77c5659dfe
The program mph tries to generate an order preserving minimal perfect hashing (MPH) function for the set of keys, one per line, on stdin. Each key can be at most 4095 characters long (see keys.h to increase this limit), and the keys must be unique. If mph terminates, it emits a language independent binary or text representation of the MPH function on stdout. To generate a usable hash function, this output should be fed to a language dependent filter, like emitc. e.g. % mph <foo | emitc >hash.c The algorithm used by mph is probabilistic - it iterates until it finds a MPH function. For each failed iteration, it prints a (cryptic) reason on stderr. There is no no guarantee that mph will terminate. In practice this is unlikely, unless the constants specified with options -c or -m are too small (see below).
16 lines
823 B
Text
16 lines
823 B
Text
The program mph tries to generate an order preserving minimal perfect
|
|
hashing (MPH) function for the set of keys, one per line, on stdin.
|
|
Each key can be at most 4095 characters long (see keys.h to increase
|
|
this limit), and the keys must be unique. If mph terminates, it emits
|
|
a language independent binary or text representation of the MPH
|
|
function on stdout. To generate a usable hash function, this output
|
|
should be fed to a language dependent filter, like emitc.
|
|
|
|
e.g.
|
|
% mph <foo | emitc >hash.c
|
|
|
|
The algorithm used by mph is probabilistic - it iterates until it
|
|
finds a MPH function. For each failed iteration, it prints a
|
|
(cryptic) reason on stderr. There is no no guarantee that mph will
|
|
terminate. In practice this is unlikely, unless the constants
|
|
specified with options -c or -m are too small (see below).
|