shalmaneser 1.2.0.rc3 → 1.2.0.rc4

data/lib/rosy/RosyTrainingTestTable.rb

@@ -33,9 +33,10 @@
 
 require "common/ruby_class_extensions"
 
-require "rosy/DBTable"
+require 'db/db_table'
 require "rosy/FeatureInfo"
 
+# @note AB: Possibly this file belongs to <lib/db>. Check it!
 ######################
 class RosyTrainingTestTable
   attr_reader :database, :maintable_name, :feature_names, :feature_info

data/lib/rosy/View.rb

@@ -11,7 +11,7 @@
 # It can be returned directly, or modified by some "dynamic feature object", 
 # and its value (modified or unmodified) will always be last in the array representation of a row.
 
-require "common/SQLQuery"
+require 'db/sql_query'
 require "common/ruby_class_extensions"
 require "common/RosyConventions"
 

data/lib/rosy/opt_parser.rb

@@ -2,7 +2,7 @@
 
 require 'getoptlong'
 
-require 'rosy/RosyConfigData'
+require 'rosy/rosy_config_data'
 
 module Rosy
 

data/lib/rosy/rosy.rb

@@ -1,7 +1,7 @@
 # AB: 2011-11-14
 # Initial import done, need to reimplement the whole interface.
 
-require 'common/DBInterface'
+require 'db/db_interface'
 require 'rosy/RosyFeaturize'
 require 'rosy/RosyTest'
 require 'rosy/RosyTrain'

data/lib/rosy/rosy_config_data.rb

@@ -0,0 +1,121 @@
+require 'common/config_data'
+
+##############################
+# Class RosyConfigData
+#
+# inherits from ConfigData,
+# sets features for ROSY
+
+class RosyConfigData < ConfigData
+  CONFIG_DEFS = { # features
+    "feature" => "list",
+    "classifier" => "list",
+    
+    "verbose" => "bool" ,
+    "enduser_mode" => "bool", 
+    
+    "experiment_ID" => "string",
+    
+    "directory_input_train" => "string",
+    "directory_input_test" => "string",
+    "directory_output" => "string", 
+    
+    "preproc_descr_file_train" => "string",
+    "preproc_descr_file_test" => "string",
+    "external_descr_file"    => "string",
+    
+    "dbtype" => "string",    # "mysql" or "sqlite"
+    
+    "host" => "string",      # DB access: sqlite only
+    "user" => "string",
+    "passwd" => "string",
+    "dbname" => "string",
+    
+    "data_dir" => "string",  # for external use
+    "rosy_dir" => "pattern", # for internal use only, set by rosy.rb
+    
+    "classifier_dir" => "string", # if present, special directory for classifiers
+    
+    "classif_column_name" => "string",
+    "main_table_name" => "pattern",
+    "test_table_name" => "pattern",
+    
+    "eval_file" => "pattern", 
+    "log_file" => "pattern",
+    "failed_file" => "pattern",
+    "classifier_file" => "pattern",
+    "classifier_output_file" => "pattern",
+    "noval" => "string",
+    
+    
+    "split_nones" => "bool",
+    "print_eval_log" => "bool",
+    "assume_argrec_perfect" => "bool", 
+    "xwise_argrec" => "string",
+    "xwise_arglab" => "string",
+    "xwise_onestep" => "string",
+    
+    "fe_syn_repair" => "bool", # map words to constituents for FEs: idealize?
+    "fe_rel_repair" => "bool", # FEs: include non-included relative clauses into FEs
+    "prune" => "string", # pruning prior to argrec?
+
+    # Imported from PrepConfigData
+    'do_postag' => 'bool',
+    'do_lemmatize' => 'bool',
+    'do_parse' => 'bool',
+    'pos_tagger' => 'string',
+    'lemmatizer' => 'string',
+    'parser' => 'string'
+    }
+
+  def initialize(filename)
+    super(filename, CONFIG_DEFS, ["exp_ID", "test_ID", "split_ID",
+                                  "feature_name", "classif", "step", 
+                                  "group", "dataset","mode"])
+
+    # set access functions for list features
+    set_list_feature_access("feature", 
+			    method("access_feature"))
+    
+    # set access functions for list features
+    set_list_feature_access("classifier", 
+			    method("access_feature"))
+    
+  end
+
+  ###
+  # protected
+  
+  #####
+  # access_feature
+  #
+  # access function for feature 'feature'
+  #
+  # assumed format in the config file:
+  #
+  #   feature = path [option]*
+  #
+  # i.e. first the name of the feature type to use, then
+  # optionally options associated with that feature,
+  # e.g. 'argrec': use that feature only when computing argrec
+  #
+  # the access function is called with parameter val_list, an array of
+  # string tuples, one string tuple for each feature defined.
+  # the first string in the tuple is the feature name, the rest are the options
+  #
+  # returns: a list of pairs [feature_name(string), options(array:string)]
+  # of defined features
+  def access_feature(val_list) # array:array:string: list of tuples defined in config file
+		               # for feature 'feature'
+    if val_list.nil?
+      []
+    else
+      val_list.map do |feature_descr_tuple|
+        [feature_descr_tuple.first, feature_descr_tuple[1..-1]]
+      end
+    end
+  end
+end
+
+
+ 

data/lib/shalmaneser/opt_parser.rb

@@ -4,6 +4,12 @@ require 'shalmaneser/version'
 
 module Shalmaneser
   class OptParser
+    
+    # Specify a default option first.
+    ENCODINGS = %w{iso utf8 hex}
+    LANGUAGES = %w{de en}
+    PARSERS   = %w{BerkeleyParser StanfordParser CollinsParser}
+
     def self.parse(cmd_args)
 
       parser = create_parser
@@ -32,10 +38,34 @@ module Shalmaneser
 
     def self.create_parser
       OptionParser.new do |opts|
-        opts.banner = 'Usage: shalmaneser OPTIONS'
+        opts.banner = "CAUTION: Shalmaneser DOES NOT work in Enduser Mode for now!\n" +
+          'Usage: shalmaneser -i path [-o path -e enc -l lang -p parser]'
+        opts.separator ''
+        opts.separator 'Mandatory options:'
+        opts.on('-i', '--input INPUTPATH', String,
+                'Path to directory with input files.')
+        opts.separator ''
+
+        opts.separator 'Facultative options:'
+        opts.on('-o', '--output OUTPUTPATH', String,
+                'Path to directory for output files.',
+                'If not set it defaults to <users home directory>.')
+        opts.on('-e', '--encoding ENCODING', ENCODINGS,
+                "Encoding of input files. Allowed encodings are: #{ENCODINGS.join(', ')}.",
+                "If not set it defaults to <#{ENCODINGS.first}>.")
+        opts.on('-l', '--language LANGUAGE', LANGUAGES,
+                "Language to be processed. Allowed language are: #{LANGUAGES.join(', ')}.",
+                "If not set it defaults to <#{LANGUAGES.first}>.")
+        opts.on('-p', '--parser PARSER', PARSERS,
+                "Parser name you want to use.",
+                "Implemented parsers are: #{PARSERS.join(', ')}.",
+                "If not set it defaults to <#{PARSERS.first}>.")
+        opts.on('--visualize', 'Open output files with SALTO.',
+                'This is ignored if SALTO is not found on your system.')
+
         opts.separator ''
         opts.separator 'Common options:'
-        
+
         opts.on_tail('-h', '--help', 'Show the help message.') do
           puts opts
           exit

data/lib/shalmaneser/version.rb

@@ -1,3 +1,3 @@
 module Shalmaneser
-  VERSION = '1.2.0.rc3'
+  VERSION = '1.2.0.rc4'
 end

metadata

@@ -1,107 +1,15 @@
 --- !ruby/object:Gem::Specification
 name: shalmaneser
 version: !ruby/object:Gem::Version
-  version: 1.2.0.rc3
+  version: 1.2.0.rc4
 platform: ruby
 authors:
 - Andrei Beliankou
 autorequire: 
 bindir: bin
-cert_chain:
-- |
-  -----BEGIN CERTIFICATE-----
-  MIIDZDCCAkygAwIBAgIBATANBgkqhkiG9w0BAQUFADA8MQ4wDAYDVQQDDAVhcmJv
-  eDEWMBQGCgmSJomT8ixkARkWBnlhbmRleDESMBAGCgmSJomT8ixkARkWAnJ1MB4X
-  DTE0MDEwNjE1NDU0MFoXDTE1MDEwNjE1NDU0MFowPDEOMAwGA1UEAwwFYXJib3gx
-  FjAUBgoJkiaJk/IsZAEZFgZ5YW5kZXgxEjAQBgoJkiaJk/IsZAEZFgJydTCCASIw
-  DQYJKoZIhvcNAQEBBQADggEPADCCAQoCggEBAKpdkXWo8sFAq/Dd+rCLRCKHpH02
-  8cZsiy3Dx5kt9qpjYn/LX4/QlJ2mc2C3QXUr++DFJjA0K3TcRS2esUVS9ZlNMDM9
-  YQnxFmPJ4tfpsMiteQMBVqU643aZrh64rqddklg8BwRec+prIIDxfQHzXalnNBad
-  YfiHhjgTh5YQsx3Q0zidhlAtsIbJljaNLuJ4DiVQUtjumEnOI0HTLTuUdpg/Hhh+
-  nPlnhwOUBGzj5hUGzf9QcbV2k99KXsKlHQVkMDn7gsXuIKsisVde07lUbhhR7YGy
-  Z3vGnZK7oNI0It0LIBm7pdx2gtB4YG9O5QKEJo0WzLY60TiY8DzDguLndIcCAwEA
-  AaNxMG8wCQYDVR0TBAIwADALBgNVHQ8EBAMCBLAwHQYDVR0OBBYEFHhWOk+TWhtU
-  KMnM8ZyfBZYcVXxDMBoGA1UdEQQTMBGBD2FyYm94QHlhbmRleC5ydTAaBgNVHRIE
-  EzARgQ9hcmJveEB5YW5kZXgucnUwDQYJKoZIhvcNAQEFBQADggEBAF2Y+mc/uTug
-  OX3ivVkD4AaPpFsB2EglJhQxivlAHkix593RpZPXNf6jeu36oRCV/vRFLkzzaZ73
-  N7MaI5Z2HczDkZvi8ZZM5L3p4wHttquranUdI3bZv4SiAVFmhkeFZLSp6pFf/Fmg
-  qmEeXWVbsCIhYI7KYQ0XKbnRuj9AmjUEoMBZPnMsM1S/R+dBQfrUszXROWqxaENA
-  728ScNHCmRYuNutDO9yRDJT1SRumpgwH4df6c0LHBCuXuQTWODYqc/CDZJJb9Tfi
-  BJreIpPMe0KFMphkN/x5cHkRDtMoY+rBGcqRe60otCEsAHdM+CXox9tAREnr/4lT
-  Jn9sRDVszy4=
-  -----END CERTIFICATE-----
-date: 2014-01-11 00:00:00.000000000 Z
-dependencies:
-- !ruby/object:Gem::Dependency
-  name: mysql
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: rdoc
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: bundler
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: yard
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: rake
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - '>='
-      - !ruby/object:Gem::Version
-        version: '0'
+cert_chain: []
+date: 2015-09-04 00:00:00.000000000 Z
+dependencies: []
 description: |
   SHALMANESER - SHALlow seMANtic parSER. This package provides a toolbox for
   Semantic Role Labeling (SRL). SHALMANESER uses supervised learning algorithms to
@@ -123,7 +31,7 @@ extra_rdoc_files:
 - doc/exp_files.md
 - doc/index.md
 files:
-- .yardopts
+- ".yardopts"
 - CHANGELOG.md
 - LICENSE.md
 - README.md
@@ -134,13 +42,10 @@ files:
 - doc/exp_files.md
 - doc/index.md
 - lib/common/AbstractSynInterface.rb
-- lib/common/ConfigData.rb
 - lib/common/Counter.rb
-- lib/common/DBInterface.rb
 - lib/common/EnduserMode.rb
 - lib/common/Eval.rb
 - lib/common/FixSynSemMapping.rb
-- lib/common/FrPrepConfigData.rb
 - lib/common/Graph.rb
 - lib/common/ISO-8859-1.rb
 - lib/common/ML.rb
@@ -150,7 +55,6 @@ files:
 - lib/common/Parser.rb
 - lib/common/RegXML.rb
 - lib/common/RosyConventions.rb
-- lib/common/SQLQuery.rb
 - lib/common/STXmlTerminalOrder.rb
 - lib/common/SalsaTigerRegXML.rb
 - lib/common/SalsaTigerXMLHelper.rb
@@ -159,16 +63,24 @@ files:
 - lib/common/Tiger.rb
 - lib/common/Timbl.rb
 - lib/common/Tree.rb
-- lib/common/frprep_helper.rb
+- lib/common/config_data.rb
+- lib/common/config_format_element.rb
 - lib/common/headz.rb
 - lib/common/option_parser.rb
+- lib/common/prep_config_data.rb
+- lib/common/prep_helper.rb
 - lib/common/ruby_class_extensions.rb
+- lib/db/db_interface.rb
+- lib/db/db_mysql.rb
+- lib/db/db_sqlite.rb
+- lib/db/db_table.rb
+- lib/db/db_wrapper.rb
+- lib/db/sql_query.rb
 - lib/ext/maxent/Classify.class
 - lib/ext/maxent/Train.class
 - lib/fred/Baseline.rb
 - lib/fred/FileZipped.rb
 - lib/fred/FredBOWContext.rb
-- lib/fred/FredConfigData.rb
 - lib/fred/FredConventions.rb
 - lib/fred/FredDetermineTargets.rb
 - lib/fred/FredEval.rb
@@ -183,6 +95,7 @@ files:
 - lib/fred/FredTrain.rb
 - lib/fred/PlotAndREval.rb
 - lib/fred/fred.rb
+- lib/fred/fred_config_data.rb
 - lib/fred/md5.rb
 - lib/fred/opt_parser.rb
 - lib/frprep/Ampersand.rb
@@ -208,17 +121,12 @@ files:
 - lib/frprep/opt_parser.rb
 - lib/frprep/ruby_class_extensions.rb
 - lib/rosy/AbstractFeatureAndExternal.rb
-- lib/rosy/DBMySQL.rb
-- lib/rosy/DBSQLite.rb
-- lib/rosy/DBTable.rb
-- lib/rosy/DBWrapper.rb
 - lib/rosy/ExternalConfigData.rb
 - lib/rosy/FailedParses.rb
 - lib/rosy/FeatureInfo.rb
 - lib/rosy/GfInduce.rb
 - lib/rosy/GfInduceFeature.rb
 - lib/rosy/InputData.rb
-- lib/rosy/RosyConfigData.rb
 - lib/rosy/RosyConfusability.rb
 - lib/rosy/RosyEval.rb
 - lib/rosy/RosyFeatureExtractors.rb
@@ -237,6 +145,7 @@ files:
 - lib/rosy/View.rb
 - lib/rosy/opt_parser.rb
 - lib/rosy/rosy.rb
+- lib/rosy/rosy_config_data.rb
 - lib/shalmaneser/opt_parser.rb
 - lib/shalmaneser/version.rb
 homepage: http://bu.chsta.be/projects/shalmaneser/
@@ -246,7 +155,7 @@ metadata:
   issue_tracker: https://github.com/arbox/shalmaneser/issues
 post_install_message: |2+
 
-  Thank you for installing Shalmaneser 1.2.0.rc3!
+  Thank you for installing Shalmaneser 1.2.0.rc4!
 
   This software package has multiple external dependencies:
   - OpenNLP Maximum Entropy Classifier;
@@ -263,24 +172,24 @@ post_install_message: |2+
   https://github.com/arbox/shalmaneser/issues
 
 rdoc_options:
-- -m
+- "-m"
 - README.md
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - '='
     - !ruby/object:Gem::Version
-      version: 1.8.7
+      version: '2.0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>'
+  - - ">"
     - !ruby/object:Gem::Version
       version: 1.3.1
 requirements:
 - mysql-server
 rubyforge_project: 
-rubygems_version: 2.2.0
+rubygems_version: 2.4.5
 signing_key: 
 specification_version: 4
 summary: SHALMANESER - SHALlow seMANtic parSER