#Generated by ResourceBundle Editor (http://eclipse-rbe.sourceforge.net)
#-------------------------------------------------------------------------------
# Copyright (c) 2012 Jakub Kovac, Katarina Kotrlova, Pavol Lukca, Viktor Tomkovic, Tatiana Tothova
#
# This program 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 .
#-------------------------------------------------------------------------------
capacity = Capacity
dynamicarray = Dynamic array
dynamicarray-empty = Dynamic array is empty.
dynamicarray-full = Array is full.
dynamicarray-insert = Insert new element.
dynamicarray-use-coin = Use this coins to allocate new array.
dynamicarray-new = New array.
dynamicarray-copy = Now copy elements from old array.
dynamicarray-insert-coin = Store coins for later.
dynamicarray-needless-first = Since element is in first half, we do not need them.
dynamicarray-needless-second = Since element is in second half, we do not need them.
dynamicarray-needless-empty = Since array is empty, we do not need them.
dynamicarray-insert-enough = Already stored, we do not need more for now.
dynamicarray-small = Since just a quarter of array is occupied, we free unnecessary memory.
dynamicarray-pop = Pop the last element.
dynamicarray-erase = Free the old memory.
dynamicarray-allocateText = Memory
dynamicarray-copyText = Copying
button-pop = Pop
234tree = 2-3-4 tree
23tree = 2-3 tree
EMPTYSTR = \
aaok = This node is OK.
aaskew = Skew: Left subtree has the same rank; we perform a right rotation.
aaskew2 = Skew2: Left subtree has the same rank; we perform a right rotation.
aaskew3 = Skew3: Left subtree has the same rank; we perform a right rotation.
aasplit = Split: The pseudonode (nodes of equal rank) is too big; we perform a left rotation and promote the middle node.
aasplit2 = Split2: The pseudonode (nodes of equal rank) is too big; we perform a left rotation and promote the middle node.
aatree = AA tree
activeheap = Active heap:
alreadythere = The key is already in the tree.
avedepth = Ave. depth
avldeletebal = Node has been deleted. We go back and update the balance information.
avlinsertbal = Node has been inserted. We go back and update the balance information.
avll = The right subtree is too high: We perform a left rotation.
avllr = The left subtree is too high, but its right subtree is higher that the left one: We perform a left-right rotation.
avlr = The left subtree is too high: We perform a right rotation.
avlrl = The right subtree is too high, but its left subtree is higher than the right one: We perform a right-left rotation.
avltree = AVL tree
avlupdatebal = New balance is #1.
badword = You have inserted an invalid word. Try to insert words containing only small and capital letters from the English alphabet. Lowercase letters will be transfered to uppercase ones.
bdelete1 = Case I: The key is in a leaf; we can delete it.
bdelete2 = Case II: The key is in an internal node; we replace it by its successor.
bfind = #1 < #2 < #3, we go along the #4. link.
bfind0 = #1 < #2, we go along the 1. link.
bfindn = #1 < #2, we go along the #3. link.
binheap = Binomial heap
binheap-add-tree = Add another tree.
binheap-bottom-empty = The bottom heap is empty, melding is trivial.
binheap-findmax = We remove the maximum and find the new one - this must be one of the roots in the list.
binheap-findmin = We remove the minimum and find the new one - this must be one of the roots in the list.
binheap-insert = Inserting a new node to a heap is the same as melding it with a 1-node heap.
binheap-link = Trees #1 and #2 have the same order so we link them (#1 becomes a child of #2).
binheap-meld-idea = We gradually add the trees from the bottom heap while linking two binomial trees of the same order to produce one tree of higher order.
binheap-meldchildren = Children of the removed node form a binomial heap which will be melded with the original one.
binheap-newmax = The new maximum is #1.
binheap-newmin = The new minimum is #1.
binheap-next = Proceed to the next node.
binheap-nochildren = The deleted node had no children.
binheap-oldmax = #1 remains the maximum.
binheap-oldmin = #1 remains the minimum.
binheap-top-empty = The top heap is empty, melding is trivial.
binsertleaf = We insert the key into this node.
bleft = Case I: The node is too small, but its left brother is big enough so we can take one key from it.
bmerge = Case III: The node is too small and so is its brother, so we merge them.
bplustree = B+ tree
bright = Case II: The node is too small, but its right brother is big enough so we can take one key from it.
bsplit = The node is too big, we have to split it.
bst = Binary search tree
bst-delete-case1 = Case I: The node is a leaf so we can simply remove it.
bst-delete-case2 = Case II: The node has only one son.
bst-delete-case3 = Case III: Node #1 has two children. In this case, we find #1's successor - the leftmost node in the right subtree - which will replace it. The successor has at most one child, so we can remove it easily.
bst-delete-go-left = Go left.
bst-delete-linkpar = Remove #1 by linking #2 to its grandparent #3.
bst-delete-newroot = Remove #1. Its son #2 becomes the new root.
bst-delete-replace = Replace #1 by #2 and remove #1.
bst-delete-succ = Node #2 is the successor of #1 (there is no element between #1 and #2 in the tree). Since #2 doesn't have the left child, we can remove it easily.
bst-delete-succ-link = We remove the successor by linking #1 under #2.
bst-delete-succ-start = Start at the root of the right subtree.
bst-delete-succ-unlink = We simply unlink the successor.
bst-delete-unlink = We unlink the node.
bst-insert-left = Since #1 < #2, we insert it in the left subtree.
bst-insert-right = Since #1 > #2, we insert it in the right subtree.
bst-insert-start = We start at the root.
bstdeletestart = First we have to find the node.
bstfindleft = Since #1 < #2, we search the left subtree.
bstfindright = Since #1 > #2, we search the right subtree.
bstfindstart = We start searching at the root.
btree = B tree
btreeorder = Order of the B Tree:
button-changekey = Change key
button-clear = Clear
button-create-st = Create suffix tree
button-decreasekey = Decrease key
button-delete = Delete
button-deletemax = Delete Maximum
button-deletemin = Delete Minimum
button-find = Find
button-findmax = Find max of interval
button-findmin = Find min of interval
button-findsum = Find sum of interval
button-increasekey = Increase key
button-insert = Insert
button-makeset = Add elements
button-meld = Meld
button-pause = Pause
button-random = Random
button-random-unions = Random unions
button-rotate = Rotate
button-save = Save
button-uffind = Find
button-union = Union
changekey = Change of value
changekeyv = We change the value of key.
control = Control
daryheap = d-ary heap
daryheaporder = Order of the d-ary heap:
datastructures = Data structures
decreasekey = Decrease key
decrkeymin = We decrease the key.
delete = Delete #1
delete-max = Delete maximum
delete-min = Delete minimum
deleted = Deleted
dictionary = Dictionaries
display = Display
done = Done.
empty = The tree is empty.
emptyheap = heap is empty
excess = Excess nodes
fibheap = Fibonacci heap
find = Find #1
findmax = Find the maximum of the interval 〈#1,#2〉
findmin = Find the minimum of the interval 〈#1,#2〉
findsum = Find sum of the interval 〈#1,#2〉
found = Found.
full = full
fullheap = heap is full
gbdeletedeleted = The node has been already deleted.
gbdeletemark = We mark the node for deletion (we will delete when we rebuild the subtree).
gbdeleterebuild = Half of the nodes has been marked for deletion. We rebuild the whole tree.
gbfinddeleted = The node has been deleted. Not found.
gbinsertunmark = The key is already in the tree but marked as deleted. We just unmark it.
gbrebuild1 = Phase I: We transform the subtree into a linear list and delete the nodes marked for deletion.
gbrebuild2 = Phase II: We transform the list into a perfectly balanced tree.
gbtoohigh = This subtree is too high. We rebuild the whole subtree.
heap = Heap
heap-insert-last = Insert as the last node.
heap-last = This is the last node, the resulting heap is empty.
heap-replace-root = Replace root with the last node.
heapchange = We exchange the root and the last vertex and then remove it.
heapempty = The heap is empty.
heapfull = The heap is full.
height = Height
implicit = Show implicit nodes
increasekey = Increase key
incrkeymax = We increase the key.
insert = Insert #1
intervalchangev = After changing the value of an element we have to adjust the values of all nodes on the path from this leaf node to the root node accordingly.
intervalempty = Interval 〈#1,#2〉 represented by this node is empty.
intervalextend = Since the number of elements has reached the number of leafs, we have to extend the tree.
intervalfind = There are three possible mutual positions of the interval represented by a node currently considered by the DFS (a) and the interval we search for (b). \r\n
1. They don't intersect - we omit the subtree rooted in this node. \r\n
2. They intersect but (a) is not nested in (b) - we continue searching in the subtree rooted in this node.\r\n
3. Interval (a) is nested in interval (b) - we found node which represents part of the interval we search for, therefore we don't continue searching in the subtree.
intervalin = Interval 〈#4,#5〉 represented by node #3 is nested in interval 〈#1,#2〉.
intervalinsert = After adding a new element we have to adjust the values of all nodes on the path from this leaf node to the root node accordingly.
intervalkeyempty = Since the right node is empty, we choose key of the left node #1.
intervalmax = We choose larger key of #2 and #1.
intervalmin = We choose smaller key of #1 and #2.
intervalout = Interval 〈#4,#5〉 represented by node #3 and interval 〈#1,#2〉 don't intersect.
intervalpart = Interval 〈#4,#5〉 represented by node #3 is not nested in interval 〈#1,#2〉, but they intersect.
intervalsum = We take the sum of left and right sons keys #2 + #1.
intervaltree = Interval tree
intervaltrees = Interval trees
keys = Keys
language = Language
layout = Layout
layout-compact = Compact
layout-simple = Simple
lazybinheap = Lazy binomial heap
leftheap = Leftist heap
leftinsertright = Since #1 > #2, we insert it in the right subtree.
leftinsertup = Since #1 ≤ #2, #2 becomes the right son of #1.
leftmeldnoson = Since #2 has no right son, we append the rest of the heap as the right son of #2.
leftmeldrightg = Since #1 > #2, nothing happens and we continue along the right path.
leftmeldrightl = Since #1 < #2, nothing happens and we continue along the right path.
leftmeldstart = We start by comparing the roots as the first elements in the right path.
leftmeldswapg = Since #1 ≥ #2, we swap the heaps and continue along the right path.
leftmeldswapl = Since #1 ≤ #2, we swap the heaps and continue along the right path.
leftrankstart = We make the heap leftist by swapping left and right children according to their ranks. If the rank of the right son is greater than the rank of the left son, we exchange them, otherwise nothing happens.
leftrankupdate = We update the node ranks.
lipsum = English lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut commodo, neque id accumsan imperdiet, metus felis pharetra est, nec molestie diam elit a lectus. Maecenas erat purus, tempor nec aliquam id, tincidunt vel dui. Cras fermentum suscipit porta. Aenean vel magna vestibulum augue hendrerit tincidunt et nec magna. Aliquam at metus et quam vehicula iaculis. Curabitur sapien est, ultrices id auctor vel, ullamcorper at ipsum. Mauris sodales adipiscing dignissim.
max = max
maxdheapbubbledown = We bubble the node down (swap it with the greatest child) until it is greater than all of its children.
maxheap = Max Heap
maxheapbubbledown = We bubble the node down (swap it with its greater child) until it is greater than both of its children.
maxheapbubbleup = We bubble the node up (swap it with its parent) until its parent has greater priority.
maximum = Maximum is #1.
meld = Meld heaps ###1 and ###2
meldable-pq = Meldable priority queues
min = min
mindheapbubbledown = We bubble the node down (swap it with the smallest child) until it is smaller than all of its children.
minheap = Min Heap
minheapbubbledown = We bubble the node down (swap it with its smaller child) until it is smaller than both of its children.
minheapbubbleup = We bubble the node up (swap it with its parent) until its parent has smaller priority.
minimum = Minimum is #1.
mode23 = 2-3 tree correspondence
mode234 = 2-3-4 tree correspondence
newroot = The tree is empty so we make a new root.
next = Next
nodes = Nodes
notfound = Not found.
opt = opt
pairdecr = After decreasing the key, the heap property could be invalidated, so we cut the vertex and re-link it to the heap.
pairincr = After increasing the key, the heap property could be invalidated, so we cut the vertex and re-link it to the heap.
pairing = Pairing
pairingbf = back-front
pairingfb = front-back
pairingheap = Pairing heap
pairinglazymulti = lazy multipass
pairinglrrl = left-to-right then right-to-left
pairingmultipass = multipass
pairingnaive = naive
pairlinkmax = Since #1 < #2, we link #1 to #2.
pairlinkmin = Since #1 < #2, we link #2 to #1.
pairlrrl1 = If we choose "left-to-right then right-to-left" pairing, first we pair trees left to right. (We link the first and second, then the third and fourth and so on.)
pairlrrl2 = Then we link each remaining tree to the last one working from the right to left.
pairnaive = If "naive" combining rule is chosen, we choose one of the trees and successively link each of the remaing ones with it.
pq = Priority queues
previous = Previous
rbdelete1 = Case I: Node's sibling is red: We recolor some nodes and transform it to Case II, III, or IV.
rbdelete2 = Case II: Node's sibling and both his children are black: the extra black is moved up the tree.
rbdelete3 = Case III: Node's sibling is black, the closer child is red, the next one is black: We transform it to Case IV.
rbdelete4 = Case IV: Node's sibling and his child closer to us is black, the other one is red: By some recoloring and one rotation we can remove the extra black.
rbinsertcase1 = Case I, red uncle: We color the father and uncle black, the grandfather red and continue from grandfather.
rbinsertcase2 = Case II, black uncle, "inner" vertex: We transform this to Case III.
rbinsertcase3 = Case III, black uncle "outer" vertex: We perform a rotation and recolor the nodes.
redblack = Red-black tree
rotate = Rotate #1.
rotate-change = Nodes #1 and #2 exchange their roles: #1 becomes a new son of #2...
rotate-change-b = ...and #1 becomes son of #2.
rotate-change-nullb = ...and #1's son [null] becomes son of #2.
rotate-change-parent = ...#1 becomes parent of #2...
rotate-changes = Nodes #1 and #3 exchange their roles: #1 becomes a new son of #4 and #3 becomes a son of #1. Node #2 exchanges its parent #1 for #3.
rotate-fall = This subtree descends.
rotate-header = Rotate #1
rotate-newroot = ...#1 becomes the new root...
rotate-preserves-order = Note that rotation preserves the order of nodes.
rotate-rise = This subtree rises.
rotate-root = Node #1 is root – we can't rotate it.
rotations = Rotations
scapegoat = Scapegoat tree
search = Search
show-order = Show order
show-subtrees = Show subtrees
size = Size
skewheap = Skew heap
skewheapswap = We swap the left and right children of every node on the right path except the lowest one.
skipdelete = Now we delete the node from each level.
skipfindstart = We start searching at the top left corner.
skipinsertafter = We insert the new node right after this one.
skipinsertnext = Next key is less than our key. We insert it right.
skipinsertstart = We start at the top left corner and find appropriate place to insert the new node.
skiplist = Skiplist
skiplist-delete-found = We have found the node to delete.
skiplist-down = Next key is > #1 so we go down.
skiplist-head = Toss ##1: HEADS. We promote the node.
skiplist-next = The next key is ≤ #1, so we go right.
skiplist-tail = Toss ##1: TAILS. We stop.
skiplist-tossing = Now we will toss a coin until it comes TAILS. For each HEAD, we promote the node up, so on average there will be 1/2^k nodes on the k-th level.
splay = Splay
splay-found = Found. We splay this node.
splay-higher = Key #1 is not in the tree. We splay #2, the smallest higher key in the tree, instead.
splay-insert-left = Note that the root and its left subtree is less than #1 while the whole root's right subtree is greater than #1.
splay-insert-left2 = We simply make #1 the new root, link the current root left and it's right subtree right.
splay-insert-right = Note that the root and its right subtree is greater than #1 while the whole root's left subtree is less than #1.
splay-insert-right2 = We simply make #1 the new root, link the current root right and it's left subtree left.
splay-lower = Key #1 is not in the tree. We splay #2, the biggest lower key in the tree, instead.
splay-root = There is no grandparent; we just rotate.
splay-start = We start by splaying #1. This will bring #1 to the root. If #1 is not in the tree the smallest higher or the biggest lower key is splayed.
splay-zig-zag-left = Zig-zag case (#1 is a left son, but #2 is a right son): we rotate #1 twice.
splay-zig-zag-right = Zig-zag case (#1 is a right son, but #2 is a left son): we rotate #1 twice.
splay-zig-zig-left = Zig-zig case (both #1 and #2 are left sons of their fathers): we first rotate #2 and then #1.
splay-zig-zig-right = Zig-zig case (both #1 and #2 are right sons of their fathers): we first rotate #2 and then #1.
splaydelete = We delete the root, splay the right subtree for minimum.
splaydeleteleft = The root has no right subtree; we just delete it and make the left child the new root.
splaydeletelink = The minimum of the right tree will be the new root; since minimum has no left son, we can just link the left tree.
splaydeleteright = The root has no left subtree; we just delete it and make the right child the new root.
splayinroot = Now our key, smallest bigger or biggest smaller is in the root.
splaytree = Splay tree
stringology = Stringology
suffixtree = Suffix Tree
suffixtree-found = Each leaf in this subtree corresponds to a suffix beginning with "#1" and stores its position. We have found #2 occurence(s).
sum = sum
sumimum = Sum of the interval is #1.
sxbaftersecondrule = We continue appending.
sxbcontinue = We start extending vertices on working path by the letter '#1'.
sxbdownwalk = We find the place to append the letter '#1'.
sxbdownwalkedge = We continue searching.
sxbexplicit = We transform the tree to explicit form.
sxbfind = We find the string '#1'.
sxbfirstrule = We extend all vertices from first rule with the letter '#1'.
sxbphase = We step into #1. phase.
sxbsecondrule = We divide the edge and append a letter '#1'.
sxbslink = We use suffix link.
sxbstart = We start inserting the word.
sxbthirdrule = Third rule is applied here. We are finished with this phase.
sxbupwalk = We go to the nearest node upward.
text = Text
treap = Treap
treapbubbledown = We bubble the node down.
treapbubbleup = We bubble the node up until its parent has greater priority.
treapdeletecase1 = Now the node is a leaf and we can simply delete it.
trie = Trie
triea = Stringology
triedelete = Delete "#1"
triedeletedbdb = We delete a letter.
triedeletedbend = The dead branch is deleted.
triedeletefindunsu = We simply cannot delete a word which is not in the tree.
triedeletenote1 = First, we have to find the given word.
triedeletenote2 = Now, we delete an "end of a word" mark and then we delete a "dead branch", that is all nodes hanging, not determined by end of a word.
triedeletewodb = Now, it's about time to delete the dead branch.
triefind = Find "#1"
triefindending1 = A letter '#1' is not appended to this node. So, the tree does not contain the given word!
triefindending2 = We read the whole given word, but in the tree no word ends here. So, the tree does not contain the given word!
triefindmovedown = We move down to a letter '#1'.
triefindnote = We traverse the tree through edges with letters from a given word. When whole word is read and we are in a node marked by end-of-word symbol, the tree contains the given word.
triefindsucc = We read the whole given word and there is also a "end of a word" mark here. So, the tree contains the given word!
triei = Trie
trieinsert = Insert "#1"
trieinserteow = We have inserted the whole word. Now, it's time to mark an end of the word.
trieinsertneow = The word is in the tree.
trieinsertnote = We divide a given word into single letters which then append in sequental order. If a letter is already appended we don't append it but simple move down to it.
trieinsertwch = We move down to a letter '#1'.
trieinsertwoch = Append a letter '#1'.
trierootstart = We start in the root.
uf-byrank = by rank
uf-compresion = path compression
uf-find-heuristic = Find:
uf-find-start = Start searching for the representative.
uf-found-root = This is the representative of the set containing #1.
uf-go-old-parent = Move to the former parent.
uf-go-up = Move toward root.
uf-halving = path halving
uf-link = Link #1 directly under #2.
uf-none = na\u00EFve
uf-path-halved = Note that the path from #1 to the root halved.
uf-path-split = Note how the path from #1 to the root split.
uf-same-set = Elements are in the same set, no linking needed.
uf-splitting = path splitting
uf-union-heuristic = Union:
ufa = Union Find
ufcompression = Let's compress the marked path.
ufdown = We are going down.
ufdownstart = We go back and link all elements to the representative.
uffind = Find a representative
ufi = Union Find
ufunion = Union
ufunionfirstsecond = This node has higher rank, so we link #2 under #1.
ufunionsamerank = Both representatives have the same rank. So we choose to link #2 under #1 and increment #1's rank.
ufunionsecondfirst = This node has higher rank, so we link #1 under #2.
ufunionsimple = Link second representative to first one.\u0009
ufupspecial = We link grandchild.
zoomio = Zoom in/out