Skip site navigation (1)Skip section navigation (2)

FreeBSD Manual Pages


home | help
GPERF(1)			      FSF			      GPERF(1)

       gperf - generate	a perfect hash function	from a key set

       gperf [OPTION]... [INPUT-FILE]

       GNU 'gperf' generates perfect hash functions.

       If  a  long option shows	an argument as mandatory, then it is mandatory
       for the equivalent short	option also.

   Output file location:
       --output-file=FILE Write	output to specified file.

       The results are written to standard output if no	output file is	speci-
       fied or if it is	-.

   Input file interpretation:
       -e, --delimiters=DELIMITER-LIST
	      Allow  user  to  provide	a string containing delimiters used to
	      separate keywords	from their attributes.	Default	is ",".

       -t, --struct-type
	      Allows the user to include a  structured	type  declaration  for
	      generated	 code.	Any  text  before %% is	considered part	of the
	      type declaration.	Key words and  additional  fields  may	follow
	      this, one	group of fields	per line.

	      Consider	upper  and  lower case ASCII characters	as equivalent.
	      Note that	locale dependent case mappings are ignored.

   Language for	the output code:
       -L, --language=LANGUAGE-NAME
	      Generates	code in	the specified language.	Languages handled  are
	      currently	C++, ANSI-C, C,	and KR-C. The default is ANSI-C.

   Details in the output code:
       -K, --slot-name=NAME
	      Select name of the keyword component in the keyword structure.

       -F, --initializer-suffix=INITIALIZERS
	      Initializers for additional components in	the keyword structure.

       -H, --hash-function-name=NAME
	      Specify name of generated	hash function. Default is 'hash'.

       -N, --lookup-function-name=NAME
	      Specify  name  of	 generated  lookup  function.  Default name is

       -Z, --class-name=NAME
	      Specify name of generated	 C++  class.  Default  name  is	 'Per-

       -7, --seven-bit
	      Assume 7-bit characters.

       -l, --compare-lengths
	      Compare  key  lengths before trying a string comparison. This is
	      necessary	if the keywords	contain	NUL bytes. It also  helps  cut
	      down on the number of string comparisons made during the lookup.

       -c, --compare-strncmp
	      Generate comparison code using strncmp rather than strcmp.

       -C, --readonly-tables
	      Make  the	 contents  of  generated lookup	tables constant, i.e.,

       -E, --enum
	      Define constant values using an enum local to the	 lookup	 func-
	      tion rather than with defines.

       -I, --includes
	      Include  the necessary system include file <string.h> at the be-
	      ginning of the code.

       -G, --global-table
	      Generate the static table	of keywords as a static	 global	 vari-
	      able, rather than	hiding it inside of the	lookup function	(which
	      is the default behavior).

       -P, --pic
	      Optimize the generated table for inclusion in shared  libraries.
	      This reduces the startup time of programs	using a	shared library
	      containing the generated code.

       -Q, --string-pool-name=NAME
	      Specify name of string pool generated by option --pic.   Default
	      name is 'stringpool'.

	      Use  NULL	strings	instead	of empty strings for empty keyword ta-
	      ble entries.

	      Specify prefix for the constants like TOTAL_KEYWORDS.

       -W, --word-array-name=NAME
	      Specify name of word list	array. Default name is 'wordlist'.

	      Specify name of length table array.  Default  name  is  'length-

       -S, --switch=COUNT
	      Causes  the  generated  C	code to	use a switch statement scheme,
	      rather than an array lookup table.  This can lead	to a reduction
	      in both time and space requirements for some keyfiles. The COUNT
	      argument determines how many switch statements are generated.  A
	      value  of	 1  generates  1 switch	containing all the elements, a
	      value of 2 generates 2 tables with 1/2 the elements in each  ta-
	      ble,  etc.  If COUNT is very large, say 1000000, the generated C
	      code does	a binary search.

       -T, --omit-struct-type
	      Prevents the transfer of the  type  declaration  to  the	output
	      file. Use	this option if the type	is already defined elsewhere.

   Algorithm employed by gperf:
       -k, --key-positions=KEYS
	      Select  the key positions	used in	the hash function.  The	allow-
	      able choices range between 1-255,	inclusive.  The	positions  are
	      separated	 by  commas, ranges may	be used, and key positions may
	      occur in any order.  Also, the  meta-character  '*'  causes  the
	      generated	hash function to consider ALL key positions, and $ in-
	      dicates the "final character" of a key, e.g., $,1,2,4,6-10.

       -D, --duplicates
	      Handle keywords that hash	to duplicate values.  This  is	useful
	      for certain highly redundant keyword sets.

       -m, --multiple-iterations=ITERATIONS
	      Perform multiple choices of the -i and -j	values,	and choose the
	      best results. This increases the running time by a factor	of IT-
	      ERATIONS	but  does  a  good  job	minimizing the generated table

       -i, --initial-asso=N
	      Provide an initial value for the associate values	array. Default
	      is  0.  Setting  this value larger helps inflate the size	of the
	      final table.

       -j, --jump=JUMP-VALUE
	      Affects the "jump	value",	i.e., how far to advance  the  associ-
	      ated character value upon	collisions. Must be an odd number, de-
	      fault is 5.

       -n, --no-strlen
	      Do not include the length	of the keyword when computing the hash

       -r, --random
	      Utilizes randomness to initialize	the associated values table.

       -s, --size-multiple=N
	      Affects  the size	of the generated hash table. The numeric argu-
	      ment N indicates "how many times larger or smaller" the  associ-
	      ated  value  range  should  be, in relationship to the number of
	      keys, e.g. a value of 3  means  "allow  the  maximum  associated
	      value to be about	3 times	larger than the	number of input	keys".
	      Conversely, a value of 1/3 means "make  the  maximum  associated
	      value  about  3  times smaller than the number of	input keys". A
	      larger table should decrease the time required for an unsuccess-
	      ful  search,  at the expense of extra table space. Default value
	      is 1.

   Informative output:
       -h, --help
	      Print this message.

       -v, --version
	      Print the	gperf version number.

       -d, --debug
	      Enables the debugging option (produces  verbose  output  to  the
	      standard error).

       Written by Douglas C. Schmidt and Bruno Haible.

       Report bugs to <>.

       Copyright (C) 1989-2017 Free Software Foundation, Inc.  License GPLv3+:
       GNU GPL version 3 or later <>
       This is free software: you are free  to	change	and  redistribute  it.
       There is	NO WARRANTY, to	the extent permitted by	law.

       The full	documentation for gperf	is maintained as a Texinfo manual.  If
       the info	and gperf programs are properly	installed at  your  site,  the

	      info gperf

       should give you access to the complete manual.

GNU gperf 3.1			 January 2017			      GPERF(1)


Want to link to this manual page? Use this URL:

home | help