Next: Introduction, Previous: (dir), Up: (dir) [Contents][Index]
This is the cl-heap Reference Manual, version 0.1.6, generated automatically by Declt version 3.0 "Montgomery Scott" on Tue Dec 22 12:23:24 2020 GMT+0.
• Introduction | What cl-heap is all about | |
• Systems | The systems documentation | |
• Files | The files documentation | |
• Packages | The packages documentation | |
• Definitions | The symbols documentation | |
• Indexes | Concepts, functions, variables and data types |
CL-HEAP is a Common Lisp library that provides efficient priority queue and heap implementations. Heaps are tree data structure in which any child, c, of a parent, p, obeys the relation p R c, for some relation R (often less-than or greater-than). Heaps can be used directly as priority queues, and are important components of many algorithms, like those for finding shortest paths and calculating minimum spanning trees.
The library is simple to use. Here's an example covering most of the priority queue functionality:
(defparameter *queue* (make-instance 'cl-heap:priority-queue))
(cl-heap:enqueue *queue* 'test 15) ; Enqueue the item with the priority of 15.
(cl-heap:enqueue *queue* 'this -5)
(cl-heap:enqueue *queue* 'a 10)
(cl-heap:enqueue *queue* 'is 5)
(cl-heap:peep-at-queue *queue*) => 'this
(cl-heap:dequeue *queue*) => 'this
(cl-heap:dequeue *queue*) => 'is
(cl-heap:dequeue *queue*) => 'a
(cl-heap:dequeue *queue*) => 'test
The library provides an implementation of an array-based binary heap - offering O(log(n)) times for most operations - and the Fibonacci heap. While any sequence of arbitrary operations on a Fibonacci heap occur in amortised logarithmic time, many common operations occur in constant and amortised constant time. See below for further details. The Fibonacci heap should generally be your heap of choice from this library.
CL-HEAP depends on XLUNIT, which is used to perform unit testing. CL-HEAP has been tested and is known to run on SBCL. If you've tried this on other compilers and it runs successfully, please let me know. The library can be installed using either ASDF-INSTALL or QUICKLISP:
(require 'asdf-install)
(asdf-install:install 'cl-heap)
(quicklisp:quickload 'cl-heap)
(quicklisp:quickload 'cl-heap-tests)
After which, the library can then be loaded using either ASDF or QUICKLISP:
(asdf:load-system 'cl-heap)
CL-HEAP comes with a test suite. The tests for the various classes can be performed as follows:
(xlunit:textui-test-run (xlunit:get-suite cl-heap:fibonacci-test))
(xlunit:textui-test-run (xlunit:get-suite cl-heap:binary-test))
(xlunit:textui-test-run (xlunit:get-suite cl-heap:priority-queue-test))
This priority queue is a container for items in which all the items are sorted by some given priority. We implement the priority queue using a Fibonacci heap.
MAKE-INSTANCE accepts the argument :sort-fun to provide the function which the items' priorities are sorted by. This defaults to #'<, which will cause the queue to always return the item of lowest priority.
(ENQUEUE queue item priority)
Adds the item to the queue at the given priority. Returns a list containing first the item's priority, then the item itself. Running time: O(1)
(DEQUEUE queue)
Removes the item of lowest priority from the queue. Running time: amortised O(log(n))
(PEEP-AT-QUEUE queue)
Returns the item of lowest priority from the queue, without modifying the queue. Running time: O(1)
(EMPTY-QUEUE queue)
Removes all items from the queue. Returns the heap. Running time: O(1)
(QUEUE-SIZE queue)
Returns the number of items in the queue. Running time: O(1)
HEAP provides functionality common to the two heap classes implement by CL-HEAP. Each heap implementation accepts at least two arguments to MAKE-INSTANCE: :key and :sort-fun. :sort-fun supplies the function to be used when comparing two objects to each other, and defaults to #'<. :key gives a function which should be first applied to the elements in the HEAP before comparing them using the sorting function. :key defaults to #'identity. These functions can be accessed using HEAP-KEY and HEAP-SORTING-FUNCTION.
Both of the heap implementations obey the same interface:
(HEAP-KEY heap)
Returns the function passed as the :key argument to the heap.
(HEAP-SORTING-FUNCTION heap)
Returns the function used to sort the elements in the heap.
(HEAP-SIZE heap)
Returns the number of elements in the heap.
(EMPTY-HEAP heap)
Removes all items from the heap, and returns the heap.
(IS-EMPTY-HEAP-P heap)
Returns true iff the heap contains no elements.
(ADD-TO-HEAP heap item)
Adds the given item to the heap. Returns two values: first the item itself, and then some index-key which can be used by DECREASE-KEY and DELETE-KEY to identify which values should be affected by these operations. See below for an example.
(ADD-ALL-TO-HEAP heap items)
Adds all items in the list to the heap, as if by repeated calls to ADD-TO-HEAP. ADD-ALL-TO-HEAP can often be implemented more efficiently, though, than by merely executing consecutive calls to ADD-TO-HEAP directly. Returns the heap object.
(PEEP-AT-HEAP heap)
Returns the minimum item in the heap, without modifying the heap.
(POP-HEAP heap)
Removes the smallest item in the heap, and returns it.
(MERGE-HEAPS first second)
Returns a new heap which contains all the items in both heaps. Only heaps which use the same HEAP-KEY and HEAP-SORTING-FUNCTION can be merged, otherwise a HEAP-ERROR is signalled.
(NMERGE-HEAPS first second)
Destructively creates a new heap which contains the items in both heaps. Only heaps which use the same HEAP-KEY and HEAP-SORTING-FUNCTION can be merged, otherwise a HEAP-ERROR is signalled.
(DECREASE-KEY heap item-index value)
Allows you to specify an item in the heap whose value should be decreased. ITEM-INDEX is the second value returned by ADD-TO-HEAP, and VALUE should be smaller than the items current value, or a KEY-ERROR will be signalled. Returns the heap. See below for an example.
(DELETE-FROM-HEAP heap item-index)
Allows you to specify an arbitrary item to remove from the heap. ITEM-INDEX is the second value returned by ADD-TO-HEAP.
DECREASE-KEY requires that HEAP-KEY is either the function IDENTITY, or a function that accepts an optional second argument, which will be the value of the new key. For instance, if the elements in the heap are lists, and they are to be sorted by the first element in the list, an appropriate HEAP-KEY could be:
(lambda (item &optional new-value)
(if new-value
(setf (first item) new-value)
(first item)))
Of course, if DECREASE-KEY is not going to be used, then #'first itself could simply be used as the HEAP-KEY.
BINARY-HEAP is constructed in the typical fashion, using an array. The BINARY-HEAP MAKE-INSTANCE call accepts an extra argument, :size, which sets the data array's initial size.
Note that DECREASE-KEY and DELETE-FROM-HEAP are not very useful operations on this heap: while they work, the required index-keys may be invalidated as soon as any heap operation is performed. If you do want to make use of DELETE-FROM-HEAP or DECREASE-KEY, consider using FIBONACCI-HEAP instead.
BINARY-HEAP provides an extra function to the above API:
(BINARY-HEAP-EXTENSION-FACTOR heap)
The percentage at which the space allocated for data in the heap should grow at once that space has been exceeded. This defaults to 50.
Here are the Big-Oh running times for the heap API, where n and m are the sizes of various heaps.
(HEAP-SIZE heap) => O(1)
(EMPTY-HEAP heap) => O(1)
(IS-EMPTY-HEAP-P heap) => O(1)
(ADD-TO-HEAP heap item) => O(log(n))
(ADD-ALL-TO-HEAP heap items) => O(n)
(PEEP-AT-HEAP heap) => O(1)
(POP-HEAP heap) => O(log(n))
(MERGE-HEAPS first second) => O(n + m + log(n + m))
(NMERGE-HEAPS first second) => O(m + log(n + m))
(DECREASE-KEY heap item-index value) => O(log(n))
(DELETE-FROM-HEAP heap item-index) => O(log(n))
The details of the Fibonacci heap are given in "Fibonacci Heaps and Their Uses in Improved Network Optimization Algorithms", by Fredman and Tarjan (see references below).
The Fibonacci heap has some interesting time constraints, and should generally be used instead of BINARY-HEAP.
Here are the Big-Oh running times for the heap API, where n and m are the sizes of various heaps.
(HEAP-SIZE heap) => O(1)
(EMPTY-HEAP heap) => O(1)
(IS-EMPTY-HEAP-P heap) => O(1)
(ADD-TO-HEAP heap item) => O(1)
(ADD-ALL-TO-HEAP heap items) => O(n)
(PEEP-AT-HEAP heap) => O(1)
(POP-HEAP heap) => amortised O(log(n))
(MERGE-HEAPS first second) => O(m + n)
(NMERGE-HEAPS first second) => O(1)
(DECREASE-KEY heap item-index value) => amortised O(1)
(DELETE-FROM-HEAP heap item-index) => amortised O(1), except where
the deleted item was the minimum
item, in which case it is
amortised O(log(n))
A heap can be created quite simply:
(defparameter *heap* (make-instance 'cl-heap:fibonacci-heap))
This will create a heap where the elements are ordered using #'<. Elements can be added one at a time using ADD-TO-HEAP:
(cl-heap:add-to-heap *heap* 1)
or, in a batch.
(cl-heap:add-all-to-heap *heap* '(6 4 7 3 2 0))
The minimum item in this heap can easily by seen using PEEP-AT-HEEP, which will not modify the heap in any way:
(cl-heap:peep-at-heap *heap*) => 0
The minimum item can be removed from the heap using POP-HEAP:
(cl-heap:pop-heap *heap*) => 0
Heaps can be used to sort items:
(let ((heap (make-instance 'cl-heap:fibonacci-heap)))
(cl-heap:add-all-to-heap heap (loop for i from 1 upto 10 collect (random 100)))
(loop while (not (cl-heap:is-empty-heap-p heap)) collect (cl-heap:pop-heap heap)))
=> (14 19 30 32 38 64 74 90 96 98)
A max-heap can be constructed as follows:
(defparameter *heap* (make-instance 'cl-heap:fibonacci-heap :sort-fun #'>))
And this will order things from most to least:
(let ((heap (make-instance 'cl-heap:fibonacci-heap :sort-fun #'>)))
(cl-heap:add-all-to-heap heap (loop for i from 1 upto 10 collect (random 100)))
(loop while (not (cl-heap:is-empty-heap-p heap)) collect (cl-heap:pop-heap heap)))
=> (69 68 64 60 37 34 30 7 6 1)
The heaps can contain arbitrary items. For instance, a heap whose elements are all lists can be constructed as follows:
(defparameter *heap* (make-instance 'cl-heap:fibonacci-heap :key #'first))
(cl-heap:add-to-heap *heap* (list 5 'some 'arbitrary 'data))
(cl-heap:add-to-heap *heap* (list 4 'other 'data))
(cl-heap:pop-heap *heap*) => (4 other data)
DECREASE-KEY and DELETE-FROM-HEAP can be used as follows:
(defparameter *heap* (make-instance 'cl-heap:fibonacci-heap))
(cl-heap:add-all-to-heap *heap* '(5 3 4 2 1))
(let ((item-index (second (multiple-value-list (cl-heap:add-to-heap *heap* 10)))))
(format t "Smallest item: ~A~%" (cl-heap:peep-at-heap *heap*))
(cl-heap:decrease-key *heap* item-index 0)
(format t "Smallest item: ~A~%" (cl-heap:peep-at-heap *heap*)))
=> nil
Smallest item: 1
Smallest item: 0
CL-HEAP is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.
M. L. Fredman and R. E. Tarjan. 1987. "Fibonacci Heaps and Their Uses in Improved Network Optimizxation Algorithms". Journal of the Association for Computing Machinery, Vol 34, No 3. pp 596 - 615
Next: Files, Previous: Introduction, Up: Top [Contents][Index]
The main system appears first, followed by any subsystem dependency.
• The cl-heap system |
Rudy Neeser <rudy.neeser@gmail.com>
GPLv3
An implementation of heap and priority queue data structures.
0.1.6
cl-heap.asd (file)
Files are sorted by type and then listed depth-first from the systems components trees.
• Lisp files |
Next: The cl-heap/package․lisp file, Previous: Lisp files, Up: Lisp files [Contents][Index]
cl-heap.asd
cl-heap (system)
Next: The cl-heap/condition․lisp file, Previous: The cl-heap․asd file, Up: Lisp files [Contents][Index]
Next: The cl-heap/heap․lisp file, Previous: The cl-heap/package․lisp file, Up: Lisp files [Contents][Index]
package.lisp (file)
cl-heap (system)
condition.lisp
heap-error-message (method)
Next: The cl-heap/binary-heap․lisp file, Previous: The cl-heap/condition․lisp file, Up: Lisp files [Contents][Index]
condition.lisp (file)
cl-heap (system)
heap.lisp
Next: The cl-heap/fibonacci-heap․lisp file, Previous: The cl-heap/heap․lisp file, Up: Lisp files [Contents][Index]
heap.lisp (file)
cl-heap (system)
binary-heap.lisp
Next: The cl-heap/priority-queue․lisp file, Previous: The cl-heap/binary-heap․lisp file, Up: Lisp files [Contents][Index]
binary-heap.lisp (file)
cl-heap (system)
fibonacci-heap.lisp
Previous: The cl-heap/fibonacci-heap․lisp file, Up: Lisp files [Contents][Index]
fibonacci-heap.lisp (file)
cl-heap (system)
priority-queue.lisp
Next: Definitions, Previous: Files, Up: Top [Contents][Index]
Packages are listed by definition order.
• The cl-heap-asdf package | ||
• The cl-heap package |
Next: The cl-heap package, Previous: Packages, Up: Packages [Contents][Index]
cl-heap.asd
Previous: The cl-heap-asdf package, Up: Packages [Contents][Index]
An implementation of heap and priority queue data structures.
package.lisp (file)
common-lisp
Definitions are sorted by export status, category, package, and then by lexicographic order.
• Exported definitions | ||
• Internal definitions |
Next: Internal definitions, Previous: Definitions, Up: Definitions [Contents][Index]
• Exported generic functions | ||
• Exported conditions | ||
• Exported classes |
Next: Exported conditions, Previous: Exported definitions, Up: Exported definitions [Contents][Index]
Adds a list of items to a HEAP. This can typically
be done more efficiently than by using repeated calls to
ADD-TO-HEAP, since the heap order only needs to be reinstated after
all of the items have been inserted, rather than maintained through
each operation. Returns the heap object.
heap.lisp (file)
Adds the following list of items into the heap. This is an O(n) operation.
fibonacci-heap.lisp (file)
Adds all the items in the list into the BINARY-HEAP. This is a O(n + log(n)) = O(n) operation. Returns the BINARY-HEAP.
binary-heap.lisp (file)
Inserts an item into the given HEAP data
structure. Returns two values: first, the item added, and then an
index-key which can be used to identify the item for DECREASE-KEY
and DELETE-FROM-HEAP operations.
heap.lisp (file)
Adds an item to a Fibonacci-heap. This is a constant time operation. Returns the item added to the heap.
fibonacci-heap.lisp (file)
Inserts an item into a BINARY-HEAP. This is O(log(n)), n being the number of items in the heap.
binary-heap.lisp (file)
The percentage at which the space
allocated for data in the heap should grow at
once that space has been exceeded.
binary-heap.lisp (file)
Decreases the value of the item represented by
ITEM-INDEX. ITEM-INDEX is the index returned by ADD-TO-HEAP for a
particular item. VALUE is the item’s new value, and this must be
"less" than its old value. Returns the heap.
heap.lisp (file)
Changes the value of an item represented by the ITEM-INDEX to VALUE. This index is returned as the second argument to ADD-TO-HEAP. This is an amortised constant time operation.
fibonacci-heap.lisp (file)
Deceases the key of the item pointed to by ITEM-INDEX. The index is returned as the second value of ADD-TO-HEAP. The value of the item at the index is changed to VALUE, which should be less than its old value. This operation is O(log(n)), with n being the number of items in the heap. Note that using a binary heap is not ideal for this operation, since the item pointed to by any given index can be changed by performing any heap operation. This function is provided mostly for completeness.
binary-heap.lisp (file)
Removes the item from the heap represented by the ITEM-INDEX. This index is returned as the second value of ADD-TO-HEAP. Returns the heap.
heap.lisp (file)
Removes an item from the heap, as pointed to by item-index. This
operation is amortised O(1), unless the item removed is the minimum item, in
which case the operation is equivalent to a POP-HEAP.
fibonacci-heap.lisp (file)
Deltes an item from the heap. ITEM-INDEX is an index representing the value to remove, and is the second value returned from ADD-TO-HEAP. Note that running most HEAP functions can modify which value is pointed to by ITEM-INDEX, so this function is given mostly for completeness. Use a Fibonacci heap if this functionality is important to you. This operation completes in O(log(n)) time, where n is the number of items in the heap.
binary-heap.lisp (file)
Removes an element from the front of the queue and returns it. This is an amortised O(log(n)) operation.
priority-queue.lisp (file)
Removes all contents from the given heap. Returns the heap.
heap.lisp (file)
Clears all items from the heap. This is a constant time operation.
fibonacci-heap.lisp (file)
Clears the heap. A constant time operation.
binary-heap.lisp (file)
Removes all items from the queue. This is a constant time operation.
priority-queue.lisp (file)
Adds an item with a given priority on to the
queue. Returns a list containing first the priority, then the
item. This is a constant time operation.
priority-queue.lisp (file)
Returns the number of elements in the heap.
heap.lisp (file)
fibonacci-heap.lisp (file)
binary-heap.lisp (file)
Returns true iff the given heap is empty.
heap.lisp (file)
fibonacci-heap.lisp (file)
binary-heap.lisp (file)
Returns a new heap that is the merged copy of those
given here. Can only merge heaps which use the same key and sorting
function. The two arguments are nt modified by this function.
heap.lisp (file)
Returns the merge of the two given heaps. This operation runs in O(n + m), where n and m are the number of items in each heap.
fibonacci-heap.lisp (file)
Non-destructively merges two BINARY-HEAPs. This is an O(n + m + log(n + m)) operation, n and m being the zies of the two heaps.
binary-heap.lisp (file)
Destructively updates the arguments to contain the
merge of both heaps, which is then returned. Can only merge heaps
which use the same key and sorting function.
heap.lisp (file)
Destructively marges the two heaps. This is a constant time operation.
fibonacci-heap.lisp (file)
Destructively merges two BINARY-HEAPs, and returns the
result. Where n and m are the sizes of each queue, this is an order
O(m + log(n + m) operation, unless an array resize is required, for
which it becomes O(n + m + log(n + m)).
binary-heap.lisp (file)
Returns the heap’s root, without modifying the heap. Returns nil if the heap is empty.
heap.lisp (file)
See the heap’s minimum value without modifying the heap. This is a constant time operation.
fibonacci-heap.lisp (file)
Returns the minimum object of the heap, without updating the heap. This is an O(1) operation.
binary-heap.lisp (file)
Returns the element at the front of the queue,
without modifying the queue. This is a constant time operation.
priority-queue.lisp (file)
Removes the top element from the heap and returns it.
heap.lisp (file)
Remove the minimum element in the tree. This has an amortised running time of O(log(n)), where n is the number of items in the heap.
fibonacci-heap.lisp (file)
Removes the minimum item from the heap. This is an O(log(n)) operation, where n is the number of items in the heap.
binary-heap.lisp (file)
Returns the number of elements in the queue.
priority-queue.lisp (file)
Next: Exported classes, Previous: Exported generic functions, Up: Exported definitions [Contents][Index]
The base condition type for all errors that should generally be thrown in the CL-HEAP package.
condition.lisp (file)
error (condition)
key-error (condition)
heap-error-message (method)
:message
(quote "an error has occured while using this data structure.")
heap-error-message (generic function)
condition.lisp (file)
heap-error (condition)
:message
(quote "when using decrease-key, the heap-key function should always take two arguments.")
Previous: Exported conditions, Up: Exported definitions [Contents][Index]
An implementation of a classic binary heap. This uses the standard array-based implementation.
binary-heap.lisp (file)
heap (class)
The heap’s stored data.
The percentage at which the space
allocated for data in the heap should grow at
once that space has been exceeded.
50
binary-heap-extension-factor (generic function)
(setf binary-heap-extension-factor) (generic function)
A heap made up of item-disjoint, heap-ordered
trees. Has some good time constraints on various heap operations.
fibonacci-heap.lisp (file)
heap (class)
The minimum element in the tree.
The number of items in the heap.
0
The base type of the two HEAP implementations.
heap.lisp (file)
standard-object (class)
The function used to apply an order to elements in the heap.
:sort-fun
(function <)
heap-sorting-function (generic function)
A function used to obtain the elements which
should be compared to give an order to the items in the
heap.
:key
(function identity)
heap-key (generic function)
A simple priority queue implementaiton, based on a Fibonacci heap.
priority-queue.lisp (file)
standard-object (class)
Previous: Exported definitions, Up: Definitions [Contents][Index]
• Internal macros | ||
• Internal functions | ||
• Internal generic functions | ||
• Internal classes |
Next: Internal functions, Previous: Internal definitions, Up: Internal definitions [Contents][Index]
fibonacci-heap.lisp (file)
Next: Internal generic functions, Previous: Internal macros, Up: Internal definitions [Contents][Index]
binary-heap.lisp (file)
binary-heap.lisp (file)
Next: Internal classes, Previous: Internal functions, Up: Internal definitions [Contents][Index]
Compares two items, using the HEAP’s SORT-FUN and KEY functions.
fibonacci-heap.lisp (file)
Cuts a child from its parent and makes and places it in the root list.
fibonacci-heap.lisp (file)
Deletes this node from the linked list that it
represents, and returns the new list. Nulls the node’s parent, and
resets its rank if appropriate.
fibonacci-heap.lisp (file)
condition.lisp (file)
fibonacci-heap.lisp (file)
Places node-two as a child of node-one if node-one’s item is smaller, or vice versa.
fibonacci-heap.lisp (file)
fibonacci-heap.lisp (file)
Joins together two fibonacci heaps.
fibonacci-heap.lisp (file)
Adds a node to the heap.
automatically generated reader method
fibonacci-heap.lisp (file)
automatically generated writer method
fibonacci-heap.lisp (file)
automatically generated reader method
fibonacci-heap.lisp (file)
automatically generated writer method
fibonacci-heap.lisp (file)
automatically generated reader method
fibonacci-heap.lisp (file)
automatically generated writer method
fibonacci-heap.lisp (file)
Used to implement cascading cuts.
fibonacci-heap.lisp (file)
automatically generated reader method
fibonacci-heap.lisp (file)
automatically generated writer method
fibonacci-heap.lisp (file)
automatically generated reader method
fibonacci-heap.lisp (file)
automatically generated writer method
fibonacci-heap.lisp (file)
The number of children the node has.
fibonacci-heap.lisp (file)
Used to move a value in the DATA array of a
BINARY-HEAP down the parent-child relationship hierarchy, and so
preserve heap-ordering.
binary-heap.lisp (file)
binary-heap.lisp (file)
fibonacci-heap.lisp (file)
Previous: Internal generic functions, Up: Internal definitions [Contents][Index]
A class used for storing data in a FIBONACCI-HEAP.
fibonacci-heap.lisp (file)
standard-object (class)
:item
node-item (generic function)
(setf node-item) (generic function)
node-parent (generic function)
(setf node-parent) (generic function)
node-child (generic function)
(setf node-child) (generic function)
The number of children the node has.
0
node-rank (generic function)
(setf node-rank) (generic function)
Used to implement cascading cuts.
node-marked-p (generic function)
(setf node-marked-p) (generic function)
node-next (generic function)
(setf node-next) (generic function)
node-last (generic function)
(setf node-last) (generic function)
Previous: Definitions, Up: Top [Contents][Index]
• Concept index | ||
• Function index | ||
• Variable index | ||
• Data type index |
Next: Function index, Previous: Indexes, Up: Indexes [Contents][Index]
Jump to: | C F L |
---|
Jump to: | C F L |
---|
Next: Variable index, Previous: Concept index, Up: Indexes [Contents][Index]
Jump to: | (
A B C D E F G H I L M N P Q U |
---|
Jump to: | (
A B C D E F G H I L M N P Q U |
---|
Next: Data type index, Previous: Function index, Up: Indexes [Contents][Index]
Jump to: | C D E H I K L M N P R S |
---|
Jump to: | C D E H I K L M N P R S |
---|
Previous: Variable index, Up: Indexes [Contents][Index]
Jump to: | B C F H K N P S |
---|
Jump to: | B C F H K N P S |
---|