The source code and dockerfile for the GSW2024 AI Lab.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
This repo is archived. You can view files and clone it, but cannot push or open issues/pull-requests.
 
 
 
 
 
 

693 lines
27 KiB

<HTML>
<HEAD><TITLE>The dddmp package</TITLE></HEAD>
<BODY>
<DL>
<dt><pre>
<A NAME="Dddmp_Bin2Text"></A>
int <I></I>
<B>Dddmp_Bin2Text</B>(
char * <b>filein</b>, <i>IN: name of binary file</i>
char * <b>fileout</b> <i>IN: name of ASCII file</i>
)
</pre>
<dd> Converts from binary to ASCII format. A BDD array is loaded and
and stored to the target file.
<p>
<dd> <b>Side Effects</b> None
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_Text2Bin()">Dddmp_Text2Bin()</a>
</code>
<dt><pre>
<A NAME="Dddmp_Text2Bin"></A>
int <I></I>
<B>Dddmp_Text2Bin</B>(
char * <b>filein</b>, <i>IN: name of ASCII file</i>
char * <b>fileout</b> <i>IN: name of binary file</i>
)
</pre>
<dd> Converts from ASCII to binary format. A BDD array is loaded and
and stored to the target file.
<p>
<dd> <b>Side Effects</b> None
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_Bin2Text()">Dddmp_Bin2Text()</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddAddArrayLoad"></A>
int <I></I>
<B>Dddmp_cuddAddArrayLoad</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
Dddmp_RootMatchType <b>rootMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>rootmatchnames</b>, <i>IN: sorted names for loaded roots</i>
Dddmp_VarMatchType <b>varMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names, by ids</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids, by ids</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids, by ids</i>
int <b>mode</b>, <i>IN: requested input file format</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: file pointer</i>
DdNode *** <b>pproots</b> <i>OUT: array of returned BDD roots</i>
)
</pre>
<dd> Reads a dump file representing the argument ADDs. See
BDD load functions for detailed explanation.
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStore">Dddmp_cuddBddArrayStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddAddArrayStore"></A>
int <I></I>
<B>Dddmp_cuddAddArrayStore</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
char * <b>ddname</b>, <i>IN: DD name (or NULL)</i>
int <b>nRoots</b>, <i>IN: number of output BDD roots to be stored</i>
DdNode ** <b>f</b>, <i>IN: array of ADD roots to be stored</i>
char ** <b>rootnames</b>, <i>IN: array of root names (or NULL)</i>
char ** <b>varnames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>auxids</b>, <i>IN: array of converted var IDs</i>
int <b>mode</b>, <i>IN: storing mode selector</i>
Dddmp_VarInfoType <b>varinfo</b>, <i>IN: extra info for variables in text mode</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument array of ADDs to file. Dumping is
either in text or binary form. see the corresponding BDD dump
function for further details.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from the unique hash
table. They are re-linked after the store operation in a
modified order.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddAddStore">Dddmp_cuddAddStore</a>
<a href="#Dddmp_cuddAddLoad">Dddmp_cuddAddLoad</a>
<a href="#Dddmp_cuddAddArrayLoad">Dddmp_cuddAddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddAddLoad"></A>
DdNode * <I></I>
<B>Dddmp_cuddAddLoad</B>(
DdManager * <b>ddMgr</b>, <i>IN: Manager</i>
Dddmp_VarMatchType <b>varMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names by IDs</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids by IDs</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids by IDs</i>
int <b>mode</b>, <i>IN: requested input file format</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b> <i>IN: file pointer</i>
)
</pre>
<dd> Reads a dump file representing the argument ADD.
Dddmp_cuddAddArrayLoad is used through a dummy array.
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddAddStore">Dddmp_cuddAddStore</a>
<a href="#Dddmp_cuddAddArrayLoad">Dddmp_cuddAddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddAddStore"></A>
int <I></I>
<B>Dddmp_cuddAddStore</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
char * <b>ddname</b>, <i>IN: DD name (or NULL)</i>
DdNode * <b>f</b>, <i>IN: ADD root to be stored</i>
char ** <b>varnames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>auxids</b>, <i>IN: array of converted var ids</i>
int <b>mode</b>, <i>IN: storing mode selector</i>
Dddmp_VarInfoType <b>varinfo</b>, <i>IN: extra info for variables in text mode</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument ADD to file. Dumping is done through
Dddmp_cuddAddArrayStore, And a dummy array of 1 ADD root is
used for this purpose.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from unique hash. They are
re-linked after the store operation in a modified order.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddAddLoad">Dddmp_cuddAddLoad</a>
<a href="#Dddmp_cuddAddArrayLoad">Dddmp_cuddAddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayLoadCnf"></A>
int <I></I>
<B>Dddmp_cuddBddArrayLoadCnf</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
Dddmp_RootMatchType <b>rootmatchmode</b>, <i>IN: storing mode selector</i>
char ** <b>rootmatchnames</b>, <i>IN: sorted names for loaded roots</i>
Dddmp_VarMatchType <b>varmatchmode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names, by IDs</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids, by IDs</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids, by IDs</i>
int <b>mode</b>, <i>IN: computation Mode</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: file pointer</i>
DdNode *** <b>rootsPtrPtr</b>, <i>OUT: array of returned BDD roots</i>
int * <b>nRoots</b> <i>OUT: number of BDDs returned</i>
)
</pre>
<dd> Reads a dump file representing the argument BDD in a
CNF formula.
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayLoad"></A>
int <I></I>
<B>Dddmp_cuddBddArrayLoad</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
Dddmp_RootMatchType <b>rootMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>rootmatchnames</b>, <i>IN: sorted names for loaded roots</i>
Dddmp_VarMatchType <b>varMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names, by ids</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids, by ids</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids, by ids</i>
int <b>mode</b>, <i>IN: requested input file format</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: file pointer</i>
DdNode *** <b>pproots</b> <i>OUT: array of returned BDD roots</i>
)
</pre>
<dd> Reads a dump file representing the argument BDDs. The header is
common to both text and binary mode. The node list is either
in text or binary format. A dynamic vector of DD pointers
is allocated to support conversion from DD indexes to pointers.
Several criteria are supported for variable match between file
and dd manager. Several changes/permutations/compositions are allowed
for variables while loading DDs. Variable of the dd manager are allowed
to match with variables on file on ids, permids, varnames,
varauxids; also direct composition between ids and
composeids is supported. More in detail:
<ol>
<li> varMatchMode=DDDMP_VAR_MATCHIDS <p>
allows the loading of a DD keeping variable IDs unchanged
(regardless of the variable ordering of the reading manager); this
is useful, for example, when swapping DDs to file and restoring them
later from file, after possible variable reordering activations.
<li> varMatchMode=DDDMP_VAR_MATCHPERMIDS <p>
is used to allow variable match according to the position in the
ordering.
<li> varMatchMode=DDDMP_VAR_MATCHNAMES <p>
requires a non NULL varmatchnames parameter; this is a vector of
strings in one-to-one correspondence with variable IDs of the
reading manager. Variables in the DD file read are matched with
manager variables according to their name (a non NULL varnames
parameter was required while storing the DD file).
<li> varMatchMode=DDDMP_VAR_MATCHIDS <p>
has a meaning similar to DDDMP_VAR_MATCHNAMES, but integer auxiliary
IDs are used instead of strings; the additional non NULL
varmatchauxids parameter is needed.
<li> varMatchMode=DDDMP_VAR_COMPOSEIDS <p>
uses the additional varcomposeids parameter is used as array of
variable ids to be composed with ids stored in file.
</ol>
In the present implementation, the array varnames (3), varauxids (4)
and composeids (5) need to have one entry for each variable in the
DD manager (NULL pointers are allowed for unused variables
in varnames). Hence variables need to be already present in the
manager. All arrays are sorted according to IDs.
All the loaded BDDs are referenced before returning them.
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStore">Dddmp_cuddBddArrayStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayStoreBlif"></A>
int <I></I>
<B>Dddmp_cuddBddArrayStoreBlif</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nroots</b>, <i>IN: number of output BDD roots to be stored</i>
DdNode ** <b>f</b>, <i>IN: array of BDD roots to be stored</i>
char ** <b>inputNames</b>, <i>IN: array of variable names (or NULL)</i>
char ** <b>outputNames</b>, <i>IN: array of root names (or NULL)</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStoreBLif.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStorePrefix">Dddmp_cuddBddArrayStorePrefix</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayStoreCnf"></A>
int <I></I>
<B>Dddmp_cuddBddArrayStoreCnf</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
DdNode ** <b>f</b>, <i>IN: array of BDD roots to be stored</i>
int <b>rootN</b>, <i>IN: # output BDD roots to be stored</i>
Dddmp_DecompCnfStoreType <b>mode</b>, <i>IN: format selection</i>
int <b>noHeader</b>, <i>IN: do not store header iff 1</i>
char ** <b>varNames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>bddIds</b>, <i>IN: array of converted var IDs</i>
int * <b>bddAuxIds</b>, <i>IN: array of BDD node Auxiliary Ids</i>
int * <b>cnfIds</b>, <i>IN: array of converted var IDs</i>
int <b>idInitial</b>, <i>IN: starting id for cutting variables</i>
int <b>edgeInTh</b>, <i>IN: Max # Incoming Edges</i>
int <b>pathLengthTh</b>, <i>IN: Max Path Length</i>
char * <b>fname</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: pointer to the store file</i>
int * <b>clauseNPtr</b>, <i>OUT: number of clause stored</i>
int * <b>varNewNPtr</b> <i>OUT: number of new variable created</i>
)
</pre>
<dd> Dumps the argument array of BDDs to file.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from the unique hash
table. They are re-linked after the store operation in a
modified order.
Three methods are allowed:
* NodeByNode method: Insert a cut-point for each BDD node (but the
terminal nodes)
* MaxtermByMaxterm method: Insert no cut-points, i.e. the off-set of
trhe function is stored
* Best method: Tradeoff between the previous two methods.
Auxiliary variables, i.e., cut points are inserted following these
criterias:
* edgeInTh
indicates the maximum number of incoming edges up to which
no cut point (auxiliary variable) is inserted.
If edgeInTh:
* is equal to -1 no cut point due to incoming edges are inserted
(MaxtermByMaxterm method.)
* is equal to 0 a cut point is inserted for each node with a single
incoming edge, i.e., each node, (NodeByNode method).
* is equal to n a cut point is inserted for each node with (n+1)
incoming edges.
* pathLengthTh
indicates the maximum length path up to which no cut points
(auxiliary variable) is inserted.
If the path length between two nodes exceeds this value, a cut point
is inserted.
If pathLengthTh:
* is equal to -1 no cut point due path length are inserted
(MaxtermByMaxterm method.)
* is equal to 0 a cut point is inserted for each node (NodeByNode
method).
* is equal to n a cut point is inserted on path whose length is
equal to (n+1).
Notice that the maximum number of literals in a clause is equal
to (pathLengthTh + 2), i.e., for each path we have to keep into
account a CNF variable for each node plus 2 added variables for
the bottom and top-path cut points.
The stored file can contain a file header or not depending on the
noHeader parameter (IFF 0, usual setting, the header is usually stored.
This option can be useful in storing multiple BDDs, as separate BDDs,
on the same file leaving the opening of the file to the caller.
<p>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayStorePrefix"></A>
int <I></I>
<B>Dddmp_cuddBddArrayStorePrefix</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nroots</b>, <i>IN: number of output BDD roots to be stored</i>
DdNode ** <b>f</b>, <i>IN: array of BDD roots to be stored</i>
char ** <b>inputNames</b>, <i>IN: array of variable names (or NULL)</i>
char ** <b>outputNames</b>, <i>IN: array of root names (or NULL)</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStorePrefix.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStore">Dddmp_cuddBddArrayStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayStoreSmv"></A>
int <I></I>
<B>Dddmp_cuddBddArrayStoreSmv</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nroots</b>, <i>IN: number of output BDD roots to be stored</i>
DdNode ** <b>f</b>, <i>IN: array of BDD roots to be stored</i>
char ** <b>inputNames</b>, <i>IN: array of variable names (or NULL)</i>
char ** <b>outputNames</b>, <i>IN: array of root names (or NULL)</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStorePrefix.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStore">Dddmp_cuddBddArrayStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddArrayStore"></A>
int <I></I>
<B>Dddmp_cuddBddArrayStore</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
char * <b>ddname</b>, <i>IN: dd name (or NULL)</i>
int <b>nRoots</b>, <i>IN: number of output BDD roots to be stored</i>
DdNode ** <b>f</b>, <i>IN: array of BDD roots to be stored</i>
char ** <b>rootnames</b>, <i>IN: array of root names (or NULL)</i>
char ** <b>varnames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>auxids</b>, <i>IN: array of converted var IDs</i>
int <b>mode</b>, <i>IN: storing mode selector</i>
Dddmp_VarInfoType <b>varinfo</b>, <i>IN: extra info for variables in text mode</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument array of BDDs to file. Dumping is either
in text or binary form. BDDs are stored to the fp (already
open) file if not NULL. Otherwise the file whose name is
fname is opened in write mode. The header has the same format
for both textual and binary dump. Names are allowed for input
variables (vnames) and for represented functions (rnames).
For sake of generality and because of dynamic variable
ordering both variable IDs and permuted IDs are included.
New IDs are also supported (auxids). Variables are identified
with incremental numbers. according with their positiom in
the support set. In text mode, an extra info may be added,
chosen among the following options: name, ID, PermID, or an
auxiliary id. Since conversion from DD pointers to integers
is required, DD nodes are temporarily removed from the unique
hash table. This allows the use of the next field to store
node IDs.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from the unique hash
table. They are re-linked after the store operation in a
modified order.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStore">Dddmp_cuddBddStore</a>
<a href="#Dddmp_cuddBddLoad">Dddmp_cuddBddLoad</a>
<a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddDisplayBinary"></A>
int <I></I>
<B>Dddmp_cuddBddDisplayBinary</B>(
char * <b>fileIn</b>, <i>IN: name of binary file</i>
char * <b>fileOut</b> <i>IN: name of text file</i>
)
</pre>
<dd> Display a binary dump file in a text file
<p>
<dd> <b>Side Effects</b> None
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStore">Dddmp_cuddBddStore</a>
<a href="#Dddmp_cuddBddLoad">Dddmp_cuddBddLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddLoadCnf"></A>
int <I></I>
<B>Dddmp_cuddBddLoadCnf</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
Dddmp_VarMatchType <b>varmatchmode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names, by IDs</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids, by IDs</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids accessed, by IDs</i>
int <b>mode</b>, <i>IN: computation mode</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: file pointer</i>
DdNode *** <b>rootsPtrPtr</b>, <i>OUT: array of returned BDD roots</i>
int * <b>nRoots</b> <i>OUT: number of BDDs returned</i>
)
</pre>
<dd> Reads a dump file representing the argument BDD in a
CNF formula.
Dddmp_cuddBddArrayLoadCnf is used through a dummy array.
The results is returned in different formats depending on the
mode selection:
IFF mode == 0 Return the Clauses without Conjunction
IFF mode == 1 Return the sets of BDDs without Quantification
IFF mode == 2 Return the sets of BDDs AFTER Existential Quantification
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddLoad">Dddmp_cuddBddLoad</a>
<a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddLoad"></A>
DdNode * <I></I>
<B>Dddmp_cuddBddLoad</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
Dddmp_VarMatchType <b>varMatchMode</b>, <i>IN: storing mode selector</i>
char ** <b>varmatchnames</b>, <i>IN: array of variable names - by IDs</i>
int * <b>varmatchauxids</b>, <i>IN: array of variable auxids - by IDs</i>
int * <b>varcomposeids</b>, <i>IN: array of new ids accessed - by IDs</i>
int <b>mode</b>, <i>IN: requested input file format</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b> <i>IN: file pointer</i>
)
</pre>
<dd> Reads a dump file representing the argument BDD.
Dddmp_cuddBddArrayLoad is used through a dummy array (see this
function's description for more details).
Mode, the requested input file format, is checked against
the file format.
The loaded BDDs is referenced before returning it.
<p>
<dd> <b>Side Effects</b> A vector of pointers to DD nodes is allocated and freed.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStore">Dddmp_cuddBddStore</a>
<a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddStoreBlif"></A>
int <I></I>
<B>Dddmp_cuddBddStoreBlif</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nRoots</b>, <i>IN: Number of BDD roots</i>
DdNode * <b>f</b>, <i>IN: BDD root to be stored</i>
char ** <b>inputNames</b>, <i>IN: Array of variable names</i>
char ** <b>outputNames</b>, <i>IN: Array of root names</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fileName</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStoreBlif.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStorePrefix">Dddmp_cuddBddStorePrefix</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddStoreCnf"></A>
int <I></I>
<B>Dddmp_cuddBddStoreCnf</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
DdNode * <b>f</b>, <i>IN: BDD root to be stored</i>
Dddmp_DecompCnfStoreType <b>mode</b>, <i>IN: format selection</i>
int <b>noHeader</b>, <i>IN: do not store header iff 1</i>
char ** <b>varNames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>bddIds</b>, <i>IN: array of var ids</i>
int * <b>bddAuxIds</b>, <i>IN: array of BDD node Auxiliary Ids</i>
int * <b>cnfIds</b>, <i>IN: array of CNF var ids</i>
int <b>idInitial</b>, <i>IN: starting id for cutting variables</i>
int <b>edgeInTh</b>, <i>IN: Max # Incoming Edges</i>
int <b>pathLengthTh</b>, <i>IN: Max Path Length</i>
char * <b>fname</b>, <i>IN: file name</i>
FILE * <b>fp</b>, <i>IN: pointer to the store file</i>
int * <b>clauseNPtr</b>, <i>OUT: number of clause stored</i>
int * <b>varNewNPtr</b> <i>OUT: number of new variable created</i>
)
</pre>
<dd> Dumps the argument BDD to file.
This task is performed by calling the function
Dddmp_cuddBddArrayStoreCnf.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from unique hash. They are
re-linked after the store operation in a modified order.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayStoreCnf">Dddmp_cuddBddArrayStoreCnf</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddStorePrefix"></A>
int <I></I>
<B>Dddmp_cuddBddStorePrefix</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nRoots</b>, <i>IN: Number of BDD roots</i>
DdNode * <b>f</b>, <i>IN: BDD root to be stored</i>
char ** <b>inputNames</b>, <i>IN: Array of variable names</i>
char ** <b>outputNames</b>, <i>IN: Array of root names</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fileName</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStorePrefix.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStore">Dddmp_cuddBddStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddStoreSmv"></A>
int <I></I>
<B>Dddmp_cuddBddStoreSmv</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
int <b>nRoots</b>, <i>IN: Number of BDD roots</i>
DdNode * <b>f</b>, <i>IN: BDD root to be stored</i>
char ** <b>inputNames</b>, <i>IN: Array of variable names</i>
char ** <b>outputNames</b>, <i>IN: Array of root names</i>
char * <b>modelName</b>, <i>IN: Model Name</i>
char * <b>fileName</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file.
Dumping is done through Dddmp_cuddBddArrayStorePrefix.
A dummy array of 1 BDD root is used for this purpose.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddStore">Dddmp_cuddBddStore</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddBddStore"></A>
int <I></I>
<B>Dddmp_cuddBddStore</B>(
DdManager * <b>ddMgr</b>, <i>IN: DD Manager</i>
char * <b>ddname</b>, <i>IN: DD name (or NULL)</i>
DdNode * <b>f</b>, <i>IN: BDD root to be stored</i>
char ** <b>varnames</b>, <i>IN: array of variable names (or NULL)</i>
int * <b>auxids</b>, <i>IN: array of converted var ids</i>
int <b>mode</b>, <i>IN: storing mode selector</i>
Dddmp_VarInfoType <b>varinfo</b>, <i>IN: extra info for variables in text mode</i>
char * <b>fname</b>, <i>IN: File name</i>
FILE * <b>fp</b> <i>IN: File pointer to the store file</i>
)
</pre>
<dd> Dumps the argument BDD to file. Dumping is done through
Dddmp_cuddBddArrayStore. A dummy array of 1 BDD root is
used for this purpose.
<p>
<dd> <b>Side Effects</b> Nodes are temporarily removed from unique hash. They are
re-linked after the store operation in a modified order.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddLoad">Dddmp_cuddBddLoad</a>
<a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddHeaderLoadCnf"></A>
int <I></I>
<B>Dddmp_cuddHeaderLoadCnf</B>(
int * <b>nVars</b>, <i>OUT: number of DD variables</i>
int * <b>nsuppvars</b>, <i>OUT: number of support variables</i>
char *** <b>suppVarNames</b>, <i>OUT: array of support variable names</i>
char *** <b>orderedVarNames</b>, <i>OUT: array of variable names</i>
int ** <b>varIds</b>, <i>OUT: array of variable ids</i>
int ** <b>varComposeIds</b>, <i>OUT: array of permids ids</i>
int ** <b>varAuxIds</b>, <i>OUT: array of variable aux ids</i>
int * <b>nRoots</b>, <i>OUT: number of root in the file</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b> <i>IN: file pointer</i>
)
</pre>
<dd> Reads the header of a dump file representing the argument BDDs.
Returns main information regarding DD type stored in the file,
the variable ordering used, the number of variables, etc.
It reads only the header of the file NOT the BDD/ADD section.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
<dt><pre>
<A NAME="Dddmp_cuddHeaderLoad"></A>
int <I></I>
<B>Dddmp_cuddHeaderLoad</B>(
Dddmp_DecompType * <b>ddType</b>, <i>OUT: selects the proper decomp type</i>
int * <b>nVars</b>, <i>OUT: number of DD variables</i>
int * <b>nsuppvars</b>, <i>OUT: number of support variables</i>
char *** <b>suppVarNames</b>, <i>OUT: array of support variable names</i>
char *** <b>orderedVarNames</b>, <i>OUT: array of variable names</i>
int ** <b>varIds</b>, <i>OUT: array of variable ids</i>
int ** <b>varComposeIds</b>, <i>OUT: array of permids ids</i>
int ** <b>varAuxIds</b>, <i>OUT: array of variable aux ids</i>
int * <b>nRoots</b>, <i>OUT: number of root in the file</i>
char * <b>file</b>, <i>IN: file name</i>
FILE * <b>fp</b> <i>IN: file pointer</i>
)
</pre>
<dd> Reads the header of a dump file representing the argument BDDs.
Returns main information regarding DD type stored in the file,
the variable ordering used, the number of variables, etc.
It reads only the header of the file NOT the BDD/ADD section.
<p>
<dd> <b>See Also</b> <code><a href="#Dddmp_cuddBddArrayLoad">Dddmp_cuddBddArrayLoad</a>
</code>
</DL>
<HR>
Last updated on 1040218 17h14
</BODY></HTML>