Roman Klyujkov's Practical Algorithm Template library
C++ CMake Other
Latest commit b158267 Jul 16, 2017 @jnorthrup jnorthrup put a placeholder in for clang, and rebase (big breaking tree change!…
…) to uxn's own repo with a bunch of branches as yet unmerged.


Major features
 * STL-compatible, high-performance template containers and auxiliary classes useful for stringology with fine-grained low level design

Library features
 * set & map containers with trie (digital/radix search tree) functionality - trie_set & trie_map;
 * set & map containers with linear time & space online suffix tree functionality - suffix_set & suffix_map - 16 bytes per symbol, alphabet-independent;
 * bunch of iterators over each type of containers - standard (const_)(reverse_)iterator, special (const_)(post|pre)order_iterator for post- and preorder traversal, partial match iterators (const_)partimator and suffix_cont-special match_iterator & (super_)maxrep_iterator for linear-time matching statistics and (super)maximal repeats;
 * partimator's can be specialized with one of the decision functors: partial_match (filter strings with joker pattern), hamming_distance (filter strings with Hamming automata), levenshtein_(tp_|ms_)distance (filter strings with Levenshtein automata with optional operations like transposition or merge-split);
 * lowest common ancestor (LCA) constant time requests after linear time & memory postprocessing of the suffix_cont - lca_oracle;
 * patricia_dot_creator - utility class to create .dot graphviz input files for internal data structure visualization.
All features of this library can be examined with small and clever demo programs delivered within.

 =Compilers: Visual C++ 2003/2005/2008, GCC G++ 3.x= 

In russian или с чего все начиналось: в сентябре 2005 ( и продолжение в году 2008 (

В текущем состоянии в библиотеке реализовано множество алгоритмов, востребованных в обработке текстов и двоичных данных, в методах сжатия информации и выделении повторяющихся структур.