The Mmap class implement memory-mapped file objects

  require 'mmap'
  
  mmap = Mmap.new(__FILE__)
  mmap.advise(Mmap::MADV_SEQUENTIAL)
  
  mmap.each do |line|
    puts line
  end

gem install mmap

rake docs

WARNING: The variables $' and $` are not available with gsub! and sub!

Object

• Comparable
• Enumerable

• lockall(flag):
  disable paging of all pages mapped. |flag| can be Mmap::MCL_CURRENT or Mmap::MCL_FUTURE

• new(file, mode = "r", protection = Mmap::MAP_SHARED, options = {})
new(nil, length, protection = Mmap::MAP_SHARED, options = {}):
  create a new Mmap object

  • file: Pathname of the file, if nil is given an anonymous map is created Mmanp::MAP_ANON

  • mode: Mode to open the file, it can be r, w, rw, a

  • protection: specify the nature of the mapping

    • Mmap::MAP_SHARED: Creates a mapping that's shared with all other processes mapping the same areas of the file. The default value is Mmap::MAP_SHARED

    • Mmap::MAP_PRIVATE: Creates a private copy-on-write mapping, so changes to the contents of the mmap object will be private to this process

  • options: Hash. If one of the options length or offset is specified it will not possible to modify the size of the mapped file.

    • length: Maps length bytes from the file
    • offset: The mapping begin at offset
    • advice: The type of the access (see #madvise)

• unlockall: reenable paging

• extend(count): add count bytes to the file (i.e. pre-extend the file)

• madvise(advice): advice can have the value Mmap::MADV_NORMAL, Mmap::MADV_RANDOM, Mmap::MADV_SEQUENTIAL, Mmap::MADV_WILLNEED, Mmap::MADV_DONTNEED

• mprotect(mode): change the mode, value must be r, w or rw

• mlock: disable paging

• msync
flush: flush the file

• munlock: reenable paging

• munmap: terminate the association

• self == other: comparison

• self > other: comparison

• self >= other: comparison

• self < other: comparison

• self <= other: comparison

• self === other: used for case comparison

• self << other: append other to self

• self =~ other: return an index of the match

• self[nth]: retrieve the nth character

• self[start..last]: return a substring from start to last

• self[start, length]: return a substring of lenght characters from start

• self[nth] = val: change the nth character with val

• self[start..last] = val: change substring from start to last with val

• self[start, len] = val: replace length characters from start with val.

• self <=> other: comparison : return -1, 0, 1

• casecmp(other) >= 1.7.1:

• concat(other): append the contents of other

• capitalize!: change the first character to uppercase letter

• chop!: chop off the last character

• chomp!([rs]): chop off the line ending character, specified by rs

• count(o1 [, o2, ...]): each parameter defines a set of character to count

• crypt(salt): crypt with salt

• delete!(str): delete every characters included in str

• downcase!: change all uppercase character to lowercase character

• each_byte {|char|...}: iterate on each byte

• each([rs]) {|line|...}: - each_line([rs]) {|line|...}: iterate on each line

• empty?: return true if the file is empty

• freeze: freeze the current file

• frozen: return true if the file is frozen

• gsub!(pattern, replace): global substitution

• gsub!(pattern) {|str|...}: global substitution

• include?(other): return true if other is found

• index(substr[, pos]): return the index of substr

• insert(index, str) >= 1.7.1: insert str at index

• length: return the size of the file

• reverse!: reverse the content of the file

• rindex(substr[, pos]): return the index of the last occurrence of substr

• scan(pattern): return an array of all occurence matched by pattern

• scan(pattern) {|str| ...}: iterate through the file, matching the pattern

• size: return the size of the file

• slice: same than []

• slice!: delete the specified portion of the file

• split([sep[, limit]]): splits into a list of strings and return this array

• squeeze!([str]): squeezes sequences of the same characters which is included in str

• strip!: removes leading and trailing whitespace

• sub!(pattern, replace): substitution

• sub!(pattern) {|str| ...}: substitution

• sum([bits]): return a checksum

• swapcase!: replaces all lowercase characters to uppercase characters, and vice-versa

• tr!(search, replace): translate the character from search to replace

• tr_s!(search, replace): translate the character from search to replace, then squeeze sequence of the same characters

• upcase!: replaces all lowercase characters to downcase characters

This extension module is copyrighted free software by Guy Decoux

You can redistribute it and/or modify it under the same term as Ruby.

Guy Decoux ts@moulon.inra.fr