mon, 28 Jul 2008 22:33:43 -0700
From: joe...@gm...
To: n3...@ho...
Subject: Re: [Javabdd-devel] Ranges
CC: jav...@li...



I think the varRange implementation could be making assumptions about the BDD variable ordering, leading to weird behavior.  If you manage to track down the issue, let me know.

-John



On Mon, Jul 28, 2008 at 2:37 PM, Ricardo Oliveira <n3...@ho...> wrote:


Hi again,

I have another issue related to ranges. It seems that there's something odd happening when I select ranges with some values. So for instance, if I ask for a range to the following values:

BDDDomain domain = bddFactory.extDomain(16);
BDD bdd = domain.varRange(2, 3);
bdd.printSet();

I get:



Which means that the possible values are from 0 to 3 and not from 2 to 3:

0 0 0 0
0 0 0 1
0 0 1 0
0 0 1 1

So, I tried from 10 to 12 and got the strange values:



which represent the interval 8 to 12.
To test it further, I tried with 11 to 12 and got the exact same set:



It seems that this is making the range from the closest 8 bit partition(?) or whatever to the value you say. This means that any range between 0 and 16 that you ask will always be between 0 or 8 (according to the lowest bit) and the upper value you give it.

Here's the output for range 2 to 12:



0 0 0 0 - 0
0 1 0 0 - 4
1 0 0 0 - 8
1 1 0 0 - 12
0 0 1 0 - 2
1 0 1 0 - 10
0 1 1 0 - 6
0 0 0 1 - 1
0 0 1 1 - 3
1 0 0 1 - 9
1 0 1 1 - 11
0 1 0 1 - 5
0 1 1 1 - 7

Is this the expected behavior?

Thank you,
Ricardo.


> From: n3...@ho...
> To: jav...@li...
> Date: Mon, 28 Jul 2008 19:26:20 +0000
> Subject: [Javabdd-devel] Ranges



>
>
>
> Hi,
>
> I'm having problem using the varRange() method of BDDDomain.
>
> If I create BDDDomain with these sizes:
>
> long sizes[] = new long[] {65535, 65535};
> BDDDomain[] domain= bddFactory.extDomain(sizes);
>
> Every time I try to create a range for the maximum values (which should be inclusive):
>
> BDD bdd = domain[0].varRange(0, 65535);
>
> I get the following error:
>
> Exception in thread "main" net.sf.javabdd.BDDException: range is invalid
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:253)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
>
> If I try with:
>
> long sizes[] = new long[] {65536, 65536};
> BDDDomain[] domain= bddFactory.extDomain(sizes);
>
> I get:
>
> Exception in thread "main" java.lang.ArithmeticException: Negative bit address
> at java.math.BigInteger.testBit(Unknown Source)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:262)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
>
> One solution is using one extra bit (increasing the size to 65535*2), but I was really hoping this wouldn't be necessary. Perhaps I'm doing something incredibly stupid, but I can't see what at this point.
>
> Thank you very much for your help,
> Ricardo.
> _________________________________________________________________
> Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy!
> http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel


_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE



-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Javabdd-devel mailing list
Jav...@li...
https://lists.sourceforge.net/lists/listinfo/javabdd-devel



________________________________

Get news, entertainment and everything you care about at Live.com. Check it out!

_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE

Re: [Javabdd-devel] Ranges

[Ricardo O. <n3...@ho...>]

Hi,
 
I tried a few more things, but couldn't figure out why it could be doing this. I didn't check your code, though. I wrote a small wrapper to handle ranges. It only determines the correct BDD for a range, by removing the extra part that the varRange() method adds at the beginning (basic set difference). It seems to be enough for now, but I'm still not very sure because sometimes varRange() seems to work correctly..
Hope you have more time than me to check this out.
 
Ricardo.
 
 



Date: Mon, 28 Jul 2008 22:33:43 -0700From: joe...@gm...To: n3...@ho...Subject: Re: [Javabdd-devel] RangesCC: jav...@li...
I think the varRange implementation could be making assumptions about the BDD variable ordering, leading to weird behavior.  If you manage to track down the issue, let me know.-John
On Mon, Jul 28, 2008 at 2:37 PM, Ricardo Oliveira <n3...@ho...> wrote:
Hi again,I have another issue related to ranges. It seems that there's something odd happening when I select ranges with some values. So for instance, if I ask for a range to the following values:BDDDomain domain = bddFactory.extDomain(16);BDD bdd = domain.varRange(2, 3);bdd.printSet();I get:Which means that the possible values are from 0 to 3 and not from 2 to 3:0 0 0 00 0 0 10 0 1 00 0 1 1So, I tried from 10 to 12 and got the strange values:which represent the interval 8 to 12.To test it further, I tried with 11 to 12 and got the exact same set:It seems that this is making the range from the closest 8 bit partition(?) or whatever to the value you say. This means that any range between 0 and 16 that you ask will always be between 0 or 8 (according to the lowest bit) and the upper value you give it.Here's the output for range 2 to 12:0 0 0 0 - 00 1 0 0 - 41 0 0 0 - 81 1 0 0 - 120 0 1 0 - 21 0 1 0 - 100 1 1 0 - 60 0 0 1 - 10 0 1 1 - 31 0 0 1 - 91 0 1 1 - 110 1 0 1 - 50 1 1 1 - 7Is this the expected behavior?Thank you,Ricardo.> From: n3...@ho...> To: jav...@li...> Date: Mon, 28 Jul 2008 19:26:20 +0000> Subject: [Javabdd-devel] Ranges


>>>> Hi,>> I'm having problem using the varRange() method of BDDDomain.>> If I create BDDDomain with these sizes:>> long sizes[] = new long[] {65535, 65535};> BDDDomain[] domain= bddFactory.extDomain(sizes);>> Every time I try to create a range for the maximum values (which should be inclusive):>> BDD bdd = domain[0].varRange(0, 65535);>> I get the following error:>> Exception in thread "main" net.sf.javabdd.BDDException: range is invalid> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:253)> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)>> If I try with:>> long sizes[] = new long[] {65536, 65536};> BDDDomain[] domain= bddFactory.extDomain(sizes);>> I get:>> Exception in thread "main" java.lang.ArithmeticException: Negative bit address> at java.math.BigInteger.testBit(Unknown Source)> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:262)> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)>> One solution is using one extra bit (increasing the size to 65535*2), but I was really hoping this wouldn't be necessary. Perhaps I'm doing something incredibly stupid, but I can't see what at this point.>> Thank you very much for your help,> Ricardo.> _________________________________________________________________> Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy!> http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us> -------------------------------------------------------------------------> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge> Build the coolest Linux based applications with Moblin SDK & win great prizes> Grand prize is a trip for two to an Open Source event anywhere in the world> http://moblin-contest.org/redirect.php?banner_id=100&url=/> _______________________________________________> Javabdd-devel mailing list> Jav...@li...> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
_________________________________________________________________Discover the new Windows Vistahttp://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE


-------------------------------------------------------------------------This SF.Net email is sponsored by the Moblin Your Move Developer's challengeBuild the coolest Linux based applications with Moblin SDK & win great prizesGrand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/_______________________________________________Javabdd-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/javabdd-devel
_________________________________________________________________
News, entertainment and everything you care about at Live.com. Get it now!
http://www.live.com/getstarted.aspx

Re: [Javabdd-devel] Ranges

[John W. <joe...@gm...>]

I think the varRange implementation could be making assumptions about the
BDD variable ordering, leading to weird behavior.  If you manage to track
down the issue, let me know.

-John


On Mon, Jul 28, 2008 at 2:37 PM, Ricardo Oliveira <n3...@ho...> wrote:

>
> Hi again,
>
> I have another issue related to ranges. It seems that there's something odd
> happening when I select ranges with some values. So for instance, if I ask
> for a range to the following values:
>
> BDDDomain domain = bddFactory.extDomain(16);
> BDD bdd = domain.varRange(2, 3);
> bdd.printSet();
>
> I get:
>
>
>
> Which means that the possible values are from 0 to 3 and not from 2 to 3:
>
> 0 0 0 0
> 0 0 0 1
> 0 0 1 0
> 0 0 1 1
>
> So, I tried from 10 to 12 and got the strange values:
>
>
>
> which represent the interval 8 to 12.
> To test it further, I tried with 11 to 12 and got the exact same set:
>
>
>
> It seems that this is making the range from the closest 8 bit partition(?)
> or whatever to the value you say. This means that any range between 0 and 16
> that you ask will always be between 0 or 8 (according to the lowest bit) and
> the upper value you give it.
>
> Here's the output for range 2 to 12:
>
>
>
> 0 0 0 0 - 0
> 0 1 0 0 - 4
> 1 0 0 0 - 8
> 1 1 0 0 - 12
> 0 0 1 0 - 2
> 1 0 1 0 - 10
> 0 1 1 0 - 6
> 0 0 0 1 - 1
> 0 0 1 1 - 3
> 1 0 0 1 - 9
> 1 0 1 1 - 11
> 0 1 0 1 - 5
> 0 1 1 1 - 7
>
> Is this the expected behavior?
>
> Thank you,
> Ricardo.
>
>
> > From: n3...@ho...
> > To: jav...@li...
> > Date: Mon, 28 Jul 2008 19:26:20 +0000
> > Subject: [Javabdd-devel] Ranges
> >
> >
> >
> > Hi,
> >
> > I'm having problem using the varRange() method of BDDDomain.
> >
> > If I create BDDDomain with these sizes:
> >
> > long sizes[] = new long[] {65535, 65535};
> > BDDDomain[] domain= bddFactory.extDomain(sizes);
> >
> > Every time I try to create a range for the maximum values (which should
> be inclusive):
> >
> > BDD bdd = domain[0].varRange(0, 65535);
> >
> > I get the following error:
> >
> > Exception in thread "main" net.sf.javabdd.BDDException: range is invalid
> > at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:253)
> > at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
> >
> > If I try with:
> >
> > long sizes[] = new long[] {65536, 65536};
> > BDDDomain[] domain= bddFactory.extDomain(sizes);
> >
> > I get:
> >
> > Exception in thread "main" java.lang.ArithmeticException: Negative bit
> address
> > at java.math.BigInteger.testBit(Unknown Source)
> > at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:262)
> > at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
> >
> > One solution is using one extra bit (increasing the size to 65535*2), but
> I was really hoping this wouldn't be necessary. Perhaps I'm doing something
> incredibly stupid, but I can't see what at this point.
> >
> > Thank you very much for your help,
> > Ricardo.
> > _________________________________________________________________
> > Invite your mail contacts to join your friends list with Windows Live
> Spaces. It's easy!
> >
> http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us
> > -------------------------------------------------------------------------
> > This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> > Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> > Grand prize is a trip for two to an Open Source event anywhere in the
> world
> > http://moblin-contest.org/redirect.php?banner_id=100&url=/
> > _______________________________________________
> > Javabdd-devel mailing list
> > Jav...@li...
> > https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>
> _________________________________________________________________
> Discover the new Windows Vista
> http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>

Re: [Javabdd-devel] Ranges

[Ricardo O. <n3...@ho...>]

Hi again,

I have another issue related to ranges. It seems that there's something odd happening when I select ranges with some values. So for instance, if I ask for a range to the following values:

BDDDomain domain = bddFactory.extDomain(16);
BDD bdd = domain.varRange(2, 3);
bdd.printSet();

I get:



Which means that the possible values are from 0 to 3 and not from 2 to 3:

0 0 0 0
0 0 0 1
0 0 1 0
0 0 1 1

So, I tried from 10 to 12 and got the strange values:



which represent the interval 8 to 12.
To test it further, I tried with 11 to 12 and got the exact same set:



It seems that this is making the range from the closest 8 bit partition(?) or whatever to the value you say. This means that any range between 0 and 16 that you ask will always be between 0 or 8 (according to the lowest bit) and the upper value you give it.

Here's the output for range 2 to 12:



0 0 0 0 - 0
0 1 0 0 - 4
1 0 0 0 - 8
1 1 0 0 - 12
0 0 1 0 - 2
1 0 1 0 - 10
0 1 1 0 - 6
0 0 0 1 - 1
0 0 1 1 - 3
1 0 0 1 - 9
1 0 1 1 - 11
0 1 0 1 - 5
0 1 1 1 - 7

Is this the expected behavior?

Thank you,
Ricardo.


> From: n3...@ho...
> To: jav...@li...
> Date: Mon, 28 Jul 2008 19:26:20 +0000
> Subject: [Javabdd-devel] Ranges
>
>
>
> Hi,
>
> I'm having problem using the varRange() method of BDDDomain.
>
> If I create BDDDomain with these sizes:
>
> long sizes[] = new long[] {65535, 65535};
> BDDDomain[] domain= bddFactory.extDomain(sizes);
>
> Every time I try to create a range for the maximum values (which should be inclusive):
>
> BDD bdd = domain[0].varRange(0, 65535);
>
> I get the following error:
>
> Exception in thread "main" net.sf.javabdd.BDDException: range is invalid
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:253)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
>
> If I try with:
>
> long sizes[] = new long[] {65536, 65536};
> BDDDomain[] domain= bddFactory.extDomain(sizes);
>
> I get:
>
> Exception in thread "main" java.lang.ArithmeticException: Negative bit address
> at java.math.BigInteger.testBit(Unknown Source)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:262)
> at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)
>
> One solution is using one extra bit (increasing the size to 65535*2), but I was really hoping this wouldn't be necessary. Perhaps I'm doing something incredibly stupid, but I can't see what at this point.
>
> Thank you very much for your help,
> Ricardo.
> _________________________________________________________________
> Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy!
> http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel

_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE

[Javabdd-devel] Ranges

[Ricardo O. <n3...@ho...>]

Hi,

I'm having problem using the varRange() method of BDDDomain.

If I create BDDDomain with these sizes:

long sizes[] = new long[] {65535, 65535};
BDDDomain[] domain= bddFactory.extDomain(sizes);

Every time I try to create a range for the maximum values (which should be inclusive):

BDD bdd = domain[0].varRange(0, 65535);

I get the following error:

Exception in thread "main" net.sf.javabdd.BDDException: range  is invalid
	at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:253)
	at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)

If I try with:

long sizes[] = new long[] {65536, 65536};
BDDDomain[] domain= bddFactory.extDomain(sizes);

I get:

Exception in thread "main" java.lang.ArithmeticException: Negative bit address
	at java.math.BigInteger.testBit(Unknown Source)
	at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:262)
	at net.sf.javabdd.BDDDomain.varRange(BDDDomain.java:249)

One solution is using one extra bit (increasing the size to 65535*2), but I was really hoping this wouldn't be necessary. Perhaps I'm doing something incredibly stupid, but I can't see what at this point.

Thank you very much for your help,
Ricardo.
_________________________________________________________________
Invite your mail contacts to join your friends list with Windows Live Spaces. It's easy!
http://spaces.live.com/spacesapi.aspx?wx_action=create&wx_url=/friends.aspx&mkt=en-us

Re: [Javabdd-devel] Verbose output

[John W. <joe...@gm...>]

You can pass in a static method and then the object can be null.  Otherwise
the class that contains the method must match the object type.

-John


On Wed, Jul 23, 2008 at 11:24 AM, Ricardo Oliveira <n3...@ho...>wrote:

>  Hi,
>
> Thanks for your quick reply. :)
>
> I tried using that, but am getting an error:
>
> Base object for callback method is the wrong type
>
> What type should the object be? I actually also tried assigning it to null,
> but it didn't work ;)
>
> Thanks,
> Ricardo.
>
>
>
> ------------------------------
>
> Date: Tue, 22 Jul 2008 19:48:12 -0700
> From: joe...@gm...
> To: n3...@ho...
> Subject: Re: [Javabdd-devel] Verbose output
> CC: jav...@li...
>
>
>
> Hi,
>
> I think you can register a GC callback using "*registerGCCallback<http://javabdd.sourceforge.net/apidocs/net/sf/javabdd/BDDFactory.html#registerGCCallback%28java.lang.Object,+java.lang.reflect.Method%29>
> "*.  That will override the default callback which just prints a message
> to stderr.
>
> -John
>
>
> On Tue, Jul 22, 2008 at 11:10 AM, Ricardo Oliveira <n3...@ho...>wrote:
>
> Hi,
>
> I'd like to know if there's a way do disable the GC verbose output that
> keeps coming out to stderr.
>
> BTW, great job on JavaBDD :)
> Thank you,
> Ricardo.
>
> ------------------------------
> Discover the new Windows Vista Learn more!<http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>
>
>
> ------------------------------
> Connect to the next generation of MSN Messenger  Get it now!
> <http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>
>

Re: [Javabdd-devel] Verbose output

[Ricardo O. <n3...@ho...>]

Hi,
 
Thanks for your quick reply. :)
 
I tried using that, but am getting an error:
 
Base object for callback method is the wrong type
 
What type should the object be? I actually also tried assigning it to null, but it didn't work ;)
 
Thanks,
Ricardo.




Date: Tue, 22 Jul 2008 19:48:12 -0700From: joe...@gm...To: n3...@ho...Subject: Re: [Javabdd-devel] Verbose outputCC: jav...@li...
Hi,I think you can register a GC callback using "registerGCCallback".  That will override the default callback which just prints a message to stderr.-John
On Tue, Jul 22, 2008 at 11:10 AM, Ricardo Oliveira <n3...@ho...> wrote:

Hi, I'd like to know if there's a way do disable the GC verbose output that keeps coming out to stderr. BTW, great job on JavaBDD :)Thank you,Ricardo.

Discover the new Windows Vista Learn more!-------------------------------------------------------------------------This SF.Net email is sponsored by the Moblin Your Move Developer's challengeBuild the coolest Linux based applications with Moblin SDK & win great prizesGrand prize is a trip for two to an Open Source event anywhere in the worldhttp://moblin-contest.org/redirect.php?banner_id=100&url=/_______________________________________________Javabdd-devel mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/javabdd-devel
_________________________________________________________________
Connect to the next generation of MSN Messenger 
http://imagine-msn.com/messenger/launch80/default.aspx?locale=en-us&source=wlmailtagline

Re: [Javabdd-devel] Verbose output

[John W. <joe...@gm...>]

Hi,

I think you can register a GC callback using
"*registerGCCallback<http://javabdd.sourceforge.net/apidocs/net/sf/javabdd/BDDFactory.html#registerGCCallback%28java.lang.Object,%20java.lang.reflect.Method%29>
"*.  That will override the default callback which just prints a message to
stderr.

-John


On Tue, Jul 22, 2008 at 11:10 AM, Ricardo Oliveira <n3...@ho...>wrote:

>  Hi,
>
> I'd like to know if there's a way do disable the GC verbose output that
> keeps coming out to stderr.
>
> BTW, great job on JavaBDD :)
> Thank you,
> Ricardo.
>
> ------------------------------
> Discover the new Windows Vista Learn more!<http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win great
> prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>
>

[Javabdd-devel] Verbose output

[Ricardo O. <n3...@ho...>]

Hi, I'd like to know if there's a way do disable the GC verbose output that keeps coming out to stderr. BTW, great job on JavaBDD :)Thank you,Ricardo.
_________________________________________________________________
Discover the new Windows Vista
http://search.msn.com/results.aspx?q=windows+vista&mkt=en-US&form=QBRE

Re: [Javabdd-devel] Release of 2.0

[John W. <joe...@gm...>]

Hi Naga,

2.0 is available in the Subversion repository but I haven't had a chance to
build and test a 2.0 package yet.  2.0 should be stable if you build it from
the Subversion repository.

-John


On Tue, Jun 24, 2008 at 4:43 AM, Naga Nalamati <na...@ya...> wrote:

> Hi,
>
> Can you please update us of the Release 2.0 status as API version on
> sourceforge.net is 2.0 but downloadable jar version is 1.0b2?
>
> Which version should be used ?
>
> Thanks and Regards
> Lakshmikant
>
>
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>

[Javabdd-devel] Release of 2.0

[Naga N. <na...@ya...>]

Hi,

Can you please update us of the Release 2.0 status as API version on sourceforge.net is 2.0 but downloadable jar version is 1.0b2?

Which version should be used ?

Thanks and Regards
Lakshmikant


      

[Javabdd-devel] BDD reordering

[shay a. <at...@gm...>]

hello everybody!

i am a student in the computer engineering department , and i am working
lately on my final project about BDDs and MTBDDs.
my first assignment was to build a BDD from a boolean function input such as
"x1+x3" , so i downloaded all of the packages exist
today(CUDD,BuDDY,JDD,JavaBDD etc.) and finally i figured out how to create a
BDD diagram using the JDD package.
The thing is that the ordinary BDD in this package doesn't use reordering,
and i haven't figured out how to
do it...so i tried to work with JavaBDD and in the BDDFactory class they
offer really there's a function
that can reorder the BDD automatically , great!! but in the JavaBdd package
i haven't figured out how to create a BDD.
what i m trying to say is that i have 2 options :
1. does anybody know how can i turn the BDD i created from the JDD package
to a reduced one ?
or
2. if i get as input a boolean function such as "x1+x2*x3" how do i create a
BDD using JavaBDD classes ?

if anyone knows i'll be greatful,
thanks alot in advance.

[Javabdd-devel] slowdown from javabdd-1.0b2 to javabdd-2.0?

[Mayur N. <mh...@cs...>]

Hi,

I was using javabdd-1.0b2 for a long time and decided to switch to
javabdd-2.0 recently but noticed a significant slowdown in the new version.
There seems to be a significant change in the source code (e.g. a lot of
functionality moved from class BDD to BDDVarSet).  Perhaps that's the reason
for the slowdown?  I'm attaching two testcases which illustrate the
slowdown; each of them creates a BDD factory and then adds 5000 random
tuples to an empty bdd.  To observe the slowdown, run:

$ javac -classpath javabdd-1.0b2.jar OldTest.java
$ javac -classpath javabdd-2.0.jar NewTest.java

$ time java -classpath javabdd-2.0.jar:. NewTest
...
1.80s user 0.03s system 98% cpu 1.857 total

$ time java -classpath javabdd-1.0b2.jar:. OldTest
...
0.34s user 0.03s system 91% cpu 0.404 total

Is there some problem with the way I'm creating the BDD factory, or the way
I'm adding tuples to the bdd being constructed?  If you cannot observe the
slowdown, try replacing the number 5000 in both test cases by say 50000
(i.e. add 50000 random tuples instead of 5000).

Thanks,

-- Mayur

Re: [Javabdd-devel] Retrieving a set from a BDD

[John W. <joe...@gm...>]

The BDD "compression" happens within the BDD library.  You can have
1000 copies of the same BDD set and it takes the same amount of space
as one copy.

-John


On Nov 29, 2007 12:29 AM, Wassim Masri <wm...@au...> wrote:
> Hi John,
>
> (Thank you for the quick response)
>
> If I keep "itemsSet1" and "itemsSet2" around, would I be achieving any
> compression?
>
> It seems to me that set1 and set2 will stored in 2 different places
> now, since I would have:
>
> 1) set1 and set2 represented by the BDDs referenced by "itemsSet1" and
> "itemsSet2" (no compression is achieved here).
>
> AND
>
> 2) set1 and set2 stored (potentially compressed) in poolOfSets.
>
> Note that in order to keep "itemsSet1" and "itemsSet2" around, they
> cannot be consumed, i.e., I must pass their clone when adding them to
> "poolOfSets":
> ...
> poolOfSets.andWith(itemsSet1.id());
> poolOfSets.andWith(itemsSet2.id());
> ...
>
>
> Finally, in order to verify my understanding above, I freed
> "poolOfSets" then tried to retrieve set1 and set2 using "itemsSet1"
> and "itemsSet2", the test was successful (please see last 3 statements):
>
> BDDFactory B;
> B = BDDFactory.init(1000, 1000);
> B.setVarNum(500);
>
> // pool of items
> BDD [] items = new BDD[100];
> for (int i = 0; i < 100; i++)
> {
>         items[i] = B.ithVar(i);
> }
>
> // set 1
> BDD itemsSet1 = B.zero();
> itemsSet1.orWith(items[0].id());
> itemsSet1.orWith(items[1].id());
> itemsSet1.orWith(items[2].id());
> itemsSet1.orWith(items[3].id());
> System.out.println("itemsSet1 " + itemsSet1);
> System.out.println("count " + itemsSet1.nodeCount());
>
> // set 2
> BDD itemsSet2 = B.zero();
> itemsSet2.orWith(items[10].id());
> itemsSet2.orWith(items[11].id());
> itemsSet2.orWith(items[2].id());
> itemsSet2.orWith(items[3].id());
> System.out.println("itemsSet2 " + itemsSet2);
> System.out.println("count " + itemsSet2.nodeCount());
>
> // pool of sets
> BDD poolOfSets;
> poolOfSets = B.one();
> poolOfSets.andWith(itemsSet1.id());
> poolOfSets.andWith(itemsSet2.id());
> System.out.println("poolOfSets " + poolOfSets);
> System.out.println("count " + poolOfSets.nodeCount());
>
> // retrieve set1 and set2
> System.out.println("itemsSet1 " + itemsSet1); // got set1 successfully
> System.out.println("itemsSet1 " + itemsSet2); // got set2 successfully
>
> // retrieve set1 and set2 after freeing poolOfSets
> poolOfSets.free();  // <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> System.out.println("itemsSet1 " + itemsSet1); // got set1 successfully
> System.out.println("itemsSet1 " + itemsSet2); // got set2 successfully
>
> To recap, my goal is:
> 1) To be able to free "itemsSet1" and "itemsSet2" right after I store
> set1 and set2 in "poolOfSets"
>
> 2) Retrieve set1 and set2 from "poolOfSets", which maintain them in
> compressed form.
>
> Thank you for your help
> Wes Masri
>
>
> ----------------------------------------------------------------
>
>

Re: [Javabdd-devel] Retrieving a set from a BDD

[Wassim M. <wm...@au...>]

Hi John,

(Thank you for the quick response)

If I keep "itemsSet1" and "itemsSet2" around, would I be achieving any =20
compression?

It seems to me that set1 and set2 will stored in 2 different places =20
now, since I would have:

1) set1 and set2 represented by the BDDs referenced by "itemsSet1" and =20
"itemsSet2" (no compression is achieved here).

AND

2) set1 and set2 stored (potentially compressed) in poolOfSets.

Note that in order to keep "itemsSet1" and "itemsSet2" around, they =20
cannot be consumed, i.e., I must pass their clone when adding them to =20
"poolOfSets":
...
poolOfSets.andWith(itemsSet1.id());
poolOfSets.andWith(itemsSet2.id());
...


Finally, in order to verify my understanding above, I freed =20
"poolOfSets" then tried to retrieve set1 and set2 using "itemsSet1" =20
and "itemsSet2", the test was successful (please see last 3 statements):

BDDFactory B;
B =3D BDDFactory.init(1000, 1000);
B.setVarNum(500);

// pool of items
BDD [] items =3D new BDD[100];
for (int i =3D 0; i < 100; i++)
{
=09items[i] =3D B.ithVar(i);
}

// set 1
BDD itemsSet1 =3D B.zero();
itemsSet1.orWith(items[0].id());
itemsSet1.orWith(items[1].id());
itemsSet1.orWith(items[2].id());
itemsSet1.orWith(items[3].id());
System.out.println("itemsSet1 " + itemsSet1);
System.out.println("count " + itemsSet1.nodeCount());

// set 2
BDD itemsSet2 =3D B.zero();
itemsSet2.orWith(items[10].id());
itemsSet2.orWith(items[11].id());
itemsSet2.orWith(items[2].id());
itemsSet2.orWith(items[3].id());
System.out.println("itemsSet2 " + itemsSet2);
System.out.println("count " + itemsSet2.nodeCount());

// pool of sets
BDD poolOfSets;
poolOfSets =3D B.one();
poolOfSets.andWith(itemsSet1.id());
poolOfSets.andWith(itemsSet2.id());
System.out.println("poolOfSets " + poolOfSets);
System.out.println("count " + poolOfSets.nodeCount());

// retrieve set1 and set2
System.out.println("itemsSet1 " + itemsSet1); // got set1 successfully
System.out.println("itemsSet1 " + itemsSet2); // got set2 successfully

// retrieve set1 and set2 after freeing poolOfSets
poolOfSets.free();  // <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
System.out.println("itemsSet1 " + itemsSet1); // got set1 successfully
System.out.println("itemsSet1 " + itemsSet2); // got set2 successfully

To recap, my goal is:
1) To be able to free "itemsSet1" and "itemsSet2" right after I store =20
set1 and set2 in "poolOfSets"

2) Retrieve set1 and set2 from "poolOfSets", which maintain them in =20
compressed form.

Thank you for your help
Wes Masri


----------------------------------------------------------------

Re: [Javabdd-devel] Retrieving a set from a BDD

[John W. <joe...@gm...>]

Hi,

The "var()" method just returns the variable at the top of the BDD.
The "itemsSet1" and "itemsSet2" objects are the unique identifiers for
the set1 and set2, respectively.

-John


On Nov 28, 2007 12:10 AM, Wassim Masri <wm...@au...> wrote:
> Dear All,
>
> My goal is to achieve the following:
> 1) Create a number of sets of ?items?
> 2) Maintain them in a BDD for compression purposes, i.e., insert a
> given set then retrieve/delete it using some unique identifier (UID)
> that is associated with it
>
> Below is the sample code that I started with. It seems that I am
> properly creating the two sets (itemsSet1, itemsSet2) and inserting
> them in the pool of sets (poolOfSets). But I am not able to properly
> retrieve them from poolOfSets: the last two statements return two
> nodes each, and not the two sets that were originally inserted.
>
> BDDFactory B;
> B = BDDFactory.init(1000, 1000);
> B.setVarNum(500);
>
> // pool of items
> BDD [] items = new BDD[100];
> for (int i = 0; i < 100; i++)
> {
>         items[i] = B.ithVar(i);
> }
>
> // set1
> BDD itemsSet1 = B.zero();
> itemsSet1.orWith(items[0].id());
> itemsSet1.orWith(items[1].id());
> itemsSet1.orWith(items[2].id());
> itemsSet1.orWith(items[3].id());
> int var1 = itemsSet1.var(); // assuming that this is the UID for set1
>
> // set2
> BDD itemsSet2 = B.zero();
> itemsSet2.orWith(items[10].id());
> itemsSet2.orWith(items[11].id());
> itemsSet2.orWith(items[2].id());
> itemsSet2.orWith(items[3].id());
> int var2 = itemsSet2.var();// assuming that this is the UID for set2
>
> // pool of sets
> BDD poolOfSets;
> poolOfSets = B.one();
> poolOfSets.andWith(itemsSet1.id());
> poolOfSets.andWith(itemsSet2.id());
>
> // ?later?in a different context? retrieve set1 and set2
> BDD set1 = B.ithVar(var1); // this contains a single node only
> (expecting itemsSet1)
>
> BDD set2 = B.ithVar(var2); // this contains a single node only
> (expecting itemsSet2)
>
>
> I appreciate your help and any links to relevant tutorials or documentation.
>
> Thanks,
> Wes
>
>
> ----------------------------------------------------------------
>
>

[Javabdd-devel] Retrieving a set from a BDD

[Wassim M. <wm...@au...>]

Dear All,

My goal is to achieve the following:
1) Create a number of sets of ?items?
2) Maintain them in a BDD for compression purposes, i.e., insert a =20
given set then retrieve/delete it using some unique identifier (UID) =20
that is associated with it

Below is the sample code that I started with. It seems that I am =20
properly creating the two sets (itemsSet1, itemsSet2) and inserting =20
them in the pool of sets (poolOfSets). But I am not able to properly =20
retrieve them from poolOfSets: the last two statements return two =20
nodes each, and not the two sets that were originally inserted.

BDDFactory B;
B =3D BDDFactory.init(1000, 1000);
B.setVarNum(500);

// pool of items
BDD [] items =3D new BDD[100];
for (int i =3D 0; i < 100; i++)
{
=09items[i] =3D B.ithVar(i);
}

// set1
BDD itemsSet1 =3D B.zero();
itemsSet1.orWith(items[0].id());
itemsSet1.orWith(items[1].id());
itemsSet1.orWith(items[2].id());
itemsSet1.orWith(items[3].id());
int var1 =3D itemsSet1.var(); // assuming that this is the UID for set1

// set2
BDD itemsSet2 =3D B.zero();
itemsSet2.orWith(items[10].id());
itemsSet2.orWith(items[11].id());
itemsSet2.orWith(items[2].id());
itemsSet2.orWith(items[3].id());
int var2 =3D itemsSet2.var();// assuming that this is the UID for set2

// pool of sets
BDD poolOfSets;
poolOfSets =3D B.one();
poolOfSets.andWith(itemsSet1.id());
poolOfSets.andWith(itemsSet2.id());

// ?later?in a different context? retrieve set1 and set2
BDD set1 =3D B.ithVar(var1); // this contains a single node only =20
(expecting itemsSet1)

BDD set2 =3D B.ithVar(var2); // this contains a single node only =20
(expecting itemsSet2)


I appreciate your help and any links to relevant tutorials or documentation.

Thanks,
Wes


----------------------------------------------------------------

Re: [Javabdd-devel] Multi-Terminal Binary Decision Diagrams

[John W. <joe...@gm...>]

Hi Florent,

I'm not aware of any Java packages that implement MTBDDs/ADDs.
However, JavaBDD does already have an interface to CUDD, so adding ADD
support through a CUDD interface wouldn't be too difficult I think.
(You would still need the CUDD native library to run your Java
application.)  Let me know if you are interested in this and I can
point you in the right direction.

-John


On 12/14/06, Florent Teichteil <flo...@ya...> wrote:
> Hi,
>
> I would like to recode a C++ application using CUDD's BDD package in
> java. JavaBDD seems to be a good candidate for replacing CUDD in the new
> java application.
>
> Unfortunately, we used Multi-Terminal Binary Decision Diagrams (ADDs -
> Algebraic Decision Diagrams - in CUDD) to code continuous functions of
> boolean variables. I am not sure that MTBDDs/ADDs are implemented in
> JavaBDD. If not, is it planned to implement it? Or does someone know
> other java packages which implement MTBDDs/ADDs?
>
> Thanks,
>
> Florent Teichteil.
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Javabdd-devel mailing list
> Jav...@li...
> https://lists.sourceforge.net/lists/listinfo/javabdd-devel
>

[Javabdd-devel] Multi-Terminal Binary Decision Diagrams

[Florent T. <flo...@ya...>]

Hi,

I would like to recode a C++ application using CUDD's BDD package in
java. JavaBDD seems to be a good candidate for replacing CUDD in the new
java application.

Unfortunately, we used Multi-Terminal Binary Decision Diagrams (ADDs -
Algebraic Decision Diagrams - in CUDD) to code continuous functions of
boolean variables. I am not sure that MTBDDs/ADDs are implemented in
JavaBDD. If not, is it planned to implement it? Or does someone know
other java packages which implement MTBDDs/ADDs?

Thanks,

Florent Teichteil.

[Javabdd-devel] Moving joeq/JavaBDD/bddbddb from CVS to Subversion

[John W. <joe...@gm...>]

Hello all,

Now that Sourceforge has Subversion support, I'm planning to move the
Java project repositories on Sourceforge from using CVS over to
Subversion.

Subversion is better that CVS in just about every possible way.  First
of all, actions on a repository are atomic.  Second, it supports
operations like renames and copies natively.  Third, it does a much
better job of working disconnected.  Fourth, with Subversion there is
no more lag between the anonymous and developer repositories.  Fifth,
it communicates using the standard https port rather than the ssh
port, so it is more firewall-friendly.  Sixth, the Eclipse Subversion
plugin (Subclipse) appears to allow you to use both the command-line
'svn' command and the Subversion commands from the UI, whereas Eclipse
CVS was incompatible with the command line 'cvs' command.  And the
list goes on and on...

The downsides to switching over to Subversion are that everyone will
have to check out a new copy of the source tree, the existing maven
scripts for autogenerating the website will break, and you can no
longer use it with Eclipse out of the box - you will need to install
Subclipse.

I've already ported over the repositories and I've tested out
committing, Eclipse integration, etc.  I've given SVN commit access to
everyone who had CVS commit access.

For the time being, the repositories are available from both CVS and
Subversion.  But at some point next week, I want to turn off CVS
access and transition entirely to Subversion.

Steps to moving over to Subversion:

 1. If you are using Eclipse, install the Subclipse plugin by going to
"Help -> Software Updates -> Find and Install", then clicking "Search
for new features to install" and "Add new remote site".  Add the site
"http://subclipse.tigris.org/update_1.0.x" and continue, then select
Subclipse to be installed.

 2. If you are using cygwin, run the cygwin setup utility and install
the subversion package.

 3. Run CVS update on your existing checkout to update to the latest
version and find any modifications you want to keep.

 4. Check out a clean copy of the tree using Eclipse.  Click on the
"Open Perspective" button, then on "SVN Repository Exploring".  Right
click and add these repositories:
    https://svn.sourceforge.net/svnroot/bddbddb
    https://svn.sourceforge.net/svnroot/javabdd
    https://svn.sourceforge.net/svnroot/joeq
Browse into the trunk directory of each repository and right-click to
"Checkout..." the projects that you need.  (If you have changes in the
existing project you want to keep, you might want to move the project
directory out of the way first.)
    You can also do it from the command line by using this command:
     svn co https://svn.sourceforge.net/svnroot/joeq/trunk joeq_core
    (Change 'joeq' to 'javabdd' or 'bddbddb' as necessary.)

 5. Move any modified files you want to keep into the newly checked
out tree, and commit them if necessary.

Command-line subversion works much like command-line CVS.  The major
difference is to see what files you have modified, you can type "svn
status" rather than "cvs update -Pd".

If you could send an email when you've successfully made the
transition, that would be great.

-John

[Javabdd-devel] Moving joeq/JavaBDD/bddbddb from CVS to Subversion

[John W. <joe...@us...>]

[Javabdd-devel] Moving joeq/JavaBDD/bddbddb from CVS to Subversion

[John W. <joe...@gm...>]

Hello all,

(I apologize in advance if you receive this email multiple times.  I'm
not sure how Sourceforge deals with sending to multiple mailing
lists.)

Now that Sourceforge has Subversion support, I'm planning to move the
Java project repositories on Sourceforge from using CVS over to
Subversion.

Subversion is better that CVS in just about every possible way.  First
of all, actions on a repository are atomic.  Second, it supports
operations like renames and copies natively.  Third, it does a much
better job of working disconnected.  Fourth, with Subversion there is
no more lag between the anonymous and developer repositories.  Fifth,
it communicates using the standard https port rather than the ssh
port, so it is more firewall-friendly.  Sixth, the Eclipse Subversion
plugin (Subclipse) appears to allow you to use both the command-line
'svn' command and the Subversion commands from the UI, whereas Eclipse
CVS was incompatible with the command line 'cvs' command.  And the
list goes on and on...

The downsides to switching over to Subversion are that everyone will
have to check out a new copy of the source tree, the existing maven
scripts for autogenerating the website will break, and you can no
longer use it with Eclipse out of the box - you will need to install
Subclipse.

I've already ported over the repositories and I've tested out
committing, Eclipse integration, etc.  I've given SVN commit access to
everyone who had CVS commit access.

For the time being, the repositories are available from both CVS and
Subversion.  But at some point next week, I want to turn off CVS
access and transition entirely to Subversion.

Steps to moving over to Subversion:

 1. If you are using Eclipse, install the Subclipse plugin by going to
"Help -> Software Updates -> Find and Install", then clicking "Search
for new features to install" and "Add new remote site".  Add the site
"http://subclipse.tigris.org/update_1.0.x" and continue, then select
Subclipse to be installed.

 2. If you are using cygwin, run the cygwin setup utility and install
the subversion package.

 3. Run CVS update on your existing checkout to update to the latest
version and find any modifications you want to keep.

 4. Check out a clean copy of the tree using Eclipse.  Click on the
"Open Perspective" button, then on "SVN Repository Exploring".  Right
click and add these repositories:
   https://svn.sourceforge.net/svnroot/bddbddb
   https://svn.sourceforge.net/svnroot/javabdd
   https://svn.sourceforge.net/svnroot/joeq
Browse into the trunk directory of each repository and right-click to
"Checkout..." the projects that you need.  (If you have changes in the
existing project you want to keep, you might want to move the project
directory out of the way first.)
   You can also do it from the command line by using this command:
    svn co https://svn.sourceforge.net/svnroot/joeq/trunk joeq
   (Change 'joeq' to 'javabdd' or 'bddbddb' as necessary.)

 5. Move any modified files you want to keep into the newly checked
out tree, and commit them if necessary.

Command-line subversion works much like command-line CVS.  The major
difference is to see what files you have modified, you can type "svn
status" rather than "cvs update -Pd".

If you could send an email when you've successfully made the
transition, that would be great.

-John

RV: [Javabdd-devel] Appling filters to variable values

[Sergio <ser...@ly...>]

 

Thank for your answer John but you misunderstood me.

 

x.satOne() return a satisfying variable assignment in which x is true but I
need set the value of several variables.

 

Imagine I have 25 variables (a,b,c,d,e,f.) with relations (and , or , biimp)
between them.

I want a satisfying variable assignment where c,d,h, and k are true at the
same time.

 

If I use c.satOne() it just works for c.!

 

I would thank you any help.

 

Sincerely

 

Sergio

 

  _____  

De: John Whaley [mailto:joe...@gm...] 
Enviado el: lunes, 23 de enero de 2006 16:52
Para: Sergio
Asunto: Re: [Javabdd-devel] Appling filters to variable values

 

See the "satOne()" method.  It will do what you want.

-John

On 1/23/06, Sergio <ser...@ly...> wrote:

Hi,

 

I am interested in setting values to variables after applying dependencies
to them. I will show what I mean with an example:

 

1.	Create variables

 

BDD x=B.ithVar(0);

BDD y=B.ithVar(1);

 ..

 

2.	Apply dependencies:

 

x.andWith(y);

 .

 

3.	Set some variables values to true or false and obtaining a
satisfying variable assignment with that restriction or filter

 

 

For instance, I would like to get a satisfying variable assignment in which
variables x, y, z, etc. is always true.

 

I have try with x=B.one() but it does not work.

 

Any help?

 

Sincerely,

 

Sergio 

 

[Javabdd-devel] Appling filters to variable values

[Sergio <ser...@ly...>]

Hi,

 

I am interested in setting values to variables after applying dependencies
to them. I will show what I mean with an example:

 

1.	Create variables

 

BDD x=B.ithVar(0);

BDD y=B.ithVar(1);

 ..

 

2.	Apply dependencies:

 

x.andWith(y);

 .

 

3.	Set some variables values to true or false and obtaining a
satisfying variable assignment with that restriction or filter

 

 

For instance, I would like to get a satisfying variable assignment in which
variables x, y, z, etc. is always true.

 

I have try with x=B.one() but it does not work.

 

Any help?

 

Sincerely,

 

Sergio