treetop-dcf 0.2.0

data/README.markdown

@@ -0,0 +1,34 @@
+# treetop-dcf
+
+Easy to use parser for
+[Debian Control
+Files](http://www.debian.org/doc/debian-policy/ch-controlfields.html).
+The format is very similar to YAML, but with some subtle differences (e.g.
+YAML requires a space after field names, DCF does not).
+
+Parsed paragraphs are returned as hashes in an array. Example:
+
+    irb(main):001:0> require "dcf"
+    => true
+    irb(main):002:0> Dcf.parse "Attr: Value"
+    => [{"Attr"=>"Value"}]
+
+The parser is currently quite slow, so if the files you're parsing allows it
+I'd recommend using this YAML based parser instead: http://gist.github.com/117293
+Its much faster but fails on certain valid files.
+
+## Installation
+
+sudo gem install treetop-dcf
+
+## Acknowledgements
+
+This project was created during Google Summer of Code 2009, as part of my project
+for the [R Foundation for Statistical
+Computing](http://www.r-project.org/foundation/main.html). (CRAN uses DCF for
+package information, and we needed a parser written in Ruby.) Therefore, a big
+thanks goes out to both the R Foundation and Google.
+
+## Author
+
+Bjørn Arild Mæland <bjorn.maeland at gmail.com>

data/lib/dcf.rb

@@ -0,0 +1,24 @@
+require "rubygems"
+require "treetop"
+require File.join(File.dirname(__FILE__) + "/dcf_grammar")
+
+module Dcf
+  # @param [String] input
+  # @return [Array, nil] An array of { attr => val } hashes or nil if failure
+  def self.parse(input)
+    parse = DcfParser.new.parse(input)
+    return if parse.nil?
+
+    parse.elements.collect do |i|
+      paragraph = {}
+      i.paragraph.elements.each do |row|
+        paragraph[row.field.attribute.text_value] = row.field.value.text_value
+      end
+      paragraph
+    end
+  end
+
+  def self.version
+    "0.2.0"
+  end
+end

data/lib/dcf_grammar.treetop

@@ -0,0 +1,43 @@
+grammar Dcf
+  rule paragraphs
+    (paragraph eol?)+
+  end
+
+  rule paragraph
+    (field eol?)+
+  end
+
+  rule field
+    attribute separator value
+  end
+
+  rule attribute
+    (!":" !eol .)+
+  end
+
+  rule value
+    eol? (!eol .)+ (eol white+ (!eol .)+)*
+    {
+      def text_value
+        # I don't really like this, but it works for the files I'm parsing, so..
+        super.strip.gsub(/\n */, ' ')
+      end
+    }
+  end
+
+  rule next_record
+    "\n" [a-zA-Z]
+  end
+
+  rule separator
+    ':' white*
+  end
+
+  rule eol
+    "\n"
+  end
+
+  rule white
+    [ \t]
+  end
+end

data/test/helper.rb

@@ -0,0 +1,6 @@
+require 'rubygems'
+
+require File.join(File.dirname(__FILE__), *%w[.. lib dcf])
+
+require 'test/unit'
+require 'shoulda'

data/test/suite.rb

@@ -0,0 +1,9 @@
+require 'test/unit'
+
+# for some reason these tests fail when run via TextMate
+# but succeed when run on the command line.
+
+tests = Dir["#{File.dirname(__FILE__)}/test_*.rb"]
+tests.each do |file|
+  require file
+end

data/test/test_dcf.rb

@@ -0,0 +1,128 @@
+# -*- coding: utf-8 -*-
+require File.dirname(__FILE__) + '/helper'
+
+class TestDcfParser < Test::Unit::TestCase
+
+  should "parse a description file" do
+    description = <<EOF
+Package: RSQLite
+Version: 0.7-1
+Title: SQLite interface for R
+Author: David A. James
+Maintainer: Seth Falcon <seth@userprimary.net>
+Description: Database Interface R driver for SQLite. This package
+        embeds the SQLite database engine in R and provides an
+        interface compliant with the DBI package. The source for the
+        SQLite engine (version 3.6.4) is included.
+LazyLoad: yes
+Depends: R (>= 2.6.0), methods, DBI (>= 0.2-3)
+Imports: methods, DBI (>= 0.2-3)
+License: LGPL (>= 2)
+Collate: zzz.R S4R.R dbObjectId.R SQLite.R SQLiteSupport.R
+Packaged: Sat Oct 25 18:54:05 2008; seth
+Repository: CRAN
+Date/Publication: 2008-10-26 18:29:06
+EOF
+    sqlite = Dcf.parse(description).first
+    assert_equal 14, sqlite.keys.length
+    assert_equal "RSQLite", sqlite["Package"]
+    assert_equal "2008-10-26 18:29:06", sqlite["Date/Publication"]
+    assert_equal "zzz.R S4R.R dbObjectId.R SQLite.R SQLiteSupport.R", sqlite["Collate"]
+  end
+
+  should "parse badly formatted description files" do
+    description = <<EOF
+Package: StatFingerprints
+Version: 1.3
+Depends:
+        rgl,car,MASS,ffmanova,akima,labdsv,RColorBrewer,vegan,tcltk2,tcltk,maptools,sciplot
+EOF
+    stat = Dcf.parse(description).first
+    assert_equal "rgl,car,MASS,ffmanova,akima,labdsv,RColorBrewer,vegan,tcltk2,tcltk,maptools,sciplot", stat["Depends"]
+  end
+
+  should "parse a simple entry" do
+    parse = Dcf.parse("Attr: Value\n").first
+    assert_equal ["Attr"], parse.keys
+    assert_equal "Value", parse["Attr"]
+  end
+
+  should "ignore trailing whitespace" do
+    parse = Dcf.parse("Attr: Value    \n").first
+    assert_equal "Value", parse["Attr"]
+  end
+
+  should "parse values that contains colons" do
+    sample = <<EOF
+Package: BCE
+Version: 1.3
+Title: Bayesian composition estimator: estimating sample (taxonomic)
+        composition from biomarker data
+Author: Karel Van den Meersche <k.vdmeersche@nioo.knaw.nl>, Karline
+        Soetaert <k.soetaert@nioo.knaw.nl>
+Maintainer: Karel Van den Meersche <k.vdmeersche@nioo.knaw.nl>
+Depends: R (>= 2.01), limSolve
+Description: Function to estimates taxonomic compositions from
+        biomarker data, using a Bayesian approach.
+License: GPL
+LazyData: yes
+Repository: CRAN
+Repository/R-Forge/Project: bce
+Repository/R-Forge/Revision: 43
+Date/Publication: 2009-06-03 20:45:44
+Packaged: 2009-06-03 02:59:04 UTC; rforge
+EOF
+    parse = Dcf.parse(sample).first
+    assert_equal "Bayesian composition estimator: estimating sample (taxonomic) " +
+                 "composition from biomarker data", parse["Title"]
+  end
+
+  should "parse multiple paragraphs" do
+    sample = <<EOF
+Package: AIGIS
+Version: 1.0
+Depends: R (>= 2.5.1), gpclib (>= 1.4)
+
+Package: AIS
+Version: 1.0
+Depends: R (>= 2.1.0)
+Suggests: R2HTML
+EOF
+    parse = Dcf.parse(sample)
+    assert_equal "AIGIS", parse[0]["Package"]
+    assert_equal "R (>= 2.5.1), gpclib (>= 1.4)", parse[0]["Depends"]
+    assert_equal "AIS", parse[1]["Package"]
+  end
+
+  should "parse multiline field values" do
+    sample = <<EOF
+Package: ADaCGH
+Version: 1.3-10
+Depends: R (>= 2.2.1), cgh, tilingArray, aCGH, cghMCR, papply, GDD,
+        waveslim, cluster, snapCGH, Hmisc
+Suggests: Rmpi, GLAD, DNAcopy
+EOF
+    parse = Dcf.parse(sample).first
+    assert_equal "R (>= 2.2.1), cgh, tilingArray, aCGH, cghMCR, papply, GDD, " +
+                 "waveslim, cluster, snapCGH, Hmisc",
+                 parse["Depends"]
+  end
+
+  should "parse file with UTF-8 data" do
+    sample = <<EOF
+Author: Vincent Goulet, Sébastien Auclair, Christophe Dutang, Xavier
+        Milhaud, Tommy Ouellet, Louis-Philippe Pouliot, Mathieu Pigeon
+Encoding: latin1
+Packaged: Tue May 26 10:40:31 2009; Grömping
+Repository: CRAN
+Date/Publication: 2009-05-26 09:23:28
+EOF
+    sample = (RUBY_VERSION =~ /1.9/) ? sample.force_encoding("utf-8") : sample
+    match = (RUBY_VERSION =~ /1.9/) ? "Grömping".force_encoding("utf-8") : "Grömping"
+    parse = Dcf.parse(sample).first
+    assert_not_nil parse
+    assert_match "Sébastien Auclair", parse["Author"]
+    assert_equal "Tue May 26 10:40:31 2009; #{match}", parse["Packaged"]
+  end
+
+end

metadata

@@ -0,0 +1,68 @@
+--- !ruby/object:Gem::Specification 
+name: treetop-dcf
+version: !ruby/object:Gem::Version 
+  version: 0.2.0
+platform: ruby
+authors: 
+- "Bj\xC3\xB8rn Arild M\xC3\xA6land"
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2009-10-28 00:00:00 +01:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: treetop
+  type: :runtime
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: 1.2.5
+    version: 
+description: Easy to use parser for Debian Control Files.
+email: bjorn.maeland@gmail.com
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- README.markdown
+files: 
+- lib/dcf.rb
+- lib/dcf_grammar.treetop
+- README.markdown
+has_rdoc: true
+homepage: http://github.com/Chrononaut/treetop-dcf
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      version: "0"
+  version: 
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.5
+signing_key: 
+specification_version: 3
+summary: Easy to use parser for Debian Control Files.
+test_files: 
+- test/helper.rb
+- test/test_dcf.rb
+- test/suite.rb