1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
|
Source: triehash
Section: devel
Priority: optional
Maintainer: Julian Andres Klode <jak@debian.org>
Rules-Requires-Root: no
Build-Depends: debhelper (>= 11)
Standards-Version: 4.3.0
Homepage: https://github.com/julian-klode/triehash
Vcs-Git: https://github.com/julian-klode/triehash -b debian/master
Vcs-Browser: https://github.com/julian-klode/triehash
Package: triehash
Architecture: all
Depends: ${misc:Depends}, ${perl:Depends}
Multi-Arch: foreign
Description: Generates perfect hash functions as native machine code
TrieHash generates perfect hash functions as C code which then gets
compiled into optimal machine code as part of the usual program compilation.
.
TrieHash works by translating a list of strings to a trie, and then converting
the trie to a set of recursive switch statements; first switching by length,
and then switching by bytes.
.
TrieHash has various optimizations such as processing multiple bytes at once
(on GNU C), and shortcuts for reducing the complexity of case-insensitive
matching (ASCII only). Generated code performs substantially faster than
gperf, but is larger.
.
TrieHash was written for use in APT.
|
