stats_package_syntax_file_generator 1.1.3 → 1.1.6

data/lib/syntax_file/maker_spss.rb

@@ -4,21 +4,21 @@
 #   https://github.com/mnpopcenter/stats_package_syntax_file_generator
 
 module SyntaxFile
-  class MakerSPSS < Maker
-
-    def initialize(sfc, syntax_type)
-      super
-      m = @sfc.max_var_name_length
-      @var_loc_format = "  %-#{m}s  %s"
-      @var_lab_format = "  %-#{m}s  %s %s"
-      @cmd_end = '.'
-      @var_label_max_leng = 120
-      @val_label_max_leng = 120
-      @segment_max_leng = 100
-    end
-
-    def syntax
-      r = [
+class MakerSPSS < Maker
+
+def initialize (sfc, syntax_type)
+    super
+    m = @sfc.max_var_name_length
+    @var_loc_format     = "  %-#{m}s  %s"
+    @var_lab_format     = "  %-#{m}s  %s %s"
+    @cmd_end            = '.'
+    @var_label_max_leng = 120
+    @val_label_max_leng = 120
+    @segment_max_leng   = 100
+end
+
+def syntax
+    r = [
         comments_start,
         syn_cd,
         blank,
@@ -28,167 +28,168 @@ module SyntaxFile
         syn_execute,
         blank,
         comments_end,
-      ]
-      r.flatten
-    end
-
-    def comments_start
-      convert_to_comments(super)
-    end
-
-    def comments_end
-      convert_to_comments(super)
-    end
-
-    def convert_to_comments(lines)
-      return [] if lines.empty?
-      lines.push @cmd_end
-      [
+    ]
+    r.flatten
+end
+
+def comments_start
+    convert_to_comments(super)
+end
+
+def comments_end
+    convert_to_comments(super)
+end
+
+def convert_to_comments (lines)
+    return [] if lines.empty?
+    lines.push @cmd_end
+    [
         lines.map { |ln| '* ' + ln },
         blank,
-      ].flatten
-    end
+    ].flatten
+end
 
-    def syn_cd
-      'cd ' + q(@sfc.data_dir_name) + @cmd_end
-    end
+def syn_cd
+    'cd ' + q(@sfc.data_dir_name) + @cmd_end
+end
 
-    def syn_df
-      @sfc.data_structure == 'hier' ? syn_dfh : syn_dfr
-    end
+def syn_df
+    @sfc.data_structure == 'hier' ? syn_dfh : syn_dfr
+end
 
-    def syn_dfr
-      r = [
+def syn_dfr
+    r = [
         syn_dfr_start,
         syn_var_locations(@sfc.variables),
         syntax_end,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_dfr_start
-      'data list file = ' + q(@sfc.data_file_name) + ' /'
-    end
+def syn_dfr_start
+    'data list file = ' + q(@sfc.data_file_name) + ' /'
+end
 
-    def syn_dfh
-      r = [
+def syn_dfh
+    r = [
         syn_dfh_file_type,
         syn_dfh_data_blocks,
         'end file type' + @cmd_end,
         blank,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_dfh_file_type
-      r = [
+def syn_dfh_file_type
+    r = [
         'file type ' + nested_or_mixed(),
         '  /file = ' + q(@sfc.data_file_name),
         '  /record = ' + var_loc_with_fmt(@sfc.record_type_var).to_s,
         syntax_end,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_dfh_data_blocks
-      r = @sfc.record_types.map { |rt|
+def syn_dfh_data_blocks
+    r = @sfc.record_types.map { |rt|
         [
-          syn_dfh_data_block_start(rt),
-          syn_var_locations(@sfc.get_vars_by_record_type(rt)),
-          syntax_end,
+            syn_dfh_data_block_start(rt),
+            syn_var_locations(@sfc.get_vars_by_record_type(rt)),
+            syntax_end,
         ]
-      }
-      r.flatten
-    end
+    }
+    r.flatten
+end
 
-    def syn_dfh_data_block_start(rt)
-      rt_var = @sfc.record_type_var
-      [
+def syn_dfh_data_block_start (rt)
+    rt_var = @sfc.record_type_var
+    [
         'record type ' + val_q(rt_var, val_as_s(rt_var, rt)) + @cmd_end,
         'data list /',
-      ]
-    end
+    ]
+end
 
-    def syn_var_locations(var_list)
-      var_list.map { |v| sprintf @var_loc_format, v.name, var_loc_with_fmt(v) }
-    end
+def syn_var_locations (var_list)
+    var_list.map { |v| sprintf @var_loc_format, v.name, var_loc_with_fmt(v) }
+end
 
-    def syn_var_labs(var_list = [])
-      var_list = @sfc.get_vars_with_var_labels if var_list.empty?
-      return [] if var_list.empty?
-      r = [
+def syn_var_labs (var_list = [])
+    var_list = @sfc.get_vars_with_var_labels if var_list.empty?
+    return [] if var_list.empty?
+    r = [
         'variable labels',
         var_list.map { |var| syn_var_lab_for_var(var) },
         syntax_end,
-      ]
-      r.flatten
-    end
-
-    def syn_var_lab_for_var(var)
-      lab = label_trunc(var.label, @var_label_max_leng)
-      lab_segments = label_segments(lab, @segment_max_leng).map { |s| q(s) }
-      weave_label_segments(@var_lab_format, [var.name], lab_segments, ' ', '+')
-    end
-
-    def syn_val_labs
-      var_list = @sfc.get_vars_with_values
-      return [] if var_list.empty?
-      r = [
+    ]
+    r.flatten
+end
+
+def syn_var_lab_for_var (var)
+    lab = label_trunc(var.label, @var_label_max_leng)
+    lab_segments = label_segments(lab, @segment_max_leng).map { |s| q(s) }
+    weave_label_segments(@var_lab_format, [var.name], lab_segments, ' ', '+')
+end
+
+def syn_val_labs
+    var_list = @sfc.get_vars_with_values
+    return [] if var_list.empty?
+    r = [
         'value labels',
         syn_val_labs_for_var_list(var_list),
         syntax_end,
-      ]
-      r.flatten
-    end
-
-    def syn_val_labs_for_var_list(var_list)
-      var_list.map { |var| syn_val_labs_for_var(var) }
-    end
-
-    def syn_val_labs_for_var(var)
-      val_list = labelable_values(var)
-      return [] if val_list.empty?
-      m = max_value_length(var, val_list)
-      m = m + 2 if var.is_string_var
-      m = @segment_max_leng + 2 if m > @segment_max_leng + 2
-      value_format = "    %-#{m}s %s %s"
-      r = [
+    ]
+    r.flatten
+end
+
+def syn_val_labs_for_var_list (var_list)
+    var_list.map { |var| syn_val_labs_for_var(var) }
+end
+
+def syn_val_labs_for_var (var)
+    val_list = labelable_values(var)
+    return [] if val_list.empty?
+    m = max_value_length(var, val_list)
+    m = m + 2 if var.is_string_var
+    m = @segment_max_leng + 2 if m > @segment_max_leng + 2
+    value_format = "    %-#{m}s %s %s"
+    r = [
         syn_val_labs_for_var_start(var),
         val_list.map { |val| syn_val_lab_for_val(var, val, value_format) },
-      ]
-      r.flatten
-    end
-
-    def syn_val_labs_for_var_start(var)
-      '  /' + var.name
-    end
-
-    def syn_val_lab_for_val(var, val, fmt)
-      lab = label_trunc(val.label, @val_label_max_leng)
-      lab = val.value if lab.nil? || (lab.strip.length == 0)
-      vs = val_as_s(var, val.value)
-      val_segments = label_segments(vs, @segment_max_leng).map { |s| val_q(var, s) }
-      lab_segments = label_segments(lab, @segment_max_leng).map { |s| q(s) }
-      weave_label_segments(fmt, val_segments, lab_segments, ' ', '+')
-    end
-
-    def syn_execute
-      'execute' + @cmd_end
-    end
-
-    def nested_or_mixed
-      @sfc.rectangularize ? 'nested' : 'mixed'
-    end
-
-    def var_loc_with_fmt(var)
-      var.column_locations_as_s + var_fmt(var)
-    end
-
-    def var_fmt(var)
-      return ' (a)' if var.is_string_var
-      return '' unless var.implied_decimals > 0
-      return ' (' + var.implied_decimals.to_s + ')'
-    end
-
-  end
+    ]
+    r.flatten
+end
+
+def syn_val_labs_for_var_start (var)
+    '  /' + var.name
+end
+
+def syn_val_lab_for_val (var, val, fmt)
+    lab = label_trunc(val.label, @val_label_max_leng)
+    lab = val.value if lab.nil? || (lab.strip.length == 0)
+    vs = val_as_s(var, val.value)
+    val_segments = label_segments(vs,  @segment_max_leng).map { |s| val_q(var, s) }
+    lab_segments = label_segments(lab, @segment_max_leng).map { |s| q(s) }
+    weave_label_segments(fmt, val_segments, lab_segments, ' ', '+')
+end
+
+def syn_execute
+    'execute' + @cmd_end
+end
+
+def nested_or_mixed
+    @sfc.rectangularize ? 'nested' : 'mixed'
+end
+
+def var_loc_with_fmt (var)
+    var.column_locations_as_s + var_fmt(var)
+end
+
+def var_fmt (var)
+    return ' (a)' if var.is_string_var
+	return ' (2) ' if var.is_double_var
+    return '' unless var.implied_decimals > 0
+    return ' (' + var.implied_decimals.to_s + ')'
+end
+
+end
 end

data/lib/syntax_file/maker_stata.rb

@@ -4,29 +4,32 @@
 #   https://github.com/mnpopcenter/stats_package_syntax_file_generator
 
 module SyntaxFile
-  class MakerSTATA < Maker
-
-    def initialize(sfc, syntax_type)
-      super
-
-      mx_var = @sfc.max_var_name_length
-      mx_col = 2 * @sfc.max_col_loc_width + 1
-      @var_loc_format = "  %-7s %-#{mx_var}s  %-#{mx_col}s %s"
-      @var_lab_format = "label var %-#{mx_var}s %s"
-      @infix_format = "%#{mx_col + mx_var + 4}s"
-      @replace_format = "replace %-#{mx_var}s = %-#{mx_var}s / %d"
-      @fixed_point_display_format = "format %-#{mx_var}s %%%d.%df"
-      @general_display_format = "format %-#{mx_var}s %%%d.%dg"
-
-      @cmd_end = ''
-      @cmd_continue = ' ///'
-      @var_label_max_leng = 80
-      @val_label_max_leng = 244
-      @sort_var_stem = '_line_num'
-    end
+class MakerSTATA < Maker
+
+def initialize (sfc, syntax_type)
+    super
+    
+
+    mx_var = @sfc.max_var_name_length
+    mx_col = 2 * @sfc.max_col_loc_width + 1
+    @var_loc_format     = "  %-7s %-#{mx_var}s  %-#{mx_col}s %s"
+    @var_lab_format     = "label var %-#{mx_var}s %s"
+    @infix_format       = "%#{mx_col + mx_var + 4}s"
+    @replace_format     = "replace %-#{mx_var}s = %-#{mx_var}s / %d"
+    @fixed_point_display_format     = "format %-#{mx_var}s %%%d.%df"
+    @general_display_format = "format %-#{mx_var}s %%%d.%dg"
+    @integer_display_format = "format %-#{mx_var}s %%%d.%df"
+	
+	
+    @cmd_end            = ''
+    @cmd_continue       = ' ///'
+    @var_label_max_leng = 80
+    @val_label_max_leng = 244
+    @sort_var_stem      = '_line_num'
+end
 
-    def syntax
-      r = [
+def syntax
+    r = [
         comments_start,
         'set more off',
         blank,
@@ -41,278 +44,275 @@ module SyntaxFile
         syn_val_labs,
         blank,
         comments_end,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def comments_start
-      convert_to_comments(super)
-    end
+def comments_start
+    convert_to_comments(super)
+end
 
-    def comments_end
-      convert_to_comments(super)
-    end
+def comments_end
+    convert_to_comments(super)
+end
 
-    def convert_to_comments(lines)
-      return [] if lines.empty?
-      [
+def convert_to_comments (lines)
+    return [] if lines.empty?
+    [
         lines.map { |ln| '* ' + ln },
         blank,
-      ].flatten
-    end
+    ].flatten
+end
 
-    def syn_df
-      @sfc.data_structure == 'hier' ? syn_dfh : syn_dfr
-    end
+def syn_df
+    @sfc.data_structure == 'hier' ? syn_dfh : syn_dfr
+end
 
-    def syn_dfr
-      syn_infix(@sfc.variables)
-    end
+def syn_dfr
+    syn_infix(@sfc.variables)
+end
 
-    def syn_dfh
-      r = [
+def syn_dfh
+    r = [
         syn_dfh_infix_blocks,
         syn_dfh_combine,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_infix(var_list)
-      r = [
+def syn_infix (var_list)
+    r = [
         syn_infix_start,
         syn_infix_var_locs(var_list),
         syn_infix_end,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_infix_start
-      [
+def syn_infix_start
+    [
         'clear',
         'quietly infix' + sprintf(@infix_format, @cmd_continue),
-      ]
-    end
+    ]
+end
 
-    def syn_infix_var_locs(var_list)
-      var_list.map { |v|
+def syn_infix_var_locs (var_list)
+    var_list.map { |v|
         sprintf @var_loc_format,
-          var_fmt(v),
-          v.name.downcase,
-          v.column_locations_as_s,
-          @cmd_continue
-      }
-    end
+                var_fmt(v),
+                v.name.downcase,
+                v.column_locations_as_s,
+                @cmd_continue
+    }
+end
 
-    def syn_infix_end
-      '  using ' + q(@sfc.data_file_name)
-    end
+def syn_infix_end
+    '  using ' + q(@sfc.data_file_name)
+end
 
-    def syn_dfh_infix_blocks
-      r = []
-      @sfc.record_types.each { |rt|
+def syn_dfh_infix_blocks
+    r = []
+    @sfc.record_types.each { |rt|
         var_list = @sfc.get_vars_by_record_type(rt)
         r.push(
-          syn_infix(var_list),
-          syn_dfh_infix_block_end(rt)
+            syn_infix(var_list),
+            syn_dfh_infix_block_end(rt)
         ) if var_list.size > 0
-      }
-      r.flatten
-    end
+    }
+    r.flatten
+end
 
-    def syn_dfh_infix_block_end(rt)
-      r = [
+def syn_dfh_infix_block_end (rt)
+    r = [
         syn_dfh_infix_gen,
         'drop if ' + rt_ne_statement(rt),
         'sort ' + sort_vars.join(' '),
         'save ' + temp_file_name(rt),
         blank,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_dfh_infix_gen
-      return ["gen  #{@sort_var_stem} = _n"] unless @sfc.rectangularize
-      sv = sort_vars()
-      r = [
+def syn_dfh_infix_gen
+    return ["gen  #{@sort_var_stem} = _n"] unless @sfc.rectangularize
+    sv = sort_vars()
+    r = [
         sv.map { |v| "gen #{v} = _n" },
         sv.zip(@sfc.record_types).map { |z|
-          'replace ' +
-            z[0] +
-            ' = ' +
-            z[0] +
+            'replace '              +
+            z[0]                    +
+            ' = '                   +
+            z[0]                    +
             '[_n - 1] if _n > 1 & ' +
             rt_ne_statement(z[1])
         }
-      ]
-      r.flatten!
-    end
+    ]
+    r.flatten!
+end
 
-    def syn_dfh_combine
-      r = [
+def syn_dfh_combine
+    r = [
         'clear',
         syn_dfh_combine_append,
         syn_dfh_combine_save,
         syn_dfh_combine_erase,
-      ]
-      r.flatten
-    end
+    ]
+    r.flatten
+end
 
-    def syn_dfh_combine_append
-      r = []
-      tf = temp_file_names()
-      if @sfc.rectangularize
+def syn_dfh_combine_append
+    r = []
+    tf = temp_file_names()
+    if @sfc.rectangularize
         sv = sort_vars.reverse
         tf = tf.reverse
         sv.shift
         r.push 'use ' + tf.shift
         sv.zip(tf).each { |z|
-          r.push 'merge m:1 ' + z[0] + ' using ' + z[1] + ', keep(master match)'
-          r.push 'drop _merge'
+            r.push 'merge m:1 ' + z[0] + ' using ' + z[1] + ', keep(master match)'
+            r.push 'drop _merge'
         }
-      else
+    else
         r.push 'use ' + tf.shift
         tf.each { |t| r.push 'append using ' + t }
-      end
-      r
     end
+    r
+end
 
-    def syn_dfh_combine_save
-      [
+def syn_dfh_combine_save
+    [
         'sort ' + sort_vars.join(' '),
         'drop ' + sort_vars.join(' '),
-      ]
-    end
+    ]
+end
 
-    def syn_dfh_combine_erase
-      temp_file_names.map { |t| 'erase ' + t }
-    end
+def syn_dfh_combine_erase
+    temp_file_names.map { |t| 'erase ' + t }
+end
 
-    def syn_convert_implied_decim
-      var_list = @sfc.variables.find_all { |var| var.implied_decimals > 0 }
-      return [] if var_list.empty?
-      var_list.map { |var|
+def syn_convert_implied_decim
+    var_list = @sfc.variables.find_all { |var| var.implied_decimals > 0 }
+    return [] if var_list.empty?
+    var_list.map { |var|
         v = var.name.downcase
         sprintf @replace_format, v, v, 10 ** var.implied_decimals
-      }
-    end
+    }
+end
 
-    def syn_display_format
-      var_list = @sfc.variables.find_all { |var|
+def syn_display_format
+    var_list = @sfc.variables.find_all { |var|
         vf = var_fmt(var)
         vf == 'double' or vf == 'float'
-      }
-      return [] if var_list.empty?
-      var_list.map { |var|
+    }
+    return [] if var_list.empty?
+    var_list.map { |var|
         v = var.name.downcase
+		
+	# If implied decimals set, it means we know exactly how much precision
+	# to show. 
+	# Otherwise we go with the underlying value and the 'g' (general) formatting rules in Stata
+	# when the variable is a double float in our metadata, or use .0f for formatting if the
+	# number is a integer in our metadata: In Stata you store large integer type data in 
+	# 'double' or 'long double' type variables. It's confusing.
+        formatting = if var.implied_decimals > 0 
+			sprintf(@fixed_point_display_format, v, var.width, var.implied_decimals) 
+		elsif  var.is_double_var
+			sprintf(@general_display_format, v, var.width, var.implied_decimals)
+		else
+			sprintf(@integer_display_format, v, var.width, 0)
+		end
+			
+	formatting
+    }
+end
 
-        # If implied decimals set, it means we know exactly how much precision
-        # to show. Otherwise we go with the underlying value and the 'g' (general) formatting rules in Stata
-        formatting = var.implied_decimals > 0 ?
-          sprintf(@fixed_point_display_format, v, var.width, var.implied_decimals) :
-          sprintf(@general_display_format, v, var.width, var.implied_decimals)
-
-        formatting
-      }
-    end
-
-    def syn_var_labs(var_list = [])
-      var_list = @sfc.get_vars_with_var_labels if var_list.empty?
-      return [] if var_list.empty?
-      var_list.map { |var|
+def syn_var_labs (var_list = [])
+    var_list = @sfc.get_vars_with_var_labels if var_list.empty?
+    return [] if var_list.empty?
+    var_list.map { |var|
         sprintf @var_lab_format,
-          var.name.downcase,
-          q(label_trunc(var.label, @var_label_max_leng))
-      }
-    end
+                var.name.downcase,
+                q( label_trunc(var.label, @var_label_max_leng) )
+    }
+end
 
-    def syn_val_labs
-      var_list = @sfc.get_vars_with_values.find_all { |var| not var.is_string_var }
-      return [] if var_list.empty?
-      r = var_list.map { |var|
-        labels = syn_val_labs_for_var(var)
-        if labels.any?
-          [
-            labels,
+def syn_val_labs
+    var_list = @sfc.get_vars_with_values.find_all { |var| not var.is_string_var }
+    return [] if var_list.empty?
+    r = var_list.map { |var|
+        [
+            syn_val_labs_for_var(var),
             "label values " + var.name.downcase + ' ' + label_handle(var),
             blank,
-          ]
-        else
-          []
-        end
-      }
-      r.flatten
-    end
-
-    def syn_val_labs_for_var(var)
-      val_list = labelable_values(var)
-      return [] if val_list.empty?
-      m = max_value_length(var, val_list)
-
-      # A Stata value 10 digits or wider cannot be labeled.
-      if m >= 10
-        return []
-      end
+        ]
+    }
+    r.flatten
+end
 
-      value_format = "label define %s %-#{m}s %s%s"
-      add_cmd = ''
-      r = []
-      val_list.each { |val|
+def syn_val_labs_for_var (var)
+    val_list = labelable_values(var)
+    return [] if val_list.empty?
+    m = max_value_length(var, val_list)
+    value_format = "label define %s %-#{m}s %s%s"
+    add_cmd = ''
+    r = []
+    val_list.each { |val|
         label_truncated = label_trunc(val.label, @val_label_max_leng)
         # stata doesn't like blank value labels
         label_truncated = val.value if label_truncated.nil? || (label_truncated.strip.length == 0)
         r.push sprintf(
-          value_format,
-          label_handle(var),
-          val.value,
-          q(label_truncated),
-          add_cmd
+            value_format,
+            label_handle(var),
+            val.value,
+            q( label_truncated ),
+            add_cmd
         )
         add_cmd = ', add'
-      }
-      r.flatten
-    end
-
-    def q(s)
-      '`"' + s.to_s + '"\''
-    end
+    }
+    r.flatten
+end
 
-    def var_fmt(var)
+def q (s)
+    '`"' + s.to_s + '"\''
+end
 
-      return 'str' if var.is_string_var
-      return 'double' if var.is_double_var || var.implied_decimals > 0
-      return 'byte' if var.width <= 2
-      return 'int' if var.width <= 4
-      return 'long' if var.width <= 7
+def var_fmt (var)
 
-      return 'double'
-    end
+    return 'str'    if var.is_string_var
+    return 'double' if var.is_double_var || var.implied_decimals > 0
+    return 'byte'   if var.width <= 2
+    return 'int'    if var.width <= 4
+    return 'long'   if var.width <= 7
+    
+    return 'double'
+end
 
-    def temp_file_names
-      tf = []
-      @sfc.record_types.each { |rt|
+def temp_file_names
+    tf = []
+    @sfc.record_types.each { |rt|
         var_list = @sfc.get_vars_by_record_type(rt)
         tf.push temp_file_name(rt) if var_list.size > 0
-      }
-      tf
-    end
+    }
+    tf
+end
 
-    def temp_file_name(rt)
-      '__temp_ipums_hier_' + rt + '.dta'
-    end
+def temp_file_name (rt)
+    '__temp_ipums_hier_' + rt + '.dta'
+end
 
-    def label_handle(var)
-      var.name.downcase + '_lbl'
-    end
+def label_handle (var)
+    var.name.downcase + '_lbl'
+end
 
-    def sort_vars
-      return [@sort_var_stem] unless @sfc.rectangularize
-      return @sfc.record_types.map { |rt| @sort_var_stem + rt }
-    end
+def sort_vars
+    return [ @sort_var_stem ] unless @sfc.rectangularize
+    return @sfc.record_types.map { |rt| @sort_var_stem + rt }
+end
 
-    def rt_ne_statement(rt)
-      rt_var = @sfc.record_type_var
-      rt_var.name.downcase + ' != ' + val_q(rt_var, val_as_s(rt_var, rt))
-    end
+def rt_ne_statement (rt)
+    rt_var = @sfc.record_type_var
+    rt_var.name.downcase + ' != ' + val_q(rt_var, val_as_s(rt_var, rt))
+end
 
-  end
+end
 end