v5 v6: ao, auto naming (segment & tag), exceptions, avoidance of name collisionssisu_5.6.6

* (i) auto naming attempts to extract a suitable numeric name from document
  headings if available, at times one appears to be available but is re-used
  elsewhere in the document, (as where the document has Parts or Sections and
  naming starts again within these sections from 1); (ii) this is treated as an
  exception and an alternative sequential naming scheme is triggered; (iii) if a
  document heading does not provide a suitable numeric name, another sequential
  scheme is used

1 files changed, 187 insertions, 60 deletions

diff --git a/lib/sisu/v5/ao_numbering.rb b/lib/sisu/v5/ao_numbering.rb
index f9257e6a..6954a286 100644
--- a/lib/sisu/v5/ao_numbering.rb
+++ b/lib/sisu/v5/ao_numbering.rb
@@ -69,6 +69,27 @@ module SiSU_AO_Numbering
       @obj=@type=@ocn=@lv=@name=@index=@comment=nil
       @chosen_seg_names=[]
     end
+    def chosen_seg_names(chosen,chosen_seg_name,dob,md,type)
+      @chosen_seg_names=if chosen.compact.uniq.length \
+      == chosen.compact.length
+        chosen
+      else
+        if md.opt.act[:maintenance][:set]==:on
+          SiSU_Utils::CodeMarker.new(__LINE__,__FILE__,:green).
+            mark(
+              "duplicated auto segment name: #{type} #{chosen}\n" \
+              + "#{chosen}\n" \
+              + " manually name level 1 segments '1~given_name'\n" \
+              + 'filename: ' + md.fns + "\n" \
+              + 'heading text: "' + dob.obj + '"' + "\n" \
+              + 'duplication: "' + chosen_seg_name + '" (level: ' + dob.lv + '; numbering type: ' + type.to_s + ')'
+            )
+        end
+        chosen=chosen[0..-2]
+        chosen_seg_name=auto_numbering_exceptions(chosen,md,dob)
+        chosen << chosen_seg_name
+      end
+    end
     def number_of_segments?
       if @@segments_count==0
         @data.each do |dob|
@@ -314,6 +335,78 @@ module SiSU_AO_Numbering
           possible_seg_name.to_s
       end
     end
+    def auto_numbering_exceptions(chosen_seg_names_,md,dob)
+      number_make=case dob.lv.to_i
+      when 1
+        @num_exc={
+          t1: @num_exc[:t1] += 1,
+          t2: 0,
+          t3: 0,
+          t4: 0
+        }
+        Mx[:segname_prefix_auto_num_other] + '_' \
+        + @num_exc[:t1].to_s
+      when 2
+        @num_exc={
+          t1: @num_exc[:t1],
+          t2: @num_exc[:t2] += 1,
+          t3: 0,
+          t4: 0
+        }
+        Mx[:segname_prefix_auto_num_other] + '_' \
+        + @num_exc[:t1].to_s + '_' \
+        + @num_exc[:t2].to_s
+      when 3
+        @num_exc={
+          t1: @num_exc[:t1],
+          t2: @num_exc[:t2],
+          t3: @num_exc[:t3] += 1,
+          t4: 0
+        }
+        Mx[:segname_prefix_auto_num_other] + '_' \
+        + @num_exc[:t1].to_s + '_' \
+        + @num_exc[:t2].to_s + '_' \
+        + @num_exc[:t3].to_s
+      when 4
+        @num_exc[:t4] += 1
+        @num_exc={
+          t1: @num_exc[:t1],
+          t2: @num_exc[:t2],
+          t3: @num_exc[:t3],
+          t4: @num_exc[:t4] += 1
+        }
+        Mx[:segname_prefix_auto_num_other] + '_' \
+        + @num_exc[:t1].to_s + '_' \
+        + @num_exc[:t2].to_s + '_' \
+        + @num_exc[:t3].to_s + '_' \
+        + @num_exc[:t4].to_s
+      end
+    end
+    def check_that_seg_names_are_unique(chosen_seg_names_,chosen_seg_name,type,md,dob)
+      begin
+        chosen_seg_names_ << chosen_seg_name
+        chosen_seg_names_=chosen_seg_names(chosen_seg_names_,chosen_seg_name,dob,md,type)
+        if chosen_seg_names_.compact.uniq.length \
+        == chosen_seg_names_.compact.length
+          #check that all auto given seg names are unique
+          chosen_seg_names_=chosen_seg_names(chosen_seg_names_,chosen_seg_name,dob,md,type)
+          chosen_seg_name
+        else
+          SiSU_Utils::CodeMarker.new(__LINE__,__FILE__,:green).
+            mark(
+              "duplicated auto segment name: #{type} #{chosen_seg_name}\n" \
+              + "#{chosen_seg_names_}\n" \
+              + " manually name level 1 segments '1~given_name'\n" \
+              + 'filename: ' + md.fns + "\n" \
+              + 'heading text: "' + dob.obj + '"' + "\n" \
+              + 'duplication: "' + chosen_seg_name + '" (level: ' + dob.lv + '; numbering type: ' + type.to_s + ')'
+            )
+          chosen_seg_name=auto_numbering_exceptions(chosen_seg_names_,md,dob)
+          check_that_seg_names_are_unique(chosen_seg_names_,chosen_seg_name,:exception,md,dob)
+        end
+      rescue
+      end
+    end
     def auto_seg_name(possible_seg_name,heading_num_is,dob,type)
       prefix=case type
       when :auto    then Mx[:segname_prefix_auto_num_provide]
@@ -324,41 +417,49 @@ module SiSU_AO_Numbering
         possible_seg_name=possible_seg_name.
           gsub(/\.$/,'')
       end
-      chosen_seg_name=if possible_seg_name.to_s =~/^[0-9]+[.]?$/m \
-      and possible_seg_name.to_i <= heading_num_is.to_i \
-      and dob.lv == '1'
-        prefix + leading_zeros_fixed_width_number(possible_seg_name)
-      elsif possible_seg_name.to_s =~/^[0-9]+[.,:-]*$/m \
-      and dob.lv == '1'
+      @chosen_seg_name=
+      if dob.lv=='4' \
+      and possible_seg_name.to_s =~/^[0-9]+(?:[.,:-][0-9]){3}/m
         possible_seg_name=possible_seg_name.to_s.
           gsub(/(?:[:,-]|\W)/,'.').
           gsub(/\.$/,'')
         prefix + possible_seg_name
-      elsif possible_seg_name.to_s =~
-      /^[0-9]+[.,:-][0-9]+[.,:-]*$/m \
-      and dob.lv == '2'
+      elsif dob.lv=='3' \
+      and possible_seg_name.to_s =~/^[0-9]+(?:[.,:-][0-9]){2}/m
         possible_seg_name=possible_seg_name.to_s.
           gsub(/(?:[:,-]|\W)/,'.').
           gsub(/\.$/,'')
         prefix + possible_seg_name
-      elsif possible_seg_name.to_s =~
-      /^[0-9]+[.,:-][0-9]+[.,:-][0-9][\d.,:-]*$/m \
-      and dob.lv == '3'
+      elsif dob.lv=='2' \
+      and possible_seg_name.to_s =~/^[0-9]+(?:[.,:-][0-9]){1}/m
         possible_seg_name=possible_seg_name.to_s.
           gsub(/(?:[:,-]|\W)/,'.').
           gsub(/\.$/,'')
         prefix + possible_seg_name
+      elsif dob.lv=='1' \
+      and possible_seg_name.to_s =~/^[0-9]+[:,-]?$/m
+        if possible_seg_name.to_i <= heading_num_is.to_i
+          prefix + leading_zeros_fixed_width_number(possible_seg_name)
+        else
+          possible_seg_name=possible_seg_name.to_s.
+            gsub(/(?:[:,-]|\W)/,'.').
+            gsub(/\.$/,'')
+          prefix + possible_seg_name
+        end
       else
-        Mx[:segname_prefix_auto_num_other]*dob.lv.to_i \
-        + possible_seg_name.to_s
+        @chosen_seg_name=auto_numbering_exceptions(@chosen_seg_names,md,dob)
       end
-      @chosen_seg_names << chosen_seg_name
-      if @chosen_seg_names.compact.uniq.length == @chosen_seg_names.compact.length #checks that all auto given seg names are unique
-        chosen_seg_name
-      else
-        SiSU_Utils::CodeMarker.new(__LINE__,__FILE__,:green).
-         mark("duplicated auto segment name: #{type} #{chosen_seg_name} - #{@chosen_seg_names}; manually name level 1 segments '1~given_name'\n  #{@md.fns}\n  #{dob.obj}")
-        exit
+      check_that_seg_names_are_unique(@chosen_seg_names,@chosen_seg_name,type,@md,dob)
+    end
+    def set_name_and_tags(dob,possible_seg_name)
+      if @md.seg_names.is_a?(Array) \
+      and not @md.seg_names.include?(possible_seg_name)
+        dob.name=possible_seg_name
+        dob.tags=set_tags(dob.tags,dob.name)
+        @md.seg_names << possible_seg_name
+      elsif (@md.opt.act[:verbose_plus][:set]==:on \
+      or @md.opt.act[:maintenance][:set]==:on)
+        puts 'warn, there may be a conflicting numbering scheme'
       end
     end
     def name_para_seg_filename(data)                                                   #segment naming, remaining
@@ -377,7 +478,7 @@ module SiSU_AO_Numbering
       #   if there is none a sequential number is designated, preceded by an underscore
       @tuned_file,@unique_auto_name=[],[]
       tags={}
-      art_filename_auto=1
+      @art_filename_auto=0
       @counter=1
       if not @md.seg_autoname_safe \
       and (@md.opt.act[:verbose_plus][:set]==:on \
@@ -385,10 +486,11 @@ module SiSU_AO_Numbering
         puts 'manual segment names, numbers used as names, risk warning (segmented html)'
       end
       ocn_html_seg=[]
+      @num_exc={ t1: 0, t2: 0, t3: 0, t4: 0 }
       data.each do |dob|
         if dob.is==:heading \
         && dob.ln \
-        and dob.ln.to_s =~/^[456]/
+        and dob.ln.to_s =~/^[4-7]/
           heading_num_is=/^\d+:(\d+);\d/m.match(dob.node)[1]
           if dob.ln==4 \
           and not dob.name \
@@ -396,47 +498,72 @@ module SiSU_AO_Numbering
             @md.set_heading_seg=true
           end
           if dob.name !~/^\S+/ \
-          and dob.obj =~/^\s*(?:\S+\s+)?([0-9][0-9.,:-]*)/m      #heading starts with a recognised numeric or word followed by a recognised numeric construct, use that as name
-            possible_seg_name=$1
-            possible_seg_name=
-              auto_seg_name(possible_seg_name,heading_num_is,dob,:extract)
-            possible_seg_name=possible_seg_name.
-              gsub(/(?:[:,-]|\W)/,'.').
-              gsub(/\.$/,'')
-            if @md.seg_names.is_a?(Array) \
-            and not @md.seg_names.include?(possible_seg_name)
-              dob.name=possible_seg_name
-              dob.tags=set_tags(dob.tags,dob.name)
-              @md.seg_names << possible_seg_name
-            elsif (@md.opt.act[:verbose_plus][:set]==:on \
-            or @md.opt.act[:maintenance][:set]==:on)
-              puts 'warn, there may be a conflicting numbering scheme'
+          and dob.ln.to_s =~/^[5-7]/ \
+          and dob.obj =~/^\s*(?:\S+\s+)?([0-9]+(?:[.,:-][0-9])+)/m
+            #heading starts with a recognised numeric
+            #or word followed by a recognised numeric construct,
+            #use that as name
+            if dob.ln==7 \
+            and dob.obj =~/^\s*(?:\S+\s+)?([0-9]+(?:[.,:-][0-9]){3})/m
+              possible_seg_name=$1.
+                gsub(/(?:[:,-]|\W)/,'.').
+                gsub(/\.$/,'')
+              possible_seg_name=
+                auto_seg_name(possible_seg_name,heading_num_is,dob,:extract)
+              set_name_and_tags(dob,possible_seg_name)
+            elsif dob.ln==6 \
+            and dob.obj =~/^\s*(?:\S+\s+)?([0-9]+(?:[.,:-][0-9]){2})/m
+              possible_seg_name=$1.
+                gsub(/(?:[:,-]|\W)/,'.').
+                gsub(/\.$/,'')
+              possible_seg_name=
+                auto_seg_name(possible_seg_name,heading_num_is,dob,:extract)
+              set_name_and_tags(dob,possible_seg_name)
+            elsif dob.ln==5 \
+            and dob.obj =~/^\s*(?:\S+\s+)?([0-9]+(?:[.,:-][0-9]){1})/m
+              possible_seg_name=$1.
+                gsub(/(?:[:,-]|\W)/,'.').
+                gsub(/\.$/,'')
+              possible_seg_name=
+                auto_seg_name(possible_seg_name,heading_num_is,dob,:extract)
+              set_name_and_tags(dob,possible_seg_name)
             end
           end
-          if dob.ln==4 \
-          and dob.name                                     #extract segment name from embedded document structure info
-            if @md.seg_names.is_a?(Array) \
-            and not @md.seg_names.include?(dob.name)
-              dob.tags=set_tags(dob.tags,dob.name)
-              @md.seg_names << dob.name
+          if dob.ln==4
+            if dob.name !~/^\S+/ \
+            and dob.obj =~/^\s*(?:\S+\s+)?([0-9]+)/m
+              #heading starts with a recognised numeric
+              #or word followed by a recognised numeric construct,
+              #use that as name
+              possible_seg_name=$1
+              possible_seg_name=
+                auto_seg_name(possible_seg_name,heading_num_is,dob,:extract)
+              set_name_and_tags(dob,possible_seg_name)
             end
-          end
-          if dob.ln==4 \
-          and not dob.name                                 #if still no segment name, provide a numerical one
-            possible_seg_name=
-              auto_seg_name(art_filename_auto,heading_num_is,dob,:auto)
-            if @md.seg_names.is_a?(Array) \
-            and not @md.seg_names.include?(possible_seg_name)
-             dob.name=possible_seg_name
-             dob.tags=set_tags(dob.tags,dob.name)
-              @md.seg_names << possible_seg_name
-            else puts 'segment name (numbering) error'
+            if dob.name
+              #extract segment name from embedded document structure info
+              if @md.seg_names.is_a?(Array) \
+              and not @md.seg_names.include?(dob.name)
+                dob.tags=set_tags(dob.tags,dob.name)
+                @md.seg_names << dob.name
+              end
+            else
+              #if no segment name,
+              #provide a numerical one
+              @art_filename_auto+=1
+              possible_seg_name=
+                auto_seg_name(@art_filename_auto,heading_num_is,dob,:auto)
+              if @md.seg_names.is_a?(Array) \
+              and not @md.seg_names.include?(possible_seg_name)
+               dob.name=possible_seg_name
+               dob.tags=set_tags(dob.tags,dob.name)
+                @md.seg_names << possible_seg_name
+              else puts 'segment name (numbering) error'
+              end
+            end
+            if not dob.name #should not occur
+              puts "e r r o r -\t#{__FILE__}::#{__LINE__}\n#{dob.inspect}"
             end
-            art_filename_auto+=1
-          end
-          if dob.ln==4 \
-          and not dob.name #should not occur
-            puts "e r r o r -\t#{__FILE__}::#{__LINE__}\n#{dob.inspect}"
           end
         end
         if (dob.is ==:heading \