Tiny Core Linux

General TC => Programming & Scripting - Unofficial => Topic started by: allthewu on September 19, 2009, 06:40:41 AM

Title: Little Ruby script for download tce
Post by: allthewu on September 19, 2009, 06:40:41 AM
Hi! May be it would be useless? I wrote this when I download tce from other comp. (first I say - my english is not so greate ^:^))

Code: [Select]
#!/usr/local/bin/ruby

# tce.rb - script for find and get .tce* Tiny Core Linux file

# end of url-path
def short_name(line)
  return line.to_s.split('/')[-1]
end

require 'open-uri'

$repository     = 'http://distro.ibiblio.org/pub/linux/distributions/tinycorelinux/'
$tce_rep_prefix = '2.x/tce/'
$tce_html       = 'tce_2x.html'

# just parse 'tce_2x.html' and puts all .tce* names to array (names $tce_base)
def build_base (repository, tce_html)
  tce_base = []
  html_list = open(repository + tce_html, 'User-Agent' => 'Ruby-Wget').read
  arr = html_list.scan(%r{<td>\s*(.*)</td>})
  for line in arr
    if line.to_s.match(%r{(.+\.[tce|tcel|tcem])}) and not line.to_s.match(%r{\s})
      tce_base.push(line)
    end
  end
  return tce_base
end

$tce_base = build_base($repository, $tce_html)

# find matches of junked-name and .tce* files
def find_tce (junk_name)
  result = []
  for line in $tce_base
    if line.to_s.scan(/#{junk_name}/i).length > 0
      result.push($repository + $tce_rep_prefix + line.to_s)
    end
  end
  return result
end

# look .dep files
def what_dep(junk_name)
  for line in find_tce(junk_name)
    begin
      puts "\n--- #{short_name(line)} depends is : ---\n\n"
      puts open(line.to_s + '.dep').read
    rescue
      puts "\n: No depends for #{short_name(line)} :)\n"
    end
  end
end

# look .info files
def info_tce(junk_name)
  for line in find_tce(junk_name)
    puts "\n--- #{short_name(line)} info ---\n\n"
    puts open(line.to_s + '.info').read
  end
end

# look .md5 files
def md5_tce(junk_name)
  for line in find_tce(junk_name)
    puts "\n--- #{short_name(line)} md5 sum : ---\n\n"
    puts open(line.to_s + '.md5.txt').read
  end
end

# download tce* files
def get_tce (junk_name)
  for f in find_tce(junk_name)
    open(f.to_s, 'rb') do |tce|
      File.open(f.to_s.split('/')[-1], "wb") do |file|
        file.write(tce.read)
      end
    end
  end
end

# look server information about file
def get_meta (junk_name)
  for f in find_tce(junk_name)
    puts "\n--- Meta for #{short_name(f)} ---\n"
    for tok in open(f.to_s).meta
      puts tok
    end
  end
end

# simple usage :

case ARGV[0]
  when '-find', '-f'
    puts find_tce(ARGV[1])
  when '-dep', '-d'
    what_dep(ARGV[1])
  when '-info', '-i'
    info_tce(ARGV[1])
  when '-md5'
    md5_tce(ARGV[1])
  when '-meta'
    get_meta(ARGV[1])
  when '-get', '-g'
    puts get_tce(ARGV[1])
  when '-help', '-h'
    puts """
This is small tools which find, wget and view info about tce files - Tiny-Core-Linux extansion files.

Usage `./tcl.rb [options] [required-tce]' or `ruby tcl.rb [options] [required-tce]' if you'r ruby in a special place.

[required-tce] - is junked-string contain the name of extension, which you want.
[options] : -f, -find for find.
-d, -dep  for see depends.
-i, -info for details information.
-md5      for look normall md5 summ
-meta     for look server information about file - size, date, etc.
-g. -get  ok, it's for wget-like.
-h, -help this message.

Example: gcc

first we find all gcc matching:

./tcl.rb -f gcc

then we can chose one and look depends and info:

./tcl.rb -i gcc.tcel
./tcl.rb -d gcc.tcel

and then reqursive process is starting - we must try it again for make, binutils, and other, while all depends not be installed. Then we can download all of it:

./tcl.rb -g gcc.tcel
./tcl.rb -g make
./tcl.rb -g m4.
...

Note: I think a .dep file must be more rightly! It would be can automating this process.
"""

end

so, it parsed main tce-html, it can w-get .dep, .info, .md5, and .tce*. What the main moral (and my [don't know word )) ] to developers) - build .dep file rightly - with truly enum dependes (oh...)

P.S. allso I prepare sbcl.tce for SBCL - binaries simply starting, but non-triviall compillation process also was success. I think SBCL work faster then other sistem.

(once more sory for en.)
Title: Re: Little Ruby script for download tce
Post by: jls on September 19, 2009, 08:00:57 AM
Code: [Select]
`scan': invalid byte sequence in UTF-8 (ArgumentError)
        from /home/tc/.local/bin/tce.rb:20:in `build_base'
        from /home/tc/.local/bin/tce.rb:29:in `<main>'
Title: Re: Little Ruby script for download tce
Post by: allthewu on September 20, 2009, 11:15:20 AM
Strange, for me working. May be replace scan(%r{<td>\s*(.*)<\/td>}), to scan(/<td>\s*(.*)<\/td>/) ? But I think problem in external system options, not in this script. Try this:

Code: [Select]
require 'open-uri'
puts open('http://tinycorelinux.com/').read

for look page in your consol. If view is bad - sorry :) Ruby stay not so good.