.fp 5 CW
.TH LIBCDT 3
.SH NAME
\fBCdt\fR \- container data types
.SH SYNOPSIS
.de Tp
.fl
.ne 2
.TP
..
.de Ss
.fl
.ne 2
.SS "\\$1"
..
.de Cs
.nf
.ft 5
..
.de Ce
.ft 1
.fi
..
.ta 1.0i 2.0i 3.0i 4.0i 5.0i
.Cs
#include <cdt.h>
.Ce
.Ss "DICTIONARY TYPES"
.Cs
Dt_t;
Dtdisc_t;
Dtmethod_t;
Dtlink_t;
Dtstat_t;
.Ce
.Ss "DICTIONARY CONTROL"
.Cs
Dt_t*       dtopen(const Dtdisc_t* disc, const Dtmethod_t* meth);
int         dtclose(Dt_t* dt);
void        dtclear(dt);
Dtmethod_t* dtmethod(Dt_t* dt, const Dtmethod_t* meth);
Dtdisc_t*   dtdisc(Dt_t* dt, const Dtdisc_t* disc);
Dt_t*       dtview(Dt_t* dt, Dt_t* view);
.Ce
.Ss "STORAGE METHODS"
.Cs
Dtmethod_t* Dtset;
Dtmethod_t* Dtoset;
Dtmethod_t* Dtobag;
Dtmethod_t* Dtqueue;
.Ce
.Ss "DISCIPLINE"
.Cs
#define DTDISC(disc,key,size,link,makef,freef,comparf)
typedef void*      (*Dtmake_f)(void*, Dtdisc_t*);
typedef void         (*Dtfree_f)(void*, Dtdisc_t*);
typedef int          (*Dtcompar_f)(Dt_t*, void*, void*, Dtdisc_t*);
.Ce
.Ss "OBJECT OPERATIONS"
.Cs
void*   dtinsert(Dt_t* dt, void* obj);
void*   dtdelete(Dt_t* dt, void* obj);
void*   dtdetach(Dt_t* dt, void* obj);
void*   dtsearch(Dt_t* dt, void* obj);
void*   dtmatch(Dt_t* dt, void* key);
void*   dtfirst(Dt_t* dt);
void*   dtnext(Dt_t* dt, void* obj);
void*   dtlast(Dt_t* dt);
void*   dtprev(Dt_t* dt, void* obj);
void*   dtfinger(Dt_t* dt);
void*   dtrenew(Dt_t* dt, void* obj);
int       dtwalk(Dt_t* dt, int (*userf)(void*, void*), void*);
Dtlink_t* dtflatten(Dt_t* dt);
Dtlink_t* dtlink(Dt_t*, Dtlink_t* link);
void*   dtobj(Dt_t* dt, Dtlink_t* link);
Dtlink_t* dtextract(Dt_t* dt);
int       dtrestore(Dt_t* dt, Dtlink_t* link);
.Ce
.Ss "DICTIONARY STATUS"
.Cs
int       dtsize(Dt_t* dt);
int       dtstat(Dt_t* dt, Dtstat_t*, int all);
.Ce
.Ss "HASH FUNCTIONS"
.Cs
unsigned int dtstrhash(void *str, int n);
.Ce
.SH DESCRIPTION
.PP
\fICdt\fP manages run-time dictionaries using standard container data types:
unordered set/multiset, ordered set/multiset, list, stack, and queue.
.PP
.Ss "DICTIONARY TYPES"
.PP
.Ss "  Dt_t"
This is the type of a dictionary handle.
.PP
.Ss "  Dtdisc_t"
This defines the type of a discipline structure which describes
object lay-out and manipulation functions.
.PP
.Ss "  Dtmethod_t"
This defines the type of a container method.
.PP
.Ss "  Dtlink_t"
This is the type of a dictionary object holder (see \f5dtdisc()\fP.)
.PP
.Ss "  Dtstat_t"
This is the type of a structure to return dictionary statistics (see \f5dtstat()\fP.)
.PP
.Ss "DICTIONARY CONTROL"
.PP
.Ss "  Dt_t* dtopen(const Dtdisc_t* disc, const Dtmethod_t* meth)"
This creates a new dictionary.
\f5disc\fP is a discipline structure to describe object format.
\f5meth\fP specifies a manipulation method.
\f5dtopen()\fP returns the new dictionary or \f5NULL\fP on error.
.PP
.Ss "  int dtclose(Dt_t* dt)"
This deletes \f5dt\fP and its objects.
Note that \f5dtclose()\fP fails if \f5dt\fP is being viewed by
some other dictionaries (see \f5dtview()\fP).
\f5dtclose()\fP returns \f50\fP on success and \f5-1\fP on error.
.PP
.Ss "  void dtclear(Dt_t* dt)"
This deletes all objects in \f5dt\fP without closing \f5dt\fP.
.PP
.Ss "  Dtmethod_t dtmethod(Dt_t* dt, const Dtmethod_t* meth)"
If \f5meth\fP is \f5NULL\fP, \f5dtmethod()\fP returns the current method.
Otherwise, it changes the storage method of \f5dt\fP to \f5meth\fP.
Object order remains the same during a
method switch for \f5Dtqueue\fP.
Switching to and from \f5Dtset\fP and \f5Dtoset/Dtobag\fP may cause
objects to be rehashed, reordered, or removed as the case requires.
\f5dtmethod()\fP returns the previous method or \f5NULL\fP on error.
.PP
.Ss "  Dtdisc_t* dtdisc(Dt_t* dt, const Dtdisc_t* disc)"
If \f5disc\fP is \f5NULL\fP, \f5dtdisc()\fP returns the current discipline.
Otherwise, it changes the discipline of \f5dt\fP to \f5disc\fP.
Objects may be rehashed, reordered, or removed as appropriate.
\f5dtdisc()\fP returns the previous discipline on success
and \f5NULL\fP on error.
.PP
.Ss "  Dt_t* dtview(Dt_t* dt, Dt_t* view)"
A viewpath allows a search or walk starting from a dictionary to continue to another.
\f5dtview()\fP first terminates any current view from \f5dt\fP to another dictionary.
Then, if \f5view\fP is \f5NULL\fP, \f5dtview\fP returns the terminated view dictionary.
If \f5view\fP is not \f5NULL\fP, a viewpath from \f5dt\fP to \f5view\fP is established.
\f5dtview()\fP returns \f5dt\fP on success and \f5NULL\fP on error.
.PP
It is an error to have dictionaries on a viewpath with different storage methods.
In addition, dictionaries on the same view path should
treat objects in a consistent manner with respect to comparison or hashing.
If not, undefined behaviors may result.
.PP
.Ss "STORAGE METHODS"
.PP
Storage methods are of type \f5Dtmethod_t*\fP.
\fICdt\fP supports the following methods:
.PP
.Ss "  Dtoset"
.Ss "  Dtobag"
Objects are ordered by comparisons.
\f5Dtoset\fP keeps unique objects.
\f5Dtobag\fP allows repeatable objects.
.PP
.Ss "  Dtset"
Objects are unordered.
\f5Dtset\fP keeps unique objects.
This method uses a hash table with chaining to manage the objects.
.PP
.Ss "  Dtqueue"
Objects are kept in a queue, i.e., in order of insertion.
Thus, the first object inserted is at queue head
and will be the first to be deleted.
.PP
.Ss "DISCIPLINE"
.PP
Object format and associated management functions are
defined in the type \f5Dtdisc_t\fP:
.Cs
    typedef struct
    { int        key, size;
      int        link;
      Dtmake_f   makef;
      Dtfree_f   freef;
      Dtcompar_f comparf;
    } Dtdisc_t;
.Ce
.Ss "  int key, size"
Each object \f5obj\fP is identified by a key used for object comparison or hashing.
\f5key\fP should be non-negative and defines an offset into \f5obj\fP.
If \f5size\fP is negative, the key is a null-terminated
string with starting address \f5*(void**)((char*)obj+key)\fP.
If \f5size\fP is zero, the key is a null-terminated string with starting address
\f5(void*)((char*)obj+key)\fP.
Finally, if \f5size\fP is positive, the key is a byte array of length \f5size\fP
starting at \f5(void*)((char*)obj+key)\fP.
.PP
.Ss "  int link"
Let \f5obj\fP be an object to be inserted into \f5dt\fP as discussed below.
If \f5link\fP is negative, an internally allocated object holder is used
to hold \f5obj\fP. Otherwise, \f5obj\fP should have
a \f5Dtlink_t\fP structure embedded \f5link\fP bytes into it,
i.e., at address \f5(Dtlink_t*)((char*)obj+link)\fP.
.PP
.Ss "  void* (*makef)(void* obj, Dtdisc_t* disc)"
If \f5makef\fP is not \f5NULL\fP,
\f5dtinsert(dt,obj)\fP will call it
to make a copy of \f5obj\fP suitable for insertion into \f5dt\fP.
If \f5makef\fP is \f5NULL\fP, \f5obj\fP itself will be inserted into \f5dt\fP.
.PP
.Ss "  void (*freef)(void* obj, Dtdisc_t* disc)"
If not \f5NULL\fP,
\f5freef\fP is used to destroy data associated with \f5obj\fP.
.PP
.Ss "int (*comparf)(Dt_t* dt, void* key1, void* key2, Dtdisc_t* disc)"
If not \f5NULL\fP, \f5comparf\fP is used to compare two keys.
Its return value should be \f5<0\fP, \f5=0\fP, or \f5>0\fP to indicate
whether \f5key1\fP is smaller, equal to, or larger than \f5key2\fP.
All three values are significant for method \f5Dtoset\fP and \f5Dtobag\fP.
For other methods, a zero value
indicates equality and a non-zero value indicates inequality.
If \f5(*comparf)()\fP is \f5NULL\fP, an internal function is used
to compare the keys as defined by the \f5Dtdisc_t.size\fP field.
.PP
.Ss "#define DTDISC(disc,key,size,link,makef,freef,comparf)"
This macro function initializes the discipline pointed to by \f5disc\fP
with the given values.
.PP
.Ss "OBJECT OPERATIONS"
.PP
.Ss "  void* dtinsert(Dt_t* dt, void* obj)"
This function adds an object prototyped by \f5obj\fP into \f5dt\fP.
\f5dtinsert()\fP performs the same function
for all methods.
If there is an existing object in \f5dt\fP matching \f5obj\fP
and the storage method is \f5Dtset\fP or \f5Dtoset\fP,
\f5dtinsert()\fP will simply return the matching object.
Otherwise, a new object is inserted according to the method in use.
See \f5Dtdisc_t.makef\fP for object construction.
The new object or a matching object as noted will be returned on success
while \f5NULL\fP is returned on error.
.PP
.Ss "  void* dtdelete(Dt_t* dt, void* obj)"
If \f5obj\fP is \f5NULL\fP, method \f5Dtqueue\fP
deletes queue head while other methods do nothing.
If \f5obj\fP is not \f5NULL\fP, there are two cases.
If the method in use is not \f5Dtobag\fP,
the first object matching \f5obj\fP is deleted.
On the other hand, if the method in use is or \f5Dtobag\fP,
the library check to see if \f5obj\fP is in the dictionary and delete it.
If \f5obj\fP is not in the dictionary, some object matching it will be deleted.
See \f5Dtdisc_t.freef\fP for object destruction.
\f5dtdelete()\fP returns the deleted object (even if it was deallocated)
or \f5NULL\fP on error.
.PP
.Ss "  void* dtdetach(Dt_t* dt, void* obj)"
This function is similar to \f5dtdelete()\fP but the object to be deleted
from \f5dt\fP will not be freed (via the discipline \f5freef\fP function).
.PP
.Ss "  void* dtsearch(Dt_t* dt, void* obj)"
.Ss "  void* dtmatch(Dt_t* dt, void* key)"
These functions find an object matching \f5obj\fP or \f5key\fP either from \f5dt\fP or
from some dictionary accessible from \f5dt\fP via a viewpath (see \f5dtview()\fP.)
\f5dtsearch()\fP and \f5dtmatch()\fP return the matching object or
\f5NULL\fP on failure.
.PP
.Ss "  void* dtfirst(Dt_t* dt)"
.Ss "  void* dtnext(Dt_t* dt, void* obj)"
\f5dtfirst()\fP returns the first object in \f5dt\fP.
\f5dtnext()\fP returns the object following \f5obj\fP.
Objects are ordered based on the storage method in use.
For \f5Dtoset\fP and \f5Dtobag\fP, objects are ordered by object comparisons.
For \f5Dtqueue\fP, objects are ordered in order of insertion.
For \f5Dtset\fP,
objects are ordered by some internal order (more below).
Thus, objects in a dictionary or a viewpath can be walked using
a \f5for(;;)\fP loop as below.
.Cs
    for(obj = dtfirst(dt); obj; obj = dtnext(dt,obj))
.Ce
When a dictionary uses \f5Dtset\fP,
the object order is determined upon a call to \f5dtfirst()\fP/\f5dtlast()\fP.
This order is frozen until a call \f5dtnext()\fP/\f5dtprev()\fP returns \f5NULL\fP
or when these same functions are called with a \f5NULL\fP object argument.
It is important that a \f5dtfirst()/dtlast()\fP call be
balanced by a \f5dtnext()/dtprev()\fP call as described.
Nested loops will require multiple balancing, once per loop.
If loop balancing is not done carefully, either performance is degraded
or unexpected behaviors may result.
.Ss "  void* dtlast(Dt_t* dt)"
.Ss "  void* dtprev(Dt_t* dt, void* obj)"
\f5dtlast()\fP and \f5dtprev()\fP are like \f5dtfirst()\fP and \f5dtnext()\fP
but work in reverse order.
Note that dictionaries on a viewpath are still walked in order
but objects in each dictionary are walked in reverse order.
.PP
.Ss "  void* dtfinger(Dt_t* dt)"
This function returns the \fIcurrent object\fP of \f5dt\fP, if any.
The current object is defined after a successful call to one of
\f5dtsearch()\fP, \f5dtmatch()\fP, \f5dtinsert()\fP,
\f5dtfirst()\fP, \f5dtnext()\fP, \f5dtlast()\fP, or \f5dtprev()\fP.
As a side effect of this implementation of \fICdt\fP,
when a dictionary is based on \f5Dtoset\fP and \f5Dtobag\fP,
the current object is always defined and is the root of the tree.
.PP
.Ss "  void* dtrenew(Dt_t* dt, void* obj)"
This function repositions and perhaps rehashes
an object \f5obj\fP after its key has been changed.
\f5dtrenew()\fP only works if \f5obj\fP is the current object (see \f5dtfinger()\fP).
.PP
.Ss "  dtwalk(Dt_t* dt, int (*userf)(void*, void*), void* data)"
This function calls \f5(*userf)(obj,data)\fP on each object in \f5dt\fP and
other dictionaries viewable from it.
If \f5userf()\fP returns a \f5<0\fP value,
\f5dtwalk()\fP terminates and returns the same value.
\f5dtwalk()\fP returns \f50\fP on completion.
.PP
.Ss "  Dtlink_t* dtflatten(Dt_t* dt)"
.Ss "  Dtlink_t* dtlink(Dt_t* dt, Dtlink_t* link)"
.Ss "  void* dtobj(Dt_t* dt, Dtlink_t* link)"
Using \f5dtfirst()/dtnext()\fP or \f5dtlast()/dtprev()\fP
to walk a single dictionary can incur significant cost due to function calls.
For efficient walking of a single directory (i.e., no viewpathing),
\f5dtflatten()\fP and \f5dtlink()\fP can be used.
Objects in \f5dt\fP are made into a linked list and walked as follows:
.Cs
    for(link = dtflatten(dt); link; link = dtlink(dt,link) )
.Ce
.PP
Note that \f5dtflatten()\fP returns a list of type \f5Dtlink_t*\fP,
not \f5void*\fP. That is, it returns a dictionary holder pointer,
not a user object pointer
(although both are the same if the discipline field \f5link\fP is zero.)
The macro function \f5dtlink()\fP
returns the dictionary holder object following \f5link\fP.
The macro function \f5dtobj(dt,link)\fP
returns the user object associated with \f5link\fP,
Beware that the flattened object list is unflattened on any
dictionary operations other than \f5dtlink()\fP.
.PP
.Ss "  Dtlink_t* dtextract(Dt_t* dt)"
.Ss "  int dtrestore(Dt_t* dt, Dtlink_t* link)"
\f5dtextract()\fP extracts all objects from \f5dt\fP and makes it appear empty.
\f5dtrestore()\fP repopulates \f5dt\fP with
objects previously obtained via \f5dtextract()\fP.
\f5dtrestore()\fP will fail if \f5dt\fP is not empty.
These functions can be used
to share a same \f5dt\fP handle among many sets of objects.
They are useful to reduce dictionary overhead
in an application that creates many concurrent dictionaries.
It is important that the same discipline and method are in use at both
extraction and restoration. Otherwise, undefined behaviors may result.
.PP
.Ss "DICTIONARY INFORMATION"
.PP
.Ss "  int dtsize(Dt_t* dt)"
This function returns the number of objects stored in \f5dt\fP.
.PP
.Ss "  int dtstat(Dt_t *dt, Dtstat_t* st, int all)"
This function reports dictionary statistics.
If \f5all\fP is non-zero, all fields of \f5st\fP are filled.
Otherwise, only the \f5dt_type\fP and \f5dt_size\fP fields are filled.
It returns \f50\fP on success and \f5-1\fP on error.
.PP
\f5Dtstat_t\fP contains the below fields:
.Tp
\f5int dt_type\fP:
This is one of \f5DT_SET\fP, \f5DT_OSET\fP, \f5DT_OBAG\fP,
and \f5DT_QUEUE\fP.
.Tp
\f5int dt_size\fP:
This contains the number of objects in the dictionary.
.Tp
\f5int dt_n\fP:
For \f5Dtset\fP,
this is the number of non-empty chains in the hash table.
For \f5Dtoset\fP and \f5Dtobag\fP,
this is the deepest level in the tree (counting from zero.)
Each level in the tree contains all nodes of equal distance from the root node.
\f5dt_n\fP and the below two fields are undefined for other methods.
.Tp
\f5int dt_max\fP:
For \f5Dtset\fP, this is the size of a largest chain.
For \f5Dtoset\fP and \f5Dtobag\fP, this is the size of a largest level.
.Tp
\f5int* dt_count\fP:
For \f5Dtset\fP,
this is the list of counts for chains of particular sizes.
For example, \f5dt_count[1]\fP is the number of chains of size \f51\fP.
For \f5Dtoset\fP and \f5Dtobag\fP, this is the list of sizes of the levels.
For example, \f5dt_count[1]\fP is the size of level \f51\fP.
.PP
.Ss "HASH FUNCTIONS"
.PP
.Ss "  unsigned int dtstrhash(void *str, int n)"
This function computes hash values from bytes or strings.
\f5dtstrhash()\fP computes a new hash value from string \f5str\fP.
If \f5n\fP is positive, \f5str\fP is a byte array of length \f5n\fP;
otherwise, \f5str\fP is a null-terminated string.
.PP
.SH IMPLEMENTATION NOTES
\f5Dtset\fP are based on hash tables with
move-to-front collision chains.
\f5Dtoset\fP and \f5Dtobag\fP are based on top-down splay trees.
\f5Dtqueue\fP is based on doubly linked list.
.PP
.SH AUTHOR
Kiem-Phong Vo, kpv@research.att.com