worlddb-models 2.3.0 → 2.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: fe2b1dd265c24760835eeefbd445c3a8302eb554
-  data.tar.gz: 35aeaaac8672937b8388825aa4c36cc11ce4651b
+  metadata.gz: e65089aa526d90b22ff73acc40835c46ef784c93
+  data.tar.gz: 07883ca048c28555a3d27d1ba9a25a9b1a5de377
 SHA512:
-  metadata.gz: 60c27ed1d0fceebca6edef19c0a50a542dafdbd66b71bdf9ee0e84aa641de7569d856b9fae697ffb844428ee24b72c505e663dae1bd9b4bc8d8b54f66e9bef9c
-  data.tar.gz: 6acf8fcfe52845533582c7dbd010f17feb834beb0110c278e2cc72607d7240be894bff150facf73749f9f5419685e8216d05db6732c351fb44df209af52aec8f
+  metadata.gz: f9c7c04d0c3d9b7d6240865abfbc523ba464ed30d90c8433afe11826ba9e7fe770ef18176b6785f922e9ca4c51c56d6a55e253813c149ef49c0ac24a62f9b523
+  data.tar.gz: 9e615c49999dc69f574b9f50c8c14d1f489e5dd279e99e772f6629fee56b50fa7dab2445ae2b351a1efb833454694544d4c7762746b1b82774913323c691fa37

data/Manifest.txt

@@ -67,6 +67,7 @@ test/test_model_state.rb
 test/test_model_states_at.rb
 test/test_model_states_de.rb
 test/test_models.rb
+test/test_name_parser.rb
 test/test_parse_city.rb
 test/test_parse_country.rb
 test/test_parse_state.rb

data/lib/worlddb/models/city_base.rb

@@ -32,11 +32,19 @@ class CityBase < ActiveRecord::Base
   def on_create
     place_rec = Place.create!( name: name, kind: place_kind )
     self.place_id = place_rec.id 
+
+    ## create name recs -- todo: add default lang etc. - move to: after_create callback!!
+    Name.parse( name, alt_names,
+                  place_id:   place_rec.id,
+                  place_kind: place_rec.kind )
   end
 
   def on_update
     ## fix/todo: check - if name or kind changed - only update if changed ?? why? why not??
     place.update_attributes!( name: name, kind: place_kind )
+
+    ## todo/fix:
+    ## update names too ??
   end
 
   validates :key,  format: { with: /#{CITY_KEY_PATTERN}/, message: CITY_KEY_PATTERN_MESSAGE }

data/lib/worlddb/models/name.rb

@@ -1,11 +1,98 @@
 # encoding: utf-8
 
+
+##
+# todo: move to textutils!!!
+
+class NameParser
+
+  include LogUtils::Logging
+
+  def parse( chunks )
+    ## todo/fix: (re)use nameparser - for now "simple" inline version
+    ##  fix!!! - note: for now lang gets ignored
+    ##  fix: add hanlde
+    ##  Leuven[nl]|Louvain[fr] Löwen[de]
+    ##  Antwerpen[nl]|Anvers[fr] [Antwerp]
+    ##  Brussel[nl]•Bruxelles[fr]   -> official bi-lingual name
+    ##  etc.
+
+    ## values - split into names (name n lang pairs)
+    ## note: assumes (default) lang from more_attribs unless otherwise marked e.g. [] assume en etc.
+
+    ## split chunks into values
+    values = []
+    chunks.each do |chunk|
+      next if chunk.nil? || chunk.blank?  ## skip nil or empty/blank chunks
+
+      parts = chunk.split( '|' )   # 1)  split |
+
+      parts.each do |part|
+        s = StringScanner.new( part )
+        s.skip( /[ \t]+/)   # skip whitespaces
+
+        while s.eos? == false
+          if s.check( /\[/ )
+            ## scan everything until the end of bracket (e.g.])
+            ##  fix!!! - note: for now lang gets ignored
+            value = s.scan( /\[[^\]]+\]/)
+            value = value[1...-1]   # strip enclosing [] e.g. [Bavaria] => Bavaria
+          else
+            ## scan everything until the begin of bracket (e.g.[)
+            value = s.scan( /[^\[]+/)
+            value = value.strip
+          end
+          values << value
+
+          s.skip( /[ \t]+/)  # skip whitespaces
+          logger.debug( "[NameParser] eos?: #{s.eos?}, rest: >#{s.rest}<" )
+        end
+      end
+    end
+
+    logger.debug( "[NameParser] values=#{values.inspect}")
+
+    names = []
+    values.each do |value|
+      name = value
+      ## todo: split by bullet ? (official multilang name) e.g. Brussel • Bruxelles
+      ## todo: process variants w/ () e.g. Krems (a. d. Donau) etc. ??
+      names << name
+    end
+
+    logger.debug( "[NameParser] names=#{names.inspect}")
+
+    names
+  end # method parse
+end # class NameParser
+
+
+
 module WorldDb
   module Model
 
 class Name < ActiveRecord::Base
 
 
+  def self.parse( *args )
+    ## remove (extract) attribs hash (if last arg is a hash n present)
+    more_attribs = args.last.is_a?(Hash) ? args.pop : {}  ## extract_options!
+    chunks       = args
+
+    names = NameParser.new.parse( chunks )
+    recs = []
+
+    names.each do |name|
+      attribs = more_attribs.merge( name: name )  ## overwrite (default) attribs (lang, etc.) plus add name
+      puts "[Name.parse] adding Name record:"
+      puts "  #{attribs.inspect}"
+      rec = Name.create!( attribs )
+      recs << rec
+    end
+
+    recs # note: returns an array of name records
+  end   # method parse
+
 end  # class Name
 
 

data/lib/worlddb/models/state.rb

@@ -13,10 +13,8 @@ class State < StateBase
   has_many   :munis,    class_name: 'Muni',    foreign_key: 'state_id'  # ADM3 / ADM4
   has_many   :cities,   class_name: 'City',    foreign_key: 'state_id'
 
-  def place_kind   # use place_kind_of_code ??
-    'ADM1'   # note: for now assumes always level 1
-  end
-  
+  def place_kind() 'STAT'; end
+
   def state_id  # return top-level (e.g. adm1) state_id
     id
   end
@@ -31,9 +29,7 @@ class Part < StateBase    ## optional ADM2 e.g. Regierungsbezirke in Bayern, etc
   has_many   :counties,  class_name: 'County', foreign_key: 'part_id'
   has_many   :cities,    class_name: 'City',   foreign_key: 'part_id'
 
-  def place_kind   # use place_kind_of_code ??
-    'ADM2'   # note: for now assumes always level 2
-  end
+  def place_kind() 'PART'; end
 
   def country_id  # return country_id via top-level (e.g. adm1) state; -- used for auto-creating cities
     state.country_id
@@ -50,9 +46,7 @@ class County < StateBase   ## note: might be ADM2 or ADM3
   has_many   :munis,  class_name: 'Muni',  foreign_key: 'county_id'
   has_many   :cities, class_name: 'City',  foreign_key: 'county_id'
 
-  def place_kind   # use place_kind_of_code ??
-    "ADM#{level}"   # note: might be ADM2 or ADM3
-  end
+  def place_kind() 'COUN'; end
 
   def country_id  # return country_id via top-level (e.g. adm1) state
     state.country_id
@@ -68,9 +62,7 @@ class Muni < StateBase   ## note: might be ADM3 or ADM4
 
   has_many   :cities, class_name: 'City',   foreign_key: 'muni_id'
 
-  def place_kind   # use place_kind_of_code ??
-    "ADM#{level}"   # note: might be ADM3 or ADM4
-  end
+  def place_kind() 'MUNI'; end
 
   def country_id  # return country_id via top-level (e.g. adm1) state
     state.country_id

data/lib/worlddb/models/state_base.rb

@@ -39,12 +39,20 @@ class StateBase < ActiveRecord::Base
 
   def on_create
     place_rec = Place.create!( name: name, kind: place_kind )
-    self.place_id = place_rec.id 
+    self.place_id = place_rec.id
+
+    ## create name recs -- todo: add default lang etc. - move to: after_create callback!!
+    Name.parse( name, alt_names,
+                  place_id:   place_rec.id,
+                  place_kind: place_rec.kind )
   end
 
   def on_update
     ## fix/todo: check - if name or kind changed - only update if changed ?? why? why not??
     place.update_attributes!( name: name, kind: place_kind )
+
+    ## todo/fix:
+    ## update names too ??
   end
 
 

data/lib/worlddb/schema.rb

@@ -18,10 +18,10 @@ create_table :places do |t|
   #  CNTY - country
   #  TERR - terr
   # state:
-  #  ADM1 - e.g. state/province
-  #  ADM2 - e.g. county/bezirk/kreis
-  #  ADM3 - e.g. municipality/gemeinde
-  #  ADM4
+  #  STAT  -- was: ADM1 - e.g. state/province/etc.
+  #  PART   e.g. regierungsbezirk ADM2
+  #  COUN -- was ADM2/ADM3 - e.g. county/bezirk/kreis/etc.
+  #  MUNI -  was ADM3/ADM4 - e.g. municipality/gemeinde/etc.
   # city:
   #  MTRO - metro
   #  CITY - city/town/village/hamlet //(de)ort/stadt/markt
@@ -50,7 +50,9 @@ end
 ### alternative names for places
 create_table :names do |t|
   t.string     :name,   null: false
-  t.references :place,  null: false
+  t.references :place,      null: false
+  t.string     :place_kind, null: false  #  add (cached) place_kind for easier search queries (w/o join)
+
   ### todo/fix: add lang_id to reference lang from lang table!!! (for now make it a duplicate; e.g. keep stirng lang for now)
   t.string     :lang,   null: false, default: 'en'   # default to english for now (or use unknown/undeterminded/unspecified???)
 

data/lib/worlddb/version.rb

@@ -5,7 +5,7 @@ module WorldDb
   # sync version w/ sport.db n friends - why? why not?
   MAJOR = 2 ## todo: namespace inside version or something - why? why not??
   MINOR = 3
-  PATCH = 0
+  PATCH = 1
   VERSION = [MAJOR,MINOR,PATCH].join('.')
 
   def self.version

data/test/test_model_states_at.rb

@@ -36,7 +36,7 @@ class TestModelStatesAt < MiniTest::Test
     assert_equal 1,          n.area 
     assert_equal at.id,      n.country_id
     assert_equal 1,          n.level
-    assert_equal 'ADM1',     n.place_kind
+    assert_equal 'STAT',     n.place_kind
 
     ### test place
     assert_equal  'Niederösterreich', n.place.name
@@ -64,7 +64,7 @@ class TestModelStatesAt < MiniTest::Test
     assert_equal 1,        tu.area 
     assert_equal n.id,     tu.state_id
     assert_equal 2,        tu.level
-    assert_equal 'ADM2',   tu.place_kind
+    assert_equal 'COUN',   tu.place_kind
 
     ### test place
     assert_equal  'Tulln', tu.place.name
@@ -94,7 +94,7 @@ class TestModelStatesAt < MiniTest::Test
     assert_equal n.id,          gr.state_id
     assert_equal tu.id,         gr.county_id
     assert_equal 3,             gr.level
-    assert_equal 'ADM3',        gr.place_kind
+    assert_equal 'MUNI',        gr.place_kind
 
     ### test place
     assert_equal  'Grafenwörth', gr.place.name

data/test/test_model_states_de.rb

@@ -36,7 +36,7 @@ class TestModelStatesDe < MiniTest::Test
     assert_equal 1,          by.area 
     assert_equal de.id,      by.country_id
     assert_equal 1,          by.level
-    assert_equal 'ADM1',     by.place_kind
+    assert_equal 'STAT',     by.place_kind
 
     ### test place
     assert_equal  'Bayern', by.place.name
@@ -64,7 +64,7 @@ class TestModelStatesDe < MiniTest::Test
     assert_equal 1,              ob.area 
     assert_equal by.id,          ob.state_id
     assert_equal 2,              ob.level
-    assert_equal 'ADM2',         ob.place_kind
+    assert_equal 'PART',         ob.place_kind
 
     ### test place
     assert_equal  'Oberbayern', ob.place.name
@@ -95,7 +95,7 @@ class TestModelStatesDe < MiniTest::Test
     assert_equal by.id,       fs.state_id
     assert_equal ob.id,       fs.part_id
     assert_equal 3,           fs.level
-    assert_equal 'ADM3',      fs.place_kind
+    assert_equal 'COUN',      fs.place_kind
 
     ### test place
     assert_equal  'Freising', fs.place.name
@@ -129,7 +129,7 @@ class TestModelStatesDe < MiniTest::Test
     assert_equal by.id,               au.state_id
     assert_equal fs.id,               au.county_id
     assert_equal 4,                   au.level
-    assert_equal 'ADM4',              au.place_kind
+    assert_equal 'MUNI',              au.place_kind
 
     ### test place
     assert_equal  'Au i.d. Hallertau', au.place.name

data/test/test_name_parser.rb

@@ -0,0 +1,38 @@
+# encoding: utf-8
+
+###
+#  to run use
+#     ruby -I ./lib -I ./test test/test_name_parser.rb
+
+
+require 'helper'
+
+class TestNameParser < MiniTest::Test
+
+  def test_bayern
+    names = NameParser.new.parse( [ 'Bayern [Bavaria]' ] )
+
+    assert_equal 2, names.size
+    assert_equal 'Bayern',  names[0]
+    assert_equal 'Bavaria', names[1]
+  end
+
+  def test_brussel
+    names = NameParser.new.parse( [ 'Brussel•Bruxelles [Brussels]' ] )
+
+    ## fix: (auto-)split into three variants e.g.Brussel•Bruxelles, Brussel, Bruxelles
+    assert_equal 2, names.size
+    assert_equal 'Brussel•Bruxelles',  names[0]
+    assert_equal 'Brussels',           names[1]
+  end
+
+  def test_antwerp
+    names = NameParser.new.parse( [ 'Antwerpen|Anvers [Antwerp]' ] )
+
+    assert_equal 3, names.size
+    assert_equal 'Antwerpen', names[0]
+    assert_equal 'Anvers',    names[1]
+    assert_equal 'Antwerp',   names[2]
+  end
+
+end # class TestNameParser

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: worlddb-models
 version: !ruby/object:Gem::Version
-  version: 2.3.0
+  version: 2.3.1
 platform: ruby
 authors:
 - Gerald Bauer
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-06-08 00:00:00.000000000 Z
+date: 2015-06-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: props
@@ -243,6 +243,7 @@ files:
 - test/test_model_states_at.rb
 - test/test_model_states_de.rb
 - test/test_models.rb
+- test/test_name_parser.rb
 - test/test_parse_city.rb
 - test/test_parse_country.rb
 - test/test_parse_state.rb
@@ -276,6 +277,7 @@ signing_key:
 specification_version: 4
 summary: worlddb -  world.db schema 'n' models for easy (re)use
 test_files:
+- test/test_name_parser.rb
 - test/test_models.rb
 - test/adm/test_read_tree.rb
 - test/adm/test_fixture_matcher_tree.rb